From 5b7c4cabbb65f5c469464da6c5f614cbd7f730f2 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Tue, 21 Feb 2023 18:24:12 -0800 Subject: Merge tag 'net-next-6.3' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next Pull networking updates from Jakub Kicinski: "Core: - Add dedicated kmem_cache for typical/small skb->head, avoid having to access struct page at kfree time, and improve memory use. - Introduce sysctl to set default RPS configuration for new netdevs. - Define Netlink protocol specification format which can be used to describe messages used by each family and auto-generate parsers. Add tools for generating kernel data structures and uAPI headers. - Expose all net/core sysctls inside netns. - Remove 4s sleep in netpoll if carrier is instantly detected on boot. - Add configurable limit of MDB entries per port, and port-vlan. - Continue populating drop reasons throughout the stack. - Retire a handful of legacy Qdiscs and classifiers. Protocols: - Support IPv4 big TCP (TSO frames larger than 64kB). - Add IP_LOCAL_PORT_RANGE socket option, to control local port range on socket by socket basis. - Track and report in procfs number of MPTCP sockets used. - Support mixing IPv4 and IPv6 flows in the in-kernel MPTCP path manager. - IPv6: don't check net.ipv6.route.max_size and rely on garbage collection to free memory (similarly to IPv4). - Support Penultimate Segment Pop (PSP) flavor in SRv6 (RFC8986). - ICMP: add per-rate limit counters. - Add support for user scanning requests in ieee802154. - Remove static WEP support. - Support minimal Wi-Fi 7 Extremely High Throughput (EHT) rate reporting. - WiFi 7 EHT channel puncturing support (client & AP). BPF: - Add a rbtree data structure following the "next-gen data structure" precedent set by recently added linked list, that is, by using kfunc + kptr instead of adding a new BPF map type. - Expose XDP hints via kfuncs with initial support for RX hash and timestamp metadata. - Add BPF_F_NO_TUNNEL_KEY extension to bpf_skb_set_tunnel_key to better support decap on GRE tunnel devices not operating in collect metadata. - Improve x86 JIT's codegen for PROBE_MEM runtime error checks. - Remove the need for trace_printk_lock for bpf_trace_printk and bpf_trace_vprintk helpers. - Extend libbpf's bpf_tracing.h support for tracing arguments of kprobes/uprobes and syscall as a special case. - Significantly reduce the search time for module symbols by livepatch and BPF. - Enable cpumasks to be used as kptrs, which is useful for tracing programs tracking which tasks end up running on which CPUs in different time intervals. - Add support for BPF trampoline on s390x and riscv64. - Add capability to export the XDP features supported by the NIC. - Add __bpf_kfunc tag for marking kernel functions as kfuncs. - Add cgroup.memory=nobpf kernel parameter option to disable BPF memory accounting for container environments. Netfilter: - Remove the CLUSTERIP target. It has been marked as obsolete for years, and we still have WARN splats wrt races of the out-of-band /proc interface installed by this target. - Add 'destroy' commands to nf_tables. They are identical to the existing 'delete' commands, but do not return an error if the referenced object (set, chain, rule...) did not exist. Driver API: - Improve cpumask_local_spread() locality to help NICs set the right IRQ affinity on AMD platforms. - Separate C22 and C45 MDIO bus transactions more clearly. - Introduce new DCB table to control DSCP rewrite on egress. - Support configuration of Physical Layer Collision Avoidance (PLCA) Reconciliation Sublayer (RS) (802.3cg-2019). Modern version of shared medium Ethernet. - Support for MAC Merge layer (IEEE 802.3-2018 clause 99). Allowing preemption of low priority frames by high priority frames. - Add support for controlling MACSec offload using netlink SET. - Rework devlink instance refcounts to allow registration and de-registration under the instance lock. Split the code into multiple files, drop some of the unnecessarily granular locks and factor out common parts of netlink operation handling. - Add TX frame aggregation parameters (for USB drivers). - Add a new attr TCA_EXT_WARN_MSG to report TC (offload) warning messages with notifications for debug. - Allow offloading of UDP NEW connections via act_ct. - Add support for per action HW stats in TC. - Support hardware miss to TC action (continue processing in SW from a specific point in the action chain). - Warn if old Wireless Extension user space interface is used with modern cfg80211/mac80211 drivers. Do not support Wireless Extensions for Wi-Fi 7 devices at all. Everyone should switch to using nl80211 interface instead. - Improve the CAN bit timing configuration. Use extack to return error messages directly to user space, update the SJW handling, including the definition of a new default value that will benefit CAN-FD controllers, by increasing their oscillator tolerance. New hardware / drivers: - Ethernet: - nVidia BlueField-3 support (control traffic driver) - Ethernet support for imx93 SoCs - Motorcomm yt8531 gigabit Ethernet PHY - onsemi NCN26000 10BASE-T1S PHY (with support for PLCA) - Microchip LAN8841 PHY (incl. cable diagnostics and PTP) - Amlogic gxl MDIO mux - WiFi: - RealTek RTL8188EU (rtl8xxxu) - Qualcomm Wi-Fi 7 devices (ath12k) - CAN: - Renesas R-Car V4H Drivers: - Bluetooth: - Set Per Platform Antenna Gain (PPAG) for Intel controllers. - Ethernet NICs: - Intel (1G, igc): - support TSN / Qbv / packet scheduling features of i226 model - Intel (100G, ice): - use GNSS subsystem instead of TTY - multi-buffer XDP support - extend support for GPIO pins to E823 devices - nVidia/Mellanox: - update the shared buffer configuration on PFC commands - implement PTP adjphase function for HW offset control - TC support for Geneve and GRE with VF tunnel offload - more efficient crypto key management method - multi-port eswitch support - Netronome/Corigine: - add DCB IEEE support - support IPsec offloading for NFP3800 - Freescale/NXP (enetc): - support XDP_REDIRECT for XDP non-linear buffers - improve reconfig, avoid link flap and waiting for idle - support MAC Merge layer - Other NICs: - sfc/ef100: add basic devlink support for ef100 - ionic: rx_push mode operation (writing descriptors via MMIO) - bnxt: use the auxiliary bus abstraction for RDMA - r8169: disable ASPM and reset bus in case of tx timeout - cpsw: support QSGMII mode for J721e CPSW9G - cpts: support pulse-per-second output - ngbe: add an mdio bus driver - usbnet: optimize usbnet_bh() by avoiding unnecessary queuing - r8152: handle devices with FW with NCM support - amd-xgbe: support 10Mbps, 2.5GbE speeds and rx-adaptation - virtio-net: support multi buffer XDP - virtio/vsock: replace virtio_vsock_pkt with sk_buff - tsnep: XDP support - Ethernet high-speed switches: - nVidia/Mellanox (mlxsw): - add support for latency TLV (in FW control messages) - Microchip (sparx5): - separate explicit and implicit traffic forwarding rules, make the implicit rules always active - add support for egress DSCP rewrite - IS0 VCAP support (Ingress Classification) - IS2 VCAP filters (protos, L3 addrs, L4 ports, flags, ToS etc.) - ES2 VCAP support (Egress Access Control) - support for Per-Stream Filtering and Policing (802.1Q, 8.6.5.1) - Ethernet embedded switches: - Marvell (mv88e6xxx): - add MAB (port auth) offload support - enable PTP receive for mv88e6390 - NXP (ocelot): - support MAC Merge layer - support for the the vsc7512 internal copper phys - Microchip: - lan9303: convert to PHYLINK - lan966x: support TC flower filter statistics - lan937x: PTP support for KSZ9563/KSZ8563 and LAN937x - lan937x: support Credit Based Shaper configuration - ksz9477: support Energy Efficient Ethernet - other: - qca8k: convert to regmap read/write API, use bulk operations - rswitch: Improve TX timestamp accuracy - Intel WiFi (iwlwifi): - EHT (Wi-Fi 7) rate reporting - STEP equalizer support: transfer some STEP (connection to radio on platforms with integrated wifi) related parameters from the BIOS to the firmware. - Qualcomm 802.11ax WiFi (ath11k): - IPQ5018 support - Fine Timing Measurement (FTM) responder role support - channel 177 support - MediaTek WiFi (mt76): - per-PHY LED support - mt7996: EHT (Wi-Fi 7) support - Wireless Ethernet Dispatch (WED) reset support - switch to using page pool allocator - RealTek WiFi (rtw89): - support new version of Bluetooth co-existance - Mobile: - rmnet: support TX aggregation" * tag 'net-next-6.3' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next: (1872 commits) page_pool: add a comment explaining the fragment counter usage net: ethtool: fix __ethtool_dev_mm_supported() implementation ethtool: pse-pd: Fix double word in comments xsk: add linux/vmalloc.h to xsk.c sefltests: netdevsim: wait for devlink instance after netns removal selftest: fib_tests: Always cleanup before exit net/mlx5e: Align IPsec ASO result memory to be as required by hardware net/mlx5e: TC, Set CT miss to the specific ct action instance net/mlx5e: Rename CHAIN_TO_REG to MAPPED_OBJ_TO_REG net/mlx5: Refactor tc miss handling to a single function net/mlx5: Kconfig: Make tc offload depend on tc skb extension net/sched: flower: Support hardware miss to tc action net/sched: flower: Move filter handle initialization earlier net/sched: cls_api: Support hardware miss to tc action net/sched: Rename user cookie and act cookie sfc: fix builds without CONFIG_RTC_LIB sfc: clean up some inconsistent indentings net/mlx4_en: Introduce flexible array to silence overflow warning net: lan966x: Fix possible deadlock inside PTP net/ulp: Remove redundant ->clone() test in inet_clone_ulp(). ... --- drivers/net/ethernet/sfc/tc_counters.c | 503 +++++++++++++++++++++++++++++++++ 1 file changed, 503 insertions(+) create mode 100644 drivers/net/ethernet/sfc/tc_counters.c (limited to 'drivers/net/ethernet/sfc/tc_counters.c') diff --git a/drivers/net/ethernet/sfc/tc_counters.c b/drivers/net/ethernet/sfc/tc_counters.c new file mode 100644 index 000000000..d1a91d54c --- /dev/null +++ b/drivers/net/ethernet/sfc/tc_counters.c @@ -0,0 +1,503 @@ +// SPDX-License-Identifier: GPL-2.0-only +/**************************************************************************** + * Driver for Solarflare network controllers and boards + * Copyright 2022 Advanced Micro Devices, Inc. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License version 2 as published + * by the Free Software Foundation, incorporated herein by reference. + */ + +#include "tc_counters.h" +#include "mae_counter_format.h" +#include "mae.h" +#include "rx_common.h" + +/* Counter-management hashtables */ + +static const struct rhashtable_params efx_tc_counter_id_ht_params = { + .key_len = offsetof(struct efx_tc_counter_index, linkage), + .key_offset = 0, + .head_offset = offsetof(struct efx_tc_counter_index, linkage), +}; + +static const struct rhashtable_params efx_tc_counter_ht_params = { + .key_len = offsetof(struct efx_tc_counter, linkage), + .key_offset = 0, + .head_offset = offsetof(struct efx_tc_counter, linkage), +}; + +static void efx_tc_counter_free(void *ptr, void *__unused) +{ + struct efx_tc_counter *cnt = ptr; + + kfree(cnt); +} + +static void efx_tc_counter_id_free(void *ptr, void *__unused) +{ + struct efx_tc_counter_index *ctr = ptr; + + WARN_ON(refcount_read(&ctr->ref)); + kfree(ctr); +} + +int efx_tc_init_counters(struct efx_nic *efx) +{ + int rc; + + rc = rhashtable_init(&efx->tc->counter_id_ht, &efx_tc_counter_id_ht_params); + if (rc < 0) + goto fail_counter_id_ht; + rc = rhashtable_init(&efx->tc->counter_ht, &efx_tc_counter_ht_params); + if (rc < 0) + goto fail_counter_ht; + return 0; +fail_counter_ht: + rhashtable_destroy(&efx->tc->counter_id_ht); +fail_counter_id_ht: + return rc; +} + +/* Only call this in init failure teardown. + * Normal exit should fini instead as there may be entries in the table. + */ +void efx_tc_destroy_counters(struct efx_nic *efx) +{ + rhashtable_destroy(&efx->tc->counter_ht); + rhashtable_destroy(&efx->tc->counter_id_ht); +} + +void efx_tc_fini_counters(struct efx_nic *efx) +{ + rhashtable_free_and_destroy(&efx->tc->counter_id_ht, efx_tc_counter_id_free, NULL); + rhashtable_free_and_destroy(&efx->tc->counter_ht, efx_tc_counter_free, NULL); +} + +/* Counter allocation */ + +static struct efx_tc_counter *efx_tc_flower_allocate_counter(struct efx_nic *efx, + int type) +{ + struct efx_tc_counter *cnt; + int rc, rc2; + + cnt = kzalloc(sizeof(*cnt), GFP_USER); + if (!cnt) + return ERR_PTR(-ENOMEM); + + spin_lock_init(&cnt->lock); + cnt->touched = jiffies; + cnt->type = type; + + rc = efx_mae_allocate_counter(efx, cnt); + if (rc) + goto fail1; + rc = rhashtable_insert_fast(&efx->tc->counter_ht, &cnt->linkage, + efx_tc_counter_ht_params); + if (rc) + goto fail2; + return cnt; +fail2: + /* If we get here, it implies that we couldn't insert into the table, + * which in turn probably means that the fw_id was already taken. + * In that case, it's unclear whether we really 'own' the fw_id; but + * the firmware seemed to think we did, so it's proper to free it. + */ + rc2 = efx_mae_free_counter(efx, cnt); + if (rc2) + netif_warn(efx, hw, efx->net_dev, + "Failed to free MAE counter %u, rc %d\n", + cnt->fw_id, rc2); +fail1: + kfree(cnt); + return ERR_PTR(rc > 0 ? -EIO : rc); +} + +static void efx_tc_flower_release_counter(struct efx_nic *efx, + struct efx_tc_counter *cnt) +{ + int rc; + + rhashtable_remove_fast(&efx->tc->counter_ht, &cnt->linkage, + efx_tc_counter_ht_params); + rc = efx_mae_free_counter(efx, cnt); + if (rc) + netif_warn(efx, hw, efx->net_dev, + "Failed to free MAE counter %u, rc %d\n", + cnt->fw_id, rc); + /* This doesn't protect counter updates coming in arbitrarily long + * after we deleted the counter. The RCU just ensures that we won't + * free the counter while another thread has a pointer to it. + * Ensuring we don't update the wrong counter if the ID gets re-used + * is handled by the generation count. + */ + synchronize_rcu(); + EFX_WARN_ON_PARANOID(spin_is_locked(&cnt->lock)); + kfree(cnt); +} + +static struct efx_tc_counter *efx_tc_flower_find_counter_by_fw_id( + struct efx_nic *efx, int type, u32 fw_id) +{ + struct efx_tc_counter key = {}; + + key.fw_id = fw_id; + key.type = type; + + return rhashtable_lookup_fast(&efx->tc->counter_ht, &key, + efx_tc_counter_ht_params); +} + +/* TC cookie to counter mapping */ + +void efx_tc_flower_put_counter_index(struct efx_nic *efx, + struct efx_tc_counter_index *ctr) +{ + if (!refcount_dec_and_test(&ctr->ref)) + return; /* still in use */ + rhashtable_remove_fast(&efx->tc->counter_id_ht, &ctr->linkage, + efx_tc_counter_id_ht_params); + efx_tc_flower_release_counter(efx, ctr->cnt); + kfree(ctr); +} + +struct efx_tc_counter_index *efx_tc_flower_get_counter_index( + struct efx_nic *efx, unsigned long cookie, + enum efx_tc_counter_type type) +{ + struct efx_tc_counter_index *ctr, *old; + struct efx_tc_counter *cnt; + + ctr = kzalloc(sizeof(*ctr), GFP_USER); + if (!ctr) + return ERR_PTR(-ENOMEM); + ctr->cookie = cookie; + old = rhashtable_lookup_get_insert_fast(&efx->tc->counter_id_ht, + &ctr->linkage, + efx_tc_counter_id_ht_params); + if (old) { + /* don't need our new entry */ + kfree(ctr); + if (!refcount_inc_not_zero(&old->ref)) + return ERR_PTR(-EAGAIN); + /* existing entry found */ + ctr = old; + } else { + cnt = efx_tc_flower_allocate_counter(efx, type); + if (IS_ERR(cnt)) { + rhashtable_remove_fast(&efx->tc->counter_id_ht, + &ctr->linkage, + efx_tc_counter_id_ht_params); + kfree(ctr); + return (void *)cnt; /* it's an ERR_PTR */ + } + ctr->cnt = cnt; + refcount_set(&ctr->ref, 1); + } + return ctr; +} + +struct efx_tc_counter_index *efx_tc_flower_find_counter_index( + struct efx_nic *efx, unsigned long cookie) +{ + struct efx_tc_counter_index key = {}; + + key.cookie = cookie; + return rhashtable_lookup_fast(&efx->tc->counter_id_ht, &key, + efx_tc_counter_id_ht_params); +} + +/* TC Channel. Counter updates are delivered on this channel's RXQ. */ + +static void efx_tc_handle_no_channel(struct efx_nic *efx) +{ + netif_warn(efx, drv, efx->net_dev, + "MAE counters require MSI-X and 1 additional interrupt vector.\n"); +} + +static int efx_tc_probe_channel(struct efx_channel *channel) +{ + struct efx_rx_queue *rx_queue = &channel->rx_queue; + + channel->irq_moderation_us = 0; + rx_queue->core_index = 0; + + INIT_WORK(&rx_queue->grant_work, efx_mae_counters_grant_credits); + + return 0; +} + +static int efx_tc_start_channel(struct efx_channel *channel) +{ + struct efx_rx_queue *rx_queue = efx_channel_get_rx_queue(channel); + struct efx_nic *efx = channel->efx; + + return efx_mae_start_counters(efx, rx_queue); +} + +static void efx_tc_stop_channel(struct efx_channel *channel) +{ + struct efx_rx_queue *rx_queue = efx_channel_get_rx_queue(channel); + struct efx_nic *efx = channel->efx; + int rc; + + rc = efx_mae_stop_counters(efx, rx_queue); + if (rc) + netif_warn(efx, drv, efx->net_dev, + "Failed to stop MAE counters streaming, rc=%d.\n", + rc); + rx_queue->grant_credits = false; + flush_work(&rx_queue->grant_work); +} + +static void efx_tc_remove_channel(struct efx_channel *channel) +{ +} + +static void efx_tc_get_channel_name(struct efx_channel *channel, + char *buf, size_t len) +{ + snprintf(buf, len, "%s-mae", channel->efx->name); +} + +static void efx_tc_counter_update(struct efx_nic *efx, + enum efx_tc_counter_type counter_type, + u32 counter_idx, u64 packets, u64 bytes, + u32 mark) +{ + struct efx_tc_counter *cnt; + + rcu_read_lock(); /* Protect against deletion of 'cnt' */ + cnt = efx_tc_flower_find_counter_by_fw_id(efx, counter_type, counter_idx); + if (!cnt) { + /* This can legitimately happen when a counter is removed, + * with updates for the counter still in-flight; however this + * should be an infrequent occurrence. + */ + if (net_ratelimit()) + netif_dbg(efx, drv, efx->net_dev, + "Got update for unwanted MAE counter %u type %u\n", + counter_idx, counter_type); + goto out; + } + + spin_lock_bh(&cnt->lock); + if ((s32)mark - (s32)cnt->gen < 0) { + /* This counter update packet is from before the counter was + * allocated; thus it must be for a previous counter with + * the same ID that has since been freed, and it should be + * ignored. + */ + } else { + /* Update latest seen generation count. This ensures that + * even a long-lived counter won't start getting ignored if + * the generation count wraps around, unless it somehow + * manages to go 1<<31 generations without an update. + */ + cnt->gen = mark; + /* update counter values */ + cnt->packets += packets; + cnt->bytes += bytes; + cnt->touched = jiffies; + } + spin_unlock_bh(&cnt->lock); +out: + rcu_read_unlock(); +} + +static void efx_tc_rx_version_1(struct efx_nic *efx, const u8 *data, u32 mark) +{ + u16 n_counters, i; + + /* Header format: + * + | 0 | 1 | 2 | 3 | + * 0 |version | reserved | + * 4 | seq_index | n_counters | + */ + + n_counters = le16_to_cpu(*(const __le16 *)(data + 6)); + + /* Counter update entry format: + * | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | a | b | c | d | e | f | + * | counter_idx | packet_count | byte_count | + */ + for (i = 0; i < n_counters; i++) { + const void *entry = data + 8 + 16 * i; + u64 packet_count, byte_count; + u32 counter_idx; + + counter_idx = le32_to_cpu(*(const __le32 *)entry); + packet_count = le32_to_cpu(*(const __le32 *)(entry + 4)) | + ((u64)le16_to_cpu(*(const __le16 *)(entry + 8)) << 32); + byte_count = le16_to_cpu(*(const __le16 *)(entry + 10)) | + ((u64)le32_to_cpu(*(const __le32 *)(entry + 12)) << 16); + efx_tc_counter_update(efx, EFX_TC_COUNTER_TYPE_AR, counter_idx, + packet_count, byte_count, mark); + } +} + +#define TCV2_HDR_PTR(pkt, field) \ + ((void)BUILD_BUG_ON_ZERO(ERF_SC_PACKETISER_HEADER_##field##_LBN & 7), \ + (pkt) + ERF_SC_PACKETISER_HEADER_##field##_LBN / 8) +#define TCV2_HDR_BYTE(pkt, field) \ + ((void)BUILD_BUG_ON_ZERO(ERF_SC_PACKETISER_HEADER_##field##_WIDTH != 8),\ + *TCV2_HDR_PTR(pkt, field)) +#define TCV2_HDR_WORD(pkt, field) \ + ((void)BUILD_BUG_ON_ZERO(ERF_SC_PACKETISER_HEADER_##field##_WIDTH != 16),\ + (void)BUILD_BUG_ON_ZERO(ERF_SC_PACKETISER_HEADER_##field##_LBN & 15), \ + *(__force const __le16 *)TCV2_HDR_PTR(pkt, field)) +#define TCV2_PKT_PTR(pkt, poff, i, field) \ + ((void)BUILD_BUG_ON_ZERO(ERF_SC_PACKETISER_PAYLOAD_##field##_LBN & 7), \ + (pkt) + ERF_SC_PACKETISER_PAYLOAD_##field##_LBN/8 + poff + \ + i * ER_RX_SL_PACKETISER_PAYLOAD_WORD_SIZE) + +/* Read a little-endian 48-bit field with 16-bit alignment */ +static u64 efx_tc_read48(const __le16 *field) +{ + u64 out = 0; + int i; + + for (i = 0; i < 3; i++) + out |= (u64)le16_to_cpu(field[i]) << (i * 16); + return out; +} + +static enum efx_tc_counter_type efx_tc_rx_version_2(struct efx_nic *efx, + const u8 *data, u32 mark) +{ + u8 payload_offset, header_offset, ident; + enum efx_tc_counter_type type; + u16 n_counters, i; + + ident = TCV2_HDR_BYTE(data, IDENTIFIER); + switch (ident) { + case ERF_SC_PACKETISER_HEADER_IDENTIFIER_AR: + type = EFX_TC_COUNTER_TYPE_AR; + break; + case ERF_SC_PACKETISER_HEADER_IDENTIFIER_CT: + type = EFX_TC_COUNTER_TYPE_CT; + break; + case ERF_SC_PACKETISER_HEADER_IDENTIFIER_OR: + type = EFX_TC_COUNTER_TYPE_OR; + break; + default: + if (net_ratelimit()) + netif_err(efx, drv, efx->net_dev, + "ignored v2 MAE counter packet (bad identifier %u" + "), counters may be inaccurate\n", ident); + return EFX_TC_COUNTER_TYPE_MAX; + } + header_offset = TCV2_HDR_BYTE(data, HEADER_OFFSET); + /* mae_counter_format.h implies that this offset is fixed, since it + * carries on with SOP-based LBNs for the fields in this header + */ + if (header_offset != ERF_SC_PACKETISER_HEADER_HEADER_OFFSET_DEFAULT) { + if (net_ratelimit()) + netif_err(efx, drv, efx->net_dev, + "choked on v2 MAE counter packet (bad header_offset %u" + "), counters may be inaccurate\n", header_offset); + return EFX_TC_COUNTER_TYPE_MAX; + } + payload_offset = TCV2_HDR_BYTE(data, PAYLOAD_OFFSET); + n_counters = le16_to_cpu(TCV2_HDR_WORD(data, COUNT)); + + for (i = 0; i < n_counters; i++) { + const void *counter_idx_p, *packet_count_p, *byte_count_p; + u64 packet_count, byte_count; + u32 counter_idx; + + /* 24-bit field with 32-bit alignment */ + counter_idx_p = TCV2_PKT_PTR(data, payload_offset, i, COUNTER_INDEX); + BUILD_BUG_ON(ERF_SC_PACKETISER_PAYLOAD_COUNTER_INDEX_WIDTH != 24); + BUILD_BUG_ON(ERF_SC_PACKETISER_PAYLOAD_COUNTER_INDEX_LBN & 31); + counter_idx = le32_to_cpu(*(const __le32 *)counter_idx_p) & 0xffffff; + /* 48-bit field with 16-bit alignment */ + packet_count_p = TCV2_PKT_PTR(data, payload_offset, i, PACKET_COUNT); + BUILD_BUG_ON(ERF_SC_PACKETISER_PAYLOAD_PACKET_COUNT_WIDTH != 48); + BUILD_BUG_ON(ERF_SC_PACKETISER_PAYLOAD_PACKET_COUNT_LBN & 15); + packet_count = efx_tc_read48((const __le16 *)packet_count_p); + /* 48-bit field with 16-bit alignment */ + byte_count_p = TCV2_PKT_PTR(data, payload_offset, i, BYTE_COUNT); + BUILD_BUG_ON(ERF_SC_PACKETISER_PAYLOAD_BYTE_COUNT_WIDTH != 48); + BUILD_BUG_ON(ERF_SC_PACKETISER_PAYLOAD_BYTE_COUNT_LBN & 15); + byte_count = efx_tc_read48((const __le16 *)byte_count_p); + + if (type == EFX_TC_COUNTER_TYPE_CT) { + /* CT counters are 1-bit saturating counters to update + * the lastuse time in CT stats. A received CT counter + * should have packet counter to 0 and only LSB bit on + * in byte counter. + */ + if (packet_count || byte_count != 1) + netdev_warn_once(efx->net_dev, + "CT counter with inconsistent state (%llu, %llu)\n", + packet_count, byte_count); + /* Do not increment the driver's byte counter */ + byte_count = 0; + } + + efx_tc_counter_update(efx, type, counter_idx, packet_count, + byte_count, mark); + } + return type; +} + +/* We always swallow the packet, whether successful or not, since it's not + * a network packet and shouldn't ever be forwarded to the stack. + * @mark is the generation count for counter allocations. + */ +static bool efx_tc_rx(struct efx_rx_queue *rx_queue, u32 mark) +{ + struct efx_channel *channel = efx_rx_queue_channel(rx_queue); + struct efx_rx_buffer *rx_buf = efx_rx_buffer(rx_queue, + channel->rx_pkt_index); + const u8 *data = efx_rx_buf_va(rx_buf); + struct efx_nic *efx = rx_queue->efx; + enum efx_tc_counter_type type; + u8 version; + + /* version is always first byte of packet */ + version = *data; + switch (version) { + case 1: + type = EFX_TC_COUNTER_TYPE_AR; + efx_tc_rx_version_1(efx, data, mark); + break; + case ERF_SC_PACKETISER_HEADER_VERSION_VALUE: // 2 + type = efx_tc_rx_version_2(efx, data, mark); + break; + default: + if (net_ratelimit()) + netif_err(efx, drv, efx->net_dev, + "choked on MAE counter packet (bad version %u" + "); counters may be inaccurate\n", + version); + goto out; + } + + if (type < EFX_TC_COUNTER_TYPE_MAX) { + /* Update seen_gen unconditionally, to avoid a missed wakeup if + * we race with efx_mae_stop_counters(). + */ + efx->tc->seen_gen[type] = mark; + if (efx->tc->flush_counters && + (s32)(efx->tc->flush_gen[type] - mark) <= 0) + wake_up(&efx->tc->flush_wq); + } +out: + efx_free_rx_buffers(rx_queue, rx_buf, 1); + channel->rx_pkt_n_frags = 0; + return true; +} + +const struct efx_channel_type efx_tc_channel_type = { + .handle_no_channel = efx_tc_handle_no_channel, + .pre_probe = efx_tc_probe_channel, + .start = efx_tc_start_channel, + .stop = efx_tc_stop_channel, + .post_remove = efx_tc_remove_channel, + .get_name = efx_tc_get_channel_name, + .receive_raw = efx_tc_rx, + .keep_eventq = true, +}; -- cgit v1.2.3