Merge tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma

Pull RDMA subsystem updates from Jason Gunthorpe:
 "Overall this cycle did not have any major excitement, and did not
  require any shared branch with netdev.

  Lots of driver updates, particularly of the scale-up and performance
  variety. The largest body of core work was Parav's patches fixing and
  restructing some of the core code to make way for future RDMA
  containerization.

  Summary:

   - misc small driver fixups to
     bnxt_re/hfi1/qib/hns/ocrdma/rdmavt/vmw_pvrdma/nes

   - several major feature adds to bnxt_re driver: SRIOV VF RoCE
     support, HugePages support, extended hardware stats support, and
     SRQ support

   - a notable number of fixes to the i40iw driver from debugging scale
     up testing

   - more work to enable the new hip08 chip in the hns driver

   - misc small ULP fixups to srp/srpt//ipoib

   - preparation for srp initiator and target to support the RDMA-CM
     protocol for connections

   - add RDMA-CM support to srp initiator, srp target is still a WIP

   - fixes for a couple of places where ipoib could spam the dmesg log

   - fix encode/decode of FDR/EDR data rates in the core

   - many patches from Parav with ongoing work to clean up
     inconsistencies and bugs in RoCE support around the rdma_cm

   - mlx5 driver support for the userspace features 'thread domain',
     'wallclock timestamps' and 'DV Direct Connected transport'. Support
     for the firmware dual port rocee capability

   - core support for more than 32 rdma devices in the char dev
     allocation

   - kernel doc updates from Randy Dunlap

   - new netlink uAPI for inspecting RDMA objects similar in spirit to 'ss'

   - one minor change to the kobject code acked by Greg KH"

* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma: (259 commits)
  RDMA/nldev: Provide detailed QP information
  RDMA/nldev: Provide global resource utilization
  RDMA/core: Add resource tracking for create and destroy PDs
  RDMA/core: Add resource tracking for create and destroy CQs
  RDMA/core: Add resource tracking for create and destroy QPs
  RDMA/restrack: Add general infrastructure to track RDMA resources
  RDMA/core: Save kernel caller name when creating PD and CQ objects
  RDMA/core: Use the MODNAME instead of the function name for pd callers
  RDMA: Move enum ib_cq_creation_flags to uapi headers
  IB/rxe: Change RDMA_RXE kconfig to use select
  IB/qib: remove qib_keys.c
  IB/mthca: remove mthca_user.h
  RDMA/cm: Fix access to uninitialized variable
  RDMA/cma: Use existing netif_is_bond_master function
  IB/core: Avoid SGID attributes query while converting GID from OPA to IB
  RDMA/mlx5: Avoid memory leak in case of XRCD dealloc failure
  IB/umad: Fix use of unprotected device pointer
  IB/iser: Combine substrings for three messages
  IB/iser: Delete an unnecessary variable initialisation in iser_send_data_out()
  IB/iser: Delete an error message for a failed memory allocation in iser_send_data_out()
  ...

11 files changed, 299 insertions, 29 deletions

diff --git a/drivers/net/ethernet/mellanox/mlx4/qp.c b/drivers/net/ethernet/mellanox/mlx4/qp.c
index 769598f7b6c8..3aaf4bad6c5a 100644
--- a/drivers/net/ethernet/mellanox/mlx4/qp.c
+++ b/drivers/net/ethernet/mellanox/mlx4/qp.c
@@ -287,6 +287,9 @@ void mlx4_qp_release_range(struct mlx4_dev *dev, int base_qpn, int cnt)
 	u64 in_param = 0;
 	int err;
 
+	if (!cnt)
+		return;
+
 	if (mlx4_is_mfunc(dev)) {
 		set_param_l(&in_param, base_qpn);
 		set_param_h(&in_param, cnt);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eq.c b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
index e7e7cef2bde4..14d57828945d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
@@ -417,7 +417,11 @@ static irqreturn_t mlx5_eq_int(int irq, void *eq_ptr)
 			cqn = be32_to_cpu(eqe->data.comp.cqn) & 0xffffff;
 			mlx5_cq_completion(dev, cqn);
 			break;
-
+		case MLX5_EVENT_TYPE_DCT_DRAINED:
+			rsn = be32_to_cpu(eqe->data.dct.dctn) & 0xffffff;
+			rsn |= (MLX5_RES_DCT << MLX5_USER_INDEX_LEN);
+			mlx5_rsc_event(dev, rsn, eqe->type);
+			break;
 		case MLX5_EVENT_TYPE_PATH_MIG:
 		case MLX5_EVENT_TYPE_COMM_EST:
 		case MLX5_EVENT_TYPE_SQ_DRAINED:
@@ -715,6 +719,9 @@ int mlx5_start_eqs(struct mlx5_core_dev *dev)
 
 	if (MLX5_CAP_GEN(dev, fpga))
 		async_event_mask |= (1ull << MLX5_EVENT_TYPE_FPGA_ERROR);
+	if (MLX5_CAP_GEN_MAX(dev, dct))
+		async_event_mask |= (1ull << MLX5_EVENT_TYPE_DCT_DRAINED);
+
 
 	err = mlx5_create_map_eq(dev, &table->cmd_eq, MLX5_EQ_VEC_CMD,
 				 MLX5_NUM_CMD_EQE, 1ull << MLX5_EVENT_TYPE_CMD,
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fpga/conn.c b/drivers/net/ethernet/mellanox/mlx5/core/fpga/conn.c
index c4392f741c5f..e6175f8ac0e4 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fpga/conn.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fpga/conn.c
@@ -688,7 +688,7 @@ static inline int mlx5_fpga_conn_init_qp(struct mlx5_fpga_conn *conn)
 	MLX5_SET(qpc, qpc, st, MLX5_QP_ST_RC);
 	MLX5_SET(qpc, qpc, pm_state, MLX5_QP_PM_MIGRATED);
 	MLX5_SET(qpc, qpc, primary_address_path.pkey_index, MLX5_FPGA_PKEY_INDEX);
-	MLX5_SET(qpc, qpc, primary_address_path.port, MLX5_FPGA_PORT_NUM);
+	MLX5_SET(qpc, qpc, primary_address_path.vhca_port_num, MLX5_FPGA_PORT_NUM);
 	MLX5_SET(qpc, qpc, pd, conn->fdev->conn_res.pdn);
 	MLX5_SET(qpc, qpc, cqn_snd, conn->cq.mcq.cqn);
 	MLX5_SET(qpc, qpc, cqn_rcv, conn->cq.mcq.cqn);
@@ -727,7 +727,7 @@ static inline int mlx5_fpga_conn_rtr_qp(struct mlx5_fpga_conn *conn)
 	MLX5_SET(qpc, qpc, next_rcv_psn,
 		 MLX5_GET(fpga_qpc, conn->fpga_qpc, next_send_psn));
 	MLX5_SET(qpc, qpc, primary_address_path.pkey_index, MLX5_FPGA_PKEY_INDEX);
-	MLX5_SET(qpc, qpc, primary_address_path.port, MLX5_FPGA_PORT_NUM);
+	MLX5_SET(qpc, qpc, primary_address_path.vhca_port_num, MLX5_FPGA_PORT_NUM);
 	ether_addr_copy(MLX5_ADDR_OF(qpc, qpc, primary_address_path.rmac_47_32),
 			MLX5_ADDR_OF(fpga_qpc, conn->fpga_qpc, fpga_mac_47_32));
 	MLX5_SET(qpc, qpc, primary_address_path.udp_sport,
@@ -888,7 +888,8 @@ struct mlx5_fpga_conn *mlx5_fpga_conn_create(struct mlx5_fpga_device *fdev,
 	err = mlx5_core_roce_gid_set(fdev->mdev, conn->qp.sgid_index,
 				     MLX5_ROCE_VERSION_2,
 				     MLX5_ROCE_L3_TYPE_IPV6,
-				     remote_ip, remote_mac, true, 0);
+				     remote_ip, remote_mac, true, 0,
+				     MLX5_FPGA_PORT_NUM);
 	if (err) {
 		mlx5_fpga_err(fdev, "Failed to set SGID: %d\n", err);
 		ret = ERR_PTR(err);
@@ -954,7 +955,7 @@ err_cq:
 	mlx5_fpga_conn_destroy_cq(conn);
 err_gid:
 	mlx5_core_roce_gid_set(fdev->mdev, conn->qp.sgid_index, 0, 0, NULL,
-			       NULL, false, 0);
+			       NULL, false, 0, MLX5_FPGA_PORT_NUM);
 err_rsvd_gid:
 	mlx5_core_reserved_gid_free(fdev->mdev, conn->qp.sgid_index);
 err:
@@ -982,7 +983,7 @@ void mlx5_fpga_conn_destroy(struct mlx5_fpga_conn *conn)
 	mlx5_fpga_conn_destroy_cq(conn);
 
 	mlx5_core_roce_gid_set(conn->fdev->mdev, conn->qp.sgid_index, 0, 0,
-			       NULL, NULL, false, 0);
+			       NULL, NULL, false, 0, MLX5_FPGA_PORT_NUM);
 	mlx5_core_reserved_gid_free(conn->fdev->mdev, conn->qp.sgid_index);
 	kfree(conn);
 }
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fw.c b/drivers/net/ethernet/mellanox/mlx5/core/fw.c
index 5ef1b56b6a96..9d11e92fb541 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fw.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fw.c
@@ -195,12 +195,20 @@ int mlx5_query_hca_caps(struct mlx5_core_dev *dev)
 	return 0;
 }
 
-int mlx5_cmd_init_hca(struct mlx5_core_dev *dev)
+int mlx5_cmd_init_hca(struct mlx5_core_dev *dev, uint32_t *sw_owner_id)
 {
 	u32 out[MLX5_ST_SZ_DW(init_hca_out)] = {0};
 	u32 in[MLX5_ST_SZ_DW(init_hca_in)]   = {0};
+	int i;
 
 	MLX5_SET(init_hca_in, in, opcode, MLX5_CMD_OP_INIT_HCA);
+
+	if (MLX5_CAP_GEN(dev, sw_owner_id)) {
+		for (i = 0; i < 4; i++)
+			MLX5_ARRAY_SET(init_hca_in, in, sw_owner_id, i,
+				       sw_owner_id[i]);
+	}
+
 	return mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c
index ee2f378c5030..a281d95ce17c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c
@@ -189,7 +189,7 @@ int mlx5i_create_underlay_qp(struct mlx5_core_dev *mdev, struct mlx5_core_qp *qp
 		 MLX5_QP_ENHANCED_ULP_STATELESS_MODE);
 
 	addr_path = MLX5_ADDR_OF(qpc, qpc, primary_address_path);
-	MLX5_SET(ads, addr_path, port, 1);
+	MLX5_SET(ads, addr_path, vhca_port_num, 1);
 	MLX5_SET(ads, addr_path, grh, 1);
 
 	ret = mlx5_core_create_qp(mdev, qp, in, inlen);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c b/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c
index 5701f125e99c..e159243e0fcf 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c
@@ -31,6 +31,8 @@
  */
 
 #include <linux/clocksource.h>
+#include <linux/highmem.h>
+#include <rdma/mlx5-abi.h>
 #include "en.h"
 
 enum {
@@ -71,6 +73,28 @@ static u64 read_internal_timer(const struct cyclecounter *cc)
 	return mlx5_read_internal_timer(mdev) & cc->mask;
 }
 
+static void mlx5_update_clock_info_page(struct mlx5_core_dev *mdev)
+{
+	struct mlx5_ib_clock_info *clock_info = mdev->clock_info;
+	struct mlx5_clock *clock = &mdev->clock;
+	u32 sign;
+
+	if (!clock_info)
+		return;
+
+	sign = smp_load_acquire(&clock_info->sign);
+	smp_store_mb(clock_info->sign,
+		     sign | MLX5_IB_CLOCK_INFO_KERNEL_UPDATING);
+
+	clock_info->cycles = clock->tc.cycle_last;
+	clock_info->mult   = clock->cycles.mult;
+	clock_info->nsec   = clock->tc.nsec;
+	clock_info->frac   = clock->tc.frac;
+
+	smp_store_release(&clock_info->sign,
+			  sign + MLX5_IB_CLOCK_INFO_KERNEL_UPDATING * 2);
+}
+
 static void mlx5_pps_out(struct work_struct *work)
 {
 	struct mlx5_pps *pps_info = container_of(work, struct mlx5_pps,
@@ -109,6 +133,7 @@ static void mlx5_timestamp_overflow(struct work_struct *work)
 
 	write_lock_irqsave(&clock->lock, flags);
 	timecounter_read(&clock->tc);
+	mlx5_update_clock_info_page(clock->mdev);
 	write_unlock_irqrestore(&clock->lock, flags);
 	schedule_delayed_work(&clock->overflow_work, clock->overflow_period);
 }
@@ -123,6 +148,7 @@ static int mlx5_ptp_settime(struct ptp_clock_info *ptp,
 
 	write_lock_irqsave(&clock->lock, flags);
 	timecounter_init(&clock->tc, &clock->cycles, ns);
+	mlx5_update_clock_info_page(clock->mdev);
 	write_unlock_irqrestore(&clock->lock, flags);
 
 	return 0;
@@ -152,6 +178,7 @@ static int mlx5_ptp_adjtime(struct ptp_clock_info *ptp, s64 delta)
 
 	write_lock_irqsave(&clock->lock, flags);
 	timecounter_adjtime(&clock->tc, delta);
+	mlx5_update_clock_info_page(clock->mdev);
 	write_unlock_irqrestore(&clock->lock, flags);
 
 	return 0;
@@ -179,6 +206,7 @@ static int mlx5_ptp_adjfreq(struct ptp_clock_info *ptp, s32 delta)
 	timecounter_read(&clock->tc);
 	clock->cycles.mult = neg_adj ? clock->nominal_c_mult - diff :
 				       clock->nominal_c_mult + diff;
+	mlx5_update_clock_info_page(clock->mdev);
 	write_unlock_irqrestore(&clock->lock, flags);
 
 	return 0;
@@ -474,6 +502,7 @@ void mlx5_init_clock(struct mlx5_core_dev *mdev)
 						  clock->cycles.shift);
 	clock->nominal_c_mult = clock->cycles.mult;
 	clock->cycles.mask = CLOCKSOURCE_MASK(41);
+	clock->mdev = mdev;
 
 	timecounter_init(&clock->tc, &clock->cycles,
 			 ktime_to_ns(ktime_get_real()));
@@ -486,6 +515,25 @@ void mlx5_init_clock(struct mlx5_core_dev *mdev)
 	do_div(ns, NSEC_PER_SEC / 2 / HZ);
 	clock->overflow_period = ns;
 
+	mdev->clock_info_page = alloc_page(GFP_KERNEL);
+	if (mdev->clock_info_page) {
+		mdev->clock_info = kmap(mdev->clock_info_page);
+		if (!mdev->clock_info) {
+			__free_page(mdev->clock_info_page);
+			mlx5_core_warn(mdev, "failed to map clock page\n");
+		} else {
+			mdev->clock_info->sign   = 0;
+			mdev->clock_info->nsec   = clock->tc.nsec;
+			mdev->clock_info->cycles = clock->tc.cycle_last;
+			mdev->clock_info->mask   = clock->cycles.mask;
+			mdev->clock_info->mult   = clock->nominal_c_mult;
+			mdev->clock_info->shift  = clock->cycles.shift;
+			mdev->clock_info->frac   = clock->tc.frac;
+			mdev->clock_info->overflow_period =
+						clock->overflow_period;
+		}
+	}
+
 	INIT_WORK(&clock->pps_info.out_work, mlx5_pps_out);
 	INIT_DELAYED_WORK(&clock->overflow_work, mlx5_timestamp_overflow);
 	if (clock->overflow_period)
@@ -525,5 +573,12 @@ void mlx5_cleanup_clock(struct mlx5_core_dev *mdev)
 
 	cancel_work_sync(&clock->pps_info.out_work);
 	cancel_delayed_work_sync(&clock->overflow_work);
+
+	if (mdev->clock_info) {
+		kunmap(mdev->clock_info_page);
+		__free_page(mdev->clock_info_page);
+		mdev->clock_info = NULL;
+	}
+
 	kfree(clock->ptp_info.pin_config);
 }
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/gid.c b/drivers/net/ethernet/mellanox/mlx5/core/lib/gid.c
index 573f59f46d41..7722a3f9bb68 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lib/gid.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/gid.c
@@ -121,7 +121,7 @@ EXPORT_SYMBOL_GPL(mlx5_core_reserved_gids_count);
 
 int mlx5_core_roce_gid_set(struct mlx5_core_dev *dev, unsigned int index,
 			   u8 roce_version, u8 roce_l3_type, const u8 *gid,
-			   const u8 *mac, bool vlan, u16 vlan_id)
+			   const u8 *mac, bool vlan, u16 vlan_id, u8 port_num)
 {
 #define MLX5_SET_RA(p, f, v) MLX5_SET(roce_addr_layout, p, f, v)
 	u32  in[MLX5_ST_SZ_DW(set_roce_address_in)] = {0};
@@ -148,6 +148,9 @@ int mlx5_core_roce_gid_set(struct mlx5_core_dev *dev, unsigned int index,
 		memcpy(addr_l3_addr, gid, gidsz);
 	}
 
+	if (MLX5_CAP_GEN(dev, num_vhca_ports) > 0)
+		MLX5_SET(set_roce_address_in, in, vhca_port_num, port_num);
+
 	MLX5_SET(set_roce_address_in, in, roce_address_index, index);
 	MLX5_SET(set_roce_address_in, in, opcode, MLX5_CMD_OP_SET_ROCE_ADDRESS);
 	return mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c
index 0f88fd30a09a..2ef641c91c26 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c
@@ -75,6 +75,8 @@ static unsigned int prof_sel = MLX5_DEFAULT_PROF;
 module_param_named(prof_sel, prof_sel, uint, 0444);
 MODULE_PARM_DESC(prof_sel, "profile selector. Valid range 0 - 2");
 
+static u32 sw_owner_id[4];
+
 enum {
 	MLX5_ATOMIC_REQ_MODE_BE = 0x0,
 	MLX5_ATOMIC_REQ_MODE_HOST_ENDIANNESS = 0x1,
@@ -551,6 +553,15 @@ static int handle_hca_cap(struct mlx5_core_dev *dev)
 			 cache_line_128byte,
 			 cache_line_size() == 128 ? 1 : 0);
 
+	if (MLX5_CAP_GEN_MAX(dev, dct))
+		MLX5_SET(cmd_hca_cap, set_hca_cap, dct, 1);
+
+	if (MLX5_CAP_GEN_MAX(dev, num_vhca_ports))
+		MLX5_SET(cmd_hca_cap,
+			 set_hca_cap,
+			 num_vhca_ports,
+			 MLX5_CAP_GEN_MAX(dev, num_vhca_ports));
+
 	err = set_caps(dev, set_ctx, set_sz,
 		       MLX5_SET_HCA_CAP_OP_MOD_GENERAL_DEVICE);
 
@@ -1107,7 +1118,7 @@ static int mlx5_load_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv,
 		goto reclaim_boot_pages;
 	}
 
-	err = mlx5_cmd_init_hca(dev);
+	err = mlx5_cmd_init_hca(dev, sw_owner_id);
 	if (err) {
 		dev_err(&pdev->dev, "init hca failed\n");
 		goto err_pagealloc_stop;
@@ -1643,6 +1654,8 @@ static int __init init(void)
 {
 	int err;
 
+	get_random_bytes(&sw_owner_id, sizeof(sw_owner_id));
+
 	mlx5_core_verify_params();
 	mlx5_register_debugfs();
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
index ff4a0b889a6f..b05868728da7 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
@@ -86,7 +86,7 @@ enum {
 
 int mlx5_query_hca_caps(struct mlx5_core_dev *dev);
 int mlx5_query_board_id(struct mlx5_core_dev *dev);
-int mlx5_cmd_init_hca(struct mlx5_core_dev *dev);
+int mlx5_cmd_init_hca(struct mlx5_core_dev *dev, uint32_t *sw_owner_id);
 int mlx5_cmd_teardown_hca(struct mlx5_core_dev *dev);
 int mlx5_cmd_force_teardown_hca(struct mlx5_core_dev *dev);
 void mlx5_core_event(struct mlx5_core_dev *dev, enum mlx5_dev_event event,
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/qp.c b/drivers/net/ethernet/mellanox/mlx5/core/qp.c
index 889130edb715..02d6c5b5d502 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/qp.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/qp.c
@@ -98,6 +98,11 @@ 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) {
@@ -107,6 +112,8 @@ static bool is_event_type_allowed(int rsc_type, int event_type)
 		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;
@@ -116,6 +123,7 @@ static bool is_event_type_allowed(int rsc_type, int event_type)
 void mlx5_rsc_event(struct mlx5_core_dev *dev, u32 rsn, int event_type)
 {
 	struct mlx5_core_rsc_common *common = mlx5_get_rsc(dev, rsn);
+	struct mlx5_core_dct *dct;
 	struct mlx5_core_qp *qp;
 
 	if (!common)
@@ -134,7 +142,11 @@ void mlx5_rsc_event(struct mlx5_core_dev *dev, u32 rsn, int event_type)
 		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:
 		mlx5_core_warn(dev, "invalid resource type for 0x%x\n", rsn);
 	}
@@ -142,9 +154,9 @@ void mlx5_rsc_event(struct mlx5_core_dev *dev, u32 rsn, int event_type)
 	mlx5_core_put_rsc(common);
 }
 
-static int create_qprqsq_common(struct mlx5_core_dev *dev,
-				struct mlx5_core_qp *qp,
-				int rsc_type)
+static int create_resource_common(struct mlx5_core_dev *dev,
+				  struct mlx5_core_qp *qp,
+				  int rsc_type)
 {
 	struct mlx5_qp_table *table = &dev->priv.qp_table;
 	int err;
@@ -165,8 +177,8 @@ static int create_qprqsq_common(struct mlx5_core_dev *dev,
 	return 0;
 }
 
-static void destroy_qprqsq_common(struct mlx5_core_dev *dev,
-				  struct mlx5_core_qp *qp)
+static void destroy_resource_common(struct mlx5_core_dev *dev,
+				    struct mlx5_core_qp *qp)
 {
 	struct mlx5_qp_table *table = &dev->priv.qp_table;
 	unsigned long flags;
@@ -179,6 +191,40 @@ static void destroy_qprqsq_common(struct mlx5_core_dev *dev,
 	wait_for_completion(&qp->common.free);
 }
 
+int mlx5_core_create_dct(struct mlx5_core_dev *dev,
+			 struct mlx5_core_dct *dct,
+			 u32 *in, int inlen)
+{
+	u32 out[MLX5_ST_SZ_DW(create_dct_out)]   = {0};
+	u32 din[MLX5_ST_SZ_DW(destroy_dct_in)]   = {0};
+	u32 dout[MLX5_ST_SZ_DW(destroy_dct_out)] = {0};
+	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, in, inlen, &out, sizeof(out));
+	if (err) {
+		mlx5_core_warn(dev, "create DCT failed, ret %d\n", err);
+		return err;
+	}
+
+	qp->qpn = MLX5_GET(create_dct_out, out, dctn);
+	err = create_resource_common(dev, qp, MLX5_RES_DCT);
+	if (err)
+		goto err_cmd;
+
+	return 0;
+err_cmd:
+	MLX5_SET(destroy_dct_in, din, opcode, MLX5_CMD_OP_DESTROY_DCT);
+	MLX5_SET(destroy_dct_in, din, dctn, qp->qpn);
+	mlx5_cmd_exec(dev, (void *)&in, sizeof(din),
+		      (void *)&out, sizeof(dout));
+	return err;
+}
+EXPORT_SYMBOL_GPL(mlx5_core_create_dct);
+
 int mlx5_core_create_qp(struct mlx5_core_dev *dev,
 			struct mlx5_core_qp *qp,
 			u32 *in, int inlen)
@@ -197,7 +243,7 @@ int mlx5_core_create_qp(struct mlx5_core_dev *dev,
 	qp->qpn = MLX5_GET(create_qp_out, out, qpn);
 	mlx5_core_dbg(dev, "qpn = 0x%x\n", qp->qpn);
 
-	err = create_qprqsq_common(dev, qp, MLX5_RES_QP);
+	err = create_resource_common(dev, qp, MLX5_RES_QP);
 	if (err)
 		goto err_cmd;
 
@@ -220,6 +266,47 @@ err_cmd:
 }
 EXPORT_SYMBOL_GPL(mlx5_core_create_qp);
 
+static int mlx5_core_drain_dct(struct mlx5_core_dev *dev,
+			       struct mlx5_core_dct *dct)
+{
+	u32 out[MLX5_ST_SZ_DW(drain_dct_out)] = {0};
+	u32 in[MLX5_ST_SZ_DW(drain_dct_in)]   = {0};
+	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);
+	return mlx5_cmd_exec(dev, (void *)&in, sizeof(in),
+			     (void *)&out, sizeof(out));
+}
+
+int mlx5_core_destroy_dct(struct mlx5_core_dev *dev,
+			  struct mlx5_core_dct *dct)
+{
+	u32 out[MLX5_ST_SZ_DW(destroy_dct_out)] = {0};
+	u32 in[MLX5_ST_SZ_DW(destroy_dct_in)]   = {0};
+	struct mlx5_core_qp *qp = &dct->mqp;
+	int err;
+
+	err = mlx5_core_drain_dct(dev, dct);
+	if (err) {
+		if (dev->state == MLX5_DEVICE_STATE_INTERNAL_ERROR) {
+			goto destroy;
+		} else {
+			mlx5_core_warn(dev, "failed drain DCT 0x%x with error 0x%x\n", qp->qpn, err);
+			return err;
+		}
+	}
+	wait_for_completion(&dct->drained);
+destroy:
+	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);
+	err = mlx5_cmd_exec(dev, (void *)&in, sizeof(in),
+			    (void *)&out, sizeof(out));
+	return err;
+}
+EXPORT_SYMBOL_GPL(mlx5_core_destroy_dct);
+
 int mlx5_core_destroy_qp(struct mlx5_core_dev *dev,
 			 struct mlx5_core_qp *qp)
 {
@@ -229,7 +316,7 @@ int mlx5_core_destroy_qp(struct mlx5_core_dev *dev,
 
 	mlx5_debug_qp_remove(dev, qp);
 
-	destroy_qprqsq_common(dev, 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);
@@ -405,6 +492,20 @@ int mlx5_core_qp_query(struct mlx5_core_dev *dev, struct mlx5_core_qp *qp,
 }
 EXPORT_SYMBOL_GPL(mlx5_core_qp_query);
 
+int mlx5_core_dct_query(struct mlx5_core_dev *dev, struct mlx5_core_dct *dct,
+			u32 *out, int outlen)
+{
+	u32 in[MLX5_ST_SZ_DW(query_dct_in)] = {0};
+	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, (void *)&in, sizeof(in),
+			     (void *)out, outlen);
+}
+EXPORT_SYMBOL_GPL(mlx5_core_dct_query);
+
 int mlx5_core_xrcd_alloc(struct mlx5_core_dev *dev, u32 *xrcdn)
 {
 	u32 out[MLX5_ST_SZ_DW(alloc_xrcd_out)] = {0};
@@ -441,7 +542,7 @@ int mlx5_core_create_rq_tracked(struct mlx5_core_dev *dev, u32 *in, int inlen,
 		return err;
 
 	rq->qpn = rqn;
-	err = create_qprqsq_common(dev, rq, MLX5_RES_RQ);
+	err = create_resource_common(dev, rq, MLX5_RES_RQ);
 	if (err)
 		goto err_destroy_rq;
 
@@ -457,7 +558,7 @@ EXPORT_SYMBOL(mlx5_core_create_rq_tracked);
 void mlx5_core_destroy_rq_tracked(struct mlx5_core_dev *dev,
 				  struct mlx5_core_qp *rq)
 {
-	destroy_qprqsq_common(dev, rq);
+	destroy_resource_common(dev, rq);
 	mlx5_core_destroy_rq(dev, rq->qpn);
 }
 EXPORT_SYMBOL(mlx5_core_destroy_rq_tracked);
@@ -473,7 +574,7 @@ int mlx5_core_create_sq_tracked(struct mlx5_core_dev *dev, u32 *in, int inlen,
 		return err;
 
 	sq->qpn = sqn;
-	err = create_qprqsq_common(dev, sq, MLX5_RES_SQ);
+	err = create_resource_common(dev, sq, MLX5_RES_SQ);
 	if (err)
 		goto err_destroy_sq;
 
@@ -489,7 +590,7 @@ EXPORT_SYMBOL(mlx5_core_create_sq_tracked);
 void mlx5_core_destroy_sq_tracked(struct mlx5_core_dev *dev,
 				  struct mlx5_core_qp *sq)
 {
-	destroy_qprqsq_common(dev, sq);
+	destroy_resource_common(dev, sq);
 	mlx5_core_destroy_sq(dev, sq->qpn);
 }
 EXPORT_SYMBOL(mlx5_core_destroy_sq_tracked);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/vport.c b/drivers/net/ethernet/mellanox/mlx5/core/vport.c
index a1296a62497d..dfe36cf6fbea 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/vport.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/vport.c
@@ -36,6 +36,9 @@
 #include <linux/mlx5/vport.h>
 #include "mlx5_core.h"
 
+/* Mutex to hold while enabling or disabling RoCE */
+static DEFINE_MUTEX(mlx5_roce_en_lock);
+
 static int _mlx5_query_vport_state(struct mlx5_core_dev *mdev, u8 opmod,
 				   u16 vport, u32 *out, int outlen)
 {
@@ -998,17 +1001,35 @@ static int mlx5_nic_vport_update_roce_state(struct mlx5_core_dev *mdev,
 
 int mlx5_nic_vport_enable_roce(struct mlx5_core_dev *mdev)
 {
-	if (atomic_inc_return(&mdev->roce.roce_en) != 1)
-		return 0;
-	return mlx5_nic_vport_update_roce_state(mdev, MLX5_VPORT_ROCE_ENABLED);
+	int err = 0;
+
+	mutex_lock(&mlx5_roce_en_lock);
+	if (!mdev->roce.roce_en)
+		err = mlx5_nic_vport_update_roce_state(mdev, MLX5_VPORT_ROCE_ENABLED);
+
+	if (!err)
+		mdev->roce.roce_en++;
+	mutex_unlock(&mlx5_roce_en_lock);
+
+	return err;
 }
 EXPORT_SYMBOL_GPL(mlx5_nic_vport_enable_roce);
 
 int mlx5_nic_vport_disable_roce(struct mlx5_core_dev *mdev)
 {
-	if (atomic_dec_return(&mdev->roce.roce_en) != 0)
-		return 0;
-	return mlx5_nic_vport_update_roce_state(mdev, MLX5_VPORT_ROCE_DISABLED);
+	int err = 0;
+
+	mutex_lock(&mlx5_roce_en_lock);
+	if (mdev->roce.roce_en) {
+		mdev->roce.roce_en--;
+		if (mdev->roce.roce_en == 0)
+			err = mlx5_nic_vport_update_roce_state(mdev, MLX5_VPORT_ROCE_DISABLED);
+
+		if (err)
+			mdev->roce.roce_en++;
+	}
+	mutex_unlock(&mlx5_roce_en_lock);
+	return err;
 }
 EXPORT_SYMBOL_GPL(mlx5_nic_vport_disable_roce);
 
@@ -1110,3 +1131,61 @@ ex:
 	return err;
 }
 EXPORT_SYMBOL_GPL(mlx5_core_modify_hca_vport_context);
+
+int mlx5_nic_vport_affiliate_multiport(struct mlx5_core_dev *master_mdev,
+				       struct mlx5_core_dev *port_mdev)
+{
+	int inlen = MLX5_ST_SZ_BYTES(modify_nic_vport_context_in);
+	void *in;
+	int err;
+
+	in = kvzalloc(inlen, GFP_KERNEL);
+	if (!in)
+		return -ENOMEM;
+
+	err = mlx5_nic_vport_enable_roce(port_mdev);
+	if (err)
+		goto free;
+
+	MLX5_SET(modify_nic_vport_context_in, in, field_select.affiliation, 1);
+	MLX5_SET(modify_nic_vport_context_in, in,
+		 nic_vport_context.affiliated_vhca_id,
+		 MLX5_CAP_GEN(master_mdev, vhca_id));
+	MLX5_SET(modify_nic_vport_context_in, in,
+		 nic_vport_context.affiliation_criteria,
+		 MLX5_CAP_GEN(port_mdev, affiliate_nic_vport_criteria));
+
+	err = mlx5_modify_nic_vport_context(port_mdev, in, inlen);
+	if (err)
+		mlx5_nic_vport_disable_roce(port_mdev);
+
+free:
+	kvfree(in);
+	return err;
+}
+EXPORT_SYMBOL_GPL(mlx5_nic_vport_affiliate_multiport);
+
+int mlx5_nic_vport_unaffiliate_multiport(struct mlx5_core_dev *port_mdev)
+{
+	int inlen = MLX5_ST_SZ_BYTES(modify_nic_vport_context_in);
+	void *in;
+	int err;
+
+	in = kvzalloc(inlen, GFP_KERNEL);
+	if (!in)
+		return -ENOMEM;
+
+	MLX5_SET(modify_nic_vport_context_in, in, field_select.affiliation, 1);
+	MLX5_SET(modify_nic_vport_context_in, in,
+		 nic_vport_context.affiliated_vhca_id, 0);
+	MLX5_SET(modify_nic_vport_context_in, in,
+		 nic_vport_context.affiliation_criteria, 0);
+
+	err = mlx5_modify_nic_vport_context(port_mdev, in, inlen);
+	if (!err)
+		mlx5_nic_vport_disable_roce(port_mdev);
+
+	kvfree(in);
+	return err;
+}
+EXPORT_SYMBOL_GPL(mlx5_nic_vport_unaffiliate_multiport);