Merge tag 'net-next-6.3' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-nextgrafted

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().
  ...

1 files changed, 930 insertions, 0 deletions

diff --git a/drivers/infiniband/hw/mlx5/counters.c b/drivers/infiniband/hw/mlx5/counters.c
new file mode 100644
index 000000000..3e1272695
--- /dev/null
+++ b/drivers/infiniband/hw/mlx5/counters.c
@@ -0,0 +1,930 @@
+// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB
+/*
+ * Copyright (c) 2013-2020, Mellanox Technologies inc. All rights reserved.
+ */
+
+#include "mlx5_ib.h"
+#include <linux/mlx5/eswitch.h>
+#include "counters.h"
+#include "ib_rep.h"
+#include "qp.h"
+
+struct mlx5_ib_counter {
+	const char *name;
+	size_t offset;
+	u32 type;
+};
+
+#define INIT_Q_COUNTER(_name)		\
+	{ .name = #_name, .offset = MLX5_BYTE_OFF(query_q_counter_out, _name)}
+
+static const struct mlx5_ib_counter basic_q_cnts[] = {
+	INIT_Q_COUNTER(rx_write_requests),
+	INIT_Q_COUNTER(rx_read_requests),
+	INIT_Q_COUNTER(rx_atomic_requests),
+	INIT_Q_COUNTER(out_of_buffer),
+};
+
+static const struct mlx5_ib_counter out_of_seq_q_cnts[] = {
+	INIT_Q_COUNTER(out_of_sequence),
+};
+
+static const struct mlx5_ib_counter retrans_q_cnts[] = {
+	INIT_Q_COUNTER(duplicate_request),
+	INIT_Q_COUNTER(rnr_nak_retry_err),
+	INIT_Q_COUNTER(packet_seq_err),
+	INIT_Q_COUNTER(implied_nak_seq_err),
+	INIT_Q_COUNTER(local_ack_timeout_err),
+};
+
+#define INIT_CONG_COUNTER(_name)		\
+	{ .name = #_name, .offset =	\
+		MLX5_BYTE_OFF(query_cong_statistics_out, _name ## _high)}
+
+static const struct mlx5_ib_counter cong_cnts[] = {
+	INIT_CONG_COUNTER(rp_cnp_ignored),
+	INIT_CONG_COUNTER(rp_cnp_handled),
+	INIT_CONG_COUNTER(np_ecn_marked_roce_packets),
+	INIT_CONG_COUNTER(np_cnp_sent),
+};
+
+static const struct mlx5_ib_counter extended_err_cnts[] = {
+	INIT_Q_COUNTER(resp_local_length_error),
+	INIT_Q_COUNTER(resp_cqe_error),
+	INIT_Q_COUNTER(req_cqe_error),
+	INIT_Q_COUNTER(req_remote_invalid_request),
+	INIT_Q_COUNTER(req_remote_access_errors),
+	INIT_Q_COUNTER(resp_remote_access_errors),
+	INIT_Q_COUNTER(resp_cqe_flush_error),
+	INIT_Q_COUNTER(req_cqe_flush_error),
+};
+
+static const struct mlx5_ib_counter roce_accl_cnts[] = {
+	INIT_Q_COUNTER(roce_adp_retrans),
+	INIT_Q_COUNTER(roce_adp_retrans_to),
+	INIT_Q_COUNTER(roce_slow_restart),
+	INIT_Q_COUNTER(roce_slow_restart_cnps),
+	INIT_Q_COUNTER(roce_slow_restart_trans),
+};
+
+#define INIT_EXT_PPCNT_COUNTER(_name)		\
+	{ .name = #_name, .offset =	\
+	MLX5_BYTE_OFF(ppcnt_reg, \
+		      counter_set.eth_extended_cntrs_grp_data_layout._name##_high)}
+
+static const struct mlx5_ib_counter ext_ppcnt_cnts[] = {
+	INIT_EXT_PPCNT_COUNTER(rx_icrc_encapsulated),
+};
+
+#define INIT_OP_COUNTER(_name, _type)		\
+	{ .name = #_name, .type = MLX5_IB_OPCOUNTER_##_type}
+
+static const struct mlx5_ib_counter basic_op_cnts[] = {
+	INIT_OP_COUNTER(cc_rx_ce_pkts, CC_RX_CE_PKTS),
+};
+
+static const struct mlx5_ib_counter rdmarx_cnp_op_cnts[] = {
+	INIT_OP_COUNTER(cc_rx_cnp_pkts, CC_RX_CNP_PKTS),
+};
+
+static const struct mlx5_ib_counter rdmatx_cnp_op_cnts[] = {
+	INIT_OP_COUNTER(cc_tx_cnp_pkts, CC_TX_CNP_PKTS),
+};
+
+static int mlx5_ib_read_counters(struct ib_counters *counters,
+				 struct ib_counters_read_attr *read_attr,
+				 struct uverbs_attr_bundle *attrs)
+{
+	struct mlx5_ib_mcounters *mcounters = to_mcounters(counters);
+	struct mlx5_read_counters_attr mread_attr = {};
+	struct mlx5_ib_flow_counters_desc *desc;
+	int ret, i;
+
+	mutex_lock(&mcounters->mcntrs_mutex);
+	if (mcounters->cntrs_max_index > read_attr->ncounters) {
+		ret = -EINVAL;
+		goto err_bound;
+	}
+
+	mread_attr.out = kcalloc(mcounters->counters_num, sizeof(u64),
+				 GFP_KERNEL);
+	if (!mread_attr.out) {
+		ret = -ENOMEM;
+		goto err_bound;
+	}
+
+	mread_attr.hw_cntrs_hndl = mcounters->hw_cntrs_hndl;
+	mread_attr.flags = read_attr->flags;
+	ret = mcounters->read_counters(counters->device, &mread_attr);
+	if (ret)
+		goto err_read;
+
+	/* do the pass over the counters data array to assign according to the
+	 * descriptions and indexing pairs
+	 */
+	desc = mcounters->counters_data;
+	for (i = 0; i < mcounters->ncounters; i++)
+		read_attr->counters_buff[desc[i].index] += mread_attr.out[desc[i].description];
+
+err_read:
+	kfree(mread_attr.out);
+err_bound:
+	mutex_unlock(&mcounters->mcntrs_mutex);
+	return ret;
+}
+
+static int mlx5_ib_destroy_counters(struct ib_counters *counters)
+{
+	struct mlx5_ib_mcounters *mcounters = to_mcounters(counters);
+
+	mlx5_ib_counters_clear_description(counters);
+	if (mcounters->hw_cntrs_hndl)
+		mlx5_fc_destroy(to_mdev(counters->device)->mdev,
+				mcounters->hw_cntrs_hndl);
+	return 0;
+}
+
+static int mlx5_ib_create_counters(struct ib_counters *counters,
+				   struct uverbs_attr_bundle *attrs)
+{
+	struct mlx5_ib_mcounters *mcounters = to_mcounters(counters);
+
+	mutex_init(&mcounters->mcntrs_mutex);
+	return 0;
+}
+
+
+static const struct mlx5_ib_counters *get_counters(struct mlx5_ib_dev *dev,
+						   u32 port_num)
+{
+	return is_mdev_switchdev_mode(dev->mdev) ? &dev->port[0].cnts :
+						   &dev->port[port_num].cnts;
+}
+
+/**
+ * mlx5_ib_get_counters_id - Returns counters id to use for device+port
+ * @dev:	Pointer to mlx5 IB device
+ * @port_num:	Zero based port number
+ *
+ * mlx5_ib_get_counters_id() Returns counters set id to use for given
+ * device port combination in switchdev and non switchdev mode of the
+ * parent device.
+ */
+u16 mlx5_ib_get_counters_id(struct mlx5_ib_dev *dev, u32 port_num)
+{
+	const struct mlx5_ib_counters *cnts = get_counters(dev, port_num);
+
+	return cnts->set_id;
+}
+
+static struct rdma_hw_stats *do_alloc_stats(const struct mlx5_ib_counters *cnts)
+{
+	struct rdma_hw_stats *stats;
+	u32 num_hw_counters;
+	int i;
+
+	num_hw_counters = cnts->num_q_counters + cnts->num_cong_counters +
+			  cnts->num_ext_ppcnt_counters;
+	stats = rdma_alloc_hw_stats_struct(cnts->descs,
+					   num_hw_counters +
+					   cnts->num_op_counters,
+					   RDMA_HW_STATS_DEFAULT_LIFESPAN);
+	if (!stats)
+		return NULL;
+
+	for (i = 0; i < cnts->num_op_counters; i++)
+		set_bit(num_hw_counters + i, stats->is_disabled);
+
+	return stats;
+}
+
+static struct rdma_hw_stats *
+mlx5_ib_alloc_hw_device_stats(struct ib_device *ibdev)
+{
+	struct mlx5_ib_dev *dev = to_mdev(ibdev);
+	const struct mlx5_ib_counters *cnts = &dev->port[0].cnts;
+
+	return do_alloc_stats(cnts);
+}
+
+static struct rdma_hw_stats *
+mlx5_ib_alloc_hw_port_stats(struct ib_device *ibdev, u32 port_num)
+{
+	struct mlx5_ib_dev *dev = to_mdev(ibdev);
+	const struct mlx5_ib_counters *cnts = &dev->port[port_num - 1].cnts;
+
+	return do_alloc_stats(cnts);
+}
+
+static int mlx5_ib_query_q_counters(struct mlx5_core_dev *mdev,
+				    const struct mlx5_ib_counters *cnts,
+				    struct rdma_hw_stats *stats,
+				    u16 set_id)
+{
+	u32 out[MLX5_ST_SZ_DW(query_q_counter_out)] = {};
+	u32 in[MLX5_ST_SZ_DW(query_q_counter_in)] = {};
+	__be32 val;
+	int ret, i;
+
+	MLX5_SET(query_q_counter_in, in, opcode, MLX5_CMD_OP_QUERY_Q_COUNTER);
+	MLX5_SET(query_q_counter_in, in, counter_set_id, set_id);
+	ret = mlx5_cmd_exec_inout(mdev, query_q_counter, in, out);
+	if (ret)
+		return ret;
+
+	for (i = 0; i < cnts->num_q_counters; i++) {
+		val = *(__be32 *)((void *)out + cnts->offsets[i]);
+		stats->value[i] = (u64)be32_to_cpu(val);
+	}
+
+	return 0;
+}
+
+static int mlx5_ib_query_ext_ppcnt_counters(struct mlx5_ib_dev *dev,
+					    const struct mlx5_ib_counters *cnts,
+					    struct rdma_hw_stats *stats)
+{
+	int offset = cnts->num_q_counters + cnts->num_cong_counters;
+	u32 in[MLX5_ST_SZ_DW(ppcnt_reg)] = {};
+	int sz = MLX5_ST_SZ_BYTES(ppcnt_reg);
+	int ret, i;
+	void *out;
+
+	out = kvzalloc(sz, GFP_KERNEL);
+	if (!out)
+		return -ENOMEM;
+
+	MLX5_SET(ppcnt_reg, in, local_port, 1);
+	MLX5_SET(ppcnt_reg, in, grp, MLX5_ETHERNET_EXTENDED_COUNTERS_GROUP);
+	ret = mlx5_core_access_reg(dev->mdev, in, sz, out, sz, MLX5_REG_PPCNT,
+				   0, 0);
+	if (ret)
+		goto free;
+
+	for (i = 0; i < cnts->num_ext_ppcnt_counters; i++)
+		stats->value[i + offset] =
+			be64_to_cpup((__be64 *)(out +
+				    cnts->offsets[i + offset]));
+free:
+	kvfree(out);
+	return ret;
+}
+
+static int do_get_hw_stats(struct ib_device *ibdev,
+			   struct rdma_hw_stats *stats,
+			   u32 port_num, int index)
+{
+	struct mlx5_ib_dev *dev = to_mdev(ibdev);
+	const struct mlx5_ib_counters *cnts = get_counters(dev, port_num - 1);
+	struct mlx5_core_dev *mdev;
+	int ret, num_counters;
+
+	if (!stats)
+		return -EINVAL;
+
+	num_counters = cnts->num_q_counters +
+		       cnts->num_cong_counters +
+		       cnts->num_ext_ppcnt_counters;
+
+	/* q_counters are per IB device, query the master mdev */
+	ret = mlx5_ib_query_q_counters(dev->mdev, cnts, stats, cnts->set_id);
+	if (ret)
+		return ret;
+
+	if (MLX5_CAP_PCAM_FEATURE(dev->mdev, rx_icrc_encapsulated_counter)) {
+		ret =  mlx5_ib_query_ext_ppcnt_counters(dev, cnts, stats);
+		if (ret)
+			return ret;
+	}
+
+	if (MLX5_CAP_GEN(dev->mdev, cc_query_allowed)) {
+		if (!port_num)
+			port_num = 1;
+		mdev = mlx5_ib_get_native_port_mdev(dev, port_num, NULL);
+		if (!mdev) {
+			/* If port is not affiliated yet, its in down state
+			 * which doesn't have any counters yet, so it would be
+			 * zero. So no need to read from the HCA.
+			 */
+			goto done;
+		}
+		ret = mlx5_lag_query_cong_counters(dev->mdev,
+						   stats->value +
+						   cnts->num_q_counters,
+						   cnts->num_cong_counters,
+						   cnts->offsets +
+						   cnts->num_q_counters);
+
+		mlx5_ib_put_native_port_mdev(dev, port_num);
+		if (ret)
+			return ret;
+	}
+
+done:
+	return num_counters;
+}
+
+static int do_get_op_stat(struct ib_device *ibdev,
+			  struct rdma_hw_stats *stats,
+			  u32 port_num, int index)
+{
+	struct mlx5_ib_dev *dev = to_mdev(ibdev);
+	const struct mlx5_ib_counters *cnts;
+	const struct mlx5_ib_op_fc *opfcs;
+	u64 packets = 0, bytes;
+	u32 type;
+	int ret;
+
+	cnts = get_counters(dev, port_num - 1);
+	opfcs = cnts->opfcs;
+	type = *(u32 *)cnts->descs[index].priv;
+	if (type >= MLX5_IB_OPCOUNTER_MAX)
+		return -EINVAL;
+
+	if (!opfcs[type].fc)
+		goto out;
+
+	ret = mlx5_fc_query(dev->mdev, opfcs[type].fc,
+			    &packets, &bytes);
+	if (ret)
+		return ret;
+
+out:
+	stats->value[index] = packets;
+	return index;
+}
+
+static int do_get_op_stats(struct ib_device *ibdev,
+			   struct rdma_hw_stats *stats,
+			   u32 port_num)
+{
+	struct mlx5_ib_dev *dev = to_mdev(ibdev);
+	const struct mlx5_ib_counters *cnts;
+	int index, ret, num_hw_counters;
+
+	cnts = get_counters(dev, port_num - 1);
+	num_hw_counters = cnts->num_q_counters + cnts->num_cong_counters +
+			  cnts->num_ext_ppcnt_counters;
+	for (index = num_hw_counters;
+	     index < (num_hw_counters + cnts->num_op_counters); index++) {
+		ret = do_get_op_stat(ibdev, stats, port_num, index);
+		if (ret != index)
+			return ret;
+	}
+
+	return cnts->num_op_counters;
+}
+
+static int mlx5_ib_get_hw_stats(struct ib_device *ibdev,
+				struct rdma_hw_stats *stats,
+				u32 port_num, int index)
+{
+	int num_counters, num_hw_counters, num_op_counters;
+	struct mlx5_ib_dev *dev = to_mdev(ibdev);
+	const struct mlx5_ib_counters *cnts;
+
+	cnts = get_counters(dev, port_num - 1);
+	num_hw_counters = cnts->num_q_counters + cnts->num_cong_counters +
+		cnts->num_ext_ppcnt_counters;
+	num_counters = num_hw_counters + cnts->num_op_counters;
+
+	if (index < 0 || index > num_counters)
+		return -EINVAL;
+	else if (index > 0 && index < num_hw_counters)
+		return do_get_hw_stats(ibdev, stats, port_num, index);
+	else if (index >= num_hw_counters && index < num_counters)
+		return do_get_op_stat(ibdev, stats, port_num, index);
+
+	num_hw_counters = do_get_hw_stats(ibdev, stats, port_num, index);
+	if (num_hw_counters < 0)
+		return num_hw_counters;
+
+	num_op_counters = do_get_op_stats(ibdev, stats, port_num);
+	if (num_op_counters < 0)
+		return num_op_counters;
+
+	return num_hw_counters + num_op_counters;
+}
+
+static struct rdma_hw_stats *
+mlx5_ib_counter_alloc_stats(struct rdma_counter *counter)
+{
+	struct mlx5_ib_dev *dev = to_mdev(counter->device);
+	const struct mlx5_ib_counters *cnts =
+		get_counters(dev, counter->port - 1);
+
+	return do_alloc_stats(cnts);
+}
+
+static int mlx5_ib_counter_update_stats(struct rdma_counter *counter)
+{
+	struct mlx5_ib_dev *dev = to_mdev(counter->device);
+	const struct mlx5_ib_counters *cnts =
+		get_counters(dev, counter->port - 1);
+
+	return mlx5_ib_query_q_counters(dev->mdev, cnts,
+					counter->stats, counter->id);
+}
+
+static int mlx5_ib_counter_dealloc(struct rdma_counter *counter)
+{
+	struct mlx5_ib_dev *dev = to_mdev(counter->device);
+	u32 in[MLX5_ST_SZ_DW(dealloc_q_counter_in)] = {};
+
+	if (!counter->id)
+		return 0;
+
+	MLX5_SET(dealloc_q_counter_in, in, opcode,
+		 MLX5_CMD_OP_DEALLOC_Q_COUNTER);
+	MLX5_SET(dealloc_q_counter_in, in, counter_set_id, counter->id);
+	return mlx5_cmd_exec_in(dev->mdev, dealloc_q_counter, in);
+}
+
+static int mlx5_ib_counter_bind_qp(struct rdma_counter *counter,
+				   struct ib_qp *qp)
+{
+	struct mlx5_ib_dev *dev = to_mdev(qp->device);
+	int err;
+
+	if (!counter->id) {
+		u32 out[MLX5_ST_SZ_DW(alloc_q_counter_out)] = {};
+		u32 in[MLX5_ST_SZ_DW(alloc_q_counter_in)] = {};
+
+		MLX5_SET(alloc_q_counter_in, in, opcode,
+			 MLX5_CMD_OP_ALLOC_Q_COUNTER);
+		MLX5_SET(alloc_q_counter_in, in, uid, MLX5_SHARED_RESOURCE_UID);
+		err = mlx5_cmd_exec_inout(dev->mdev, alloc_q_counter, in, out);
+		if (err)
+			return err;
+		counter->id =
+			MLX5_GET(alloc_q_counter_out, out, counter_set_id);
+	}
+
+	err = mlx5_ib_qp_set_counter(qp, counter);
+	if (err)
+		goto fail_set_counter;
+
+	return 0;
+
+fail_set_counter:
+	mlx5_ib_counter_dealloc(counter);
+	counter->id = 0;
+
+	return err;
+}
+
+static int mlx5_ib_counter_unbind_qp(struct ib_qp *qp)
+{
+	return mlx5_ib_qp_set_counter(qp, NULL);
+}
+
+static void mlx5_ib_fill_counters(struct mlx5_ib_dev *dev,
+				  struct rdma_stat_desc *descs, size_t *offsets)
+{
+	int i;
+	int j = 0;
+
+	for (i = 0; i < ARRAY_SIZE(basic_q_cnts); i++, j++) {
+		descs[j].name = basic_q_cnts[i].name;
+		offsets[j] = basic_q_cnts[i].offset;
+	}
+
+	if (MLX5_CAP_GEN(dev->mdev, out_of_seq_cnt)) {
+		for (i = 0; i < ARRAY_SIZE(out_of_seq_q_cnts); i++, j++) {
+			descs[j].name = out_of_seq_q_cnts[i].name;
+			offsets[j] = out_of_seq_q_cnts[i].offset;
+		}
+	}
+
+	if (MLX5_CAP_GEN(dev->mdev, retransmission_q_counters)) {
+		for (i = 0; i < ARRAY_SIZE(retrans_q_cnts); i++, j++) {
+			descs[j].name = retrans_q_cnts[i].name;
+			offsets[j] = retrans_q_cnts[i].offset;
+		}
+	}
+
+	if (MLX5_CAP_GEN(dev->mdev, enhanced_error_q_counters)) {
+		for (i = 0; i < ARRAY_SIZE(extended_err_cnts); i++, j++) {
+			descs[j].name = extended_err_cnts[i].name;
+			offsets[j] = extended_err_cnts[i].offset;
+		}
+	}
+
+	if (MLX5_CAP_GEN(dev->mdev, roce_accl)) {
+		for (i = 0; i < ARRAY_SIZE(roce_accl_cnts); i++, j++) {
+			descs[j].name = roce_accl_cnts[i].name;
+			offsets[j] = roce_accl_cnts[i].offset;
+		}
+	}
+
+	if (MLX5_CAP_GEN(dev->mdev, cc_query_allowed)) {
+		for (i = 0; i < ARRAY_SIZE(cong_cnts); i++, j++) {
+			descs[j].name = cong_cnts[i].name;
+			offsets[j] = cong_cnts[i].offset;
+		}
+	}
+
+	if (MLX5_CAP_PCAM_FEATURE(dev->mdev, rx_icrc_encapsulated_counter)) {
+		for (i = 0; i < ARRAY_SIZE(ext_ppcnt_cnts); i++, j++) {
+			descs[j].name = ext_ppcnt_cnts[i].name;
+			offsets[j] = ext_ppcnt_cnts[i].offset;
+		}
+	}
+
+	for (i = 0; i < ARRAY_SIZE(basic_op_cnts); i++, j++) {
+		descs[j].name = basic_op_cnts[i].name;
+		descs[j].flags |= IB_STAT_FLAG_OPTIONAL;
+		descs[j].priv = &basic_op_cnts[i].type;
+	}
+
+	if (MLX5_CAP_FLOWTABLE(dev->mdev,
+			       ft_field_support_2_nic_receive_rdma.bth_opcode)) {
+		for (i = 0; i < ARRAY_SIZE(rdmarx_cnp_op_cnts); i++, j++) {
+			descs[j].name = rdmarx_cnp_op_cnts[i].name;
+			descs[j].flags |= IB_STAT_FLAG_OPTIONAL;
+			descs[j].priv = &rdmarx_cnp_op_cnts[i].type;
+		}
+	}
+
+	if (MLX5_CAP_FLOWTABLE(dev->mdev,
+			       ft_field_support_2_nic_transmit_rdma.bth_opcode)) {
+		for (i = 0; i < ARRAY_SIZE(rdmatx_cnp_op_cnts); i++, j++) {
+			descs[j].name = rdmatx_cnp_op_cnts[i].name;
+			descs[j].flags |= IB_STAT_FLAG_OPTIONAL;
+			descs[j].priv = &rdmatx_cnp_op_cnts[i].type;
+		}
+	}
+}
+
+
+static int __mlx5_ib_alloc_counters(struct mlx5_ib_dev *dev,
+				    struct mlx5_ib_counters *cnts)
+{
+	u32 num_counters, num_op_counters;
+
+	num_counters = ARRAY_SIZE(basic_q_cnts);
+
+	if (MLX5_CAP_GEN(dev->mdev, out_of_seq_cnt))
+		num_counters += ARRAY_SIZE(out_of_seq_q_cnts);
+
+	if (MLX5_CAP_GEN(dev->mdev, retransmission_q_counters))
+		num_counters += ARRAY_SIZE(retrans_q_cnts);
+
+	if (MLX5_CAP_GEN(dev->mdev, enhanced_error_q_counters))
+		num_counters += ARRAY_SIZE(extended_err_cnts);
+
+	if (MLX5_CAP_GEN(dev->mdev, roce_accl))
+		num_counters += ARRAY_SIZE(roce_accl_cnts);
+
+	cnts->num_q_counters = num_counters;
+
+	if (MLX5_CAP_GEN(dev->mdev, cc_query_allowed)) {
+		cnts->num_cong_counters = ARRAY_SIZE(cong_cnts);
+		num_counters += ARRAY_SIZE(cong_cnts);
+	}
+	if (MLX5_CAP_PCAM_FEATURE(dev->mdev, rx_icrc_encapsulated_counter)) {
+		cnts->num_ext_ppcnt_counters = ARRAY_SIZE(ext_ppcnt_cnts);
+		num_counters += ARRAY_SIZE(ext_ppcnt_cnts);
+	}
+
+	num_op_counters = ARRAY_SIZE(basic_op_cnts);
+
+	if (MLX5_CAP_FLOWTABLE(dev->mdev,
+			       ft_field_support_2_nic_receive_rdma.bth_opcode))
+		num_op_counters += ARRAY_SIZE(rdmarx_cnp_op_cnts);
+
+	if (MLX5_CAP_FLOWTABLE(dev->mdev,
+			       ft_field_support_2_nic_transmit_rdma.bth_opcode))
+		num_op_counters += ARRAY_SIZE(rdmatx_cnp_op_cnts);
+
+	cnts->num_op_counters = num_op_counters;
+	num_counters += num_op_counters;
+	cnts->descs = kcalloc(num_counters,
+			      sizeof(struct rdma_stat_desc), GFP_KERNEL);
+	if (!cnts->descs)
+		return -ENOMEM;
+
+	cnts->offsets = kcalloc(num_counters,
+				sizeof(*cnts->offsets), GFP_KERNEL);
+	if (!cnts->offsets)
+		goto err;
+
+	return 0;
+
+err:
+	kfree(cnts->descs);
+	cnts->descs = NULL;
+	return -ENOMEM;
+}
+
+static void mlx5_ib_dealloc_counters(struct mlx5_ib_dev *dev)
+{
+	u32 in[MLX5_ST_SZ_DW(dealloc_q_counter_in)] = {};
+	int num_cnt_ports;
+	int i, j;
+
+	num_cnt_ports = is_mdev_switchdev_mode(dev->mdev) ? 1 : dev->num_ports;
+
+	MLX5_SET(dealloc_q_counter_in, in, opcode,
+		 MLX5_CMD_OP_DEALLOC_Q_COUNTER);
+
+	for (i = 0; i < num_cnt_ports; i++) {
+		if (dev->port[i].cnts.set_id) {
+			MLX5_SET(dealloc_q_counter_in, in, counter_set_id,
+				 dev->port[i].cnts.set_id);
+			mlx5_cmd_exec_in(dev->mdev, dealloc_q_counter, in);
+		}
+		kfree(dev->port[i].cnts.descs);
+		kfree(dev->port[i].cnts.offsets);
+
+		for (j = 0; j < MLX5_IB_OPCOUNTER_MAX; j++) {
+			if (!dev->port[i].cnts.opfcs[j].fc)
+				continue;
+
+			if (IS_ENABLED(CONFIG_INFINIBAND_USER_ACCESS))
+				mlx5_ib_fs_remove_op_fc(dev,
+					&dev->port[i].cnts.opfcs[j], j);
+			mlx5_fc_destroy(dev->mdev,
+					dev->port[i].cnts.opfcs[j].fc);
+			dev->port[i].cnts.opfcs[j].fc = NULL;
+		}
+	}
+}
+
+static int mlx5_ib_alloc_counters(struct mlx5_ib_dev *dev)
+{
+	u32 out[MLX5_ST_SZ_DW(alloc_q_counter_out)] = {};
+	u32 in[MLX5_ST_SZ_DW(alloc_q_counter_in)] = {};
+	int num_cnt_ports;
+	int err = 0;
+	int i;
+	bool is_shared;
+
+	MLX5_SET(alloc_q_counter_in, in, opcode, MLX5_CMD_OP_ALLOC_Q_COUNTER);
+	is_shared = MLX5_CAP_GEN(dev->mdev, log_max_uctx) != 0;
+	num_cnt_ports = is_mdev_switchdev_mode(dev->mdev) ? 1 : dev->num_ports;
+
+	for (i = 0; i < num_cnt_ports; i++) {
+		err = __mlx5_ib_alloc_counters(dev, &dev->port[i].cnts);
+		if (err)
+			goto err_alloc;
+
+		mlx5_ib_fill_counters(dev, dev->port[i].cnts.descs,
+				      dev->port[i].cnts.offsets);
+
+		MLX5_SET(alloc_q_counter_in, in, uid,
+			 is_shared ? MLX5_SHARED_RESOURCE_UID : 0);
+
+		err = mlx5_cmd_exec_inout(dev->mdev, alloc_q_counter, in, out);
+		if (err) {
+			mlx5_ib_warn(dev,
+				     "couldn't allocate queue counter for port %d, err %d\n",
+				     i + 1, err);
+			goto err_alloc;
+		}
+
+		dev->port[i].cnts.set_id =
+			MLX5_GET(alloc_q_counter_out, out, counter_set_id);
+	}
+	return 0;
+
+err_alloc:
+	mlx5_ib_dealloc_counters(dev);
+	return err;
+}
+
+static int read_flow_counters(struct ib_device *ibdev,
+			      struct mlx5_read_counters_attr *read_attr)
+{
+	struct mlx5_fc *fc = read_attr->hw_cntrs_hndl;
+	struct mlx5_ib_dev *dev = to_mdev(ibdev);
+
+	return mlx5_fc_query(dev->mdev, fc,
+			     &read_attr->out[IB_COUNTER_PACKETS],
+			     &read_attr->out[IB_COUNTER_BYTES]);
+}
+
+/* flow counters currently expose two counters packets and bytes */
+#define FLOW_COUNTERS_NUM 2
+static int counters_set_description(
+	struct ib_counters *counters, enum mlx5_ib_counters_type counters_type,
+	struct mlx5_ib_flow_counters_desc *desc_data, u32 ncounters)
+{
+	struct mlx5_ib_mcounters *mcounters = to_mcounters(counters);
+	u32 cntrs_max_index = 0;
+	int i;
+
+	if (counters_type != MLX5_IB_COUNTERS_FLOW)
+		return -EINVAL;
+
+	/* init the fields for the object */
+	mcounters->type = counters_type;
+	mcounters->read_counters = read_flow_counters;
+	mcounters->counters_num = FLOW_COUNTERS_NUM;
+	mcounters->ncounters = ncounters;
+	/* each counter entry have both description and index pair */
+	for (i = 0; i < ncounters; i++) {
+		if (desc_data[i].description > IB_COUNTER_BYTES)
+			return -EINVAL;
+
+		if (cntrs_max_index <= desc_data[i].index)
+			cntrs_max_index = desc_data[i].index + 1;
+	}
+
+	mutex_lock(&mcounters->mcntrs_mutex);
+	mcounters->counters_data = desc_data;
+	mcounters->cntrs_max_index = cntrs_max_index;
+	mutex_unlock(&mcounters->mcntrs_mutex);
+
+	return 0;
+}
+
+#define MAX_COUNTERS_NUM (USHRT_MAX / (sizeof(u32) * 2))
+int mlx5_ib_flow_counters_set_data(struct ib_counters *ibcounters,
+				   struct mlx5_ib_create_flow *ucmd)
+{
+	struct mlx5_ib_mcounters *mcounters = to_mcounters(ibcounters);
+	struct mlx5_ib_flow_counters_data *cntrs_data = NULL;
+	struct mlx5_ib_flow_counters_desc *desc_data = NULL;
+	bool hw_hndl = false;
+	int ret = 0;
+
+	if (ucmd && ucmd->ncounters_data != 0) {
+		cntrs_data = ucmd->data;
+		if (cntrs_data->ncounters > MAX_COUNTERS_NUM)
+			return -EINVAL;
+
+		desc_data = kcalloc(cntrs_data->ncounters,
+				    sizeof(*desc_data),
+				    GFP_KERNEL);
+		if (!desc_data)
+			return  -ENOMEM;
+
+		if (copy_from_user(desc_data,
+				   u64_to_user_ptr(cntrs_data->counters_data),
+				   sizeof(*desc_data) * cntrs_data->ncounters)) {
+			ret = -EFAULT;
+			goto free;
+		}
+	}
+
+	if (!mcounters->hw_cntrs_hndl) {
+		mcounters->hw_cntrs_hndl = mlx5_fc_create(
+			to_mdev(ibcounters->device)->mdev, false);
+		if (IS_ERR(mcounters->hw_cntrs_hndl)) {
+			ret = PTR_ERR(mcounters->hw_cntrs_hndl);
+			goto free;
+		}
+		hw_hndl = true;
+	}
+
+	if (desc_data) {
+		/* counters already bound to at least one flow */
+		if (mcounters->cntrs_max_index) {
+			ret = -EINVAL;
+			goto free_hndl;
+		}
+
+		ret = counters_set_description(ibcounters,
+					       MLX5_IB_COUNTERS_FLOW,
+					       desc_data,
+					       cntrs_data->ncounters);
+		if (ret)
+			goto free_hndl;
+
+	} else if (!mcounters->cntrs_max_index) {
+		/* counters not bound yet, must have udata passed */
+		ret = -EINVAL;
+		goto free_hndl;
+	}
+
+	return 0;
+
+free_hndl:
+	if (hw_hndl) {
+		mlx5_fc_destroy(to_mdev(ibcounters->device)->mdev,
+				mcounters->hw_cntrs_hndl);
+		mcounters->hw_cntrs_hndl = NULL;
+	}
+free:
+	kfree(desc_data);
+	return ret;
+}
+
+void mlx5_ib_counters_clear_description(struct ib_counters *counters)
+{
+	struct mlx5_ib_mcounters *mcounters;
+
+	if (!counters || atomic_read(&counters->usecnt) != 1)
+		return;
+
+	mcounters = to_mcounters(counters);
+
+	mutex_lock(&mcounters->mcntrs_mutex);
+	kfree(mcounters->counters_data);
+	mcounters->counters_data = NULL;
+	mcounters->cntrs_max_index = 0;
+	mutex_unlock(&mcounters->mcntrs_mutex);
+}
+
+static int mlx5_ib_modify_stat(struct ib_device *device, u32 port,
+			       unsigned int index, bool enable)
+{
+	struct mlx5_ib_dev *dev = to_mdev(device);
+	struct mlx5_ib_counters *cnts;
+	struct mlx5_ib_op_fc *opfc;
+	u32 num_hw_counters, type;
+	int ret;
+
+	cnts = &dev->port[port - 1].cnts;
+	num_hw_counters = cnts->num_q_counters + cnts->num_cong_counters +
+		cnts->num_ext_ppcnt_counters;
+	if (index < num_hw_counters ||
+	    index >= (num_hw_counters + cnts->num_op_counters))
+		return -EINVAL;
+
+	if (!(cnts->descs[index].flags & IB_STAT_FLAG_OPTIONAL))
+		return -EINVAL;
+
+	type = *(u32 *)cnts->descs[index].priv;
+	if (type >= MLX5_IB_OPCOUNTER_MAX)
+		return -EINVAL;
+
+	opfc = &cnts->opfcs[type];
+
+	if (enable) {
+		if (opfc->fc)
+			return -EEXIST;
+
+		opfc->fc = mlx5_fc_create(dev->mdev, false);
+		if (IS_ERR(opfc->fc))
+			return PTR_ERR(opfc->fc);
+
+		ret = mlx5_ib_fs_add_op_fc(dev, port, opfc, type);
+		if (ret) {
+			mlx5_fc_destroy(dev->mdev, opfc->fc);
+			opfc->fc = NULL;
+		}
+		return ret;
+	}
+
+	if (!opfc->fc)
+		return -EINVAL;
+
+	mlx5_ib_fs_remove_op_fc(dev, opfc, type);
+	mlx5_fc_destroy(dev->mdev, opfc->fc);
+	opfc->fc = NULL;
+	return 0;
+}
+
+static const struct ib_device_ops hw_stats_ops = {
+	.alloc_hw_port_stats = mlx5_ib_alloc_hw_port_stats,
+	.get_hw_stats = mlx5_ib_get_hw_stats,
+	.counter_bind_qp = mlx5_ib_counter_bind_qp,
+	.counter_unbind_qp = mlx5_ib_counter_unbind_qp,
+	.counter_dealloc = mlx5_ib_counter_dealloc,
+	.counter_alloc_stats = mlx5_ib_counter_alloc_stats,
+	.counter_update_stats = mlx5_ib_counter_update_stats,
+	.modify_hw_stat = IS_ENABLED(CONFIG_INFINIBAND_USER_ACCESS) ?
+			  mlx5_ib_modify_stat : NULL,
+};
+
+static const struct ib_device_ops hw_switchdev_stats_ops = {
+	.alloc_hw_device_stats = mlx5_ib_alloc_hw_device_stats,
+	.get_hw_stats = mlx5_ib_get_hw_stats,
+	.counter_bind_qp = mlx5_ib_counter_bind_qp,
+	.counter_unbind_qp = mlx5_ib_counter_unbind_qp,
+	.counter_dealloc = mlx5_ib_counter_dealloc,
+	.counter_alloc_stats = mlx5_ib_counter_alloc_stats,
+	.counter_update_stats = mlx5_ib_counter_update_stats,
+};
+
+static const struct ib_device_ops counters_ops = {
+	.create_counters = mlx5_ib_create_counters,
+	.destroy_counters = mlx5_ib_destroy_counters,
+	.read_counters = mlx5_ib_read_counters,
+
+	INIT_RDMA_OBJ_SIZE(ib_counters, mlx5_ib_mcounters, ibcntrs),
+};
+
+int mlx5_ib_counters_init(struct mlx5_ib_dev *dev)
+{
+	ib_set_device_ops(&dev->ib_dev, &counters_ops);
+
+	if (!MLX5_CAP_GEN(dev->mdev, max_qp_cnt))
+		return 0;
+
+	if (is_mdev_switchdev_mode(dev->mdev))
+		ib_set_device_ops(&dev->ib_dev, &hw_switchdev_stats_ops);
+	else
+		ib_set_device_ops(&dev->ib_dev, &hw_stats_ops);
+	return mlx5_ib_alloc_counters(dev);
+}
+
+void mlx5_ib_counters_cleanup(struct mlx5_ib_dev *dev)
+{
+	if (!MLX5_CAP_GEN(dev->mdev, max_qp_cnt))
+		return;
+
+	mlx5_ib_dealloc_counters(dev);
+}