Merge tag 'mmc-v6.7' of git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/mmc

Pull MMC updates from Ulf Hansson:
" MMC core:
   - Enable host caps to be modified via debugfs to test speed-modes
   - Improve random I/O writes for 4k buffers for hsq enabled hosts

  MMC host:
   - atmel-mci/sdhci-of-at91: Aubin Constans takes over as maintainer
   - dw_mmc-starfive: Re-work tuning support
   - meson-gx: Fix bogus IRQ when using CMD_CFG_ERROR
   - mmci: Use peripheral flow control for the STM32 variant
   - renesas,sdhi: Add support for the RZ/G3S variant
   - sdhci-esdhc-imx: Optimize the manual tuning logic
   - sdhci-msm: Add support for the SM8650 variant
   - sdhci-npcm: Add driver to support the Nuvoton NPCM BMC variant
   - sdhci-pci-gli: Add workaround to allow GL9750 to enter ASPM L1.2"

* tag 'mmc-v6.7' of git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/mmc: (25 commits)
  dt-bindings: mmc: sdhci-msm: document the SM8650 SDHCI Controller
  mmc: meson-gx: Remove setting of CMD_CFG_ERROR
  MAINTAINERS: mmc: take over as maintainer of MCI & SDHCI MICROCHIP DRIVERS
  mmc: jz4740: Use device_get_match_data()
  mmc: sdhci-npcm: Add NPCM SDHCI driver
  dt-bindings: mmc: npcm,sdhci: Document NPCM SDHCI controller
  mmc: sdhci-pltfm: Make driver OF independent
  mmc: sdhci-pltfm: Drop unnecessary error messages in sdhci_pltfm_init()
  mmc: sdhci-pci: Switch to use acpi_evaluate_dsm_typed()
  mmc: debugfs: Allow host caps to be modified
  mmc: core: Always reselect card type
  mmc: mmci: use peripheral flow control for STM32
  mmc: vub300: replace deprecated strncpy with strscpy
  memstick: jmb38x_ms: Annotate struct jmb38x_ms with __counted_by
  mmc: starfive: Change tuning implementation
  dt-bindings: mmc: starfive: Remove properties from required
  mmc: hsq: Improve random I/O write performance for 4k buffers
  mmc: core: Allow dynamical updates of the number of requests for hsq
  mmc: sdhci-pci-gli: A workaround to allow GL9750 to enter ASPM L1.2
  dt-bindings: mmc: renesas,sdhi: Document RZ/G3S support
  ...

20 files changed, 328 insertions, 176 deletions

diff --git a/drivers/memstick/host/jmb38x_ms.c b/drivers/memstick/host/jmb38x_ms.c
index 21cb2a786058..e77eb8b0eb12 100644
--- a/drivers/memstick/host/jmb38x_ms.c
+++ b/drivers/memstick/host/jmb38x_ms.c
@@ -66,7 +66,7 @@ struct jmb38x_ms_host {
 struct jmb38x_ms {
 	struct pci_dev        *pdev;
 	int                   host_cnt;
-	struct memstick_host  *hosts[];
+	struct memstick_host  *hosts[] __counted_by(host_cnt);
 };
 
 #define BLOCK_COUNT_MASK       0xffff0000
diff --git a/drivers/mmc/core/debugfs.c b/drivers/mmc/core/debugfs.c
index 2c97b94aab23..1642ea72d22c 100644
--- a/drivers/mmc/core/debugfs.c
+++ b/drivers/mmc/core/debugfs.c
@@ -12,9 +12,12 @@
 #include <linux/slab.h>
 #include <linux/stat.h>
 #include <linux/fault-inject.h>
+#include <linux/time.h>
 
 #include <linux/mmc/card.h>
 #include <linux/mmc/host.h>
+#include <linux/mmc/mmc.h>
+#include <linux/mmc/sd.h>
 
 #include "core.h"
 #include "card.h"
@@ -298,6 +301,49 @@ static const struct file_operations mmc_err_stats_fops = {
 	.release = single_release,
 };
 
+static int mmc_caps_get(void *data, u64 *val)
+{
+	*val = *(u32 *)data;
+	return 0;
+}
+
+static int mmc_caps_set(void *data, u64 val)
+{
+	u32 *caps = data;
+	u32 diff = *caps ^ val;
+	u32 allowed = MMC_CAP_AGGRESSIVE_PM |
+		      MMC_CAP_SD_HIGHSPEED |
+		      MMC_CAP_MMC_HIGHSPEED |
+		      MMC_CAP_UHS |
+		      MMC_CAP_DDR;
+
+	if (diff & ~allowed)
+		return -EINVAL;
+
+	*caps = val;
+
+	return 0;
+}
+
+static int mmc_caps2_set(void *data, u64 val)
+{
+	u32 allowed = MMC_CAP2_HSX00_1_8V | MMC_CAP2_HSX00_1_2V;
+	u32 *caps = data;
+	u32 diff = *caps ^ val;
+
+	if (diff & ~allowed)
+		return -EINVAL;
+
+	*caps = val;
+
+	return 0;
+}
+
+DEFINE_DEBUGFS_ATTRIBUTE(mmc_caps_fops, mmc_caps_get, mmc_caps_set,
+			 "0x%08llx\n");
+DEFINE_DEBUGFS_ATTRIBUTE(mmc_caps2_fops, mmc_caps_get, mmc_caps2_set,
+			 "0x%08llx\n");
+
 void mmc_add_host_debugfs(struct mmc_host *host)
 {
 	struct dentry *root;
@@ -306,8 +352,9 @@ void mmc_add_host_debugfs(struct mmc_host *host)
 	host->debugfs_root = root;
 
 	debugfs_create_file("ios", S_IRUSR, root, host, &mmc_ios_fops);
-	debugfs_create_x32("caps", S_IRUSR, root, &host->caps);
-	debugfs_create_x32("caps2", S_IRUSR, root, &host->caps2);
+	debugfs_create_file("caps", 0600, root, &host->caps, &mmc_caps_fops);
+	debugfs_create_file("caps2", 0600, root, &host->caps2,
+			    &mmc_caps2_fops);
 	debugfs_create_file_unsafe("clock", S_IRUSR | S_IWUSR, root, host,
 				   &mmc_clock_fops);
 
diff --git a/drivers/mmc/core/mmc.c b/drivers/mmc/core/mmc.c
index 4a4bab9aa726..8180983bd402 100644
--- a/drivers/mmc/core/mmc.c
+++ b/drivers/mmc/core/mmc.c
@@ -419,7 +419,6 @@ static int mmc_decode_ext_csd(struct mmc_card *card, u8 *ext_csd)
 
 	card->ext_csd.strobe_support = ext_csd[EXT_CSD_STROBE_SUPPORT];
 	card->ext_csd.raw_card_type = ext_csd[EXT_CSD_CARD_TYPE];
-	mmc_select_card_type(card);
 
 	card->ext_csd.raw_s_a_timeout = ext_csd[EXT_CSD_S_A_TIMEOUT];
 	card->ext_csd.raw_erase_timeout_mult =
@@ -1732,6 +1731,12 @@ static int mmc_init_card(struct mmc_host *host, u32 ocr,
 		mmc_set_erase_size(card);
 	}
 
+	/*
+	 * Reselect the card type since host caps could have been changed when
+	 * debugging even if the card is not new.
+	 */
+	mmc_select_card_type(card);
+
 	/* Enable ERASE_GRP_DEF. This bit is lost after a reset or power off. */
 	if (card->ext_csd.rev >= 3) {
 		err = mmc_switch(card, EXT_CSD_CMD_SET_NORMAL,
diff --git a/drivers/mmc/core/queue.c b/drivers/mmc/core/queue.c
index b396e3900717..a0a2412f62a7 100644
--- a/drivers/mmc/core/queue.c
+++ b/drivers/mmc/core/queue.c
@@ -260,11 +260,7 @@ static blk_status_t mmc_mq_queue_rq(struct blk_mq_hw_ctx *hctx,
 		}
 		break;
 	case MMC_ISSUE_ASYNC:
-		/*
-		 * For MMC host software queue, we only allow 2 requests in
-		 * flight to avoid a long latency.
-		 */
-		if (host->hsq_enabled && mq->in_flight[issue_type] > 2) {
+		if (host->hsq_enabled && mq->in_flight[issue_type] > host->hsq_depth) {
 			spin_unlock_irq(&mq->lock);
 			return BLK_STS_RESOURCE;
 		}
diff --git a/drivers/mmc/host/Kconfig b/drivers/mmc/host/Kconfig
index 554e67103c1a..58bd5fe4cd25 100644
--- a/drivers/mmc/host/Kconfig
+++ b/drivers/mmc/host/Kconfig
@@ -429,6 +429,14 @@ config MMC_SDHCI_IPROC
 
 	  If unsure, say N.
 
+config MMC_SDHCI_NPCM
+	tristate "Secure Digital Host Controller Interface support for NPCM"
+	depends on ARCH_NPCM || COMPILE_TEST
+	depends on MMC_SDHCI_PLTFM
+	help
+	  This provides support for the SD/eMMC controller found in
+	  NPCM BMC family SoCs.
+
 config MMC_MESON_GX
 	tristate "Amlogic S905/GX*/AXG SD/MMC Host Controller support"
 	depends on ARCH_MESON|| COMPILE_TEST
@@ -677,9 +685,9 @@ config MMC_SDHI_SYS_DMAC
 
 config MMC_SDHI_INTERNAL_DMAC
 	tristate "DMA for SDHI SD/SDIO controllers using on-chip bus mastering"
-	depends on ARM64 || ARCH_R7S9210 || ARCH_R8A77470 || COMPILE_TEST
+	depends on ARCH_RENESAS || COMPILE_TEST
 	depends on MMC_SDHI
-	default MMC_SDHI if (ARM64 || ARCH_R7S9210 || ARCH_R8A77470)
+	default MMC_SDHI if ARCH_RENESAS
 	help
 	  This provides DMA support for SDHI SD/SDIO controllers
 	  using on-chip bus mastering. This supports the controllers
diff --git a/drivers/mmc/host/Makefile b/drivers/mmc/host/Makefile
index a693fa3d3f1c..d0be4465f3ec 100644
--- a/drivers/mmc/host/Makefile
+++ b/drivers/mmc/host/Makefile
@@ -89,6 +89,7 @@ obj-$(CONFIG_MMC_SDHCI_OF_DWCMSHC)	+= sdhci-of-dwcmshc.o
 obj-$(CONFIG_MMC_SDHCI_OF_SPARX5)	+= sdhci-of-sparx5.o
 obj-$(CONFIG_MMC_SDHCI_BCM_KONA)	+= sdhci-bcm-kona.o
 obj-$(CONFIG_MMC_SDHCI_IPROC)		+= sdhci-iproc.o
+obj-$(CONFIG_MMC_SDHCI_NPCM)		+= sdhci-npcm.o
 obj-$(CONFIG_MMC_SDHCI_MSM)		+= sdhci-msm.o
 obj-$(CONFIG_MMC_SDHCI_ST)		+= sdhci-st.o
 obj-$(CONFIG_MMC_SDHCI_MICROCHIP_PIC32)	+= sdhci-pic32.o
diff --git a/drivers/mmc/host/atmel-mci.c b/drivers/mmc/host/atmel-mci.c
index 535783c43105..dba826db739a 100644
--- a/drivers/mmc/host/atmel-mci.c
+++ b/drivers/mmc/host/atmel-mci.c
@@ -227,6 +227,7 @@ struct mci_slot_pdata {
 /**
  * struct mci_platform_data - board-specific MMC/SDcard configuration
  * @dma_slave: DMA slave interface to use in data transfers.
+ * @dma_filter: Filtering function to filter the DMA channel
  * @slot: Per-slot configuration data.
  */
 struct mci_platform_data {
@@ -674,8 +675,10 @@ atmci_of_init(struct platform_device *pdev)
 					      "cd", GPIOD_IN, "cd-gpios");
 		err = PTR_ERR_OR_ZERO(pdata->slot[slot_id].detect_pin);
 		if (err) {
-			if (err != -ENOENT)
+			if (err != -ENOENT) {
+				of_node_put(cnp);
 				return ERR_PTR(err);
+			}
 			pdata->slot[slot_id].detect_pin = NULL;
 		}
 
@@ -687,8 +690,10 @@ atmci_of_init(struct platform_device *pdev)
 					      "wp", GPIOD_IN, "wp-gpios");
 		err = PTR_ERR_OR_ZERO(pdata->slot[slot_id].wp_pin);
 		if (err) {
-			if (err != -ENOENT)
+			if (err != -ENOENT) {
+				of_node_put(cnp);
 				return ERR_PTR(err);
+			}
 			pdata->slot[slot_id].wp_pin = NULL;
 		}
 	}
diff --git a/drivers/mmc/host/dw_mmc-starfive.c b/drivers/mmc/host/dw_mmc-starfive.c
index fd05a648a8bb..b4d81ef0f3af 100644
--- a/drivers/mmc/host/dw_mmc-starfive.c
+++ b/drivers/mmc/host/dw_mmc-starfive.c
@@ -5,6 +5,7 @@
  * Copyright (c) 2022 StarFive Technology Co., Ltd.
  */
 
+#include <linux/bitfield.h>
 #include <linux/clk.h>
 #include <linux/delay.h>
 #include <linux/mfd/syscon.h>
@@ -20,13 +21,7 @@
 #define ALL_INT_CLR		0x1ffff
 #define MAX_DELAY_CHAIN		32
 
-struct starfive_priv {
-	struct device *dev;
-	struct regmap *reg_syscon;
-	u32 syscon_offset;
-	u32 syscon_shift;
-	u32 syscon_mask;
-};
+#define STARFIVE_SMPL_PHASE     GENMASK(20, 16)
 
 static void dw_mci_starfive_set_ios(struct dw_mci *host, struct mmc_ios *ios)
 {
@@ -44,117 +39,65 @@ static void dw_mci_starfive_set_ios(struct dw_mci *host, struct mmc_ios *ios)
 	}
 }
 
+static void dw_mci_starfive_set_sample_phase(struct dw_mci *host, u32 smpl_phase)
+{
+	/* change driver phase and sample phase */
+	u32 reg_value = mci_readl(host, UHS_REG_EXT);
+
+	/* In UHS_REG_EXT, only 5 bits valid in DRV_PHASE and SMPL_PHASE */
+	reg_value &= ~STARFIVE_SMPL_PHASE;
+	reg_value |= FIELD_PREP(STARFIVE_SMPL_PHASE, smpl_phase);
+	mci_writel(host, UHS_REG_EXT, reg_value);
+
+	/* We should delay 1ms wait for timing setting finished. */
+	mdelay(1);
+}
+
 static int dw_mci_starfive_execute_tuning(struct dw_mci_slot *slot,
 					     u32 opcode)
 {
 	static const int grade  = MAX_DELAY_CHAIN;
 	struct dw_mci *host = slot->host;
-	struct starfive_priv *priv = host->priv;
-	int rise_point = -1, fall_point = -1;
-	int err, prev_err = 0;
-	int i;
-	bool found = 0;
-	u32 regval;
-
-	/*
-	 * Use grade as the max delay chain, and use the rise_point and
-	 * fall_point to ensure the best sampling point of a data input
-	 * signals.
-	 */
-	for (i = 0; i < grade; i++) {
-		regval = i << priv->syscon_shift;
-		err = regmap_update_bits(priv->reg_syscon, priv->syscon_offset,
-						priv->syscon_mask, regval);
-		if (err)
-			return err;
+	int smpl_phase, smpl_raise = -1, smpl_fall = -1;
+	int ret;
+
+	for (smpl_phase = 0; smpl_phase < grade; smpl_phase++) {
+		dw_mci_starfive_set_sample_phase(host, smpl_phase);
 		mci_writel(host, RINTSTS, ALL_INT_CLR);
 
-		err = mmc_send_tuning(slot->mmc, opcode, NULL);
-		if (!err)
-			found = 1;
+		ret = mmc_send_tuning(slot->mmc, opcode, NULL);
 
-		if (i > 0) {
-			if (err && !prev_err)
-				fall_point = i - 1;
-			if (!err && prev_err)
-				rise_point = i;
+		if (!ret && smpl_raise < 0) {
+			smpl_raise = smpl_phase;
+		} else if (ret && smpl_raise >= 0) {
+			smpl_fall = smpl_phase - 1;
+			break;
 		}
-
-		if (rise_point != -1 && fall_point != -1)
-			goto tuning_out;
-
-		prev_err = err;
-		err = 0;
 	}
 
-tuning_out:
-	if (found) {
-		if (rise_point == -1)
-			rise_point = 0;
-		if (fall_point == -1)
-			fall_point = grade - 1;
-		if (fall_point < rise_point) {
-			if ((rise_point + fall_point) >
-			    (grade - 1))
-				i = fall_point / 2;
-			else
-				i = (rise_point + grade - 1) / 2;
-		} else {
-			i = (rise_point + fall_point) / 2;
-		}
-
-		regval = i << priv->syscon_shift;
-		err = regmap_update_bits(priv->reg_syscon, priv->syscon_offset,
-						priv->syscon_mask, regval);
-		if (err)
-			return err;
-		mci_writel(host, RINTSTS, ALL_INT_CLR);
+	if (smpl_phase >= grade)
+		smpl_fall = grade - 1;
 
-		dev_info(host->dev, "Found valid delay chain! use it [delay=%d]\n", i);
-	} else {
+	if (smpl_raise < 0) {
+		smpl_phase = 0;
 		dev_err(host->dev, "No valid delay chain! use default\n");
-		err = -EINVAL;
+		ret = -EINVAL;
+		goto out;
 	}
 
-	mci_writel(host, RINTSTS, ALL_INT_CLR);
-	return err;
-}
-
-static int dw_mci_starfive_parse_dt(struct dw_mci *host)
-{
-	struct of_phandle_args args;
-	struct starfive_priv *priv;
-	int ret;
-
-	priv = devm_kzalloc(host->dev, sizeof(*priv), GFP_KERNEL);
-	if (!priv)
-		return -ENOMEM;
+	smpl_phase = (smpl_raise + smpl_fall) / 2;
+	dev_dbg(host->dev, "Found valid delay chain! use it [delay=%d]\n", smpl_phase);
+	ret = 0;
 
-	ret = of_parse_phandle_with_fixed_args(host->dev->of_node,
-						"starfive,sysreg", 3, 0, &args);
-	if (ret) {
-		dev_err(host->dev, "Failed to parse starfive,sysreg\n");
-		return -EINVAL;
-	}
-
-	priv->reg_syscon = syscon_node_to_regmap(args.np);
-	of_node_put(args.np);
-	if (IS_ERR(priv->reg_syscon))
-		return PTR_ERR(priv->reg_syscon);
-
-	priv->syscon_offset = args.args[0];
-	priv->syscon_shift  = args.args[1];
-	priv->syscon_mask   = args.args[2];
-
-	host->priv = priv;
-
-	return 0;
+out:
+	dw_mci_starfive_set_sample_phase(host, smpl_phase);
+	mci_writel(host, RINTSTS, ALL_INT_CLR);
+	return ret;
 }
 
 static const struct dw_mci_drv_data starfive_data = {
 	.common_caps		= MMC_CAP_CMD23,
 	.set_ios		= dw_mci_starfive_set_ios,
-	.parse_dt		= dw_mci_starfive_parse_dt,
 	.execute_tuning		= dw_mci_starfive_execute_tuning,
 };
 
diff --git a/drivers/mmc/host/jz4740_mmc.c b/drivers/mmc/host/jz4740_mmc.c
index f379ce5b582d..6a45991ca056 100644
--- a/drivers/mmc/host/jz4740_mmc.c
+++ b/drivers/mmc/host/jz4740_mmc.c
@@ -18,9 +18,10 @@
 #include <linux/mmc/host.h>
 #include <linux/mmc/slot-gpio.h>
 #include <linux/module.h>
-#include <linux/of_device.h>
+#include <linux/of.h>
 #include <linux/pinctrl/consumer.h>
 #include <linux/platform_device.h>
+#include <linux/property.h>
 #include <linux/regulator/consumer.h>
 #include <linux/scatterlist.h>
 
@@ -1040,7 +1041,6 @@ static int jz4740_mmc_probe(struct platform_device* pdev)
 	int ret;
 	struct mmc_host *mmc;
 	struct jz4740_mmc_host *host;
-	const struct of_device_id *match;
 
 	mmc = mmc_alloc_host(sizeof(struct jz4740_mmc_host), &pdev->dev);
 	if (!mmc) {
@@ -1050,13 +1050,8 @@ static int jz4740_mmc_probe(struct platform_device* pdev)
 
 	host = mmc_priv(mmc);
 
-	match = of_match_device(jz4740_mmc_of_match, &pdev->dev);
-	if (match) {
-		host->version = (enum jz4740_mmc_version)match->data;
-	} else {
-		/* JZ4740 should be the only one using legacy probe */
-		host->version = JZ_MMC_JZ4740;
-	}
+	/* Default if no match is JZ4740 */
+	host->version = (enum jz4740_mmc_version)device_get_match_data(&pdev->dev);
 
 	ret = mmc_of_parse(mmc);
 	if (ret) {
@@ -1200,7 +1195,7 @@ static struct platform_driver jz4740_mmc_driver = {
 	.driver = {
 		.name = "jz4740-mmc",
 		.probe_type = PROBE_PREFER_ASYNCHRONOUS,
-		.of_match_table = of_match_ptr(jz4740_mmc_of_match),
+		.of_match_table = jz4740_mmc_of_match,
 		.pm = pm_sleep_ptr(&jz4740_mmc_pm_ops),
 	},
 };
diff --git a/drivers/mmc/host/meson-gx-mmc.c b/drivers/mmc/host/meson-gx-mmc.c
index 9837dab096e6..c7c067b9415a 100644
--- a/drivers/mmc/host/meson-gx-mmc.c
+++ b/drivers/mmc/host/meson-gx-mmc.c
@@ -801,7 +801,6 @@ static void meson_mmc_start_cmd(struct mmc_host *mmc, struct mmc_command *cmd)
 
 	cmd_cfg |= FIELD_PREP(CMD_CFG_CMD_INDEX_MASK, cmd->opcode);
 	cmd_cfg |= CMD_CFG_OWNER;  /* owned by CPU */
-	cmd_cfg |= CMD_CFG_ERROR; /* stop in case of error */
 
 	meson_mmc_set_response_bits(cmd, &cmd_cfg);
 
diff --git a/drivers/mmc/host/mmc_hsq.c b/drivers/mmc/host/mmc_hsq.c
index 424dc7b07858..79836705c176 100644
--- a/drivers/mmc/host/mmc_hsq.c
+++ b/drivers/mmc/host/mmc_hsq.c
@@ -21,6 +21,25 @@ static void mmc_hsq_retry_handler(struct work_struct *work)
 	mmc->ops->request(mmc, hsq->mrq);
 }
 
+static void mmc_hsq_modify_threshold(struct mmc_hsq *hsq)
+{
+	struct mmc_host *mmc = hsq->mmc;
+	struct mmc_request *mrq;
+	unsigned int tag, need_change = 0;
+
+	mmc->hsq_depth = HSQ_NORMAL_DEPTH;
+	for (tag = 0; tag < HSQ_NUM_SLOTS; tag++) {
+		mrq = hsq->slot[tag].mrq;
+		if (mrq && mrq->data &&
+		   (mrq->data->blksz * mrq->data->blocks == 4096) &&
+		   (mrq->data->flags & MMC_DATA_WRITE) &&
+		   (++need_change == 2)) {
+			mmc->hsq_depth = HSQ_PERFORMANCE_DEPTH;
+			break;
+		}
+	}
+}
+
 static void mmc_hsq_pump_requests(struct mmc_hsq *hsq)
 {
 	struct mmc_host *mmc = hsq->mmc;
@@ -42,6 +61,8 @@ static void mmc_hsq_pump_requests(struct mmc_hsq *hsq)
 		return;
 	}
 
+	mmc_hsq_modify_threshold(hsq);
+
 	slot = &hsq->slot[hsq->next_tag];
 	hsq->mrq = slot->mrq;
 	hsq->qcnt--;
@@ -337,6 +358,7 @@ int mmc_hsq_init(struct mmc_hsq *hsq, struct mmc_host *mmc)
 	hsq->mmc = mmc;
 	hsq->mmc->cqe_private = hsq;
 	mmc->cqe_ops = &mmc_hsq_ops;
+	mmc->hsq_depth = HSQ_NORMAL_DEPTH;
 
 	for (i = 0; i < HSQ_NUM_SLOTS; i++)
 		hsq->tag_slot[i] = HSQ_INVALID_TAG;
diff --git a/drivers/mmc/host/mmc_hsq.h b/drivers/mmc/host/mmc_hsq.h
index 1808024fc6c5..dd352a6ac32a 100644
--- a/drivers/mmc/host/mmc_hsq.h
+++ b/drivers/mmc/host/mmc_hsq.h
@@ -5,6 +5,17 @@
 #define HSQ_NUM_SLOTS	64
 #define HSQ_INVALID_TAG	HSQ_NUM_SLOTS
 
+/*
+ * For MMC host software queue, we only allow 2 requests in
+ * flight to avoid a long latency.
+ */
+#define HSQ_NORMAL_DEPTH	2
+/*
+ * For 4k random writes, we allow hsq_depth to increase to 5
+ * for better performance.
+ */
+#define HSQ_PERFORMANCE_DEPTH	5
+
 struct hsq_slot {
 	struct mmc_request *mrq;
 };
diff --git a/drivers/mmc/host/mmci.c b/drivers/mmc/host/mmci.c
index dda756a56379..e967cca7a16f 100644
--- a/drivers/mmc/host/mmci.c
+++ b/drivers/mmc/host/mmci.c
@@ -249,6 +249,7 @@ static struct variant_data variant_stm32 = {
 	.f_max			= 48000000,
 	.pwrreg_clkgate		= true,
 	.pwrreg_nopower		= true,
+	.dma_flow_controller	= true,
 	.init			= mmci_variant_init,
 };
 
@@ -1015,7 +1016,7 @@ static int _mmci_dmae_prep_data(struct mmci_host *host, struct mmc_data *data,
 		.dst_addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES,
 		.src_maxburst = variant->fifohalfsize >> 2, /* # of words */
 		.dst_maxburst = variant->fifohalfsize >> 2, /* # of words */
-		.device_fc = false,
+		.device_fc = variant->dma_flow_controller,
 	};
 	struct dma_chan *chan;
 	struct dma_device *device;
diff --git a/drivers/mmc/host/mmci.h b/drivers/mmc/host/mmci.h
index 253197f132fc..34d9897c289b 100644
--- a/drivers/mmc/host/mmci.h
+++ b/drivers/mmc/host/mmci.h
@@ -332,6 +332,7 @@ enum mmci_busy_state {
  * @opendrain: bitmask identifying the OPENDRAIN bit inside MMCIPOWER register
  * @dma_lli: true if variant has dma link list feature.
  * @stm32_idmabsize_mask: stm32 sdmmc idma buffer size.
+ * @dma_flow_controller: use peripheral as flow controller for DMA.
  */
 struct variant_data {
 	unsigned int		clkreg;
@@ -378,6 +379,7 @@ struct variant_data {
 	u8			dma_lli:1;
 	u32			stm32_idmabsize_mask;
 	u32			stm32_idmabsize_align;
+	bool			dma_flow_controller;
 	void (*init)(struct mmci_host *host);
 };
 
diff --git a/drivers/mmc/host/sdhci-esdhc-imx.c b/drivers/mmc/host/sdhci-esdhc-imx.c
index 3b8030f3552a..40a6e2f8145a 100644
--- a/drivers/mmc/host/sdhci-esdhc-imx.c
+++ b/drivers/mmc/host/sdhci-esdhc-imx.c
@@ -1154,32 +1154,52 @@ static void esdhc_post_tuning(struct sdhci_host *host)
 	writel(reg, host->ioaddr + ESDHC_MIX_CTRL);
 }
 
+/*
+ * find the largest pass window, and use the average delay of this
+ * largest window to get the best timing.
+ */
 static int esdhc_executing_tuning(struct sdhci_host *host, u32 opcode)
 {
 	int min, max, avg, ret;
+	int win_length, target_min, target_max, target_win_length;
 
-	/* find the mininum delay first which can pass tuning */
 	min = ESDHC_TUNE_CTRL_MIN;
-	while (min < ESDHC_TUNE_CTRL_MAX) {
-		esdhc_prepare_tuning(host, min);
-		if (!mmc_send_tuning(host->mmc, opcode, NULL))
-			break;
-		min += ESDHC_TUNE_CTRL_STEP;
-	}
-
-	/* find the maxinum delay which can not pass tuning */
-	max = min + ESDHC_TUNE_CTRL_STEP;
+	max = ESDHC_TUNE_CTRL_MIN;
+	target_win_length = 0;
 	while (max < ESDHC_TUNE_CTRL_MAX) {
-		esdhc_prepare_tuning(host, max);
-		if (mmc_send_tuning(host->mmc, opcode, NULL)) {
-			max -= ESDHC_TUNE_CTRL_STEP;
-			break;
+		/* find the mininum delay first which can pass tuning */
+		while (min < ESDHC_TUNE_CTRL_MAX) {
+			esdhc_prepare_tuning(host, min);
+			if (!mmc_send_tuning(host->mmc, opcode, NULL))
+				break;
+			min += ESDHC_TUNE_CTRL_STEP;
 		}
-		max += ESDHC_TUNE_CTRL_STEP;
+
+		/* find the maxinum delay which can not pass tuning */
+		max = min + ESDHC_TUNE_CTRL_STEP;
+		while (max < ESDHC_TUNE_CTRL_MAX) {
+			esdhc_prepare_tuning(host, max);
+			if (mmc_send_tuning(host->mmc, opcode, NULL)) {
+				max -= ESDHC_TUNE_CTRL_STEP;
+				break;
+			}
+			max += ESDHC_TUNE_CTRL_STEP;
+		}
+
+		win_length = max - min + 1;
+		/* get the largest pass window */
+		if (win_length > target_win_length) {
+			target_win_length = win_length;
+			target_min = min;
+			target_max = max;
+		}
+
+		/* continue to find the next pass window */
+		min = max + ESDHC_TUNE_CTRL_STEP;
 	}
 
 	/* use average delay to get the best timing */
-	avg = (min + max) / 2;
+	avg = (target_min + target_max) / 2;
 	esdhc_prepare_tuning(host, avg);
 	ret = mmc_send_tuning(host->mmc, opcode, NULL);
 	esdhc_post_tuning(host);
diff --git a/drivers/mmc/host/sdhci-npcm.c b/drivers/mmc/host/sdhci-npcm.c
new file mode 100644
index 000000000000..5bf9d18f364e
--- /dev/null
+++ b/drivers/mmc/host/sdhci-npcm.c
@@ -0,0 +1,94 @@
+// SPDX-License-Identifier: GPL-2.0+
+/*
+ * NPCM SDHC MMC host controller driver.
+ *
+ * Copyright (c) 2023 Nuvoton Technology corporation.
+ */
+
+#include <linux/clk.h>
+#include <linux/err.h>
+#include <linux/io.h>
+#include <linux/mmc/host.h>
+#include <linux/mmc/mmc.h>
+#include <linux/mod_devicetable.h>
+#include <linux/module.h>
+#include <linux/of.h>
+
+#include "sdhci-pltfm.h"
+
+static const struct sdhci_pltfm_data npcm7xx_sdhci_pdata = {
+	.quirks  = SDHCI_QUIRK_DELAY_AFTER_POWER,
+	.quirks2 = SDHCI_QUIRK2_STOP_WITH_TC |
+		   SDHCI_QUIRK2_NO_1_8_V,
+};
+
+static const struct sdhci_pltfm_data npcm8xx_sdhci_pdata = {
+	.quirks  = SDHCI_QUIRK_DELAY_AFTER_POWER,
+	.quirks2 = SDHCI_QUIRK2_STOP_WITH_TC,
+};
+
+static int npcm_sdhci_probe(struct platform_device *pdev)
+{
+	const struct sdhci_pltfm_data *data;
+	struct sdhci_pltfm_host *pltfm_host;
+	struct device *dev = &pdev->dev;
+	struct sdhci_host *host;
+	u32 caps;
+	int ret;
+
+	data = of_device_get_match_data(dev);
+	if (!data)
+		return -EINVAL;
+
+	host = sdhci_pltfm_init(pdev, data, 0);
+	if (IS_ERR(host))
+		return PTR_ERR(host);
+
+	pltfm_host = sdhci_priv(host);
+
+	pltfm_host->clk = devm_clk_get_optional_enabled(dev, NULL);
+	if (IS_ERR(pltfm_host->clk)) {
+		ret = PTR_ERR(pltfm_host->clk);
+		goto err_sdhci;
+	}
+
+	caps = sdhci_readl(host, SDHCI_CAPABILITIES);
+	if (caps & SDHCI_CAN_DO_8BIT)
+		host->mmc->caps |= MMC_CAP_8_BIT_DATA;
+
+	ret = mmc_of_parse(host->mmc);
+	if (ret)
+		goto err_sdhci;
+
+	ret = sdhci_add_host(host);
+	if (ret)
+		goto err_sdhci;
+
+	return 0;
+
+err_sdhci:
+	sdhci_pltfm_free(pdev);
+	return ret;
+}
+
+static const struct of_device_id npcm_sdhci_of_match[] = {
+	{ .compatible = "nuvoton,npcm750-sdhci", .data = &npcm7xx_sdhci_pdata },
+	{ .compatible = "nuvoton,npcm845-sdhci", .data = &npcm8xx_sdhci_pdata },
+	{ }
+};
+MODULE_DEVICE_TABLE(of, npcm_sdhci_of_match);
+
+static struct platform_driver npcm_sdhci_driver = {
+	.driver = {
+		.name	= "npcm-sdhci",
+		.of_match_table = npcm_sdhci_of_match,
+		.pm	= &sdhci_pltfm_pmops,
+	},
+	.probe		= npcm_sdhci_probe,
+	.remove_new	= sdhci_pltfm_remove,
+};
+module_platform_driver(npcm_sdhci_driver);
+
+MODULE_DESCRIPTION("NPCM Secure Digital Host Controller Interface driver");
+MODULE_AUTHOR("Tomer Maimon <tomer.maimon@nuvoton.com>");
+MODULE_LICENSE("GPL");
diff --git a/drivers/mmc/host/sdhci-pci-core.c b/drivers/mmc/host/sdhci-pci-core.c
index 7c14feb5db77..025b31aa712c 100644
--- a/drivers/mmc/host/sdhci-pci-core.c
+++ b/drivers/mmc/host/sdhci-pci-core.c
@@ -483,11 +483,12 @@ static int __intel_dsm(struct intel_host *intel_host, struct device *dev,
 	int err = 0;
 	size_t len;
 
-	obj = acpi_evaluate_dsm(ACPI_HANDLE(dev), &intel_dsm_guid, 0, fn, NULL);
+	obj = acpi_evaluate_dsm_typed(ACPI_HANDLE(dev), &intel_dsm_guid, 0, fn, NULL,
+				      ACPI_TYPE_BUFFER);
 	if (!obj)
 		return -EOPNOTSUPP;
 
-	if (obj->type != ACPI_TYPE_BUFFER || obj->buffer.length < 1) {
+	if (obj->buffer.length < 1) {
 		err = -EINVAL;
 		goto out;
 	}
diff --git a/drivers/mmc/host/sdhci-pci-gli.c b/drivers/mmc/host/sdhci-pci-gli.c
index 109d4b010f97..d83261e857a5 100644
--- a/drivers/mmc/host/sdhci-pci-gli.c
+++ b/drivers/mmc/host/sdhci-pci-gli.c
@@ -25,6 +25,9 @@
 #define   GLI_9750_WT_EN_ON	    0x1
 #define   GLI_9750_WT_EN_OFF	    0x0
 
+#define PCI_GLI_9750_PM_CTRL	0xFC
+#define   PCI_GLI_9750_PM_STATE	  GENMASK(1, 0)
+
 #define SDHCI_GLI_9750_CFG2          0x848
 #define   SDHCI_GLI_9750_CFG2_L1DLY    GENMASK(28, 24)
 #define   GLI_9750_CFG2_L1DLY_VALUE    0x1F
@@ -536,8 +539,12 @@ static void sdhci_gl9750_set_clock(struct sdhci_host *host, unsigned int clock)
 
 static void gl9750_hw_setting(struct sdhci_host *host)
 {
+	struct sdhci_pci_slot *slot = sdhci_priv(host);
+	struct pci_dev *pdev;
 	u32 value;
 
+	pdev = slot->chip->pdev;
+
 	gl9750_wt_on(host);
 
 	value = sdhci_readl(host, SDHCI_GLI_9750_CFG2);
@@ -547,6 +554,13 @@ static void gl9750_hw_setting(struct sdhci_host *host)
 			    GLI_9750_CFG2_L1DLY_VALUE);
 	sdhci_writel(host, value, SDHCI_GLI_9750_CFG2);
 
+	/* toggle PM state to allow GL9750 to enter ASPM L1.2 */
+	pci_read_config_dword(pdev, PCI_GLI_9750_PM_CTRL, &value);
+	value |= PCI_GLI_9750_PM_STATE;
+	pci_write_config_dword(pdev, PCI_GLI_9750_PM_CTRL, value);
+	value &= ~PCI_GLI_9750_PM_STATE;
+	pci_write_config_dword(pdev, PCI_GLI_9750_PM_CTRL, value);
+
 	gl9750_wt_off(host);
 }
 
diff --git a/drivers/mmc/host/sdhci-pltfm.c b/drivers/mmc/host/sdhci-pltfm.c
index a72e123a585d..62753d72198a 100644
--- a/drivers/mmc/host/sdhci-pltfm.c
+++ b/drivers/mmc/host/sdhci-pltfm.c
@@ -19,7 +19,6 @@
 #include <linux/err.h>
 #include <linux/module.h>
 #include <linux/property.h>
-#include <linux/of.h>
 #ifdef CONFIG_PPC
 #include <asm/machdep.h>
 #endif
@@ -56,19 +55,16 @@ static bool sdhci_wp_inverted(struct device *dev)
 
 static void sdhci_get_compatibility(struct platform_device *pdev)
 {
+	struct device *dev = &pdev->dev;
 	struct sdhci_host *host = platform_get_drvdata(pdev);
-	struct device_node *np = pdev->dev.of_node;
-
-	if (!np)
-		return;
 
-	if (of_device_is_compatible(np, "fsl,p2020-rev1-esdhc"))
+	if (device_is_compatible(dev, "fsl,p2020-rev1-esdhc"))
 		host->quirks |= SDHCI_QUIRK_BROKEN_DMA;
 
-	if (of_device_is_compatible(np, "fsl,p2020-esdhc") ||
-	    of_device_is_compatible(np, "fsl,p1010-esdhc") ||
-	    of_device_is_compatible(np, "fsl,t4240-esdhc") ||
-	    of_device_is_compatible(np, "fsl,mpc8536-esdhc"))
+	if (device_is_compatible(dev, "fsl,p2020-esdhc") ||
+	    device_is_compatible(dev, "fsl,p1010-esdhc") ||
+	    device_is_compatible(dev, "fsl,t4240-esdhc") ||
+	    device_is_compatible(dev, "fsl,mpc8536-esdhc"))
 		host->quirks |= SDHCI_QUIRK_BROKEN_TIMEOUT_VAL;
 }
 
@@ -115,26 +111,21 @@ struct sdhci_host *sdhci_pltfm_init(struct platform_device *pdev,
 {
 	struct sdhci_host *host;
 	void __iomem *ioaddr;
-	int irq, ret;
+	int irq;
 
 	ioaddr = devm_platform_ioremap_resource(pdev, 0);
-	if (IS_ERR(ioaddr)) {
-		ret = PTR_ERR(ioaddr);
-		goto err;
-	}
+	if (IS_ERR(ioaddr))
+		return ERR_CAST(ioaddr);
 
 	irq = platform_get_irq(pdev, 0);
-	if (irq < 0) {
-		ret = irq;
-		goto err;
-	}
+	if (irq < 0)
+		return ERR_PTR(irq);
 
 	host = sdhci_alloc_host(&pdev->dev,
 		sizeof(struct sdhci_pltfm_host) + priv_size);
-
 	if (IS_ERR(host)) {
-		ret = PTR_ERR(host);
-		goto err;
+		dev_err(&pdev->dev, "%s failed %pe\n", __func__, host);
+		return ERR_CAST(host);
 	}
 
 	host->ioaddr = ioaddr;
@@ -152,9 +143,6 @@ struct sdhci_host *sdhci_pltfm_init(struct platform_device *pdev,
 	platform_set_drvdata(pdev, host);
 
 	return host;
-err:
-	dev_err(&pdev->dev, "%s failed %d\n", __func__, ret);
-	return ERR_PTR(ret);
 }
 EXPORT_SYMBOL_GPL(sdhci_pltfm_init);
 
diff --git a/drivers/mmc/host/vub300.c b/drivers/mmc/host/vub300.c
index 9ec593d52f0f..de3f443f5fdc 100644
--- a/drivers/mmc/host/vub300.c
+++ b/drivers/mmc/host/vub300.c
@@ -512,7 +512,7 @@ static void new_system_port_status(struct vub300_mmc_host *vub300)
 		vub300->card_present = 1;
 		vub300->bus_width = 0;
 		if (disable_offload_processing)
-			strncpy(vub300->vub_name, "EMPTY Processing Disabled",
+			strscpy(vub300->vub_name, "EMPTY Processing Disabled",
 				sizeof(vub300->vub_name));
 		else
 			vub300->vub_name[0] = 0;
@@ -1216,7 +1216,7 @@ static void __download_offload_pseudocode(struct vub300_mmc_host *vub300,
 		dev_err(&vub300->udev->dev,
 			"corrupt offload pseudocode in firmware %s\n",
 			vub300->vub_name);
-		strncpy(vub300->vub_name, "corrupt offload pseudocode",
+		strscpy(vub300->vub_name, "corrupt offload pseudocode",
 			sizeof(vub300->vub_name));
 		return;
 	}
@@ -1250,7 +1250,7 @@ static void __download_offload_pseudocode(struct vub300_mmc_host *vub300,
 				"not enough memory for xfer buffer to send"
 				" INTERRUPT_PSEUDOCODE for %s %s\n", fw->data,
 				vub300->vub_name);
-			strncpy(vub300->vub_name,
+			strscpy(vub300->vub_name,
 				"SDIO interrupt pseudocode download failed",
 				sizeof(vub300->vub_name));
 			return;
@@ -1259,7 +1259,7 @@ static void __download_offload_pseudocode(struct vub300_mmc_host *vub300,
 		dev_err(&vub300->udev->dev,
 			"corrupt interrupt pseudocode in firmware %s %s\n",
 			fw->data, vub300->vub_name);
-		strncpy(vub300->vub_name, "corrupt interrupt pseudocode",
+		strscpy(vub300->vub_name, "corrupt interrupt pseudocode",
 			sizeof(vub300->vub_name));
 		return;
 	}
@@ -1293,7 +1293,7 @@ static void __download_offload_pseudocode(struct vub300_mmc_host *vub300,
 				"not enough memory for xfer buffer to send"
 				" TRANSFER_PSEUDOCODE for %s %s\n", fw->data,
 				vub300->vub_name);
-			strncpy(vub300->vub_name,
+			strscpy(vub300->vub_name,
 				"SDIO transfer pseudocode download failed",
 				sizeof(vub300->vub_name));
 			return;
@@ -1302,7 +1302,7 @@ static void __download_offload_pseudocode(struct vub300_mmc_host *vub300,
 		dev_err(&vub300->udev->dev,
 			"corrupt transfer pseudocode in firmware %s %s\n",
 			fw->data, vub300->vub_name);
-		strncpy(vub300->vub_name, "corrupt transfer pseudocode",
+		strscpy(vub300->vub_name, "corrupt transfer pseudocode",
 			sizeof(vub300->vub_name));
 		return;
 	}
@@ -1336,13 +1336,13 @@ static void __download_offload_pseudocode(struct vub300_mmc_host *vub300,
 		dev_err(&vub300->udev->dev,
 			"corrupt dynamic registers in firmware %s\n",
 			vub300->vub_name);
-		strncpy(vub300->vub_name, "corrupt dynamic registers",
+		strscpy(vub300->vub_name, "corrupt dynamic registers",
 			sizeof(vub300->vub_name));
 		return;
 	}
 
 copy_error_message:
-	strncpy(vub300->vub_name, "SDIO pseudocode download failed",
+	strscpy(vub300->vub_name, "SDIO pseudocode download failed",
 		sizeof(vub300->vub_name));
 }
 
@@ -1370,11 +1370,11 @@ static void download_offload_pseudocode(struct vub300_mmc_host *vub300)
 		 vub300->vub_name);
 	retval = request_firmware(&fw, vub300->vub_name, &card->dev);
 	if (retval < 0) {
-		strncpy(vub300->vub_name, "vub_default.bin",
+		strscpy(vub300->vub_name, "vub_default.bin",
 			sizeof(vub300->vub_name));
 		retval = request_firmware(&fw, vub300->vub_name, &card->dev);
 		if (retval < 0) {
-			strncpy(vub300->vub_name,
+			strscpy(vub300->vub_name,
 				"no SDIO offload firmware found",
 				sizeof(vub300->vub_name));
 		} else {
@@ -1758,7 +1758,7 @@ static void vub300_cmndwork_thread(struct work_struct *work)
 			 * has been already downloaded to the VUB300 chip
 			 */
 		} else if (0 == vub300->mmc->card->sdio_funcs) {
-			strncpy(vub300->vub_name, "SD memory device",
+			strscpy(vub300->vub_name, "SD memory device",
 				sizeof(vub300->vub_name));
 		} else {
 			download_offload_pseudocode(vub300);