diff options
author | Linus Torvalds <torvalds@linux-foundation.org> | 2024-07-16 19:28:34 -0700 |
---|---|---|
committer | Linus Torvalds <torvalds@linux-foundation.org> | 2024-07-16 19:28:34 -0700 |
commit | 51835949dda3783d4639cfa74ce13a3c9829de00 (patch) | |
tree | 2b593de5eba6ecc73f7c58fc65fdaffae45c7323 /drivers/net/ethernet/wangxun | |
parent | 0434dbe32053d07d658165be681505120c6b1abc (diff) | |
parent | 77ae5e5b00720372af2860efdc4bc652ac682696 (diff) |
Merge tag 'net-next-6.11' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-nextHEADmaster
Pull networking updates from Jakub Kicinski:
"Not much excitement - a handful of large patchsets (devmem among them)
did not make it in time.
Core & protocols:
- Use local_lock in addition to local_bh_disable() to protect per-CPU
resources in networking, a step closer for local_bh_disable() not
to act as a big lock on PREEMPT_RT
- Use flex array for netdevice priv area, ensure its cache alignment
- Add a sysctl knob to allow user to specify a default rto_min at
socket init time. Bit of a big hammer but multiple companies were
independently carrying such patch downstream so clearly it's useful
- Support scheduling transmission of packets based on CLOCK_TAI
- Un-pin TCP TIMEWAIT timer to avoid it firing on CPUs later cordoned
off using cpusets
- Support multiple L2TPv3 UDP tunnels using the same 5-tuple address
- Allow configuration of multipath hash seed, to both allow
synchronizing hashing of two routers, and preventing partial
accidental sync
- Improve TCP compliance with RFC 9293 for simultaneous connect()
- Support sending NAT keepalives in IPsec ESP in UDP states.
Userspace IKE daemon had to do this before, but the kernel can
better keep track of it
- Support sending supervision HSR frames with MAC addresses stored in
ProxyNodeTable when RedBox (i.e. HSR-SAN) is enabled
- Introduce IPPROTO_SMC for selecting SMC when socket is created
- Allow UDP GSO transmit from devices with no checksum offload
- openvswitch: add packet sampling via psample, separating the
sampled traffic from "upcall" packets sent to user space for
forwarding
- nf_tables: shrink memory consumption for transaction objects
Things we sprinkled into general kernel code:
- Power Sequencing subsystem (used by Qualcomm Bluetooth driver for
QCA6390) [ Already merged separately - Linus ]
- Add IRQ information in sysfs for auxiliary bus
- Introduce guard definition for local_lock
- Add aligned flavor of __cacheline_group_{begin, end}() markings for
grouping fields in structures
BPF:
- Notify user space (via epoll) when a struct_ops object is getting
detached/unregistered
- Add new kfuncs for a generic, open-coded bits iterator
- Enable BPF programs to declare arrays of kptr, bpf_rb_root, and
bpf_list_head
- Support resilient split BTF which cuts down on duplication and
makes BTF as compact as possible WRT BTF from modules
- Add support for dumping kfunc prototypes from BTF which enables
both detecting as well as dumping compilable prototypes for kfuncs
- riscv64 BPF JIT improvements in particular to add 12-argument
support for BPF trampolines and to utilize bpf_prog_pack for the
latter
- Add the capability to offload the netfilter flowtable in XDP layer
through kfuncs
Driver API:
- Allow users to configure IRQ tresholds between which automatic IRQ
moderation can choose
- Expand Power Sourcing (PoE) status with power, class and failure
reason. Support setting power limits
- Track additional RSS contexts in the core, make sure configuration
changes don't break them
- Support IPsec crypto offload for IPv6 ESP and IPv4 UDP-encapsulated
ESP data paths
- Support updating firmware on SFP modules
Tests and tooling:
- mptcp: use net/lib.sh to manage netns
- TCP-AO and TCP-MD5: replace debug prints used by tests with
tracepoints
- openvswitch: make test self-contained (don't depend on OvS CLI
tools)
Drivers:
- Ethernet high-speed NICs:
- Broadcom (bnxt):
- increase the max total outstanding PTP TX packets to 4
- add timestamping statistics support
- implement netdev_queue_mgmt_ops
- support new RSS context API
- Intel (100G, ice, idpf):
- implement FEC statistics and dumping signal quality indicators
- support E825C products (with 56Gbps PHYs)
- nVidia/Mellanox:
- support HW-GRO
- mlx4/mlx5: support per-queue statistics via netlink
- obey the max number of EQs setting in sub-functions
- AMD/Solarflare:
- support new RSS context API
- AMD/Pensando:
- ionic: rework fix for doorbell miss to lower overhead and
skip it on new HW
- Wangxun:
- txgbe: support Flow Director perfect filters
- Ethernet NICs consumer, embedded and virtual:
- Add driver for Tehuti Networks TN40xx chips
- Add driver for Meta's internal NIC chips
- Add driver for Ethernet MAC on Airoha EN7581 SoCs
- Add driver for Renesas Ethernet-TSN devices
- Google cloud vNIC:
- flow steering support
- Microsoft vNIC:
- support page sizes other than 4KB on ARM64
- vmware vNIC:
- support latency measurement (update to version 9)
- VirtIO net:
- support for Byte Queue Limits
- support configuring thresholds for automatic IRQ moderation
- support for AF_XDP Rx zero-copy
- Synopsys (stmmac):
- support for STM32MP13 SoC
- let platforms select the right PCS implementation
- TI:
- icssg-prueth: add multicast filtering support
- icssg-prueth: enable PTP timestamping and PPS
- Renesas:
- ravb: improve Rx performance 30-400% by using page pool,
theaded NAPI and timer-based IRQ coalescing
- ravb: add MII support for R-Car V4M
- Cadence (macb):
- macb: add ARP support to Wake-On-LAN
- Cortina:
- use phylib for RX and TX pause configuration
- Ethernet switches:
- nVidia/Mellanox:
- support configuration of multipath hash seed
- report more accurate max MTU
- use page_pool to improve Rx performance
- MediaTek:
- mt7530: add support for bridge port isolation
- Qualcomm:
- qca8k: add support for bridge port isolation
- Microchip:
- lan9371/2: add 100BaseTX PHY support
- NXP:
- vsc73xx: implement VLAN operations
- Ethernet PHYs:
- aquantia: enable support for aqr115c
- aquantia: add support for PHY LEDs
- realtek: add support for rtl8224 2.5Gbps PHY
- xpcs: add memory-mapped device support
- add BroadR-Reach link mode and support in Broadcom's PHY driver
- CAN:
- add document for ISO 15765-2 protocol support
- mcp251xfd: workaround for erratum DS80000789E, use timestamps to
catch when device returns incorrect FIFO status
- WiFi:
- mac80211/cfg80211:
- parse Transmit Power Envelope (TPE) data in mac80211 instead
of in drivers
- improvements for 6 GHz regulatory flexibility
- multi-link improvements
- support multiple radios per wiphy
- remove DEAUTH_NEED_MGD_TX_PREP flag
- Intel (iwlwifi):
- bump FW API to 91 for BZ/SC devices
- report 64-bit radiotap timestamp
- enable P2P low latency by default
- handle Transmit Power Envelope (TPE) advertised by AP
- remove support for older FW for new devices
- fast resume (keeping the device configured)
- mvm: re-enable Multi-Link Operation (MLO)
- aggregation (A-MSDU) optimizations
- MediaTek (mt76):
- mt7925 Multi-Link Operation (MLO) support
- Qualcomm (ath10k):
- LED support for various chipsets
- Qualcomm (ath12k):
- remove unsupported Tx monitor handling
- support channel 2 in 6 GHz band
- support Spatial Multiplexing Power Save (SMPS) in 6 GHz band
- supprt multiple BSSID (MBSSID) and Enhanced Multi-BSSID
Advertisements (EMA)
- support dynamic VLAN
- add panic handler for resetting the firmware state
- DebugFS support for datapath statistics
- WCN7850: support for Wake on WLAN
- Microchip (wilc1000):
- read MAC address during probe to make it visible to user space
- suspend/resume improvements
- TI (wl18xx):
- support newer firmware versions
- RealTek (rtw89):
- preparation for RTL8852BE-VT support
- Wake on WLAN support for WiFi 6 chips
- 36-bit PCI DMA support
- RealTek (rtlwifi):
- RTL8192DU support
- Broadcom (brcmfmac):
- Management Frame Protection support (to enable WPA3)
- Bluetooth:
- qualcomm: use the power sequencer for QCA6390
- btusb: mediatek: add ISO data transmission functions
- hci_bcm4377: add BCM4388 support
- btintel: add support for BlazarU core
- btintel: add support for Whale Peak2
- btnxpuart: add support for AW693 A1 chipset
- btnxpuart: add support for IW615 chipset
- btusb: add Realtek RTL8852BE support ID 0x13d3:0x3591"
* tag 'net-next-6.11' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next: (1589 commits)
eth: fbnic: Fix spelling mistake "tiggerring" -> "triggering"
tcp: Replace strncpy() with strscpy()
wifi: ath12k: fix build vs old compiler
tcp: Don't access uninit tcp_rsk(req)->ao_keyid in tcp_create_openreq_child().
eth: fbnic: Write the TCAM tables used for RSS control and Rx to host
eth: fbnic: Add L2 address programming
eth: fbnic: Add basic Rx handling
eth: fbnic: Add basic Tx handling
eth: fbnic: Add link detection
eth: fbnic: Add initial messaging to notify FW of our presence
eth: fbnic: Implement Rx queue alloc/start/stop/free
eth: fbnic: Implement Tx queue alloc/start/stop/free
eth: fbnic: Allocate a netdevice and napi vectors with queues
eth: fbnic: Add FW communication mechanism
eth: fbnic: Add message parsing for FW messages
eth: fbnic: Add register init to set PCIe/Ethernet device config
eth: fbnic: Allocate core device specific structures and devlink interface
eth: fbnic: Add scaffolding for Meta's NIC driver
PCI: Add Meta Platforms vendor ID
net/sched: cls_flower: propagate tca[TCA_OPTIONS] to NL_REQ_ATTR_CHECK
...
Diffstat (limited to 'drivers/net/ethernet/wangxun')
-rw-r--r-- | drivers/net/ethernet/wangxun/libwx/wx_ethtool.c | 39 | ||||
-rw-r--r-- | drivers/net/ethernet/wangxun/libwx/wx_hw.c | 32 | ||||
-rw-r--r-- | drivers/net/ethernet/wangxun/libwx/wx_hw.h | 2 | ||||
-rw-r--r-- | drivers/net/ethernet/wangxun/libwx/wx_lib.c | 62 | ||||
-rw-r--r-- | drivers/net/ethernet/wangxun/libwx/wx_lib.h | 1 | ||||
-rw-r--r-- | drivers/net/ethernet/wangxun/libwx/wx_type.h | 56 | ||||
-rw-r--r-- | drivers/net/ethernet/wangxun/ngbe/ngbe_ethtool.c | 4 | ||||
-rw-r--r-- | drivers/net/ethernet/wangxun/ngbe/ngbe_main.c | 2 | ||||
-rw-r--r-- | drivers/net/ethernet/wangxun/txgbe/Makefile | 1 | ||||
-rw-r--r-- | drivers/net/ethernet/wangxun/txgbe/txgbe_ethtool.c | 427 | ||||
-rw-r--r-- | drivers/net/ethernet/wangxun/txgbe/txgbe_fdir.c | 643 | ||||
-rw-r--r-- | drivers/net/ethernet/wangxun/txgbe/txgbe_fdir.h | 20 | ||||
-rw-r--r-- | drivers/net/ethernet/wangxun/txgbe/txgbe_main.c | 18 | ||||
-rw-r--r-- | drivers/net/ethernet/wangxun/txgbe/txgbe_type.h | 147 |
14 files changed, 1439 insertions, 15 deletions
diff --git a/drivers/net/ethernet/wangxun/libwx/wx_ethtool.c b/drivers/net/ethernet/wangxun/libwx/wx_ethtool.c index cc3bec42ed8e..abe5921dde02 100644 --- a/drivers/net/ethernet/wangxun/libwx/wx_ethtool.c +++ b/drivers/net/ethernet/wangxun/libwx/wx_ethtool.c @@ -43,6 +43,11 @@ static const struct wx_stats wx_gstrings_stats[] = { WX_STAT("alloc_rx_buff_failed", alloc_rx_buff_failed), }; +static const struct wx_stats wx_gstrings_fdir_stats[] = { + WX_STAT("fdir_match", stats.fdirmatch), + WX_STAT("fdir_miss", stats.fdirmiss), +}; + /* drivers allocates num_tx_queues and num_rx_queues symmetrically so * we set the num_rx_queues to evaluate to num_tx_queues. This is * used because we do not have a good way to get the max number of @@ -55,13 +60,17 @@ static const struct wx_stats wx_gstrings_stats[] = { (WX_NUM_TX_QUEUES + WX_NUM_RX_QUEUES) * \ (sizeof(struct wx_queue_stats) / sizeof(u64))) #define WX_GLOBAL_STATS_LEN ARRAY_SIZE(wx_gstrings_stats) +#define WX_FDIR_STATS_LEN ARRAY_SIZE(wx_gstrings_fdir_stats) #define WX_STATS_LEN (WX_GLOBAL_STATS_LEN + WX_QUEUE_STATS_LEN) int wx_get_sset_count(struct net_device *netdev, int sset) { + struct wx *wx = netdev_priv(netdev); + switch (sset) { case ETH_SS_STATS: - return WX_STATS_LEN; + return (wx->mac.type == wx_mac_sp) ? + WX_STATS_LEN + WX_FDIR_STATS_LEN : WX_STATS_LEN; default: return -EOPNOTSUPP; } @@ -70,6 +79,7 @@ EXPORT_SYMBOL(wx_get_sset_count); void wx_get_strings(struct net_device *netdev, u32 stringset, u8 *data) { + struct wx *wx = netdev_priv(netdev); u8 *p = data; int i; @@ -77,6 +87,10 @@ void wx_get_strings(struct net_device *netdev, u32 stringset, u8 *data) case ETH_SS_STATS: for (i = 0; i < WX_GLOBAL_STATS_LEN; i++) ethtool_puts(&p, wx_gstrings_stats[i].stat_string); + if (wx->mac.type == wx_mac_sp) { + for (i = 0; i < WX_FDIR_STATS_LEN; i++) + ethtool_puts(&p, wx_gstrings_fdir_stats[i].stat_string); + } for (i = 0; i < netdev->num_tx_queues; i++) { ethtool_sprintf(&p, "tx_queue_%u_packets", i); ethtool_sprintf(&p, "tx_queue_%u_bytes", i); @@ -96,7 +110,7 @@ void wx_get_ethtool_stats(struct net_device *netdev, struct wx *wx = netdev_priv(netdev); struct wx_ring *ring; unsigned int start; - int i, j; + int i, j, k; char *p; wx_update_stats(wx); @@ -107,6 +121,13 @@ void wx_get_ethtool_stats(struct net_device *netdev, sizeof(u64)) ? *(u64 *)p : *(u32 *)p; } + if (wx->mac.type == wx_mac_sp) { + for (k = 0; k < WX_FDIR_STATS_LEN; k++) { + p = (char *)wx + wx_gstrings_fdir_stats[k].stat_offset; + data[i++] = *(u64 *)p; + } + } + for (j = 0; j < netdev->num_tx_queues; j++) { ring = wx->tx_ring[j]; if (!ring) { @@ -172,17 +193,21 @@ EXPORT_SYMBOL(wx_get_pause_stats); void wx_get_drvinfo(struct net_device *netdev, struct ethtool_drvinfo *info) { + unsigned int stats_len = WX_STATS_LEN; struct wx *wx = netdev_priv(netdev); + if (wx->mac.type == wx_mac_sp) + stats_len += WX_FDIR_STATS_LEN; + strscpy(info->driver, wx->driver_name, sizeof(info->driver)); strscpy(info->fw_version, wx->eeprom_id, sizeof(info->fw_version)); strscpy(info->bus_info, pci_name(wx->pdev), sizeof(info->bus_info)); if (wx->num_tx_queues <= WX_NUM_TX_QUEUES) { - info->n_stats = WX_STATS_LEN - + info->n_stats = stats_len - (WX_NUM_TX_QUEUES - wx->num_tx_queues) * (sizeof(struct wx_queue_stats) / sizeof(u64)) * 2; } else { - info->n_stats = WX_STATS_LEN; + info->n_stats = stats_len; } } EXPORT_SYMBOL(wx_get_drvinfo); @@ -383,6 +408,9 @@ void wx_get_channels(struct net_device *dev, /* record RSS queues */ ch->combined_count = wx->ring_feature[RING_F_RSS].indices; + + if (test_bit(WX_FLAG_FDIR_CAPABLE, wx->flags)) + ch->combined_count = wx->ring_feature[RING_F_FDIR].indices; } EXPORT_SYMBOL(wx_get_channels); @@ -400,6 +428,9 @@ int wx_set_channels(struct net_device *dev, if (count > wx_max_channels(wx)) return -EINVAL; + if (test_bit(WX_FLAG_FDIR_CAPABLE, wx->flags)) + wx->ring_feature[RING_F_FDIR].limit = count; + wx->ring_feature[RING_F_RSS].limit = count; return 0; diff --git a/drivers/net/ethernet/wangxun/libwx/wx_hw.c b/drivers/net/ethernet/wangxun/libwx/wx_hw.c index d1b682ce9c6d..1bf9c38e4125 100644 --- a/drivers/net/ethernet/wangxun/libwx/wx_hw.c +++ b/drivers/net/ethernet/wangxun/libwx/wx_hw.c @@ -1147,8 +1147,15 @@ static void wx_enable_rx(struct wx *wx) static void wx_set_rxpba(struct wx *wx) { u32 rxpktsize, txpktsize, txpbthresh; + u32 pbsize = wx->mac.rx_pb_size; - rxpktsize = wx->mac.rx_pb_size << WX_RDB_PB_SZ_SHIFT; + if (test_bit(WX_FLAG_FDIR_CAPABLE, wx->flags)) { + if (test_bit(WX_FLAG_FDIR_HASH, wx->flags) || + test_bit(WX_FLAG_FDIR_PERFECT, wx->flags)) + pbsize -= 64; /* Default 64KB */ + } + + rxpktsize = pbsize << WX_RDB_PB_SZ_SHIFT; wr32(wx, WX_RDB_PB_SZ(0), rxpktsize); /* Only support an equally distributed Tx packet buffer strategy. */ @@ -1261,7 +1268,7 @@ static void wx_configure_port(struct wx *wx) * Stops the receive data path and waits for the HW to internally empty * the Rx security block **/ -static int wx_disable_sec_rx_path(struct wx *wx) +int wx_disable_sec_rx_path(struct wx *wx) { u32 secrx; @@ -1271,6 +1278,7 @@ static int wx_disable_sec_rx_path(struct wx *wx) return read_poll_timeout(rd32, secrx, secrx & WX_RSC_ST_RSEC_RDY, 1000, 40000, false, wx, WX_RSC_ST); } +EXPORT_SYMBOL(wx_disable_sec_rx_path); /** * wx_enable_sec_rx_path - Enables the receive data path @@ -1278,11 +1286,12 @@ static int wx_disable_sec_rx_path(struct wx *wx) * * Enables the receive data path. **/ -static void wx_enable_sec_rx_path(struct wx *wx) +void wx_enable_sec_rx_path(struct wx *wx) { wr32m(wx, WX_RSC_CTL, WX_RSC_CTL_RX_DIS, 0); WX_WRITE_FLUSH(wx); } +EXPORT_SYMBOL(wx_enable_sec_rx_path); static void wx_vlan_strip_control(struct wx *wx, bool enable) { @@ -1499,6 +1508,13 @@ static void wx_configure_tx_ring(struct wx *wx, txdctl |= ring->count / 128 << WX_PX_TR_CFG_TR_SIZE_SHIFT; txdctl |= 0x20 << WX_PX_TR_CFG_WTHRESH_SHIFT; + ring->atr_count = 0; + if (test_bit(WX_FLAG_FDIR_CAPABLE, wx->flags) && + test_bit(WX_FLAG_FDIR_HASH, wx->flags)) + ring->atr_sample_rate = wx->atr_sample_rate; + else + ring->atr_sample_rate = 0; + /* reinitialize tx_buffer_info */ memset(ring->tx_buffer_info, 0, sizeof(struct wx_tx_buffer) * ring->count); @@ -1732,7 +1748,9 @@ void wx_configure(struct wx *wx) wx_set_rx_mode(wx->netdev); wx_restore_vlan(wx); - wx_enable_sec_rx_path(wx); + + if (test_bit(WX_FLAG_FDIR_CAPABLE, wx->flags)) + wx->configure_fdir(wx); wx_configure_tx(wx); wx_configure_rx(wx); @@ -1959,6 +1977,7 @@ int wx_sw_init(struct wx *wx) } bitmap_zero(wx->state, WX_STATE_NBITS); + bitmap_zero(wx->flags, WX_PF_FLAGS_NBITS); wx->misc_irq_domain = false; return 0; @@ -2334,6 +2353,11 @@ void wx_update_stats(struct wx *wx) hwstats->b2ogprc += rd32(wx, WX_RDM_BMC2OS_CNT); hwstats->rdmdrop += rd32(wx, WX_RDM_DRP_PKT); + if (wx->mac.type == wx_mac_sp) { + hwstats->fdirmatch += rd32(wx, WX_RDB_FDIR_MATCH); + hwstats->fdirmiss += rd32(wx, WX_RDB_FDIR_MISS); + } + for (i = 0; i < wx->mac.max_rx_queues; i++) hwstats->qmprc += rd32(wx, WX_PX_MPRC(i)); } diff --git a/drivers/net/ethernet/wangxun/libwx/wx_hw.h b/drivers/net/ethernet/wangxun/libwx/wx_hw.h index 9e219fa717a2..11fb33349482 100644 --- a/drivers/net/ethernet/wangxun/libwx/wx_hw.h +++ b/drivers/net/ethernet/wangxun/libwx/wx_hw.h @@ -28,6 +28,8 @@ void wx_mac_set_default_filter(struct wx *wx, u8 *addr); void wx_flush_sw_mac_table(struct wx *wx); int wx_set_mac(struct net_device *netdev, void *p); void wx_disable_rx(struct wx *wx); +int wx_disable_sec_rx_path(struct wx *wx); +void wx_enable_sec_rx_path(struct wx *wx); void wx_set_rx_mode(struct net_device *netdev); int wx_change_mtu(struct net_device *netdev, int new_mtu); void wx_disable_rx_queue(struct wx *wx, struct wx_ring *ring); diff --git a/drivers/net/ethernet/wangxun/libwx/wx_lib.c b/drivers/net/ethernet/wangxun/libwx/wx_lib.c index 81bedc8ee8d4..1eecba984f3b 100644 --- a/drivers/net/ethernet/wangxun/libwx/wx_lib.c +++ b/drivers/net/ethernet/wangxun/libwx/wx_lib.c @@ -148,10 +148,11 @@ static struct wx_dec_ptype wx_ptype_lookup[256] = { [0xFD] = WX_PTT(IP, IPV6, IGMV, IPV6, SCTP, PAY4), }; -static struct wx_dec_ptype wx_decode_ptype(const u8 ptype) +struct wx_dec_ptype wx_decode_ptype(const u8 ptype) { return wx_ptype_lookup[ptype]; } +EXPORT_SYMBOL(wx_decode_ptype); /* wx_test_staterr - tests bits in Rx descriptor status and error fields */ static __le32 wx_test_staterr(union wx_rx_desc *rx_desc, @@ -1453,6 +1454,7 @@ static void wx_tx_csum(struct wx_ring *tx_ring, struct wx_tx_buffer *first, static netdev_tx_t wx_xmit_frame_ring(struct sk_buff *skb, struct wx_ring *tx_ring) { + struct wx *wx = netdev_priv(tx_ring->netdev); u16 count = TXD_USE_COUNT(skb_headlen(skb)); struct wx_tx_buffer *first; u8 hdr_len = 0, ptype; @@ -1498,6 +1500,10 @@ static netdev_tx_t wx_xmit_frame_ring(struct sk_buff *skb, goto out_drop; else if (!tso) wx_tx_csum(tx_ring, first, ptype); + + if (test_bit(WX_FLAG_FDIR_CAPABLE, wx->flags) && tx_ring->atr_sample_rate) + wx->atr(tx_ring, first, ptype); + wx_tx_map(tx_ring, first, hdr_len); return NETDEV_TX_OK; @@ -1574,8 +1580,27 @@ static void wx_set_rss_queues(struct wx *wx) f = &wx->ring_feature[RING_F_RSS]; f->indices = f->limit; - wx->num_rx_queues = f->limit; - wx->num_tx_queues = f->limit; + if (!(test_bit(WX_FLAG_FDIR_CAPABLE, wx->flags))) + goto out; + + clear_bit(WX_FLAG_FDIR_HASH, wx->flags); + + /* Use Flow Director in addition to RSS to ensure the best + * distribution of flows across cores, even when an FDIR flow + * isn't matched. + */ + if (f->indices > 1) { + f = &wx->ring_feature[RING_F_FDIR]; + + f->indices = f->limit; + + if (!(test_bit(WX_FLAG_FDIR_PERFECT, wx->flags))) + set_bit(WX_FLAG_FDIR_HASH, wx->flags); + } + +out: + wx->num_rx_queues = f->indices; + wx->num_tx_queues = f->indices; } static void wx_set_num_queues(struct wx *wx) @@ -2684,6 +2709,7 @@ int wx_set_features(struct net_device *netdev, netdev_features_t features) { netdev_features_t changed = netdev->features ^ features; struct wx *wx = netdev_priv(netdev); + bool need_reset = false; if (features & NETIF_F_RXHASH) { wr32m(wx, WX_RDB_RA_CTL, WX_RDB_RA_CTL_RSS_EN, @@ -2701,6 +2727,36 @@ int wx_set_features(struct net_device *netdev, netdev_features_t features) else if (changed & (NETIF_F_HW_VLAN_CTAG_RX | NETIF_F_HW_VLAN_CTAG_FILTER)) wx_set_rx_mode(netdev); + if (!(test_bit(WX_FLAG_FDIR_CAPABLE, wx->flags))) + return 0; + + /* Check if Flow Director n-tuple support was enabled or disabled. If + * the state changed, we need to reset. + */ + switch (features & NETIF_F_NTUPLE) { + case NETIF_F_NTUPLE: + /* turn off ATR, enable perfect filters and reset */ + if (!(test_and_set_bit(WX_FLAG_FDIR_PERFECT, wx->flags))) + need_reset = true; + + clear_bit(WX_FLAG_FDIR_HASH, wx->flags); + break; + default: + /* turn off perfect filters, enable ATR and reset */ + if (test_and_clear_bit(WX_FLAG_FDIR_PERFECT, wx->flags)) + need_reset = true; + + /* We cannot enable ATR if RSS is disabled */ + if (wx->ring_feature[RING_F_RSS].limit <= 1) + break; + + set_bit(WX_FLAG_FDIR_HASH, wx->flags); + break; + } + + if (need_reset) + wx->do_reset(netdev); + return 0; } EXPORT_SYMBOL(wx_set_features); diff --git a/drivers/net/ethernet/wangxun/libwx/wx_lib.h b/drivers/net/ethernet/wangxun/libwx/wx_lib.h index c41b29ea812f..fdeb0c315b75 100644 --- a/drivers/net/ethernet/wangxun/libwx/wx_lib.h +++ b/drivers/net/ethernet/wangxun/libwx/wx_lib.h @@ -7,6 +7,7 @@ #ifndef _WX_LIB_H_ #define _WX_LIB_H_ +struct wx_dec_ptype wx_decode_ptype(const u8 ptype); void wx_alloc_rx_buffers(struct wx_ring *rx_ring, u16 cleaned_count); u16 wx_desc_unused(struct wx_ring *ring); netdev_tx_t wx_xmit_frame(struct sk_buff *skb, diff --git a/drivers/net/ethernet/wangxun/libwx/wx_type.h b/drivers/net/ethernet/wangxun/libwx/wx_type.h index 0df7f5712b6f..1d57b047817b 100644 --- a/drivers/net/ethernet/wangxun/libwx/wx_type.h +++ b/drivers/net/ethernet/wangxun/libwx/wx_type.h @@ -157,6 +157,8 @@ #define WX_RDB_RA_CTL_RSS_IPV6_TCP BIT(21) #define WX_RDB_RA_CTL_RSS_IPV4_UDP BIT(22) #define WX_RDB_RA_CTL_RSS_IPV6_UDP BIT(23) +#define WX_RDB_FDIR_MATCH 0x19558 +#define WX_RDB_FDIR_MISS 0x1955C /******************************* PSR Registers *******************************/ /* psr control */ @@ -503,6 +505,34 @@ enum WX_MSCA_CMD_value { #define WX_PTYPE_TYP_TCP 0x04 #define WX_PTYPE_TYP_SCTP 0x05 +/* Packet type non-ip values */ +enum wx_l2_ptypes { + WX_PTYPE_L2_ABORTED = (WX_PTYPE_PKT_MAC), + WX_PTYPE_L2_MAC = (WX_PTYPE_PKT_MAC | WX_PTYPE_TYP_MAC), + + WX_PTYPE_L2_IPV4_FRAG = (WX_PTYPE_PKT_IP | WX_PTYPE_TYP_IPFRAG), + WX_PTYPE_L2_IPV4 = (WX_PTYPE_PKT_IP | WX_PTYPE_TYP_IP), + WX_PTYPE_L2_IPV4_UDP = (WX_PTYPE_PKT_IP | WX_PTYPE_TYP_UDP), + WX_PTYPE_L2_IPV4_TCP = (WX_PTYPE_PKT_IP | WX_PTYPE_TYP_TCP), + WX_PTYPE_L2_IPV4_SCTP = (WX_PTYPE_PKT_IP | WX_PTYPE_TYP_SCTP), + WX_PTYPE_L2_IPV6_FRAG = (WX_PTYPE_PKT_IP | WX_PTYPE_PKT_IPV6 | + WX_PTYPE_TYP_IPFRAG), + WX_PTYPE_L2_IPV6 = (WX_PTYPE_PKT_IP | WX_PTYPE_PKT_IPV6 | + WX_PTYPE_TYP_IP), + WX_PTYPE_L2_IPV6_UDP = (WX_PTYPE_PKT_IP | WX_PTYPE_PKT_IPV6 | + WX_PTYPE_TYP_UDP), + WX_PTYPE_L2_IPV6_TCP = (WX_PTYPE_PKT_IP | WX_PTYPE_PKT_IPV6 | + WX_PTYPE_TYP_TCP), + WX_PTYPE_L2_IPV6_SCTP = (WX_PTYPE_PKT_IP | WX_PTYPE_PKT_IPV6 | + WX_PTYPE_TYP_SCTP), + + WX_PTYPE_L2_TUN4_MAC = (WX_PTYPE_TUN_IPV4 | WX_PTYPE_PKT_IGM), + WX_PTYPE_L2_TUN6_MAC = (WX_PTYPE_TUN_IPV6 | WX_PTYPE_PKT_IGM), +}; + +#define WX_PTYPE_PKT(_pt) ((_pt) & 0x30) +#define WX_PTYPE_TYPL4(_pt) ((_pt) & 0x07) + #define WX_RXD_PKTTYPE(_rxd) \ ((le32_to_cpu((_rxd)->wb.lower.lo_dword.data) >> 9) & 0xFF) #define WX_RXD_IPV6EX(_rxd) \ @@ -552,6 +582,9 @@ enum wx_tx_flags { WX_TX_FLAGS_OUTER_IPV4 = 0x100, WX_TX_FLAGS_LINKSEC = 0x200, WX_TX_FLAGS_IPSEC = 0x400, + + /* software defined flags */ + WX_TX_FLAGS_SW_VLAN = 0x40, }; /* VLAN info */ @@ -900,7 +933,13 @@ struct wx_ring { */ u16 next_to_use; u16 next_to_clean; - u16 next_to_alloc; + union { + u16 next_to_alloc; + struct { + u8 atr_sample_rate; + u8 atr_count; + }; + }; struct wx_queue_stats stats; struct u64_stats_sync syncp; @@ -939,6 +978,7 @@ struct wx_ring_feature { enum wx_ring_f_enum { RING_F_NONE = 0, RING_F_RSS, + RING_F_FDIR, RING_F_ARRAY_SIZE /* must be last in enum set */ }; @@ -980,15 +1020,26 @@ struct wx_hw_stats { u64 crcerrs; u64 rlec; u64 qmprc; + u64 fdirmatch; + u64 fdirmiss; }; enum wx_state { WX_STATE_RESETTING, WX_STATE_NBITS, /* must be last */ }; + +enum wx_pf_flags { + WX_FLAG_FDIR_CAPABLE, + WX_FLAG_FDIR_HASH, + WX_FLAG_FDIR_PERFECT, + WX_PF_FLAGS_NBITS /* must be last */ +}; + struct wx { unsigned long active_vlans[BITS_TO_LONGS(VLAN_N_VID)]; DECLARE_BITMAP(state, WX_STATE_NBITS); + DECLARE_BITMAP(flags, WX_PF_FLAGS_NBITS); void *priv; u8 __iomem *hw_addr; @@ -1078,6 +1129,9 @@ struct wx { u64 hw_csum_rx_error; u64 alloc_rx_buff_failed; + u32 atr_sample_rate; + void (*atr)(struct wx_ring *ring, struct wx_tx_buffer *first, u8 ptype); + void (*configure_fdir)(struct wx *wx); void (*do_reset)(struct net_device *netdev); }; diff --git a/drivers/net/ethernet/wangxun/ngbe/ngbe_ethtool.c b/drivers/net/ethernet/wangxun/ngbe/ngbe_ethtool.c index 46a5a3e95202..e868f7ef4920 100644 --- a/drivers/net/ethernet/wangxun/ngbe/ngbe_ethtool.c +++ b/drivers/net/ethernet/wangxun/ngbe/ngbe_ethtool.c @@ -37,9 +37,9 @@ static int ngbe_set_wol(struct net_device *netdev, wx->wol = 0; if (wol->wolopts & WAKE_MAGIC) wx->wol = WX_PSR_WKUP_CTL_MAG; - netdev->wol_enabled = !!(wx->wol); + netdev->ethtool->wol_enabled = !!(wx->wol); wr32(wx, WX_PSR_WKUP_CTL, wx->wol); - device_set_wakeup_enable(&pdev->dev, netdev->wol_enabled); + device_set_wakeup_enable(&pdev->dev, netdev->ethtool->wol_enabled); return 0; } diff --git a/drivers/net/ethernet/wangxun/ngbe/ngbe_main.c b/drivers/net/ethernet/wangxun/ngbe/ngbe_main.c index af30ca0312b8..53aeae2f884b 100644 --- a/drivers/net/ethernet/wangxun/ngbe/ngbe_main.c +++ b/drivers/net/ethernet/wangxun/ngbe/ngbe_main.c @@ -652,7 +652,7 @@ static int ngbe_probe(struct pci_dev *pdev, if (wx->wol_hw_supported) wx->wol = NGBE_PSR_WKUP_CTL_MAG; - netdev->wol_enabled = !!(wx->wol); + netdev->ethtool->wol_enabled = !!(wx->wol); wr32(wx, NGBE_PSR_WKUP_CTL, wx->wol); device_set_wakeup_enable(&pdev->dev, wx->wol); diff --git a/drivers/net/ethernet/wangxun/txgbe/Makefile b/drivers/net/ethernet/wangxun/txgbe/Makefile index 42718875277c..f74576fe7062 100644 --- a/drivers/net/ethernet/wangxun/txgbe/Makefile +++ b/drivers/net/ethernet/wangxun/txgbe/Makefile @@ -10,4 +10,5 @@ txgbe-objs := txgbe_main.o \ txgbe_hw.o \ txgbe_phy.o \ txgbe_irq.o \ + txgbe_fdir.o \ txgbe_ethtool.o diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_ethtool.c b/drivers/net/ethernet/wangxun/txgbe/txgbe_ethtool.c index 31fde3fa7c6b..d98314b26c19 100644 --- a/drivers/net/ethernet/wangxun/txgbe/txgbe_ethtool.c +++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_ethtool.c @@ -9,6 +9,7 @@ #include "../libwx/wx_type.h" #include "../libwx/wx_lib.h" #include "txgbe_type.h" +#include "txgbe_fdir.h" #include "txgbe_ethtool.h" static int txgbe_set_ringparam(struct net_device *netdev, @@ -79,6 +80,430 @@ static int txgbe_set_channels(struct net_device *dev, return txgbe_setup_tc(dev, netdev_get_num_tc(dev)); } +static int txgbe_get_ethtool_fdir_entry(struct txgbe *txgbe, + struct ethtool_rxnfc *cmd) +{ + struct ethtool_rx_flow_spec *fsp = + (struct ethtool_rx_flow_spec *)&cmd->fs; + union txgbe_atr_input *mask = &txgbe->fdir_mask; + struct txgbe_fdir_filter *rule = NULL; + struct hlist_node *node; + + /* report total rule count */ + cmd->data = (1024 << TXGBE_FDIR_PBALLOC_64K) - 2; + + hlist_for_each_entry_safe(rule, node, &txgbe->fdir_filter_list, + fdir_node) { + if (fsp->location <= rule->sw_idx) + break; + } + + if (!rule || fsp->location != rule->sw_idx) + return -EINVAL; + + /* set flow type field */ + switch (rule->filter.formatted.flow_type) { + case TXGBE_ATR_FLOW_TYPE_TCPV4: + fsp->flow_type = TCP_V4_FLOW; + break; + case TXGBE_ATR_FLOW_TYPE_UDPV4: + fsp->flow_type = UDP_V4_FLOW; + break; + case TXGBE_ATR_FLOW_TYPE_SCTPV4: + fsp->flow_type = SCTP_V4_FLOW; + break; + case TXGBE_ATR_FLOW_TYPE_IPV4: + fsp->flow_type = IP_USER_FLOW; + fsp->h_u.usr_ip4_spec.ip_ver = ETH_RX_NFC_IP4; + fsp->h_u.usr_ip4_spec.proto = 0; + fsp->m_u.usr_ip4_spec.proto = 0; + break; + default: + return -EINVAL; + } + + fsp->h_u.tcp_ip4_spec.psrc = rule->filter.formatted.src_port; + fsp->m_u.tcp_ip4_spec.psrc = mask->formatted.src_port; + fsp->h_u.tcp_ip4_spec.pdst = rule->filter.formatted.dst_port; + fsp->m_u.tcp_ip4_spec.pdst = mask->formatted.dst_port; + fsp->h_u.tcp_ip4_spec.ip4src = rule->filter.formatted.src_ip[0]; + fsp->m_u.tcp_ip4_spec.ip4src = mask->formatted.src_ip[0]; + fsp->h_u.tcp_ip4_spec.ip4dst = rule->filter.formatted.dst_ip[0]; + fsp->m_u.tcp_ip4_spec.ip4dst = mask->formatted.dst_ip[0]; + fsp->h_ext.vlan_etype = rule->filter.formatted.flex_bytes; + fsp->m_ext.vlan_etype = mask->formatted.flex_bytes; + fsp->h_ext.data[1] = htonl(rule->filter.formatted.vm_pool); + fsp->m_ext.data[1] = htonl(mask->formatted.vm_pool); + fsp->flow_type |= FLOW_EXT; + + /* record action */ + if (rule->action == TXGBE_RDB_FDIR_DROP_QUEUE) + fsp->ring_cookie = RX_CLS_FLOW_DISC; + else + fsp->ring_cookie = rule->action; + + return 0; +} + +static int txgbe_get_ethtool_fdir_all(struct txgbe *txgbe, + struct ethtool_rxnfc *cmd, + u32 *rule_locs) +{ + struct txgbe_fdir_filter *rule; + struct hlist_node *node; + int cnt = 0; + + /* report total rule count */ + cmd->data = (1024 << TXGBE_FDIR_PBALLOC_64K) - 2; + + hlist_for_each_entry_safe(rule, node, &txgbe->fdir_filter_list, + fdir_node) { + if (cnt == cmd->rule_cnt) + return -EMSGSIZE; + rule_locs[cnt] = rule->sw_idx; + cnt++; + } + + cmd->rule_cnt = cnt; + + return 0; +} + +static int txgbe_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd, + u32 *rule_locs) +{ + struct wx *wx = netdev_priv(dev); + struct txgbe *txgbe = wx->priv; + int ret = -EOPNOTSUPP; + + switch (cmd->cmd) { + case ETHTOOL_GRXRINGS: + cmd->data = wx->num_rx_queues; + ret = 0; + break; + case ETHTOOL_GRXCLSRLCNT: + cmd->rule_cnt = txgbe->fdir_filter_count; + ret = 0; + break; + case ETHTOOL_GRXCLSRULE: + ret = txgbe_get_ethtool_fdir_entry(txgbe, cmd); + break; + case ETHTOOL_GRXCLSRLALL: + ret = txgbe_get_ethtool_fdir_all(txgbe, cmd, (u32 *)rule_locs); + break; + default: + break; + } + + return ret; +} + +static int txgbe_flowspec_to_flow_type(struct ethtool_rx_flow_spec *fsp, + u8 *flow_type) +{ + switch (fsp->flow_type & ~FLOW_EXT) { + case TCP_V4_FLOW: + *flow_type = TXGBE_ATR_FLOW_TYPE_TCPV4; + break; + case UDP_V4_FLOW: + *flow_type = TXGBE_ATR_FLOW_TYPE_UDPV4; + break; + case SCTP_V4_FLOW: + *flow_type = TXGBE_ATR_FLOW_TYPE_SCTPV4; + break; + case IP_USER_FLOW: + switch (fsp->h_u.usr_ip4_spec.proto) { + case IPPROTO_TCP: + *flow_type = TXGBE_ATR_FLOW_TYPE_TCPV4; + break; + case IPPROTO_UDP: + *flow_type = TXGBE_ATR_FLOW_TYPE_UDPV4; + break; + case IPPROTO_SCTP: + *flow_type = TXGBE_ATR_FLOW_TYPE_SCTPV4; + break; + case 0: + if (!fsp->m_u.usr_ip4_spec.proto) { + *flow_type = TXGBE_ATR_FLOW_TYPE_IPV4; + break; + } + fallthrough; + default: + return -EINVAL; + } + break; + default: + return -EINVAL; + } + + return 0; +} + +static bool txgbe_match_ethtool_fdir_entry(struct txgbe *txgbe, + struct txgbe_fdir_filter *input) +{ + struct txgbe_fdir_filter *rule = NULL; + struct hlist_node *node2; + + hlist_for_each_entry_safe(rule, node2, &txgbe->fdir_filter_list, + fdir_node) { + if (rule->filter.formatted.bkt_hash == + input->filter.formatted.bkt_hash && + rule->action == input->action) { + wx_dbg(txgbe->wx, "FDIR entry already exist\n"); + return true; + } + } + return false; +} + +static int txgbe_update_ethtool_fdir_entry(struct txgbe *txgbe, + struct txgbe_fdir_filter *input, + u16 sw_idx) +{ + struct hlist_node *node = NULL, *parent = NULL; + struct txgbe_fdir_filter *rule; + struct wx *wx = txgbe->wx; + bool deleted = false; + int err; + + hlist_for_each_entry_safe(rule, node, &txgbe->fdir_filter_list, + fdir_node) { + /* hash found, or no matching entry */ + if (rule->sw_idx >= sw_idx) + break; + parent = node; + } + + /* if there is an old rule occupying our place remove it */ + if (rule && rule->sw_idx == sw_idx) { + /* hardware filters are only configured when interface is up, + * and we should not issue filter commands while the interface + * is down + */ + if (netif_running(wx->netdev) && + (!input || rule->filter.formatted.bkt_hash != + input->filter.formatted.bkt_hash)) { + err = txgbe_fdir_erase_perfect_filter(wx, + &rule->filter, + sw_idx); + if (err) + return -EINVAL; + } + + hlist_del(&rule->fdir_node); + kfree(rule); + txgbe->fdir_filter_count--; + deleted = true; + } + + /* If we weren't given an input, then this was a request to delete a + * filter. We should return -EINVAL if the filter wasn't found, but + * return 0 if the rule was successfully deleted. + */ + if (!input) + return deleted ? 0 : -EINVAL; + + /* initialize node and set software index */ + INIT_HLIST_NODE(&input->fdir_node); + + /* add filter to the list */ + if (parent) + hlist_add_behind(&input->fdir_node, parent); + else + hlist_add_head(&input->fdir_node, + &txgbe->fdir_filter_list); + + /* update counts */ + txgbe->fdir_filter_count++; + + return 0; +} + +static int txgbe_add_ethtool_fdir_entry(struct txgbe *txgbe, + struct ethtool_rxnfc *cmd) +{ + struct ethtool_rx_flow_spec *fsp = + (struct ethtool_rx_flow_spec *)&cmd->fs; + struct txgbe_fdir_filter *input; + union txgbe_atr_input mask; + struct wx *wx = txgbe->wx; + int err = -EINVAL; + u16 ptype = 0; + u8 queue; + + if (!(test_bit(WX_FLAG_FDIR_PERFECT, wx->flags))) + return -EOPNOTSUPP; + + /* ring_cookie is a masked into a set of queues and txgbe pools or + * we use drop index + */ + if (fsp->ring_cookie == RX_CLS_FLOW_DISC) { + queue = TXGBE_RDB_FDIR_DROP_QUEUE; + } else { + u32 ring = ethtool_get_flow_spec_ring(fsp->ring_cookie); + + if (ring >= wx->num_rx_queues) + return -EINVAL; + + /* Map the ring onto the absolute queue index */ + queue = wx->rx_ring[ring]->reg_idx; + } + + /* Don't allow indexes to exist outside of available space */ + if (fsp->location >= ((1024 << TXGBE_FDIR_PBALLOC_64K) - 2)) { + wx_err(wx, "Location out of range\n"); + return -EINVAL; + } + + input = kzalloc(sizeof(*input), GFP_ATOMIC); + if (!input) + return -ENOMEM; + + memset(&mask, 0, sizeof(union txgbe_atr_input)); + + /* set SW index */ + input->sw_idx = fsp->location; + + /* record flow type */ + if (txgbe_flowspec_to_flow_type(fsp, + &input->filter.formatted.flow_type)) { + wx_err(wx, "Unrecognized flow type\n"); + goto err_out; + } + + mask.formatted.flow_type = TXGBE_ATR_L4TYPE_IPV6_MASK | + TXGBE_ATR_L4TYPE_MASK; + + if (input->filter.formatted.flow_type == TXGBE_ATR_FLOW_TYPE_IPV4) + mask.formatted.flow_type &= TXGBE_ATR_L4TYPE_IPV6_MASK; + + /* Copy input into formatted structures */ + input->filter.formatted.src_ip[0] = fsp->h_u.tcp_ip4_spec.ip4src; + mask.formatted.src_ip[0] = fsp->m_u.tcp_ip4_spec.ip4src; + input->filter.formatted.dst_ip[0] = fsp->h_u.tcp_ip4_spec.ip4dst; + mask.formatted.dst_ip[0] = fsp->m_u.tcp_ip4_spec.ip4dst; + input->filter.formatted.src_port = fsp->h_u.tcp_ip4_spec.psrc; + mask.formatted.src_port = fsp->m_u.tcp_ip4_spec.psrc; + input->filter.formatted.dst_port = fsp->h_u.tcp_ip4_spec.pdst; + mask.formatted.dst_port = fsp->m_u.tcp_ip4_spec.pdst; + + if (fsp->flow_type & FLOW_EXT) { + input->filter.formatted.vm_pool = + (unsigned char)ntohl(fsp->h_ext.data[1]); + mask.formatted.vm_pool = + (unsigned char)ntohl(fsp->m_ext.data[1]); + input->filter.formatted.flex_bytes = + fsp->h_ext.vlan_etype; + mask.formatted.flex_bytes = fsp->m_ext.vlan_etype; + } + + switch (input->filter.formatted.flow_type) { + case TXGBE_ATR_FLOW_TYPE_TCPV4: + ptype = WX_PTYPE_L2_IPV4_TCP; + break; + case TXGBE_ATR_FLOW_TYPE_UDPV4: + ptype = WX_PTYPE_L2_IPV4_UDP; + break; + case TXGBE_ATR_FLOW_TYPE_SCTPV4: + ptype = WX_PTYPE_L2_IPV4_SCTP; + break; + case TXGBE_ATR_FLOW_TYPE_IPV4: + ptype = WX_PTYPE_L2_IPV4; + break; + default: + break; + } + + input->filter.formatted.vlan_id = htons(ptype); + if (mask.formatted.flow_type & TXGBE_ATR_L4TYPE_MASK) + mask.formatted.vlan_id = htons(0xFFFF); + else + mask.formatted.vlan_id = htons(0xFFF8); + + /* determine if we need to drop or route the packet */ + if (fsp->ring_cookie == RX_CLS_FLOW_DISC) + input->action = TXGBE_RDB_FDIR_DROP_QUEUE; + else + input->action = fsp->ring_cookie; + + spin_lock(&txgbe->fdir_perfect_lock); + + if (hlist_empty(&txgbe->fdir_filter_list)) { + /* save mask and program input mask into HW */ + memcpy(&txgbe->fdir_mask, &mask, sizeof(mask)); + err = txgbe_fdir_set_input_mask(wx, &mask); + if (err) + goto err_unlock; + } else if (memcmp(&txgbe->fdir_mask, &mask, sizeof(mask))) { + wx_err(wx, "Hardware only supports one mask per port. To change the mask you must first delete all the rules.\n"); + goto err_unlock; + } + + /* apply mask and compute/store hash */ + txgbe_atr_compute_perfect_hash(&input->filter, &mask); + + /* check if new entry does not exist on filter list */ + if (txgbe_match_ethtool_fdir_entry(txgbe, input)) + goto err_unlock; + + /* only program filters to hardware if the net device is running, as + * we store the filters in the Rx buffer which is not allocated when + * the device is down + */ + if (netif_running(wx->netdev)) { + err = txgbe_fdir_write_perfect_filter(wx, &input->filter, + input->sw_idx, queue); + if (err) + goto err_unlock; + } + + txgbe_update_ethtool_fdir_entry(txgbe, input, input->sw_idx); + + spin_unlock(&txgbe->fdir_perfect_lock); + + return 0; +err_unlock: + spin_unlock(&txgbe->fdir_perfect_lock); +err_out: + kfree(input); + return err; +} + +static int txgbe_del_ethtool_fdir_entry(struct txgbe *txgbe, + struct ethtool_rxnfc *cmd) +{ + struct ethtool_rx_flow_spec *fsp = + (struct ethtool_rx_flow_spec *)&cmd->fs; + int err = 0; + + spin_lock(&txgbe->fdir_perfect_lock); + err = txgbe_update_ethtool_fdir_entry(txgbe, NULL, fsp->location); + spin_unlock(&txgbe->fdir_perfect_lock); + + return err; +} + +static int txgbe_set_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd) +{ + struct wx *wx = netdev_priv(dev); + struct txgbe *txgbe = wx->priv; + int ret = -EOPNOTSUPP; + + switch (cmd->cmd) { + case ETHTOOL_SRXCLSRLINS: + ret = txgbe_add_ethtool_fdir_entry(txgbe, cmd); + break; + case ETHTOOL_SRXCLSRLDEL: + ret = txgbe_del_ethtool_fdir_entry(txgbe, cmd); + break; + default: + break; + } + + return ret; +} + static const struct ethtool_ops txgbe_ethtool_ops = { .supported_coalesce_params = ETHTOOL_COALESCE_USECS | ETHTOOL_COALESCE_TX_MAX_FRAMES_IRQ, @@ -100,6 +525,8 @@ static const struct ethtool_ops txgbe_ethtool_ops = { .set_coalesce = wx_set_coalesce, .get_channels = wx_get_channels, .set_channels = txgbe_set_channels, + .get_rxnfc = txgbe_get_rxnfc, + .set_rxnfc = txgbe_set_rxnfc, .get_msglevel = wx_get_msglevel, .set_msglevel = wx_set_msglevel, }; diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_fdir.c b/drivers/net/ethernet/wangxun/txgbe/txgbe_fdir.c new file mode 100644 index 000000000000..ef50efbaec0f --- /dev/null +++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_fdir.c @@ -0,0 +1,643 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) 2015 - 2024 Beijing WangXun Technology Co., Ltd. */ + +#include <linux/string.h> +#include <linux/types.h> +#include <linux/pci.h> + +#include "../libwx/wx_type.h" +#include "../libwx/wx_lib.h" +#include "../libwx/wx_hw.h" +#include "txgbe_type.h" +#include "txgbe_fdir.h" + +/* These defines allow us to quickly generate all of the necessary instructions + * in the function below by simply calling out TXGBE_COMPUTE_SIG_HASH_ITERATION + * for values 0 through 15 + */ +#define TXGBE_ATR_COMMON_HASH_KEY \ + (TXGBE_ATR_BUCKET_HASH_KEY & TXGBE_ATR_SIGNATURE_HASH_KEY) +#define TXGBE_COMPUTE_SIG_HASH_ITERATION(_n) \ +do { \ + u32 n = (_n); \ + if (TXGBE_ATR_COMMON_HASH_KEY & (0x01 << n)) \ + common_hash ^= lo_hash_dword >> n; \ + else if (TXGBE_ATR_BUCKET_HASH_KEY & (0x01 << n)) \ + bucket_hash ^= lo_hash_dword >> n; \ + else if (TXGBE_ATR_SIGNATURE_HASH_KEY & (0x01 << n)) \ + sig_hash ^= lo_hash_dword << (16 - n); \ + if (TXGBE_ATR_COMMON_HASH_KEY & (0x01 << (n + 16))) \ + common_hash ^= hi_hash_dword >> n; \ + else if (TXGBE_ATR_BUCKET_HASH_KEY & (0x01 << (n + 16))) \ + bucket_hash ^= hi_hash_dword >> n; \ + else if (TXGBE_ATR_SIGNATURE_HASH_KEY & (0x01 << (n + 16))) \ + sig_hash ^= hi_hash_dword << (16 - n); \ +} while (0) + +/** + * txgbe_atr_compute_sig_hash - Compute the signature hash + * @input: input bitstream to compute the hash on + * @common: compressed common input dword + * @hash: pointer to the computed hash + * + * This function is almost identical to the function above but contains + * several optimizations such as unwinding all of the loops, letting the + * compiler work out all of the conditional ifs since the keys are static + * defines, and computing two keys at once since the hashed dword stream + * will be the same for both keys. + **/ +static void txgbe_atr_compute_sig_hash(union txgbe_atr_hash_dword input, + union txgbe_atr_hash_dword common, + u32 *hash) +{ + u32 sig_hash = 0, bucket_hash = 0, common_hash = 0; + u32 hi_hash_dword, lo_hash_dword, flow_vm_vlan; + u32 i; + + /* record the flow_vm_vlan bits as they are a key part to the hash */ + flow_vm_vlan = ntohl(input.dword); + + /* generate common hash dword */ + hi_hash_dword = ntohl(common.dword); + + /* low dword is word swapped version of common */ + lo_hash_dword = (hi_hash_dword >> 16) | (hi_hash_dword << 16); + + /* apply flow ID/VM pool/VLAN ID bits to hash words */ + hi_hash_dword ^= flow_vm_vlan ^ (flow_vm_vlan >> 16); + + /* Process bits 0 and 16 */ + TXGBE_COMPUTE_SIG_HASH_ITERATION(0); + + /* apply flow ID/VM pool/VLAN ID bits to lo hash dword, we had to + * delay this because bit 0 of the stream should not be processed + * so we do not add the VLAN until after bit 0 was processed + */ + lo_hash_dword ^= flow_vm_vlan ^ (flow_vm_vlan << 16); + + /* Process remaining 30 bit of the key */ + for (i = 1; i <= 15; i++) + TXGBE_COMPUTE_SIG_HASH_ITERATION(i); + + /* combine common_hash result with signature and bucket hashes */ + bucket_hash ^= common_hash; + bucket_hash &= TXGBE_ATR_HASH_MASK; + + sig_hash ^= common_hash << 16; + sig_hash &= TXGBE_ATR_HASH_MASK << 16; + + /* return completed signature hash */ + *hash = sig_hash ^ bucket_hash; +} + +#define TXGBE_COMPUTE_BKT_HASH_ITERATION(_n) \ +do { \ + u32 n = (_n); \ + if (TXGBE_ATR_BUCKET_HASH_KEY & (0x01 << n)) \ + bucket_hash ^= lo_hash_dword >> n; \ + if (TXGBE_ATR_BUCKET_HASH_KEY & (0x01 << (n + 16))) \ + bucket_hash ^= hi_hash_dword >> n; \ +} while (0) + +/** + * txgbe_atr_compute_perfect_hash - Compute the perfect filter hash + * @input: input bitstream to compute the hash on + * @input_mask: mask for the input bitstream + * + * This function serves two main purposes. First it applies the input_mask + * to the atr_input resulting in a cleaned up atr_input data stream. + * Secondly it computes the hash and stores it in the bkt_hash field at + * the end of the input byte stream. This way it will be available for + * future use without needing to recompute the hash. + **/ +void txgbe_atr_compute_perfect_hash(union txgbe_atr_input *input, + union txgbe_atr_input *input_mask) +{ + u32 hi_hash_dword, lo_hash_dword, flow_vm_vlan; + u32 bucket_hash = 0; + __be32 hi_dword = 0; + u32 i = 0; + + /* Apply masks to input data */ + for (i = 0; i < 11; i++) + input->dword_stream[i] &= input_mask->dword_stream[i]; + + /* record the flow_vm_vlan bits as they are a key part to the hash */ + flow_vm_vlan = ntohl(input->dword_stream[0]); + + /* generate common hash dword */ + for (i = 1; i <= 10; i++) + hi_dword ^= input->dword_stream[i]; + hi_hash_dword = ntohl(hi_dword); + + /* low dword is word swapped version of common */ + lo_hash_dword = (hi_hash_dword >> 16) | (hi_hash_dword << 16); + + /* apply flow ID/VM pool/VLAN ID bits to hash words */ + hi_hash_dword ^= flow_vm_vlan ^ (flow_vm_vlan >> 16); + + /* Process bits 0 and 16 */ + TXGBE_COMPUTE_BKT_HASH_ITERATION(0); + + /* apply flow ID/VM pool/VLAN ID bits to lo hash dword, we had to + * delay this because bit 0 of the stream should not be processed + * so we do not add the VLAN until after bit 0 was processed + */ + lo_hash_dword ^= flow_vm_vlan ^ (flow_vm_vlan << 16); + + /* Process remaining 30 bit of the key */ + for (i = 1; i <= 15; i++) + TXGBE_COMPUTE_BKT_HASH_ITERATION(i); + + /* Limit hash to 13 bits since max bucket count is 8K. + * Store result at the end of the input stream. + */ + input->formatted.bkt_hash = (__force __be16)(bucket_hash & 0x1FFF); +} + +static int txgbe_fdir_check_cmd_complete(struct wx *wx) +{ + u32 val; + + return read_poll_timeout_atomic(rd32, val, + !(val & TXGBE_RDB_FDIR_CMD_CMD_MASK), + 10, 100, false, + wx, TXGBE_RDB_FDIR_CMD); +} + +/** + * txgbe_fdir_add_signature_filter - Adds a signature hash filter + * @wx: pointer to hardware structure + * @input: unique input dword + * @common: compressed common input dword + * @queue: queue index to direct traffic to + * + * @return: 0 on success and negative on failure + **/ +static int txgbe_fdir_add_signature_filter(struct wx *wx, + union txgbe_atr_hash_dword input, + union txgbe_atr_hash_dword common, + u8 queue) +{ + u32 fdirhashcmd, fdircmd; + u8 flow_type; + int err; + + /* Get the flow_type in order to program FDIRCMD properly + * lowest 2 bits are FDIRCMD.L4TYPE, third lowest bit is FDIRCMD.IPV6 + * fifth is FDIRCMD.TUNNEL_FILTER + */ + flow_type = input.formatted.flow_type; + switch (flow_type) { + case TXGBE_ATR_FLOW_TYPE_TCPV4: + case TXGBE_ATR_FLOW_TYPE_UDPV4: + case TXGBE_ATR_FLOW_TYPE_SCTPV4: + case TXGBE_ATR_FLOW_TYPE_TCPV6: + case TXGBE_ATR_FLOW_TYPE_UDPV6: + case TXGBE_ATR_FLOW_TYPE_SCTPV6: + break; + default: + wx_err(wx, "Error on flow type input\n"); + return -EINVAL; + } + + /* configure FDIRCMD register */ + fdircmd = TXGBE_RDB_FDIR_CMD_CMD_ADD_FLOW | + TXGBE_RDB_FDIR_CMD_FILTER_UPDATE | + TXGBE_RDB_FDIR_CMD_LAST | TXGBE_RDB_FDIR_CMD_QUEUE_EN; + fdircmd |= TXGBE_RDB_FDIR_CMD_FLOW_TYPE(flow_type); + fdircmd |= TXGBE_RDB_FDIR_CMD_RX_QUEUE(queue); + + txgbe_atr_compute_sig_hash(input, common, &fdirhashcmd); + fdirhashcmd |= TXGBE_RDB_FDIR_HASH_BUCKET_VALID; + wr32(wx, TXGBE_RDB_FDIR_HASH, fdirhashcmd); + wr32(wx, TXGBE_RDB_FDIR_CMD, fdircmd); + + wx_dbg(wx, "Tx Queue=%x hash=%x\n", queue, (u32)fdirhashcmd); + + err = txgbe_fdir_check_cmd_complete(wx); + if (err) + wx_err(wx, "Flow Director command did not complete!\n"); + + return err; +} + +void txgbe_atr(struct wx_ring *ring, struct wx_tx_buffer *first, u8 ptype) +{ + union txgbe_atr_hash_dword common = { .dword = 0 }; + union txgbe_atr_hash_dword input = { .dword = 0 }; + struct wx_q_vector *q_vector = ring->q_vector; + struct wx_dec_ptype dptype; + union network_header { + struct ipv6hdr *ipv6; + struct iphdr *ipv4; + void *raw; + } hdr; + struct tcphdr *th; + + /* if ring doesn't have a interrupt vector, cannot perform ATR */ + if (!q_vector) + return; + + ring->atr_count++; + dptype = wx_decode_ptype(ptype); + if (dptype.etype) { + if (WX_PTYPE_TYPL4(ptype) != WX_PTYPE_TYP_TCP) + return; + hdr.raw = (void *)skb_inner_network_header(first->skb); + th = inner_tcp_hdr(first->skb); + } else { + if (WX_PTYPE_PKT(ptype) != WX_PTYPE_PKT_IP || + WX_PTYPE_TYPL4(ptype) != WX_PTYPE_TYP_TCP) + return; + hdr.raw = (void *)skb_network_header(first->skb); + th = tcp_hdr(first->skb); + } + + /* skip this packet since it is invalid or the socket is closing */ + if (!th || th->fin) + return; + + /* sample on all syn packets or once every atr sample count */ + if (!th->syn && ring->atr_count < ring->atr_sample_rate) + return; + + /* reset sample count */ + ring->atr_count = 0; + + /* src and dst are inverted, think how the receiver sees them + * + * The input is broken into two sections, a non-compressed section + * containing vm_pool, vlan_id, and flow_type. The rest of the data + * is XORed together and stored in the compressed dword. + */ + input.formatted.vlan_id = htons((u16)ptype); + + /* since src port and flex bytes occupy the same word XOR them together + * and write the value to source port portion of compressed dword + */ + if (first->tx_flags & WX_TX_FLAGS_SW_VLAN) + common.port.src ^= th->dest ^ first->skb->protocol; + else if (first->tx_flags & WX_TX_FLAGS_HW_VLAN) + common.port.src ^= th->dest ^ first->skb->vlan_proto; + else + common.port.src ^= th->dest ^ first->protocol; + common.port.dst ^= th->source; + + if (WX_PTYPE_PKT_IPV6 & WX_PTYPE_PKT(ptype)) { + input.formatted.flow_type = TXGBE_ATR_FLOW_TYPE_TCPV6; + common.ip ^= hdr.ipv6->saddr.s6_addr32[0] ^ + hdr.ipv6->saddr.s6_addr32[1] ^ + hdr.ipv6->saddr.s6_addr32[2] ^ + hdr.ipv6->saddr.s6_addr32[3] ^ + hdr.ipv6->daddr.s6_addr32[0] ^ + hdr.ipv6->daddr.s6_addr32[1] ^ + hdr.ipv6->daddr.s6_addr32[2] ^ + hdr.ipv6->daddr.s6_addr32[3]; + } else { + input.formatted.flow_type = TXGBE_ATR_FLOW_TYPE_TCPV4; + common.ip ^= hdr.ipv4->saddr ^ hdr.ipv4->daddr; + } + + /* This assumes the Rx queue and Tx queue are bound to the same CPU */ + txgbe_fdir_add_signature_filter(q_vector->wx, input, common, + ring->queue_index); +} + +int txgbe_fdir_set_input_mask(struct wx *wx, union txgbe_atr_input *input_mask) +{ + u32 fdirm = 0, fdirtcpm = 0, flex = 0; + + /* Program the relevant mask registers. If src/dst_port or src/dst_addr + * are zero, then assume a full mask for that field. Also assume that + * a VLAN of 0 is unspecified, so mask that out as well. L4type + * cannot be masked out in this implementation. + * + * This also assumes IPv4 only. IPv6 masking isn't supported at this + * point in time. + */ + + /* verify bucket hash is cleared on hash generation */ + if (input_mask->formatted.bkt_hash) + wx_dbg(wx, "bucket hash should always be 0 in mask\n"); + + /* Program FDIRM and verify partial masks */ + switch (input_mask->formatted.vm_pool & 0x7F) { + case 0x0: + fdirm |= TXGBE_RDB_FDIR_OTHER_MSK_POOL; + break; + case 0x7F: + break; + default: + wx_err(wx, "Error on vm pool mask\n"); + return -EINVAL; + } + + switch (input_mask->formatted.flow_type & TXGBE_ATR_L4TYPE_MASK) { + case 0x0: + fdirm |= TXGBE_RDB_FDIR_OTHER_MSK_L4P; + if (input_mask->formatted.dst_port || + input_mask->formatted.src_port) { + wx_err(wx, "Error on src/dst port mask\n"); + return -EINVAL; + } + break; + case TXGBE_ATR_L4TYPE_MASK: + break; + default: + wx_err(wx, "Error on flow type mask\n"); + return -EINVAL; + } + + /* Now mask VM pool and destination IPv6 - bits 5 and 2 */ + wr32(wx, TXGBE_RDB_FDIR_OTHER_MSK, fdirm); + + flex = rd32(wx, TXGBE_RDB_FDIR_FLEX_CFG(0)); + flex &= ~TXGBE_RDB_FDIR_FLEX_CFG_FIELD0; + flex |= (TXGBE_RDB_FDIR_FLEX_CFG_BASE_MAC | + TXGBE_RDB_FDIR_FLEX_CFG_OFST(0x6)); + + switch ((__force u16)input_mask->formatted.flex_bytes & 0xFFFF) { + case 0x0000: + /* Mask Flex Bytes */ + flex |= TXGBE_RDB_FDIR_FLEX_CFG_MSK; + break; + case 0xFFFF: + break; + default: + wx_err(wx, "Error on flexible byte mask\n"); + return -EINVAL; + } + wr32(wx, TXGBE_RDB_FDIR_FLEX_CFG(0), flex); + + /* store the TCP/UDP port masks, bit reversed from port layout */ + fdirtcpm = ntohs(input_mask->formatted.dst_port); + fdirtcpm <<= TXGBE_RDB_FDIR_PORT_DESTINATION_SHIFT; + fdirtcpm |= ntohs(input_mask->formatted.src_port); + + /* write both the same so that UDP and TCP use the same mask */ + wr32(wx, TXGBE_RDB_FDIR_TCP_MSK, ~fdirtcpm); + wr32(wx, TXGBE_RDB_FDIR_UDP_MSK, ~fdirtcpm); + wr32(wx, TXGBE_RDB_FDIR_SCTP_MSK, ~fdirtcpm); + + /* store source and destination IP masks (little-enian) */ + wr32(wx, TXGBE_RDB_FDIR_SA4_MSK, + ntohl(~input_mask->formatted.src_ip[0])); + wr32(wx, TXGBE_RDB_FDIR_DA4_MSK, + ntohl(~input_mask->formatted.dst_ip[0])); + + return 0; +} + +int txgbe_fdir_write_perfect_filter(struct wx *wx, + union txgbe_atr_input *input, + u16 soft_id, u8 queue) +{ + u32 fdirport, fdirvlan, fdirhash, fdircmd; + int err = 0; + + /* currently IPv6 is not supported, must be programmed with 0 */ + wr32(wx, TXGBE_RDB_FDIR_IP6(2), ntohl(input->formatted.src_ip[0])); + wr32(wx, TXGBE_RDB_FDIR_IP6(1), ntohl(input->formatted.src_ip[1])); + wr32(wx, TXGBE_RDB_FDIR_IP6(0), ntohl(input->formatted.src_ip[2])); + + /* record the source address (little-endian) */ + wr32(wx, TXGBE_RDB_FDIR_SA, ntohl(input->formatted.src_ip[0])); + + /* record the first 32 bits of the destination address + * (little-endian) + */ + wr32(wx, TXGBE_RDB_FDIR_DA, ntohl(input->formatted.dst_ip[0])); + + /* record source and destination port (little-endian)*/ + fdirport = ntohs(input->formatted.dst_port); + fdirport <<= TXGBE_RDB_FDIR_PORT_DESTINATION_SHIFT; + fdirport |= ntohs(input->formatted.src_port); + wr32(wx, TXGBE_RDB_FDIR_PORT, fdirport); + + /* record packet type and flex_bytes (little-endian) */ + fdirvlan = ntohs(input->formatted.flex_bytes); + fdirvlan <<= TXGBE_RDB_FDIR_FLEX_FLEX_SHIFT; + fdirvlan |= ntohs(input->formatted.vlan_id); + wr32(wx, TXGBE_RDB_FDIR_FLEX, fdirvlan); + + /* configure FDIRHASH register */ + fdirhash = (__force u32)input->formatted.bkt_hash | + TXGBE_RDB_FDIR_HASH_BUCKET_VALID | + TXGBE_RDB_FDIR_HASH_SIG_SW_INDEX(soft_id); + wr32(wx, TXGBE_RDB_FDIR_HASH, fdirhash); + + /* flush all previous writes to make certain registers are + * programmed prior to issuing the command + */ + WX_WRITE_FLUSH(wx); + + /* configure FDIRCMD register */ + fdircmd = TXGBE_RDB_FDIR_CMD_CMD_ADD_FLOW | + TXGBE_RDB_FDIR_CMD_FILTER_UPDATE | + TXGBE_RDB_FDIR_CMD_LAST | TXGBE_RDB_FDIR_CMD_QUEUE_EN; + if (queue == TXGBE_RDB_FDIR_DROP_QUEUE) + fdircmd |= TXGBE_RDB_FDIR_CMD_DROP; + fdircmd |= TXGBE_RDB_FDIR_CMD_FLOW_TYPE(input->formatted.flow_type); + fdircmd |= TXGBE_RDB_FDIR_CMD_RX_QUEUE(queue); + fdircmd |= TXGBE_RDB_FDIR_CMD_VT_POOL(input->formatted.vm_pool); + + wr32(wx, TXGBE_RDB_FDIR_CMD, fdircmd); + err = txgbe_fdir_check_cmd_complete(wx); + if (err) + wx_err(wx, "Flow Director command did not complete!\n"); + + return err; +} + +int txgbe_fdir_erase_perfect_filter(struct wx *wx, + union txgbe_atr_input *input, + u16 soft_id) +{ + u32 fdirhash, fdircmd; + int err = 0; + + /* configure FDIRHASH register */ + fdirhash = (__force u32)input->formatted.bkt_hash; + fdirhash |= TXGBE_RDB_FDIR_HASH_SIG_SW_INDEX(soft_id); + wr32(wx, TXGBE_RDB_FDIR_HASH, fdirhash); + + /* flush hash to HW */ + WX_WRITE_FLUSH(wx); + + /* Query if filter is present */ + wr32(wx, TXGBE_RDB_FDIR_CMD, TXGBE_RDB_FDIR_CMD_CMD_QUERY_REM_FILT); + + err = txgbe_fdir_check_cmd_complete(wx); + if (err) { + wx_err(wx, "Flow Director command did not complete!\n"); + return err; + } + + fdircmd = rd32(wx, TXGBE_RDB_FDIR_CMD); + /* if filter exists in hardware then remove it */ + if (fdircmd & TXGBE_RDB_FDIR_CMD_FILTER_VALID) { + wr32(wx, TXGBE_RDB_FDIR_HASH, fdirhash); + WX_WRITE_FLUSH(wx); + wr32(wx, TXGBE_RDB_FDIR_CMD, + TXGBE_RDB_FDIR_CMD_CMD_REMOVE_FLOW); + } + + return 0; +} + +/** + * txgbe_fdir_enable - Initialize Flow Director control registers + * @wx: pointer to hardware structure + * @fdirctrl: value to write to flow director control register + **/ +static void txgbe_fdir_enable(struct wx *wx, u32 fdirctrl) +{ + u32 val; + int ret; + + /* Prime the keys for hashing */ + wr32(wx, TXGBE_RDB_FDIR_HKEY, TXGBE_ATR_BUCKET_HASH_KEY); + wr32(wx, TXGBE_RDB_FDIR_SKEY, TXGBE_ATR_SIGNATURE_HASH_KEY); + + wr32(wx, TXGBE_RDB_FDIR_CTL, fdirctrl); + WX_WRITE_FLUSH(wx); + ret = read_poll_timeout(rd32, val, val & TXGBE_RDB_FDIR_CTL_INIT_DONE, + 1000, 10000, false, wx, TXGBE_RDB_FDIR_CTL); + + if (ret < 0) + wx_dbg(wx, "Flow Director poll time exceeded!\n"); +} + +/** + * txgbe_init_fdir_signature -Initialize Flow Director sig filters + * @wx: pointer to hardware structure + **/ +static void txgbe_init_fdir_signature(struct wx *wx) +{ + u32 fdirctrl = TXGBE_FDIR_PBALLOC_64K; + u32 flex = 0; + + flex = rd32(wx, TXGBE_RDB_FDIR_FLEX_CFG(0)); + flex &= ~TXGBE_RDB_FDIR_FLEX_CFG_FIELD0; + + flex |= (TXGBE_RDB_FDIR_FLEX_CFG_BASE_MAC | + TXGBE_RDB_FDIR_FLEX_CFG_OFST(0x6)); + wr32(wx, TXGBE_RDB_FDIR_FLEX_CFG(0), flex); + + /* Continue setup of fdirctrl register bits: + * Move the flexible bytes to use the ethertype - shift 6 words + * Set the maximum length per hash bucket to 0xA filters + * Send interrupt when 64 filters are left + */ + fdirctrl |= TXGBE_RDB_FDIR_CTL_HASH_BITS(0xF) | + TXGBE_RDB_FDIR_CTL_MAX_LENGTH(0xA) | + TXGBE_RDB_FDIR_CTL_FULL_THRESH(4); + + /* write hashes and fdirctrl register, poll for completion */ + txgbe_fdir_enable(wx, fdirctrl); +} + +/** + * txgbe_init_fdir_perfect - Initialize Flow Director perfect filters + * @wx: pointer to hardware structure + **/ +static void txgbe_init_fdir_perfect(struct wx *wx) +{ + u32 fdirctrl = TXGBE_FDIR_PBALLOC_64K; + + /* Continue setup of fdirctrl register bits: + * Turn perfect match filtering on + * Report hash in RSS field of Rx wb descriptor + * Initialize the drop queue + * Move the flexible bytes to use the ethertype - shift 6 words + * Set the maximum length per hash bucket to 0xA filters + * Send interrupt when 64 (0x4 * 16) filters are left + */ + fdirctrl |= TXGBE_RDB_FDIR_CTL_PERFECT_MATCH | + TXGBE_RDB_FDIR_CTL_DROP_Q(TXGBE_RDB_FDIR_DROP_QUEUE) | + TXGBE_RDB_FDIR_CTL_HASH_BITS(0xF) | + TXGBE_RDB_FDIR_CTL_MAX_LENGTH(0xA) | + TXGBE_RDB_FDIR_CTL_FULL_THRESH(4); + + /* write hashes and fdirctrl register, poll for completion */ + txgbe_fdir_enable(wx, fdirctrl); +} + +static void txgbe_fdir_filter_restore(struct wx *wx) +{ + struct txgbe_fdir_filter *filter; + struct txgbe *txgbe = wx->priv; + struct hlist_node *node; + u8 queue = 0; + int ret = 0; + + spin_lock(&txgbe->fdir_perfect_lock); + + if (!hlist_empty(&txgbe->fdir_filter_list)) + ret = txgbe_fdir_set_input_mask(wx, &txgbe->fdir_mask); + + if (ret) + goto unlock; + + hlist_for_each_entry_safe(filter, node, + &txgbe->fdir_filter_list, fdir_node) { + if (filter->action == TXGBE_RDB_FDIR_DROP_QUEUE) { + queue = TXGBE_RDB_FDIR_DROP_QUEUE; + } else { + u32 ring = ethtool_get_flow_spec_ring(filter->action); + + if (ring >= wx->num_rx_queues) { + wx_err(wx, "FDIR restore failed, ring:%u\n", + ring); + continue; + } + + /* Map the ring onto the absolute queue index */ + queue = wx->rx_ring[ring]->reg_idx; + } + + ret = txgbe_fdir_write_perfect_filter(wx, + &filter->filter, + filter->sw_idx, + queue); + if (ret) + wx_err(wx, "FDIR restore failed, index:%u\n", + filter->sw_idx); + } + +unlock: + spin_unlock(&txgbe->fdir_perfect_lock); +} + +void txgbe_configure_fdir(struct wx *wx) +{ + wx_disable_sec_rx_path(wx); + + if (test_bit(WX_FLAG_FDIR_HASH, wx->flags)) { + txgbe_init_fdir_signature(wx); + } else if (test_bit(WX_FLAG_FDIR_PERFECT, wx->flags)) { + txgbe_init_fdir_perfect(wx); + txgbe_fdir_filter_restore(wx); + } + + wx_enable_sec_rx_path(wx); +} + +void txgbe_fdir_filter_exit(struct wx *wx) +{ + struct txgbe_fdir_filter *filter; + struct txgbe *txgbe = wx->priv; + struct hlist_node *node; + + spin_lock(&txgbe->fdir_perfect_lock); + + hlist_for_each_entry_safe(filter, node, + &txgbe->fdir_filter_list, fdir_node) { + hlist_del(&filter->fdir_node); + kfree(filter); + } + txgbe->fdir_filter_count = 0; + + spin_unlock(&txgbe->fdir_perfect_lock); +} diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_fdir.h b/drivers/net/ethernet/wangxun/txgbe/txgbe_fdir.h new file mode 100644 index 000000000000..1f44ce60becb --- /dev/null +++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_fdir.h @@ -0,0 +1,20 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright (c) 2015 - 2024 Beijing WangXun Technology Co., Ltd. */ + +#ifndef _TXGBE_FDIR_H_ +#define _TXGBE_FDIR_H_ + +void txgbe_atr_compute_perfect_hash(union txgbe_atr_input *input, + union txgbe_atr_input *input_mask); +void txgbe_atr(struct wx_ring *ring, struct wx_tx_buffer *first, u8 ptype); +int txgbe_fdir_set_input_mask(struct wx *wx, union txgbe_atr_input *input_mask); +int txgbe_fdir_write_perfect_filter(struct wx *wx, + union txgbe_atr_input *input, + u16 soft_id, u8 queue); +int txgbe_fdir_erase_perfect_filter(struct wx *wx, + union txgbe_atr_input *input, + u16 soft_id); +void txgbe_configure_fdir(struct wx *wx); +void txgbe_fdir_filter_exit(struct wx *wx); + +#endif /* _TXGBE_FDIR_H_ */ diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c b/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c index ca74d9422065..93180225a6f1 100644 --- a/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c +++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c @@ -18,6 +18,7 @@ #include "txgbe_hw.h" #include "txgbe_phy.h" #include "txgbe_irq.h" +#include "txgbe_fdir.h" #include "txgbe_ethtool.h" char txgbe_driver_name[] = "txgbe"; @@ -257,6 +258,14 @@ static int txgbe_sw_init(struct wx *wx) num_online_cpus()); wx->rss_enabled = true; + wx->ring_feature[RING_F_FDIR].limit = min_t(int, TXGBE_MAX_FDIR_INDICES, + num_online_cpus()); + set_bit(WX_FLAG_FDIR_CAPABLE, wx->flags); + set_bit(WX_FLAG_FDIR_HASH, wx->flags); + wx->atr_sample_rate = TXGBE_DEFAULT_ATR_SAMPLE_RATE; + wx->atr = txgbe_atr; + wx->configure_fdir = txgbe_configure_fdir; + /* enable itr by default in dynamic mode */ wx->rx_itr_setting = 1; wx->tx_itr_setting = 1; @@ -274,6 +283,12 @@ static int txgbe_sw_init(struct wx *wx) return 0; } +static void txgbe_init_fdir(struct txgbe *txgbe) +{ + txgbe->fdir_filter_count = 0; + spin_lock_init(&txgbe->fdir_perfect_lock); +} + /** * txgbe_open - Called when a network interface is made active * @netdev: network interface device structure @@ -352,6 +367,7 @@ static int txgbe_close(struct net_device *netdev) txgbe_down(wx); wx_free_irq(wx); wx_free_resources(wx); + txgbe_fdir_filter_exit(wx); wx_control_hw(wx, false); return 0; @@ -660,6 +676,8 @@ static int txgbe_probe(struct pci_dev *pdev, txgbe->wx = wx; wx->priv = txgbe; + txgbe_init_fdir(txgbe); + err = txgbe_setup_misc_irq(txgbe); if (err) goto err_release_hw; diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h b/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h index f434a7865cb7..959102c4c379 100644 --- a/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h +++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h @@ -89,6 +89,55 @@ #define TXGBE_XPCS_IDA_ADDR 0x13000 #define TXGBE_XPCS_IDA_DATA 0x13004 +/********************************* Flow Director *****************************/ +#define TXGBE_RDB_FDIR_DROP_QUEUE 127 +#define TXGBE_RDB_FDIR_CTL 0x19500 +#define TXGBE_RDB_FDIR_CTL_INIT_DONE BIT(3) +#define TXGBE_RDB_FDIR_CTL_PERFECT_MATCH BIT(4) +#define TXGBE_RDB_FDIR_CTL_DROP_Q(v) FIELD_PREP(GENMASK(14, 8), v) +#define TXGBE_RDB_FDIR_CTL_HASH_BITS(v) FIELD_PREP(GENMASK(23, 20), v) +#define TXGBE_RDB_FDIR_CTL_MAX_LENGTH(v) FIELD_PREP(GENMASK(27, 24), v) +#define TXGBE_RDB_FDIR_CTL_FULL_THRESH(v) FIELD_PREP(GENMASK(31, 28), v) +#define TXGBE_RDB_FDIR_IP6(_i) (0x1950C + ((_i) * 4)) /* 0-2 */ +#define TXGBE_RDB_FDIR_SA 0x19518 +#define TXGBE_RDB_FDIR_DA 0x1951C +#define TXGBE_RDB_FDIR_PORT 0x19520 +#define TXGBE_RDB_FDIR_PORT_DESTINATION_SHIFT 16 +#define TXGBE_RDB_FDIR_FLEX 0x19524 +#define TXGBE_RDB_FDIR_FLEX_FLEX_SHIFT 16 +#define TXGBE_RDB_FDIR_HASH 0x19528 +#define TXGBE_RDB_FDIR_HASH_SIG_SW_INDEX(v) FIELD_PREP(GENMASK(31, 16), v) +#define TXGBE_RDB_FDIR_HASH_BUCKET_VALID BIT(15) +#define TXGBE_RDB_FDIR_CMD 0x1952C +#define TXGBE_RDB_FDIR_CMD_CMD_MASK GENMASK(1, 0) +#define TXGBE_RDB_FDIR_CMD_CMD(v) FIELD_PREP(GENMASK(1, 0), v) +#define TXGBE_RDB_FDIR_CMD_CMD_ADD_FLOW TXGBE_RDB_FDIR_CMD_CMD(1) +#define TXGBE_RDB_FDIR_CMD_CMD_REMOVE_FLOW TXGBE_RDB_FDIR_CMD_CMD(2) +#define TXGBE_RDB_FDIR_CMD_CMD_QUERY_REM_FILT TXGBE_RDB_FDIR_CMD_CMD(3) +#define TXGBE_RDB_FDIR_CMD_FILTER_VALID BIT(2) +#define TXGBE_RDB_FDIR_CMD_FILTER_UPDATE BIT(3) +#define TXGBE_RDB_FDIR_CMD_FLOW_TYPE(v) FIELD_PREP(GENMASK(6, 5), v) +#define TXGBE_RDB_FDIR_CMD_DROP BIT(9) +#define TXGBE_RDB_FDIR_CMD_LAST BIT(11) +#define TXGBE_RDB_FDIR_CMD_QUEUE_EN BIT(15) +#define TXGBE_RDB_FDIR_CMD_RX_QUEUE(v) FIELD_PREP(GENMASK(22, 16), v) +#define TXGBE_RDB_FDIR_CMD_VT_POOL(v) FIELD_PREP(GENMASK(29, 24), v) +#define TXGBE_RDB_FDIR_DA4_MSK 0x1953C +#define TXGBE_RDB_FDIR_SA4_MSK 0x19540 +#define TXGBE_RDB_FDIR_TCP_MSK 0x19544 +#define TXGBE_RDB_FDIR_UDP_MSK 0x19548 +#define TXGBE_RDB_FDIR_SCTP_MSK 0x19560 +#define TXGBE_RDB_FDIR_HKEY 0x19568 +#define TXGBE_RDB_FDIR_SKEY 0x1956C +#define TXGBE_RDB_FDIR_OTHER_MSK 0x19570 +#define TXGBE_RDB_FDIR_OTHER_MSK_POOL BIT(2) +#define TXGBE_RDB_FDIR_OTHER_MSK_L4P BIT(3) +#define TXGBE_RDB_FDIR_FLEX_CFG(_i) (0x19580 + ((_i) * 4)) +#define TXGBE_RDB_FDIR_FLEX_CFG_FIELD0 GENMASK(7, 0) +#define TXGBE_RDB_FDIR_FLEX_CFG_BASE_MAC FIELD_PREP(GENMASK(1, 0), 0) +#define TXGBE_RDB_FDIR_FLEX_CFG_MSK BIT(2) +#define TXGBE_RDB_FDIR_FLEX_CFG_OFST(v) FIELD_PREP(GENMASK(7, 3), v) + /* Checksum and EEPROM pointers */ #define TXGBE_EEPROM_LAST_WORD 0x800 #define TXGBE_EEPROM_CHECKSUM 0x2F @@ -112,6 +161,98 @@ #define TXGBE_SP_RX_PB_SIZE 512 #define TXGBE_SP_TDB_PB_SZ (160 * 1024) /* 160KB Packet Buffer */ +#define TXGBE_DEFAULT_ATR_SAMPLE_RATE 20 + +/* Software ATR hash keys */ +#define TXGBE_ATR_BUCKET_HASH_KEY 0x3DAD14E2 +#define TXGBE_ATR_SIGNATURE_HASH_KEY 0x174D3614 + +/* Software ATR input stream values and masks */ +#define TXGBE_ATR_HASH_MASK 0x7fff +#define TXGBE_ATR_L4TYPE_MASK 0x3 +#define TXGBE_ATR_L4TYPE_UDP 0x1 +#define TXGBE_ATR_L4TYPE_TCP 0x2 +#define TXGBE_ATR_L4TYPE_SCTP 0x3 +#define TXGBE_ATR_L4TYPE_IPV6_MASK 0x4 +#define TXGBE_ATR_L4TYPE_TUNNEL_MASK 0x10 + +enum txgbe_atr_flow_type { + TXGBE_ATR_FLOW_TYPE_IPV4 = 0x0, + TXGBE_ATR_FLOW_TYPE_UDPV4 = 0x1, + TXGBE_ATR_FLOW_TYPE_TCPV4 = 0x2, + TXGBE_ATR_FLOW_TYPE_SCTPV4 = 0x3, + TXGBE_ATR_FLOW_TYPE_IPV6 = 0x4, + TXGBE_ATR_FLOW_TYPE_UDPV6 = 0x5, + TXGBE_ATR_FLOW_TYPE_TCPV6 = 0x6, + TXGBE_ATR_FLOW_TYPE_SCTPV6 = 0x7, + TXGBE_ATR_FLOW_TYPE_TUNNELED_IPV4 = 0x10, + TXGBE_ATR_FLOW_TYPE_TUNNELED_UDPV4 = 0x11, + TXGBE_ATR_FLOW_TYPE_TUNNELED_TCPV4 = 0x12, + TXGBE_ATR_FLOW_TYPE_TUNNELED_SCTPV4 = 0x13, + TXGBE_ATR_FLOW_TYPE_TUNNELED_IPV6 = 0x14, + TXGBE_ATR_FLOW_TYPE_TUNNELED_UDPV6 = 0x15, + TXGBE_ATR_FLOW_TYPE_TUNNELED_TCPV6 = 0x16, + TXGBE_ATR_FLOW_TYPE_TUNNELED_SCTPV6 = 0x17, +}; + +/* Flow Director ATR input struct. */ +union txgbe_atr_input { + /* Byte layout in order, all values with MSB first: + * + * vm_pool - 1 byte + * flow_type - 1 byte + * vlan_id - 2 bytes + * dst_ip - 16 bytes + * src_ip - 16 bytes + * src_port - 2 bytes + * dst_port - 2 bytes + * flex_bytes - 2 bytes + * bkt_hash - 2 bytes + */ + struct { + u8 vm_pool; + u8 flow_type; + __be16 vlan_id; + __be32 dst_ip[4]; + __be32 src_ip[4]; + __be16 src_port; + __be16 dst_port; + __be16 flex_bytes; + __be16 bkt_hash; + } formatted; + __be32 dword_stream[11]; +}; + +/* Flow Director compressed ATR hash input struct */ +union txgbe_atr_hash_dword { + struct { + u8 vm_pool; + u8 flow_type; + __be16 vlan_id; + } formatted; + __be32 ip; + struct { + __be16 src; + __be16 dst; + } port; + __be16 flex_bytes; + __be32 dword; +}; + +enum txgbe_fdir_pballoc_type { + TXGBE_FDIR_PBALLOC_NONE = 0, + TXGBE_FDIR_PBALLOC_64K = 1, + TXGBE_FDIR_PBALLOC_128K = 2, + TXGBE_FDIR_PBALLOC_256K = 3, +}; + +struct txgbe_fdir_filter { + struct hlist_node fdir_node; + union txgbe_atr_input filter; + u16 sw_idx; + u16 action; +}; + /* TX/RX descriptor defines */ #define TXGBE_DEFAULT_TXD 512 #define TXGBE_DEFAULT_TX_WORK 256 @@ -196,6 +337,12 @@ struct txgbe { struct gpio_chip *gpio; unsigned int gpio_irq; unsigned int link_irq; + + /* flow director */ + struct hlist_head fdir_filter_list; + union txgbe_atr_input fdir_mask; + int fdir_filter_count; + spinlock_t fdir_perfect_lock; /* spinlock for FDIR */ }; #endif /* _TXGBE_TYPE_H_ */ |