Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next

Pull networking updates from David Miller:

 1) Allow setting bluetooth L2CAP modes via socket option, from Luiz
    Augusto von Dentz.

 2) Add GSO partial support to igc, from Sasha Neftin.

 3) Several cleanups and improvements to r8169 from Heiner Kallweit.

 4) Add IF_OPER_TESTING link state and use it when ethtool triggers a
    device self-test. From Andrew Lunn.

 5) Start moving away from custom driver versions, use the globally
    defined kernel version instead, from Leon Romanovsky.

 6) Support GRO vis gro_cells in DSA layer, from Alexander Lobakin.

 7) Allow hard IRQ deferral during NAPI, from Eric Dumazet.

 8) Add sriov and vf support to hinic, from Luo bin.

 9) Support Media Redundancy Protocol (MRP) in the bridging code, from
    Horatiu Vultur.

10) Support netmap in the nft_nat code, from Pablo Neira Ayuso.

11) Allow UDPv6 encapsulation of ESP in the ipsec code, from Sabrina
    Dubroca. Also add ipv6 support for espintcp.

12) Lots of ReST conversions of the networking documentation, from Mauro
    Carvalho Chehab.

13) Support configuration of ethtool rxnfc flows in bcmgenet driver,
    from Doug Berger.

14) Allow to dump cgroup id and filter by it in inet_diag code, from
    Dmitry Yakunin.

15) Add infrastructure to export netlink attribute policies to
    userspace, from Johannes Berg.

16) Several optimizations to sch_fq scheduler, from Eric Dumazet.

17) Fallback to the default qdisc if qdisc init fails because otherwise
    a packet scheduler init failure will make a device inoperative. From
    Jesper Dangaard Brouer.

18) Several RISCV bpf jit optimizations, from Luke Nelson.

19) Correct the return type of the ->ndo_start_xmit() method in several
    drivers, it's netdev_tx_t but many drivers were using
    'int'. From Yunjian Wang.

20) Add an ethtool interface for PHY master/slave config, from Oleksij
    Rempel.

21) Add BPF iterators, from Yonghang Song.

22) Add cable test infrastructure, including ethool interfaces, from
    Andrew Lunn. Marvell PHY driver is the first to support this
    facility.

23) Remove zero-length arrays all over, from Gustavo A. R. Silva.

24) Calculate and maintain an explicit frame size in XDP, from Jesper
    Dangaard Brouer.

25) Add CAP_BPF, from Alexei Starovoitov.

26) Support terse dumps in the packet scheduler, from Vlad Buslov.

27) Support XDP_TX bulking in dpaa2 driver, from Ioana Ciornei.

28) Add devm_register_netdev(), from Bartosz Golaszewski.

29) Minimize qdisc resets, from Cong Wang.

30) Get rid of kernel_getsockopt and kernel_setsockopt in order to
    eliminate set_fs/get_fs calls. From Christoph Hellwig.

* git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next: (2517 commits)
  selftests: net: ip_defrag: ignore EPERM
  net_failover: fixed rollback in net_failover_open()
  Revert "tipc: Fix potential tipc_aead refcnt leak in tipc_crypto_rcv"
  Revert "tipc: Fix potential tipc_node refcnt leak in tipc_rcv"
  vmxnet3: allow rx flow hash ops only when rss is enabled
  hinic: add set_channels ethtool_ops support
  selftests/bpf: Add a default $(CXX) value
  tools/bpf: Don't use $(COMPILE.c)
  bpf, selftests: Use bpf_probe_read_kernel
  s390/bpf: Use bcr 0,%0 as tail call nop filler
  s390/bpf: Maintain 8-byte stack alignment
  selftests/bpf: Fix verifier test
  selftests/bpf: Fix sample_cnt shared between two threads
  bpf, selftests: Adapt cls_redirect to call csum_level helper
  bpf: Add csum_level helper for fixing up csum levels
  bpf: Fix up bpf_skb_adjust_room helper's skb csum setting
  sfc: add missing annotation for efx_ef10_try_update_nic_stats_vf()
  crypto/chtls: IPv6 support for inline TLS
  Crypto/chcr: Fixes a coccinile check error
  Crypto/chcr: Fixes compilations warnings
  ...

16 files changed, 813 insertions, 194 deletions

diff --git a/drivers/infiniband/hw/mlx5/Makefile b/drivers/infiniband/hw/mlx5/Makefile
index 2a334800f109..228be05fbaf8 100644
--- a/drivers/infiniband/hw/mlx5/Makefile
+++ b/drivers/infiniband/hw/mlx5/Makefile
@@ -1,11 +1,25 @@
 # SPDX-License-Identifier: GPL-2.0-only
-obj-$(CONFIG_MLX5_INFINIBAND)	+= mlx5_ib.o
+obj-$(CONFIG_MLX5_INFINIBAND) += mlx5_ib.o
+
+mlx5_ib-y := ah.o \
+	     cmd.o \
+	     cong.o \
+	     cq.o \
+	     doorbell.o \
+	     gsi.o \
+	     ib_virt.o \
+	     mad.o \
+	     main.o \
+	     mem.o \
+	     mr.o \
+	     qp.o \
+	     qpc.o \
+	     restrack.o \
+	     srq.o \
+	     srq_cmd.o
 
-mlx5_ib-y :=	main.o cq.o doorbell.o qp.o mem.o srq_cmd.o \
-		srq.o mr.o ah.o mad.o gsi.o ib_virt.o cmd.o \
-		cong.o restrack.o
 mlx5_ib-$(CONFIG_INFINIBAND_ON_DEMAND_PAGING) += odp.o
 mlx5_ib-$(CONFIG_MLX5_ESWITCH) += ib_rep.o
-mlx5_ib-$(CONFIG_INFINIBAND_USER_ACCESS) += devx.o
-mlx5_ib-$(CONFIG_INFINIBAND_USER_ACCESS) += flow.o
-mlx5_ib-$(CONFIG_INFINIBAND_USER_ACCESS) += qos.o
+mlx5_ib-$(CONFIG_INFINIBAND_USER_ACCESS) += devx.o \
+					    flow.o \
+					    qos.o
diff --git a/drivers/infiniband/hw/mlx5/cmd.c b/drivers/infiniband/hw/mlx5/cmd.c
index 4c26492ab8a3..a2fcbc49131e 100644
--- a/drivers/infiniband/hw/mlx5/cmd.c
+++ b/drivers/infiniband/hw/mlx5/cmd.c
@@ -327,23 +327,6 @@ int mlx5_cmd_xrcd_dealloc(struct mlx5_core_dev *dev, u32 xrcdn, u16 uid)
 	return mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
 }
 
-int mlx5_cmd_alloc_q_counter(struct mlx5_core_dev *dev, u16 *counter_id,
-			     u16 uid)
-{
-	u32 in[MLX5_ST_SZ_DW(alloc_q_counter_in)]   = {0};
-	u32 out[MLX5_ST_SZ_DW(alloc_q_counter_out)] = {0};
-	int err;
-
-	MLX5_SET(alloc_q_counter_in, in, opcode, MLX5_CMD_OP_ALLOC_Q_COUNTER);
-	MLX5_SET(alloc_q_counter_in, in, uid, uid);
-
-	err = mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
-	if (!err)
-		*counter_id = MLX5_GET(alloc_q_counter_out, out,
-				       counter_set_id);
-	return err;
-}
-
 int mlx5_cmd_mad_ifc(struct mlx5_core_dev *dev, const void *inb, void *outb,
 		     u16 opmod, u8 port)
 {
diff --git a/drivers/infiniband/hw/mlx5/cmd.h b/drivers/infiniband/hw/mlx5/cmd.h
index 945ebce73613..43079b18d9b4 100644
--- a/drivers/infiniband/hw/mlx5/cmd.h
+++ b/drivers/infiniband/hw/mlx5/cmd.h
@@ -61,8 +61,6 @@ int mlx5_cmd_detach_mcg(struct mlx5_core_dev *dev, union ib_gid *mgid,
 			u32 qpn, u16 uid);
 int mlx5_cmd_xrcd_alloc(struct mlx5_core_dev *dev, u32 *xrcdn, u16 uid);
 int mlx5_cmd_xrcd_dealloc(struct mlx5_core_dev *dev, u32 xrcdn, u16 uid);
-int mlx5_cmd_alloc_q_counter(struct mlx5_core_dev *dev, u16 *counter_id,
-			     u16 uid);
 int mlx5_cmd_mad_ifc(struct mlx5_core_dev *dev, const void *inb, void *outb,
 		     u16 opmod, u8 port);
 #endif /* MLX5_IB_CMD_H */
diff --git a/drivers/infiniband/hw/mlx5/cq.c b/drivers/infiniband/hw/mlx5/cq.c
index 146ba2966744..0c18cb6a2f14 100644
--- a/drivers/infiniband/hw/mlx5/cq.c
+++ b/drivers/infiniband/hw/mlx5/cq.c
@@ -36,6 +36,7 @@
 #include <rdma/ib_cache.h>
 #include "mlx5_ib.h"
 #include "srq.h"
+#include "qp.h"
 
 static void mlx5_ib_cq_comp(struct mlx5_core_cq *cq, struct mlx5_eqe *eqe)
 {
@@ -201,7 +202,7 @@ static void handle_responder(struct ib_wc *wc, struct mlx5_cqe64 *cqe,
 	case MLX5_CQE_RESP_WR_IMM:
 		wc->opcode	= IB_WC_RECV_RDMA_WITH_IMM;
 		wc->wc_flags	= IB_WC_WITH_IMM;
-		wc->ex.imm_data = cqe->imm_inval_pkey;
+		wc->ex.imm_data = cqe->immediate;
 		break;
 	case MLX5_CQE_RESP_SEND:
 		wc->opcode   = IB_WC_RECV;
@@ -213,12 +214,12 @@ static void handle_responder(struct ib_wc *wc, struct mlx5_cqe64 *cqe,
 	case MLX5_CQE_RESP_SEND_IMM:
 		wc->opcode	= IB_WC_RECV;
 		wc->wc_flags	= IB_WC_WITH_IMM;
-		wc->ex.imm_data = cqe->imm_inval_pkey;
+		wc->ex.imm_data = cqe->immediate;
 		break;
 	case MLX5_CQE_RESP_SEND_INV:
 		wc->opcode	= IB_WC_RECV;
 		wc->wc_flags	= IB_WC_WITH_INVALIDATE;
-		wc->ex.invalidate_rkey = be32_to_cpu(cqe->imm_inval_pkey);
+		wc->ex.invalidate_rkey = be32_to_cpu(cqe->inval_rkey);
 		break;
 	}
 	wc->src_qp	   = be32_to_cpu(cqe->flags_rqpn) & 0xffffff;
@@ -226,7 +227,7 @@ static void handle_responder(struct ib_wc *wc, struct mlx5_cqe64 *cqe,
 	g = (be32_to_cpu(cqe->flags_rqpn) >> 28) & 3;
 	wc->wc_flags |= g ? IB_WC_GRH : 0;
 	if (unlikely(is_qp1(qp->ibqp.qp_type))) {
-		u16 pkey = be32_to_cpu(cqe->imm_inval_pkey) & 0xffff;
+		u16 pkey = be32_to_cpu(cqe->pkey) & 0xffff;
 
 		ib_find_cached_pkey(&dev->ib_dev, qp->port, pkey,
 				    &wc->pkey_index);
@@ -484,7 +485,7 @@ repoll:
 		 * because CQs will be locked while QPs are removed
 		 * from the table.
 		 */
-		mqp = __mlx5_qp_lookup(dev->mdev, qpn);
+		mqp = radix_tree_lookup(&dev->qp_table.tree, qpn);
 		*cur_qp = to_mibqp(mqp);
 	}
 
diff --git a/drivers/infiniband/hw/mlx5/devx.c b/drivers/infiniband/hw/mlx5/devx.c
index 46e1ab771f10..35b98c2d64d5 100644
--- a/drivers/infiniband/hw/mlx5/devx.c
+++ b/drivers/infiniband/hw/mlx5/devx.c
@@ -14,6 +14,7 @@
 #include <linux/mlx5/driver.h>
 #include <linux/mlx5/fs.h>
 #include "mlx5_ib.h"
+#include "qp.h"
 #include <linux/xarray.h>
 
 #define UVERBS_MODULE_NAME mlx5_ib
@@ -1356,7 +1357,7 @@ static int devx_obj_cleanup(struct ib_uobject *uobject,
 	}
 
 	if (obj->flags & DEVX_OBJ_FLAGS_DCT)
-		ret = mlx5_core_destroy_dct(obj->ib_dev->mdev, &obj->core_dct);
+		ret = mlx5_core_destroy_dct(obj->ib_dev, &obj->core_dct);
 	else if (obj->flags & DEVX_OBJ_FLAGS_CQ)
 		ret = mlx5_core_destroy_cq(obj->ib_dev->mdev, &obj->core_cq);
 	else
@@ -1450,9 +1451,8 @@ static int UVERBS_HANDLER(MLX5_IB_METHOD_DEVX_OBJ_CREATE)(
 
 	if (opcode == MLX5_CMD_OP_CREATE_DCT) {
 		obj->flags |= DEVX_OBJ_FLAGS_DCT;
-		err = mlx5_core_create_dct(dev->mdev, &obj->core_dct,
-					   cmd_in, cmd_in_len,
-					   cmd_out, cmd_out_len);
+		err = mlx5_core_create_dct(dev, &obj->core_dct, cmd_in,
+					   cmd_in_len, cmd_out, cmd_out_len);
 	} else if (opcode == MLX5_CMD_OP_CREATE_CQ) {
 		obj->flags |= DEVX_OBJ_FLAGS_CQ;
 		obj->core_cq.comp = devx_cq_comp;
@@ -1499,7 +1499,7 @@ static int UVERBS_HANDLER(MLX5_IB_METHOD_DEVX_OBJ_CREATE)(
 
 obj_destroy:
 	if (obj->flags & DEVX_OBJ_FLAGS_DCT)
-		mlx5_core_destroy_dct(obj->ib_dev->mdev, &obj->core_dct);
+		mlx5_core_destroy_dct(obj->ib_dev, &obj->core_dct);
 	else if (obj->flags & DEVX_OBJ_FLAGS_CQ)
 		mlx5_core_destroy_cq(obj->ib_dev->mdev, &obj->core_cq);
 	else
diff --git a/drivers/infiniband/hw/mlx5/flow.c b/drivers/infiniband/hw/mlx5/flow.c
index 862b7bf3e646..3a0601c2052c 100644
--- a/drivers/infiniband/hw/mlx5/flow.c
+++ b/drivers/infiniband/hw/mlx5/flow.c
@@ -404,7 +404,10 @@ static bool mlx5_ib_modify_header_supported(struct mlx5_ib_dev *dev)
 {
 	return MLX5_CAP_FLOWTABLE_NIC_RX(dev->mdev,
 					 max_modify_header_actions) ||
-	       MLX5_CAP_FLOWTABLE_NIC_TX(dev->mdev, max_modify_header_actions);
+	       MLX5_CAP_FLOWTABLE_NIC_TX(dev->mdev,
+					 max_modify_header_actions) ||
+	       MLX5_CAP_FLOWTABLE_RDMA_TX(dev->mdev,
+					 max_modify_header_actions);
 }
 
 static int UVERBS_HANDLER(MLX5_IB_METHOD_FLOW_ACTION_CREATE_MODIFY_HEADER)(
@@ -427,7 +430,7 @@ static int UVERBS_HANDLER(MLX5_IB_METHOD_FLOW_ACTION_CREATE_MODIFY_HEADER)(
 
 	num_actions = uverbs_attr_ptr_get_array_size(
 		attrs, MLX5_IB_ATTR_CREATE_MODIFY_HEADER_ACTIONS_PRM,
-		MLX5_UN_SZ_BYTES(set_action_in_add_action_in_auto));
+		MLX5_UN_SZ_BYTES(set_add_copy_action_in_auto));
 	if (num_actions < 0)
 		return num_actions;
 
@@ -648,7 +651,7 @@ DECLARE_UVERBS_NAMED_METHOD(
 			UA_MANDATORY),
 	UVERBS_ATTR_PTR_IN(MLX5_IB_ATTR_CREATE_MODIFY_HEADER_ACTIONS_PRM,
 			   UVERBS_ATTR_MIN_SIZE(MLX5_UN_SZ_BYTES(
-				   set_action_in_add_action_in_auto)),
+				   set_add_copy_action_in_auto)),
 			   UA_MANDATORY,
 			   UA_ALLOC_AND_COPY),
 	UVERBS_ATTR_CONST_IN(MLX5_IB_ATTR_CREATE_MODIFY_HEADER_FT_TYPE,
diff --git a/drivers/infiniband/hw/mlx5/ib_virt.c b/drivers/infiniband/hw/mlx5/ib_virt.c
index b61165359954..46b2d370fb3f 100644
--- a/drivers/infiniband/hw/mlx5/ib_virt.c
+++ b/drivers/infiniband/hw/mlx5/ib_virt.c
@@ -134,7 +134,7 @@ int mlx5_ib_get_vf_stats(struct ib_device *device, int vf,
 	if (!out)
 		return -ENOMEM;
 
-	err = mlx5_core_query_vport_counter(mdev, true, vf, port, out, out_sz);
+	err = mlx5_core_query_vport_counter(mdev, true, vf, port, out);
 	if (err)
 		goto ex;
 
diff --git a/drivers/infiniband/hw/mlx5/mad.c b/drivers/infiniband/hw/mlx5/mad.c
index 14e0c17de6a9..454ce5de2de7 100644
--- a/drivers/infiniband/hw/mlx5/mad.c
+++ b/drivers/infiniband/hw/mlx5/mad.c
@@ -30,7 +30,6 @@
  * SOFTWARE.
  */
 
-#include <linux/mlx5/cmd.h>
 #include <linux/mlx5/vport.h>
 #include <rdma/ib_mad.h>
 #include <rdma/ib_smi.h>
@@ -188,8 +187,8 @@ static int process_pma_cmd(struct mlx5_ib_dev *dev, u8 port_num,
 			goto done;
 		}
 
-		err = mlx5_core_query_vport_counter(mdev, 0, 0,
-						    mdev_port_num, out_cnt, sz);
+		err = mlx5_core_query_vport_counter(mdev, 0, 0, mdev_port_num,
+						    out_cnt);
 		if (!err)
 			pma_cnt_ext_assign(pma_cnt_ext, out_cnt);
 	} else {
diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c
index 6679756506e6..566b42f3fb18 100644
--- a/drivers/infiniband/hw/mlx5/main.c
+++ b/drivers/infiniband/hw/mlx5/main.c
@@ -59,6 +59,7 @@
 #include "ib_rep.h"
 #include "cmd.h"
 #include "srq.h"
+#include "qp.h"
 #include <linux/mlx5/fs_helpers.h>
 #include <linux/mlx5/accel.h>
 #include <rdma/uverbs_std_types.h>
@@ -2443,7 +2444,7 @@ static int handle_alloc_dm_sw_icm(struct ib_ucontext *ctx,
 	act_size = roundup_pow_of_two(act_size);
 
 	dm->size = act_size;
-	err = mlx5_dm_sw_icm_alloc(dev, type, act_size,
+	err = mlx5_dm_sw_icm_alloc(dev, type, act_size, attr->alignment,
 				   to_mucontext(ctx)->devx_uid, &dm->dev_addr,
 				   &dm->icm_dm.obj_id);
 	if (err)
@@ -3697,12 +3698,13 @@ static struct mlx5_ib_flow_handler *_create_flow_rule(struct mlx5_ib_dev *dev,
 		if (!dest_num)
 			rule_dst = NULL;
 	} else {
+		if (flow_attr->flags & IB_FLOW_ATTR_FLAGS_DONT_TRAP)
+			flow_act.action |=
+				MLX5_FLOW_CONTEXT_ACTION_FWD_NEXT_PRIO;
 		if (is_egress)
 			flow_act.action |= MLX5_FLOW_CONTEXT_ACTION_ALLOW;
-		else
-			flow_act.action |=
-				dest_num ?  MLX5_FLOW_CONTEXT_ACTION_FWD_DEST :
-					MLX5_FLOW_CONTEXT_ACTION_FWD_NEXT_PRIO;
+		else if (dest_num)
+			flow_act.action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;
 	}
 
 	if ((spec->flow_context.flags & FLOW_CONTEXT_HAS_TAG)  &&
@@ -3746,30 +3748,6 @@ static struct mlx5_ib_flow_handler *create_flow_rule(struct mlx5_ib_dev *dev,
 	return _create_flow_rule(dev, ft_prio, flow_attr, dst, 0, NULL);
 }
 
-static struct mlx5_ib_flow_handler *create_dont_trap_rule(struct mlx5_ib_dev *dev,
-							  struct mlx5_ib_flow_prio *ft_prio,
-							  struct ib_flow_attr *flow_attr,
-							  struct mlx5_flow_destination *dst)
-{
-	struct mlx5_ib_flow_handler *handler_dst = NULL;
-	struct mlx5_ib_flow_handler *handler = NULL;
-
-	handler = create_flow_rule(dev, ft_prio, flow_attr, NULL);
-	if (!IS_ERR(handler)) {
-		handler_dst = create_flow_rule(dev, ft_prio,
-					       flow_attr, dst);
-		if (IS_ERR(handler_dst)) {
-			mlx5_del_flow_rules(handler->rule);
-			ft_prio->refcount--;
-			kfree(handler);
-			handler = handler_dst;
-		} else {
-			list_add(&handler_dst->list, &handler->list);
-		}
-	}
-
-	return handler;
-}
 enum {
 	LEFTOVERS_MC,
 	LEFTOVERS_UC,
@@ -3973,15 +3951,11 @@ static struct ib_flow *mlx5_ib_create_flow(struct ib_qp *qp,
 	}
 
 	if (flow_attr->type == IB_FLOW_ATTR_NORMAL) {
-		if (flow_attr->flags & IB_FLOW_ATTR_FLAGS_DONT_TRAP)  {
-			handler = create_dont_trap_rule(dev, ft_prio,
-							flow_attr, dst);
-		} else {
-			underlay_qpn = (mqp->flags & MLX5_IB_QP_UNDERLAY) ?
-					mqp->underlay_qpn : 0;
-			handler = _create_flow_rule(dev, ft_prio, flow_attr,
-						    dst, underlay_qpn, ucmd);
-		}
+		underlay_qpn = (mqp->flags & IB_QP_CREATE_SOURCE_QPN) ?
+				       mqp->underlay_qpn :
+				       0;
+		handler = _create_flow_rule(dev, ft_prio, flow_attr, dst,
+					    underlay_qpn, ucmd);
 	} else if (flow_attr->type == IB_FLOW_ATTR_ALL_DEFAULT ||
 		   flow_attr->type == IB_FLOW_ATTR_MC_DEFAULT) {
 		handler = create_leftovers_rule(dev, ft_prio, flow_attr,
@@ -4632,8 +4606,7 @@ static void delay_drop_handler(struct work_struct *work)
 	atomic_inc(&delay_drop->events_cnt);
 
 	mutex_lock(&delay_drop->lock);
-	err = mlx5_core_set_delay_drop(delay_drop->dev->mdev,
-				       delay_drop->timeout);
+	err = mlx5_core_set_delay_drop(delay_drop->dev, delay_drop->timeout);
 	if (err) {
 		mlx5_ib_warn(delay_drop->dev, "Failed to set delay drop, timeout=%u\n",
 			     delay_drop->timeout);
@@ -5439,15 +5412,21 @@ static bool is_mdev_switchdev_mode(const struct mlx5_core_dev *mdev)
 
 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;
 
 	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_valid)
-			mlx5_core_dealloc_q_counter(dev->mdev,
-						    dev->port[i].cnts.set_id);
+		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.names);
 		kfree(dev->port[i].cnts.offsets);
 	}
@@ -5556,11 +5535,14 @@ static void mlx5_ib_fill_counters(struct mlx5_ib_dev *dev,
 
 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;
 
@@ -5572,17 +5554,19 @@ static int mlx5_ib_alloc_counters(struct mlx5_ib_dev *dev)
 		mlx5_ib_fill_counters(dev, dev->port[i].cnts.names,
 				      dev->port[i].cnts.offsets);
 
-		err = mlx5_cmd_alloc_q_counter(dev->mdev,
-					       &dev->port[i].cnts.set_id,
-					       is_shared ?
-					       MLX5_SHARED_RESOURCE_UID : 0);
+		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_valid = true;
+
+		dev->port[i].cnts.set_id =
+			MLX5_GET(alloc_q_counter_out, out, counter_set_id);
 	}
 	return 0;
 
@@ -5638,27 +5622,23 @@ static int mlx5_ib_query_q_counters(struct mlx5_core_dev *mdev,
 				    struct rdma_hw_stats *stats,
 				    u16 set_id)
 {
-	int outlen = MLX5_ST_SZ_BYTES(query_q_counter_out);
-	void *out;
+	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;
 
-	out = kvzalloc(outlen, GFP_KERNEL);
-	if (!out)
-		return -ENOMEM;
-
-	ret = mlx5_core_query_q_counter(mdev, set_id, 0, out, outlen);
+	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)
-		goto free;
+		return ret;
 
 	for (i = 0; i < cnts->num_q_counters; i++) {
-		val = *(__be32 *)(out + cnts->offsets[i]);
+		val = *(__be32 *)((void *)out + cnts->offsets[i]);
 		stats->value[i] = (u64)be32_to_cpu(val);
 	}
 
-free:
-	kvfree(out);
-	return ret;
+	return 0;
 }
 
 static int mlx5_ib_query_ext_ppcnt_counters(struct mlx5_ib_dev *dev,
@@ -5765,20 +5745,38 @@ static int mlx5_ib_counter_update_stats(struct rdma_counter *counter)
 					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);
-	u16 cnt_set_id = 0;
 	int err;
 
 	if (!counter->id) {
-		err = mlx5_cmd_alloc_q_counter(dev->mdev,
-					       &cnt_set_id,
-					       MLX5_SHARED_RESOURCE_UID);
+		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 = cnt_set_id;
+		counter->id =
+			MLX5_GET(alloc_q_counter_out, out, counter_set_id);
 	}
 
 	err = mlx5_ib_qp_set_counter(qp, counter);
@@ -5788,7 +5786,7 @@ static int mlx5_ib_counter_bind_qp(struct rdma_counter *counter,
 	return 0;
 
 fail_set_counter:
-	mlx5_core_dealloc_q_counter(dev->mdev, cnt_set_id);
+	mlx5_ib_counter_dealloc(counter);
 	counter->id = 0;
 
 	return err;
@@ -5799,13 +5797,6 @@ static int mlx5_ib_counter_unbind_qp(struct ib_qp *qp)
 	return mlx5_ib_qp_set_counter(qp, NULL);
 }
 
-static int mlx5_ib_counter_dealloc(struct rdma_counter *counter)
-{
-	struct mlx5_ib_dev *dev = to_mdev(counter->device);
-
-	return mlx5_core_dealloc_q_counter(dev->mdev, counter->id);
-}
-
 static int mlx5_ib_rn_get_params(struct ib_device *device, u8 port_num,
 				 enum rdma_netdev_t type,
 				 struct rdma_netdev_alloc_params *params)
@@ -7175,6 +7166,9 @@ static const struct mlx5_ib_profile pf_profile = {
 	STAGE_CREATE(MLX5_IB_STAGE_ROCE,
 		     mlx5_ib_stage_roce_init,
 		     mlx5_ib_stage_roce_cleanup),
+	STAGE_CREATE(MLX5_IB_STAGE_QP,
+		     mlx5_init_qp_table,
+		     mlx5_cleanup_qp_table),
 	STAGE_CREATE(MLX5_IB_STAGE_SRQ,
 		     mlx5_init_srq_table,
 		     mlx5_cleanup_srq_table),
@@ -7232,6 +7226,9 @@ const struct mlx5_ib_profile raw_eth_profile = {
 	STAGE_CREATE(MLX5_IB_STAGE_ROCE,
 		     mlx5_ib_stage_raw_eth_roce_init,
 		     mlx5_ib_stage_raw_eth_roce_cleanup),
+	STAGE_CREATE(MLX5_IB_STAGE_QP,
+		     mlx5_init_qp_table,
+		     mlx5_cleanup_qp_table),
 	STAGE_CREATE(MLX5_IB_STAGE_SRQ,
 		     mlx5_init_srq_table,
 		     mlx5_cleanup_srq_table),
diff --git a/drivers/infiniband/hw/mlx5/mlx5_ib.h b/drivers/infiniband/hw/mlx5/mlx5_ib.h
index a4e522385de0..aaabb8a98eed 100644
--- a/drivers/infiniband/hw/mlx5/mlx5_ib.h
+++ b/drivers/infiniband/hw/mlx5/mlx5_ib.h
@@ -780,7 +780,6 @@ struct mlx5_ib_counters {
 	u32 num_cong_counters;
 	u32 num_ext_ppcnt_counters;
 	u16 set_id;
-	bool set_id_valid;
 };
 
 struct mlx5_ib_multiport_info;
@@ -870,6 +869,7 @@ enum mlx5_ib_stages {
 	MLX5_IB_STAGE_CAPS,
 	MLX5_IB_STAGE_NON_DEFAULT_CB,
 	MLX5_IB_STAGE_ROCE,
+	MLX5_IB_STAGE_QP,
 	MLX5_IB_STAGE_SRQ,
 	MLX5_IB_STAGE_DEVICE_RESOURCES,
 	MLX5_IB_STAGE_DEVICE_NOTIFIER,
@@ -1065,6 +1065,7 @@ struct mlx5_ib_dev {
 	struct mlx5_dm		dm;
 	u16			devx_whitelist_uid;
 	struct mlx5_srq_table   srq_table;
+	struct mlx5_qp_table    qp_table;
 	struct mlx5_async_ctx   async_ctx;
 	struct mlx5_devx_event_table devx_event_table;
 	struct mlx5_var_table var_table;
diff --git a/drivers/infiniband/hw/mlx5/odp.c b/drivers/infiniband/hw/mlx5/odp.c
index 3de7606d4a1a..16af1105cfcf 100644
--- a/drivers/infiniband/hw/mlx5/odp.c
+++ b/drivers/infiniband/hw/mlx5/odp.c
@@ -36,6 +36,7 @@
 
 #include "mlx5_ib.h"
 #include "cmd.h"
+#include "qp.h"
 
 #include <linux/mlx5/eq.h>
 
@@ -1219,7 +1220,7 @@ static inline struct mlx5_core_rsc_common *odp_get_rsc(struct mlx5_ib_dev *dev,
 	case MLX5_WQE_PF_TYPE_REQ_SEND_OR_WRITE:
 	case MLX5_WQE_PF_TYPE_RESP:
 	case MLX5_WQE_PF_TYPE_REQ_READ_OR_ATOMIC:
-		common = mlx5_core_res_hold(dev->mdev, wq_num, MLX5_RES_QP);
+		common = mlx5_core_res_hold(dev, wq_num, MLX5_RES_QP);
 		break;
 	default:
 		break;
diff --git a/drivers/infiniband/hw/mlx5/qp.c b/drivers/infiniband/hw/mlx5/qp.c
index 2210759843ba..d93eec5d3277 100644
--- a/drivers/infiniband/hw/mlx5/qp.c
+++ b/drivers/infiniband/hw/mlx5/qp.c
@@ -39,6 +39,7 @@
 #include "mlx5_ib.h"
 #include "ib_rep.h"
 #include "cmd.h"
+#include "qp.h"
 
 /* not supported currently */
 static int wq_signature;
@@ -1254,7 +1255,7 @@ static int create_raw_packet_qp_tis(struct mlx5_ib_dev *dev,
 				    struct mlx5_ib_sq *sq, u32 tdn,
 				    struct ib_pd *pd)
 {
-	u32 in[MLX5_ST_SZ_DW(create_tis_in)] = {0};
+	u32 in[MLX5_ST_SZ_DW(create_tis_in)] = {};
 	void *tisc = MLX5_ADDR_OF(create_tis_in, in, ctx);
 
 	MLX5_SET(create_tis_in, in, uid, to_mpd(pd)->uid);
@@ -1262,7 +1263,7 @@ static int create_raw_packet_qp_tis(struct mlx5_ib_dev *dev,
 	if (qp->flags & MLX5_IB_QP_UNDERLAY)
 		MLX5_SET(tisc, tisc, underlay_qpn, qp->underlay_qpn);
 
-	return mlx5_core_create_tis(dev->mdev, in, sizeof(in), &sq->tisn);
+	return mlx5_core_create_tis(dev->mdev, in, &sq->tisn);
 }
 
 static void destroy_raw_packet_qp_tis(struct mlx5_ib_dev *dev,
@@ -1336,7 +1337,7 @@ static int create_raw_packet_qp_sq(struct mlx5_ib_dev *dev,
 	pas = (__be64 *)MLX5_ADDR_OF(wq, wq, pas);
 	mlx5_ib_populate_pas(dev, sq->ubuffer.umem, page_shift, pas, 0);
 
-	err = mlx5_core_create_sq_tracked(dev->mdev, in, inlen, &sq->base.mqp);
+	err = mlx5_core_create_sq_tracked(dev, in, inlen, &sq->base.mqp);
 
 	kvfree(in);
 
@@ -1356,7 +1357,7 @@ static void destroy_raw_packet_qp_sq(struct mlx5_ib_dev *dev,
 				     struct mlx5_ib_sq *sq)
 {
 	destroy_flow_rule_vport_sq(sq);
-	mlx5_core_destroy_sq_tracked(dev->mdev, &sq->base.mqp);
+	mlx5_core_destroy_sq_tracked(dev, &sq->base.mqp);
 	ib_umem_release(sq->ubuffer.umem);
 }
 
@@ -1426,7 +1427,7 @@ static int create_raw_packet_qp_rq(struct mlx5_ib_dev *dev,
 	qp_pas = (__be64 *)MLX5_ADDR_OF(create_qp_in, qpin, pas);
 	memcpy(pas, qp_pas, rq_pas_size);
 
-	err = mlx5_core_create_rq_tracked(dev->mdev, in, inlen, &rq->base.mqp);
+	err = mlx5_core_create_rq_tracked(dev, in, inlen, &rq->base.mqp);
 
 	kvfree(in);
 
@@ -1436,7 +1437,7 @@ static int create_raw_packet_qp_rq(struct mlx5_ib_dev *dev,
 static void destroy_raw_packet_qp_rq(struct mlx5_ib_dev *dev,
 				     struct mlx5_ib_rq *rq)
 {
-	mlx5_core_destroy_rq_tracked(dev->mdev, &rq->base.mqp);
+	mlx5_core_destroy_rq_tracked(dev, &rq->base.mqp);
 }
 
 static bool tunnel_offload_supported(struct mlx5_core_dev *dev)
@@ -1459,9 +1460,8 @@ static void destroy_raw_packet_qp_tir(struct mlx5_ib_dev *dev,
 
 static int create_raw_packet_qp_tir(struct mlx5_ib_dev *dev,
 				    struct mlx5_ib_rq *rq, u32 tdn,
-				    u32 *qp_flags_en,
-				    struct ib_pd *pd,
-				    u32 *out, int outlen)
+				    u32 *qp_flags_en, struct ib_pd *pd,
+				    u32 *out)
 {
 	u8 lb_flag = 0;
 	u32 *in;
@@ -1494,9 +1494,8 @@ static int create_raw_packet_qp_tir(struct mlx5_ib_dev *dev,
 	}
 
 	MLX5_SET(tirc, tirc, self_lb_block, lb_flag);
-
-	err = mlx5_core_create_tir_out(dev->mdev, in, inlen, out, outlen);
-
+	MLX5_SET(create_tir_in, in, opcode, MLX5_CMD_OP_CREATE_TIR);
+	err = mlx5_cmd_exec_inout(dev->mdev, create_tir, in, out);
 	rq->tirn = MLX5_GET(create_tir_out, out, tirn);
 	if (!err && MLX5_GET(tirc, tirc, self_lb_block)) {
 		err = mlx5_ib_enable_lb(dev, false, true);
@@ -1556,9 +1555,8 @@ static int create_raw_packet_qp(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp,
 		if (err)
 			goto err_destroy_sq;
 
-		err = create_raw_packet_qp_tir(
-			dev, rq, tdn, &qp->flags_en, pd, out,
-			MLX5_ST_SZ_BYTES(create_tir_out));
+		err = create_raw_packet_qp_tir(dev, rq, tdn, &qp->flags_en, pd,
+					       out);
 		if (err)
 			goto err_destroy_rq;
 
@@ -1853,7 +1851,8 @@ static int create_rss_raw_qp_tir(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp,
 	MLX5_SET(rx_hash_field_select, hfso, selected_fields, selected_fields);
 
 create_tir:
-	err = mlx5_core_create_tir_out(dev->mdev, in, inlen, out, outlen);
+	MLX5_SET(create_tir_in, in, opcode, MLX5_CMD_OP_CREATE_TIR);
+	err = mlx5_cmd_exec_inout(dev->mdev, create_tir, in, out);
 
 	qp->rss_qp.tirn = MLX5_GET(create_tir_out, out, tirn);
 	if (!err && MLX5_GET(tirc, tirc, self_lb_block)) {
@@ -2347,7 +2346,7 @@ static int create_qp_common(struct mlx5_ib_dev *dev, struct ib_pd *pd,
 		err = create_raw_packet_qp(dev, qp, in, inlen, pd, udata,
 					   &resp);
 	} else {
-		err = mlx5_core_create_qp(dev->mdev, &base->mqp, in, inlen);
+		err = mlx5_core_create_qp(dev, &base->mqp, in, inlen);
 	}
 
 	if (err) {
@@ -2513,8 +2512,7 @@ static void destroy_qp_common(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp,
 	if (qp->state != IB_QPS_RESET) {
 		if (qp->ibqp.qp_type != IB_QPT_RAW_PACKET &&
 		    !(qp->flags & MLX5_IB_QP_UNDERLAY)) {
-			err = mlx5_core_qp_modify(dev->mdev,
-						  MLX5_CMD_OP_2RST_QP, 0,
+			err = mlx5_core_qp_modify(dev, MLX5_CMD_OP_2RST_QP, 0,
 						  NULL, &base->mqp);
 		} else {
 			struct mlx5_modify_raw_qp_param raw_qp_param = {
@@ -2555,7 +2553,7 @@ static void destroy_qp_common(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp,
 	    qp->flags & MLX5_IB_QP_UNDERLAY) {
 		destroy_raw_packet_qp(dev, qp);
 	} else {
-		err = mlx5_core_destroy_qp(dev->mdev, &base->mqp);
+		err = mlx5_core_destroy_qp(dev, &base->mqp);
 		if (err)
 			mlx5_ib_warn(dev, "failed to destroy QP 0x%x\n",
 				     base->mqp.qpn);
@@ -2818,7 +2816,7 @@ static int mlx5_ib_destroy_dct(struct mlx5_ib_qp *mqp)
 	if (mqp->state == IB_QPS_RTR) {
 		int err;
 
-		err = mlx5_core_destroy_dct(dev->mdev, &mqp->dct.mdct);
+		err = mlx5_core_destroy_dct(dev, &mqp->dct.mdct);
 		if (err) {
 			mlx5_ib_warn(dev, "failed to destroy DCT %d\n", err);
 			return err;
@@ -2933,7 +2931,7 @@ static int modify_raw_packet_eth_prio(struct mlx5_core_dev *dev,
 	tisc = MLX5_ADDR_OF(modify_tis_in, in, ctx);
 	MLX5_SET(tisc, tisc, prio, ((sl & 0x7) << 1));
 
-	err = mlx5_core_modify_tis(dev, sq->tisn, in, inlen);
+	err = mlx5_core_modify_tis(dev, sq->tisn, in);
 
 	kvfree(in);
 
@@ -2960,7 +2958,7 @@ static int modify_raw_packet_tx_affinity(struct mlx5_core_dev *dev,
 	tisc = MLX5_ADDR_OF(modify_tis_in, in, ctx);
 	MLX5_SET(tisc, tisc, lag_tx_port_affinity, tx_affinity);
 
-	err = mlx5_core_modify_tis(dev, sq->tisn, in, inlen);
+	err = mlx5_core_modify_tis(dev, sq->tisn, in);
 
 	kvfree(in);
 
@@ -3240,7 +3238,7 @@ static int modify_raw_packet_qp_rq(
 				"RAW PACKET QP counters are not supported on current FW\n");
 	}
 
-	err = mlx5_core_modify_rq(dev->mdev, rq->base.mqp.qpn, in, inlen);
+	err = mlx5_core_modify_rq(dev->mdev, rq->base.mqp.qpn, in);
 	if (err)
 		goto out;
 
@@ -3303,7 +3301,7 @@ static int modify_raw_packet_qp_sq(
 		MLX5_SET(sqc, sqc, packet_pacing_rate_limit_index, rl_index);
 	}
 
-	err = mlx5_core_modify_sq(dev, sq->base.mqp.qpn, in, inlen);
+	err = mlx5_core_modify_sq(dev, sq->base.mqp.qpn, in);
 	if (err) {
 		/* Remove new rate from table if failed */
 		if (new_rate_added)
@@ -3462,10 +3460,9 @@ static int __mlx5_ib_qp_set_counter(struct ib_qp *qp,
 	base = &mqp->trans_qp.base;
 	context.qp_counter_set_usr_page &= cpu_to_be32(0xffffff);
 	context.qp_counter_set_usr_page |= cpu_to_be32(set_id << 24);
-	return mlx5_core_qp_modify(dev->mdev,
-				   MLX5_CMD_OP_RTS2RTS_QP,
-				   MLX5_QP_OPTPAR_COUNTER_SET_ID,
-				   &context, &base->mqp);
+	return mlx5_core_qp_modify(dev, MLX5_CMD_OP_RTS2RTS_QP,
+				   MLX5_QP_OPTPAR_COUNTER_SET_ID, &context,
+				   &base->mqp);
 }
 
 static int __mlx5_ib_modify_qp(struct ib_qp *ibqp,
@@ -3752,8 +3749,7 @@ static int __mlx5_ib_modify_qp(struct ib_qp *ibqp,
 
 		err = modify_raw_packet_qp(dev, qp, &raw_qp_param, tx_affinity);
 	} else {
-		err = mlx5_core_qp_modify(dev->mdev, op, optpar, context,
-					  &base->mqp);
+		err = mlx5_core_qp_modify(dev, op, optpar, context, &base->mqp);
 	}
 
 	if (err)
@@ -3927,7 +3923,7 @@ static int mlx5_ib_modify_dct(struct ib_qp *ibqp, struct ib_qp_attr *attr,
 		MLX5_SET(dctc, dctc, my_addr_index, attr->ah_attr.grh.sgid_index);
 		MLX5_SET(dctc, dctc, hop_limit, attr->ah_attr.grh.hop_limit);
 
-		err = mlx5_core_create_dct(dev->mdev, &qp->dct.mdct, qp->dct.in,
+		err = mlx5_core_create_dct(dev, &qp->dct.mdct, qp->dct.in,
 					   MLX5_ST_SZ_BYTES(create_dct_in), out,
 					   sizeof(out));
 		if (err)
@@ -3935,7 +3931,7 @@ static int mlx5_ib_modify_dct(struct ib_qp *ibqp, struct ib_qp_attr *attr,
 		resp.dctn = qp->dct.mdct.mqp.qpn;
 		err = ib_copy_to_udata(udata, &resp, resp.response_length);
 		if (err) {
-			mlx5_core_destroy_dct(dev->mdev, &qp->dct.mdct);
+			mlx5_core_destroy_dct(dev, &qp->dct.mdct);
 			return err;
 		}
 	} else {
@@ -5699,8 +5695,7 @@ static int query_qp_attr(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp,
 	if (!outb)
 		return -ENOMEM;
 
-	err = mlx5_core_qp_query(dev->mdev, &qp->trans_qp.base.mqp, outb,
-				 outlen);
+	err = mlx5_core_qp_query(dev, &qp->trans_qp.base.mqp, outb, outlen);
 	if (err)
 		goto out;
 
@@ -5778,7 +5773,7 @@ static int mlx5_ib_dct_query_qp(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *mqp,
 	if (!out)
 		return -ENOMEM;
 
-	err = mlx5_core_dct_query(dev->mdev, dct, out, outlen);
+	err = mlx5_core_dct_query(dev, dct, out, outlen);
 	if (err)
 		goto out;
 
@@ -5964,7 +5959,7 @@ static int set_delay_drop(struct mlx5_ib_dev *dev)
 	if (dev->delay_drop.activate)
 		goto out;
 
-	err = mlx5_core_set_delay_drop(dev->mdev, dev->delay_drop.timeout);
+	err = mlx5_core_set_delay_drop(dev, dev->delay_drop.timeout);
 	if (err)
 		goto out;
 
@@ -6070,13 +6065,13 @@ static int  create_rq(struct mlx5_ib_rwq *rwq, struct ib_pd *pd,
 	}
 	rq_pas0 = (__be64 *)MLX5_ADDR_OF(wq, wq, pas);
 	mlx5_ib_populate_pas(dev, rwq->umem, rwq->page_shift, rq_pas0, 0);
-	err = mlx5_core_create_rq_tracked(dev->mdev, in, inlen, &rwq->core_qp);
+	err = mlx5_core_create_rq_tracked(dev, in, inlen, &rwq->core_qp);
 	if (!err && init_attr->create_flags & IB_WQ_FLAGS_DELAY_DROP) {
 		err = set_delay_drop(dev);
 		if (err) {
 			mlx5_ib_warn(dev, "Failed to enable delay drop err=%d\n",
 				     err);
-			mlx5_core_destroy_rq_tracked(dev->mdev, &rwq->core_qp);
+			mlx5_core_destroy_rq_tracked(dev, &rwq->core_qp);
 		} else {
 			rwq->create_flags |= MLX5_IB_WQ_FLAGS_DELAY_DROP;
 		}
@@ -6258,7 +6253,7 @@ struct ib_wq *mlx5_ib_create_wq(struct ib_pd *pd,
 	return &rwq->ibwq;
 
 err_copy:
-	mlx5_core_destroy_rq_tracked(dev->mdev, &rwq->core_qp);
+	mlx5_core_destroy_rq_tracked(dev, &rwq->core_qp);
 err_user_rq:
 	destroy_user_rq(dev, pd, rwq, udata);
 err:
@@ -6271,7 +6266,7 @@ void mlx5_ib_destroy_wq(struct ib_wq *wq, struct ib_udata *udata)
 	struct mlx5_ib_dev *dev = to_mdev(wq->device);
 	struct mlx5_ib_rwq *rwq = to_mrwq(wq);
 
-	mlx5_core_destroy_rq_tracked(dev->mdev, &rwq->core_qp);
+	mlx5_core_destroy_rq_tracked(dev, &rwq->core_qp);
 	destroy_user_rq(dev, wq->pd, rwq, udata);
 	kfree(rwq);
 }
@@ -6449,7 +6444,7 @@ int mlx5_ib_modify_wq(struct ib_wq *wq, struct ib_wq_attr *wq_attr,
 				"Receive WQ counters are not supported on current FW\n");
 	}
 
-	err = mlx5_core_modify_rq(dev->mdev, rwq->core_qp.qpn, in, inlen);
+	err = mlx5_core_modify_rq(dev->mdev, rwq->core_qp.qpn, in);
 	if (!err)
 		rwq->ibwq.state = (wq_state == MLX5_RQC_STATE_ERR) ? IB_WQS_ERR : wq_state;
 
diff --git a/drivers/infiniband/hw/mlx5/qp.h b/drivers/infiniband/hw/mlx5/qp.h
new file mode 100644
index 000000000000..ad9d76e3e18a
--- /dev/null
+++ b/drivers/infiniband/hw/mlx5/qp.h
@@ -0,0 +1,46 @@
+/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */
+/*
+ * Copyright (c) 2013-2020, Mellanox Technologies inc. All rights reserved.
+ */
+
+#ifndef _MLX5_IB_QP_H
+#define _MLX5_IB_QP_H
+
+#include "mlx5_ib.h"
+
+int mlx5_init_qp_table(struct mlx5_ib_dev *dev);
+void mlx5_cleanup_qp_table(struct mlx5_ib_dev *dev);
+
+int mlx5_core_create_dct(struct mlx5_ib_dev *dev, struct mlx5_core_dct *qp,
+			 u32 *in, int inlen, u32 *out, int outlen);
+int mlx5_core_create_qp(struct mlx5_ib_dev *dev, struct mlx5_core_qp *qp,
+			u32 *in, int inlen);
+int mlx5_core_qp_modify(struct mlx5_ib_dev *dev, u16 opcode, u32 opt_param_mask,
+			void *qpc, struct mlx5_core_qp *qp);
+int mlx5_core_destroy_qp(struct mlx5_ib_dev *dev, struct mlx5_core_qp *qp);
+int mlx5_core_destroy_dct(struct mlx5_ib_dev *dev, struct mlx5_core_dct *dct);
+int mlx5_core_qp_query(struct mlx5_ib_dev *dev, struct mlx5_core_qp *qp,
+		       u32 *out, int outlen);
+int mlx5_core_dct_query(struct mlx5_ib_dev *dev, struct mlx5_core_dct *dct,
+			u32 *out, int outlen);
+
+int mlx5_core_set_delay_drop(struct mlx5_ib_dev *dev, u32 timeout_usec);
+
+void mlx5_core_destroy_rq_tracked(struct mlx5_ib_dev *dev,
+				  struct mlx5_core_qp *rq);
+int mlx5_core_create_sq_tracked(struct mlx5_ib_dev *dev, u32 *in, int inlen,
+				struct mlx5_core_qp *sq);
+void mlx5_core_destroy_sq_tracked(struct mlx5_ib_dev *dev,
+				  struct mlx5_core_qp *sq);
+
+int mlx5_core_create_rq_tracked(struct mlx5_ib_dev *dev, u32 *in, int inlen,
+				struct mlx5_core_qp *rq);
+
+struct mlx5_core_rsc_common *mlx5_core_res_hold(struct mlx5_ib_dev *dev,
+						int res_num,
+						enum mlx5_res_type res_type);
+void mlx5_core_res_put(struct mlx5_core_rsc_common *res);
+
+int mlx5_core_xrcd_alloc(struct mlx5_ib_dev *dev, u32 *xrcdn);
+int mlx5_core_xrcd_dealloc(struct mlx5_ib_dev *dev, u32 xrcdn);
+#endif /* _MLX5_IB_QP_H */
diff --git a/drivers/infiniband/hw/mlx5/qpc.c b/drivers/infiniband/hw/mlx5/qpc.c
new file mode 100644
index 000000000000..ea62735042f0
--- /dev/null
+++ b/drivers/infiniband/hw/mlx5/qpc.c
@@ -0,0 +1,605 @@
+// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB
+/*
+ * Copyright (c) 2013-2020, Mellanox Technologies inc. All rights reserved.
+ */
+
+#include <linux/gfp.h>
+#include <linux/mlx5/qp.h>
+#include <linux/mlx5/driver.h>
+#include "mlx5_ib.h"
+#include "qp.h"
+
+static int mlx5_core_drain_dct(struct mlx5_ib_dev *dev,
+			       struct mlx5_core_dct *dct);
+
+static struct mlx5_core_rsc_common *
+mlx5_get_rsc(struct mlx5_qp_table *table, u32 rsn)
+{
+	struct mlx5_core_rsc_common *common;
+	unsigned long flags;
+
+	spin_lock_irqsave(&table->lock, flags);
+
+	common = radix_tree_lookup(&table->tree, rsn);
+	if (common)
+		refcount_inc(&common->refcount);
+
+	spin_unlock_irqrestore(&table->lock, flags);
+
+	return common;
+}
+
+void mlx5_core_put_rsc(struct mlx5_core_rsc_common *common)
+{
+	if (refcount_dec_and_test(&common->refcount))
+		complete(&common->free);
+}
+
+static u64 qp_allowed_event_types(void)
+{
+	u64 mask;
+
+	mask = BIT(MLX5_EVENT_TYPE_PATH_MIG) |
+	       BIT(MLX5_EVENT_TYPE_COMM_EST) |
+	       BIT(MLX5_EVENT_TYPE_SQ_DRAINED) |
+	       BIT(MLX5_EVENT_TYPE_SRQ_LAST_WQE) |
+	       BIT(MLX5_EVENT_TYPE_WQ_CATAS_ERROR) |
+	       BIT(MLX5_EVENT_TYPE_PATH_MIG_FAILED) |
+	       BIT(MLX5_EVENT_TYPE_WQ_INVAL_REQ_ERROR) |
+	       BIT(MLX5_EVENT_TYPE_WQ_ACCESS_ERROR);
+
+	return mask;
+}
+
+static u64 rq_allowed_event_types(void)
+{
+	u64 mask;
+
+	mask = BIT(MLX5_EVENT_TYPE_SRQ_LAST_WQE) |
+	       BIT(MLX5_EVENT_TYPE_WQ_CATAS_ERROR);
+
+	return mask;
+}
+
+static u64 sq_allowed_event_types(void)
+{
+	return BIT(MLX5_EVENT_TYPE_WQ_CATAS_ERROR);
+}
+
+static u64 dct_allowed_event_types(void)
+{
+	return BIT(MLX5_EVENT_TYPE_DCT_DRAINED);
+}
+
+static bool is_event_type_allowed(int rsc_type, int event_type)
+{
+	switch (rsc_type) {
+	case MLX5_EVENT_QUEUE_TYPE_QP:
+		return BIT(event_type) & qp_allowed_event_types();
+	case MLX5_EVENT_QUEUE_TYPE_RQ:
+		return BIT(event_type) & rq_allowed_event_types();
+	case MLX5_EVENT_QUEUE_TYPE_SQ:
+		return BIT(event_type) & sq_allowed_event_types();
+	case MLX5_EVENT_QUEUE_TYPE_DCT:
+		return BIT(event_type) & dct_allowed_event_types();
+	default:
+		WARN(1, "Event arrived for unknown resource type");
+		return false;
+	}
+}
+
+static int rsc_event_notifier(struct notifier_block *nb,
+			      unsigned long type, void *data)
+{
+	struct mlx5_core_rsc_common *common;
+	struct mlx5_qp_table *table;
+	struct mlx5_core_dct *dct;
+	u8 event_type = (u8)type;
+	struct mlx5_core_qp *qp;
+	struct mlx5_eqe *eqe;
+	u32 rsn;
+
+	switch (event_type) {
+	case MLX5_EVENT_TYPE_DCT_DRAINED:
+		eqe = data;
+		rsn = be32_to_cpu(eqe->data.dct.dctn) & 0xffffff;
+		rsn |= (MLX5_RES_DCT << MLX5_USER_INDEX_LEN);
+		break;
+	case MLX5_EVENT_TYPE_PATH_MIG:
+	case MLX5_EVENT_TYPE_COMM_EST:
+	case MLX5_EVENT_TYPE_SQ_DRAINED:
+	case MLX5_EVENT_TYPE_SRQ_LAST_WQE:
+	case MLX5_EVENT_TYPE_WQ_CATAS_ERROR:
+	case MLX5_EVENT_TYPE_PATH_MIG_FAILED:
+	case MLX5_EVENT_TYPE_WQ_INVAL_REQ_ERROR:
+	case MLX5_EVENT_TYPE_WQ_ACCESS_ERROR:
+		eqe = data;
+		rsn = be32_to_cpu(eqe->data.qp_srq.qp_srq_n) & 0xffffff;
+		rsn |= (eqe->data.qp_srq.type << MLX5_USER_INDEX_LEN);
+		break;
+	default:
+		return NOTIFY_DONE;
+	}
+
+	table = container_of(nb, struct mlx5_qp_table, nb);
+	common = mlx5_get_rsc(table, rsn);
+	if (!common)
+		return NOTIFY_OK;
+
+	if (!is_event_type_allowed((rsn >> MLX5_USER_INDEX_LEN), event_type))
+		goto out;
+
+	switch (common->res) {
+	case MLX5_RES_QP:
+	case MLX5_RES_RQ:
+	case MLX5_RES_SQ:
+		qp = (struct mlx5_core_qp *)common;
+		qp->event(qp, event_type);
+		break;
+	case MLX5_RES_DCT:
+		dct = (struct mlx5_core_dct *)common;
+		if (event_type == MLX5_EVENT_TYPE_DCT_DRAINED)
+			complete(&dct->drained);
+		break;
+	default:
+		break;
+	}
+out:
+	mlx5_core_put_rsc(common);
+
+	return NOTIFY_OK;
+}
+
+static int create_resource_common(struct mlx5_ib_dev *dev,
+				  struct mlx5_core_qp *qp, int rsc_type)
+{
+	struct mlx5_qp_table *table = &dev->qp_table;
+	int err;
+
+	qp->common.res = rsc_type;
+	spin_lock_irq(&table->lock);
+	err = radix_tree_insert(&table->tree,
+				qp->qpn | (rsc_type << MLX5_USER_INDEX_LEN),
+				qp);
+	spin_unlock_irq(&table->lock);
+	if (err)
+		return err;
+
+	refcount_set(&qp->common.refcount, 1);
+	init_completion(&qp->common.free);
+	qp->pid = current->pid;
+
+	return 0;
+}
+
+static void destroy_resource_common(struct mlx5_ib_dev *dev,
+				    struct mlx5_core_qp *qp)
+{
+	struct mlx5_qp_table *table = &dev->qp_table;
+	unsigned long flags;
+
+	spin_lock_irqsave(&table->lock, flags);
+	radix_tree_delete(&table->tree,
+			  qp->qpn | (qp->common.res << MLX5_USER_INDEX_LEN));
+	spin_unlock_irqrestore(&table->lock, flags);
+	mlx5_core_put_rsc((struct mlx5_core_rsc_common *)qp);
+	wait_for_completion(&qp->common.free);
+}
+
+static int _mlx5_core_destroy_dct(struct mlx5_ib_dev *dev,
+				  struct mlx5_core_dct *dct, bool need_cleanup)
+{
+	u32 in[MLX5_ST_SZ_DW(destroy_dct_in)] = {};
+	struct mlx5_core_qp *qp = &dct->mqp;
+	int err;
+
+	err = mlx5_core_drain_dct(dev, dct);
+	if (err) {
+		if (dev->mdev->state == MLX5_DEVICE_STATE_INTERNAL_ERROR)
+			goto destroy;
+
+		return err;
+	}
+	wait_for_completion(&dct->drained);
+destroy:
+	if (need_cleanup)
+		destroy_resource_common(dev, &dct->mqp);
+	MLX5_SET(destroy_dct_in, in, opcode, MLX5_CMD_OP_DESTROY_DCT);
+	MLX5_SET(destroy_dct_in, in, dctn, qp->qpn);
+	MLX5_SET(destroy_dct_in, in, uid, qp->uid);
+	err = mlx5_cmd_exec_in(dev->mdev, destroy_dct, in);
+	return err;
+}
+
+int mlx5_core_create_dct(struct mlx5_ib_dev *dev, struct mlx5_core_dct *dct,
+			 u32 *in, int inlen, u32 *out, int outlen)
+{
+	struct mlx5_core_qp *qp = &dct->mqp;
+	int err;
+
+	init_completion(&dct->drained);
+	MLX5_SET(create_dct_in, in, opcode, MLX5_CMD_OP_CREATE_DCT);
+
+	err = mlx5_cmd_exec(dev->mdev, in, inlen, out, outlen);
+	if (err)
+		return err;
+
+	qp->qpn = MLX5_GET(create_dct_out, out, dctn);
+	qp->uid = MLX5_GET(create_dct_in, in, uid);
+	err = create_resource_common(dev, qp, MLX5_RES_DCT);
+	if (err)
+		goto err_cmd;
+
+	return 0;
+err_cmd:
+	_mlx5_core_destroy_dct(dev, dct, false);
+	return err;
+}
+
+int mlx5_core_create_qp(struct mlx5_ib_dev *dev, struct mlx5_core_qp *qp,
+			u32 *in, int inlen)
+{
+	u32 out[MLX5_ST_SZ_DW(create_qp_out)] = {};
+	u32 din[MLX5_ST_SZ_DW(destroy_qp_in)] = {};
+	int err;
+
+	MLX5_SET(create_qp_in, in, opcode, MLX5_CMD_OP_CREATE_QP);
+
+	err = mlx5_cmd_exec(dev->mdev, in, inlen, out, sizeof(out));
+	if (err)
+		return err;
+
+	qp->uid = MLX5_GET(create_qp_in, in, uid);
+	qp->qpn = MLX5_GET(create_qp_out, out, qpn);
+
+	err = create_resource_common(dev, qp, MLX5_RES_QP);
+	if (err)
+		goto err_cmd;
+
+	mlx5_debug_qp_add(dev->mdev, qp);
+
+	return 0;
+
+err_cmd:
+	MLX5_SET(destroy_qp_in, din, opcode, MLX5_CMD_OP_DESTROY_QP);
+	MLX5_SET(destroy_qp_in, din, qpn, qp->qpn);
+	MLX5_SET(destroy_qp_in, din, uid, qp->uid);
+	mlx5_cmd_exec_in(dev->mdev, destroy_qp, din);
+	return err;
+}
+
+static int mlx5_core_drain_dct(struct mlx5_ib_dev *dev,
+			       struct mlx5_core_dct *dct)
+{
+	u32 in[MLX5_ST_SZ_DW(drain_dct_in)] = {};
+	struct mlx5_core_qp *qp = &dct->mqp;
+
+	MLX5_SET(drain_dct_in, in, opcode, MLX5_CMD_OP_DRAIN_DCT);
+	MLX5_SET(drain_dct_in, in, dctn, qp->qpn);
+	MLX5_SET(drain_dct_in, in, uid, qp->uid);
+	return mlx5_cmd_exec_in(dev->mdev, drain_dct, in);
+}
+
+int mlx5_core_destroy_dct(struct mlx5_ib_dev *dev,
+			  struct mlx5_core_dct *dct)
+{
+	return _mlx5_core_destroy_dct(dev, dct, true);
+}
+
+int mlx5_core_destroy_qp(struct mlx5_ib_dev *dev, struct mlx5_core_qp *qp)
+{
+	u32 in[MLX5_ST_SZ_DW(destroy_qp_in)] = {};
+
+	mlx5_debug_qp_remove(dev->mdev, qp);
+
+	destroy_resource_common(dev, qp);
+
+	MLX5_SET(destroy_qp_in, in, opcode, MLX5_CMD_OP_DESTROY_QP);
+	MLX5_SET(destroy_qp_in, in, qpn, qp->qpn);
+	MLX5_SET(destroy_qp_in, in, uid, qp->uid);
+	mlx5_cmd_exec_in(dev->mdev, destroy_qp, in);
+	return 0;
+}
+
+int mlx5_core_set_delay_drop(struct mlx5_ib_dev *dev,
+			     u32 timeout_usec)
+{
+	u32 in[MLX5_ST_SZ_DW(set_delay_drop_params_in)] = {};
+
+	MLX5_SET(set_delay_drop_params_in, in, opcode,
+		 MLX5_CMD_OP_SET_DELAY_DROP_PARAMS);
+	MLX5_SET(set_delay_drop_params_in, in, delay_drop_timeout,
+		 timeout_usec / 100);
+	return mlx5_cmd_exec_in(dev->mdev, set_delay_drop_params, in);
+}
+
+struct mbox_info {
+	u32 *in;
+	u32 *out;
+	int inlen;
+	int outlen;
+};
+
+static int mbox_alloc(struct mbox_info *mbox, int inlen, int outlen)
+{
+	mbox->inlen  = inlen;
+	mbox->outlen = outlen;
+	mbox->in = kzalloc(mbox->inlen, GFP_KERNEL);
+	mbox->out = kzalloc(mbox->outlen, GFP_KERNEL);
+	if (!mbox->in || !mbox->out) {
+		kfree(mbox->in);
+		kfree(mbox->out);
+		return -ENOMEM;
+	}
+
+	return 0;
+}
+
+static void mbox_free(struct mbox_info *mbox)
+{
+	kfree(mbox->in);
+	kfree(mbox->out);
+}
+
+static int modify_qp_mbox_alloc(struct mlx5_core_dev *dev, u16 opcode, int qpn,
+				u32 opt_param_mask, void *qpc,
+				struct mbox_info *mbox, u16 uid)
+{
+	mbox->out = NULL;
+	mbox->in = NULL;
+
+#define MBOX_ALLOC(mbox, typ)  \
+	mbox_alloc(mbox, MLX5_ST_SZ_BYTES(typ##_in), MLX5_ST_SZ_BYTES(typ##_out))
+
+#define MOD_QP_IN_SET(typ, in, _opcode, _qpn, _uid)                            \
+	do {                                                                   \
+		MLX5_SET(typ##_in, in, opcode, _opcode);                       \
+		MLX5_SET(typ##_in, in, qpn, _qpn);                             \
+		MLX5_SET(typ##_in, in, uid, _uid);                             \
+	} while (0)
+
+#define MOD_QP_IN_SET_QPC(typ, in, _opcode, _qpn, _opt_p, _qpc, _uid)          \
+	do {                                                                   \
+		MOD_QP_IN_SET(typ, in, _opcode, _qpn, _uid);                   \
+		MLX5_SET(typ##_in, in, opt_param_mask, _opt_p);                \
+		memcpy(MLX5_ADDR_OF(typ##_in, in, qpc), _qpc,                  \
+		       MLX5_ST_SZ_BYTES(qpc));                                 \
+	} while (0)
+
+	switch (opcode) {
+	/* 2RST & 2ERR */
+	case MLX5_CMD_OP_2RST_QP:
+		if (MBOX_ALLOC(mbox, qp_2rst))
+			return -ENOMEM;
+		MOD_QP_IN_SET(qp_2rst, mbox->in, opcode, qpn, uid);
+		break;
+	case MLX5_CMD_OP_2ERR_QP:
+		if (MBOX_ALLOC(mbox, qp_2err))
+			return -ENOMEM;
+		MOD_QP_IN_SET(qp_2err, mbox->in, opcode, qpn, uid);
+		break;
+
+	/* MODIFY with QPC */
+	case MLX5_CMD_OP_RST2INIT_QP:
+		if (MBOX_ALLOC(mbox, rst2init_qp))
+			return -ENOMEM;
+		MOD_QP_IN_SET_QPC(rst2init_qp, mbox->in, opcode, qpn,
+				  opt_param_mask, qpc, uid);
+		break;
+	case MLX5_CMD_OP_INIT2RTR_QP:
+		if (MBOX_ALLOC(mbox, init2rtr_qp))
+			return -ENOMEM;
+		MOD_QP_IN_SET_QPC(init2rtr_qp, mbox->in, opcode, qpn,
+				  opt_param_mask, qpc, uid);
+		break;
+	case MLX5_CMD_OP_RTR2RTS_QP:
+		if (MBOX_ALLOC(mbox, rtr2rts_qp))
+			return -ENOMEM;
+		MOD_QP_IN_SET_QPC(rtr2rts_qp, mbox->in, opcode, qpn,
+				  opt_param_mask, qpc, uid);
+		break;
+	case MLX5_CMD_OP_RTS2RTS_QP:
+		if (MBOX_ALLOC(mbox, rts2rts_qp))
+			return -ENOMEM;
+		MOD_QP_IN_SET_QPC(rts2rts_qp, mbox->in, opcode, qpn,
+				  opt_param_mask, qpc, uid);
+		break;
+	case MLX5_CMD_OP_SQERR2RTS_QP:
+		if (MBOX_ALLOC(mbox, sqerr2rts_qp))
+			return -ENOMEM;
+		MOD_QP_IN_SET_QPC(sqerr2rts_qp, mbox->in, opcode, qpn,
+				  opt_param_mask, qpc, uid);
+		break;
+	case MLX5_CMD_OP_INIT2INIT_QP:
+		if (MBOX_ALLOC(mbox, init2init_qp))
+			return -ENOMEM;
+		MOD_QP_IN_SET_QPC(init2init_qp, mbox->in, opcode, qpn,
+				  opt_param_mask, qpc, uid);
+		break;
+	default:
+		return -EINVAL;
+	}
+	return 0;
+}
+
+int mlx5_core_qp_modify(struct mlx5_ib_dev *dev, u16 opcode, u32 opt_param_mask,
+			void *qpc, struct mlx5_core_qp *qp)
+{
+	struct mbox_info mbox;
+	int err;
+
+	err = modify_qp_mbox_alloc(dev->mdev, opcode, qp->qpn,
+				   opt_param_mask, qpc, &mbox, qp->uid);
+	if (err)
+		return err;
+
+	err = mlx5_cmd_exec(dev->mdev, mbox.in, mbox.inlen, mbox.out,
+			    mbox.outlen);
+	mbox_free(&mbox);
+	return err;
+}
+
+int mlx5_init_qp_table(struct mlx5_ib_dev *dev)
+{
+	struct mlx5_qp_table *table = &dev->qp_table;
+
+	spin_lock_init(&table->lock);
+	INIT_RADIX_TREE(&table->tree, GFP_ATOMIC);
+	mlx5_qp_debugfs_init(dev->mdev);
+
+	table->nb.notifier_call = rsc_event_notifier;
+	mlx5_notifier_register(dev->mdev, &table->nb);
+
+	return 0;
+}
+
+void mlx5_cleanup_qp_table(struct mlx5_ib_dev *dev)
+{
+	struct mlx5_qp_table *table = &dev->qp_table;
+
+	mlx5_notifier_unregister(dev->mdev, &table->nb);
+	mlx5_qp_debugfs_cleanup(dev->mdev);
+}
+
+int mlx5_core_qp_query(struct mlx5_ib_dev *dev, struct mlx5_core_qp *qp,
+		       u32 *out, int outlen)
+{
+	u32 in[MLX5_ST_SZ_DW(query_qp_in)] = {};
+
+	MLX5_SET(query_qp_in, in, opcode, MLX5_CMD_OP_QUERY_QP);
+	MLX5_SET(query_qp_in, in, qpn, qp->qpn);
+	return mlx5_cmd_exec(dev->mdev, in, sizeof(in), out, outlen);
+}
+
+int mlx5_core_dct_query(struct mlx5_ib_dev *dev, struct mlx5_core_dct *dct,
+			u32 *out, int outlen)
+{
+	u32 in[MLX5_ST_SZ_DW(query_dct_in)] = {};
+	struct mlx5_core_qp *qp = &dct->mqp;
+
+	MLX5_SET(query_dct_in, in, opcode, MLX5_CMD_OP_QUERY_DCT);
+	MLX5_SET(query_dct_in, in, dctn, qp->qpn);
+
+	return mlx5_cmd_exec(dev->mdev, (void *)&in, sizeof(in), (void *)out,
+			     outlen);
+}
+
+int mlx5_core_xrcd_alloc(struct mlx5_ib_dev *dev, u32 *xrcdn)
+{
+	u32 out[MLX5_ST_SZ_DW(alloc_xrcd_out)] = {};
+	u32 in[MLX5_ST_SZ_DW(alloc_xrcd_in)] = {};
+	int err;
+
+	MLX5_SET(alloc_xrcd_in, in, opcode, MLX5_CMD_OP_ALLOC_XRCD);
+	err = mlx5_cmd_exec_inout(dev->mdev, alloc_xrcd, in, out);
+	if (!err)
+		*xrcdn = MLX5_GET(alloc_xrcd_out, out, xrcd);
+	return err;
+}
+
+int mlx5_core_xrcd_dealloc(struct mlx5_ib_dev *dev, u32 xrcdn)
+{
+	u32 in[MLX5_ST_SZ_DW(dealloc_xrcd_in)] = {};
+
+	MLX5_SET(dealloc_xrcd_in, in, opcode, MLX5_CMD_OP_DEALLOC_XRCD);
+	MLX5_SET(dealloc_xrcd_in, in, xrcd, xrcdn);
+	return mlx5_cmd_exec_in(dev->mdev, dealloc_xrcd, in);
+}
+
+static void destroy_rq_tracked(struct mlx5_ib_dev *dev, u32 rqn, u16 uid)
+{
+	u32 in[MLX5_ST_SZ_DW(destroy_rq_in)] = {};
+
+	MLX5_SET(destroy_rq_in, in, opcode, MLX5_CMD_OP_DESTROY_RQ);
+	MLX5_SET(destroy_rq_in, in, rqn, rqn);
+	MLX5_SET(destroy_rq_in, in, uid, uid);
+	mlx5_cmd_exec_in(dev->mdev, destroy_rq, in);
+}
+
+int mlx5_core_create_rq_tracked(struct mlx5_ib_dev *dev, u32 *in, int inlen,
+				struct mlx5_core_qp *rq)
+{
+	int err;
+	u32 rqn;
+
+	err = mlx5_core_create_rq(dev->mdev, in, inlen, &rqn);
+	if (err)
+		return err;
+
+	rq->uid = MLX5_GET(create_rq_in, in, uid);
+	rq->qpn = rqn;
+	err = create_resource_common(dev, rq, MLX5_RES_RQ);
+	if (err)
+		goto err_destroy_rq;
+
+	return 0;
+
+err_destroy_rq:
+	destroy_rq_tracked(dev, rq->qpn, rq->uid);
+
+	return err;
+}
+
+void mlx5_core_destroy_rq_tracked(struct mlx5_ib_dev *dev,
+				  struct mlx5_core_qp *rq)
+{
+	destroy_resource_common(dev, rq);
+	destroy_rq_tracked(dev, rq->qpn, rq->uid);
+}
+
+static void destroy_sq_tracked(struct mlx5_ib_dev *dev, u32 sqn, u16 uid)
+{
+	u32 in[MLX5_ST_SZ_DW(destroy_sq_in)] = {};
+
+	MLX5_SET(destroy_sq_in, in, opcode, MLX5_CMD_OP_DESTROY_SQ);
+	MLX5_SET(destroy_sq_in, in, sqn, sqn);
+	MLX5_SET(destroy_sq_in, in, uid, uid);
+	mlx5_cmd_exec_in(dev->mdev, destroy_sq, in);
+}
+
+int mlx5_core_create_sq_tracked(struct mlx5_ib_dev *dev, u32 *in, int inlen,
+				struct mlx5_core_qp *sq)
+{
+	u32 out[MLX5_ST_SZ_DW(create_sq_out)] = {};
+	int err;
+
+	MLX5_SET(create_sq_in, in, opcode, MLX5_CMD_OP_CREATE_SQ);
+	err = mlx5_cmd_exec(dev->mdev, in, inlen, out, sizeof(out));
+	if (err)
+		return err;
+
+	sq->qpn = MLX5_GET(create_sq_out, out, sqn);
+	sq->uid = MLX5_GET(create_sq_in, in, uid);
+	err = create_resource_common(dev, sq, MLX5_RES_SQ);
+	if (err)
+		goto err_destroy_sq;
+
+	return 0;
+
+err_destroy_sq:
+	destroy_sq_tracked(dev, sq->qpn, sq->uid);
+
+	return err;
+}
+
+void mlx5_core_destroy_sq_tracked(struct mlx5_ib_dev *dev,
+				  struct mlx5_core_qp *sq)
+{
+	destroy_resource_common(dev, sq);
+	destroy_sq_tracked(dev, sq->qpn, sq->uid);
+}
+
+struct mlx5_core_rsc_common *mlx5_core_res_hold(struct mlx5_ib_dev *dev,
+						int res_num,
+						enum mlx5_res_type res_type)
+{
+	u32 rsn = res_num | (res_type << MLX5_USER_INDEX_LEN);
+	struct mlx5_qp_table *table = &dev->qp_table;
+
+	return mlx5_get_rsc(table, rsn);
+}
+
+void mlx5_core_res_put(struct mlx5_core_rsc_common *res)
+{
+	mlx5_core_put_rsc(res);
+}
diff --git a/drivers/infiniband/hw/mlx5/srq_cmd.c b/drivers/infiniband/hw/mlx5/srq_cmd.c
index 8fc3630a9d4c..c851570791af 100644
--- a/drivers/infiniband/hw/mlx5/srq_cmd.c
+++ b/drivers/infiniband/hw/mlx5/srq_cmd.c
@@ -5,9 +5,9 @@
 
 #include <linux/kernel.h>
 #include <linux/mlx5/driver.h>
-#include <linux/mlx5/cmd.h>
 #include "mlx5_ib.h"
 #include "srq.h"
+#include "qp.h"
 
 static int get_pas_size(struct mlx5_srq_attr *in)
 {
diff --git a/drivers/infiniband/sw/siw/siw_cm.c b/drivers/infiniband/sw/siw/siw_cm.c
index 559e5fd3bad8..1662216be66d 100644
--- a/drivers/infiniband/sw/siw/siw_cm.c
+++ b/drivers/infiniband/sw/siw/siw_cm.c
@@ -947,16 +947,8 @@ static void siw_accept_newconn(struct siw_cep *cep)
 	siw_cep_get(new_cep);
 	new_s->sk->sk_user_data = new_cep;
 
-	if (siw_tcp_nagle == false) {
-		int val = 1;
-
-		rv = kernel_setsockopt(new_s, SOL_TCP, TCP_NODELAY,
-				       (char *)&val, sizeof(val));
-		if (rv) {
-			siw_dbg_cep(cep, "setsockopt NODELAY error: %d\n", rv);
-			goto error;
-		}
-	}
+	if (siw_tcp_nagle == false)
+		tcp_sock_set_nodelay(new_s->sk);
 	new_cep->state = SIW_EPSTATE_AWAIT_MPAREQ;
 
 	rv = siw_cm_queue_work(new_cep, SIW_CM_WORK_MPATIMEOUT);
@@ -1312,17 +1304,14 @@ static void siw_cm_llp_state_change(struct sock *sk)
 static int kernel_bindconnect(struct socket *s, struct sockaddr *laddr,
 			      struct sockaddr *raddr)
 {
-	int rv, flags = 0, s_val = 1;
+	int rv, flags = 0;
 	size_t size = laddr->sa_family == AF_INET ?
 		sizeof(struct sockaddr_in) : sizeof(struct sockaddr_in6);
 
 	/*
 	 * Make address available again asap.
 	 */
-	rv = kernel_setsockopt(s, SOL_SOCKET, SO_REUSEADDR, (char *)&s_val,
-			       sizeof(s_val));
-	if (rv < 0)
-		return rv;
+	sock_set_reuseaddr(s->sk);
 
 	rv = s->ops->bind(s, laddr, size);
 	if (rv < 0)
@@ -1389,16 +1378,8 @@ int siw_connect(struct iw_cm_id *id, struct iw_cm_conn_param *params)
 		siw_dbg_qp(qp, "kernel_bindconnect: error %d\n", rv);
 		goto error;
 	}
-	if (siw_tcp_nagle == false) {
-		int val = 1;
-
-		rv = kernel_setsockopt(s, SOL_TCP, TCP_NODELAY, (char *)&val,
-				       sizeof(val));
-		if (rv) {
-			siw_dbg_qp(qp, "setsockopt NODELAY error: %d\n", rv);
-			goto error;
-		}
-	}
+	if (siw_tcp_nagle == false)
+		tcp_sock_set_nodelay(s->sk);
 	cep = siw_cep_alloc(sdev);
 	if (!cep) {
 		rv = -ENOMEM;
@@ -1781,7 +1762,7 @@ int siw_create_listen(struct iw_cm_id *id, int backlog)
 	struct siw_cep *cep = NULL;
 	struct siw_device *sdev = to_siw_dev(id->device);
 	int addr_family = id->local_addr.ss_family;
-	int rv = 0, s_val;
+	int rv = 0;
 
 	if (addr_family != AF_INET && addr_family != AF_INET6)
 		return -EAFNOSUPPORT;
@@ -1793,13 +1774,8 @@ int siw_create_listen(struct iw_cm_id *id, int backlog)
 	/*
 	 * Allow binding local port when still in TIME_WAIT from last close.
 	 */
-	s_val = 1;
-	rv = kernel_setsockopt(s, SOL_SOCKET, SO_REUSEADDR, (char *)&s_val,
-			       sizeof(s_val));
-	if (rv) {
-		siw_dbg(id->device, "setsockopt error: %d\n", rv);
-		goto error;
-	}
+	sock_set_reuseaddr(s->sk);
+
 	if (addr_family == AF_INET) {
 		struct sockaddr_in *laddr = &to_sockaddr_in(id->local_addr);