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

Pull rdma updates from Jason Gunthorpe:
 "This has been a large cycle for RDMA, with several major patch series
  reworking parts of the core code.

   - Rework the so-called 'gid cache' and internal APIs to use a kref'd
     pointer to a struct instead of copying, push this upwards into the
     callers and add more stuff to the struct. The new design avoids
     some ugly races the old one suffered with. This is part of the
     namespace enablement work as the new struct is learning to be
     namespace aware.

   - Various uapi cleanups, moving more stuff to include/uapi and fixing
     some long standing bugs that have recently been discovered.

   - Driver updates for mlx5, mlx4 i40iw, rxe, cxgb4, hfi1, usnic,
     pvrdma, and hns

   - Provide max_send_sge and max_recv_sge attributes to better support
     HW where these values are asymmetric.

   - mlx5 user API 'devx' allows sending commands directly to the device
     FW, instead of trying to cram every wild and niche feature into the
     common API. Sort of like what GPU does.

   - Major write() and ioctl() API rework to cleanly support PCI device
     hot unplug and advance the ioctl conversion work

   - Sparse and compile warning cleanups

   - Add 'const' to the ib_poll_cq() signature, and permit a NULL
     'bad_wr', which is the common use case

   - Various patches to avoid high order allocations across the stack

   - SRQ support for cxgb4, hns and qedr

   - Changes to IPoIB to better follow the netdev model for working with
     struct net_device liftime"

* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma: (312 commits)
  Revert "net/smc: Replace ib_query_gid with rdma_get_gid_attr"
  RDMA/hns: Fix usage of bitmap allocation functions return values
  IB/core: Change filter function return type from int to bool
  IB/core: Update GID entries for netdevice whose mac address changes
  IB/core: Add default GIDs of the bond master netdev
  IB/core: Consider adding default GIDs of bond device
  IB/core: Delete lower netdevice default GID entries in bonding scenario
  IB/core: Avoid confusing del_netdev_default_ips
  IB/core: Add comment for change upper netevent handling
  qedr: Add user space support for SRQ
  qedr: Add support for kernel mode SRQ's
  qedr: Add wrapping generic structure for qpidr and adjust idr routines.
  IB/mlx5: Fix leaking stack memory to userspace
  Update the e-mail address of Bart Van Assche
  IB/ucm: Fix compiling ucm.c
  IB/uverbs: Do not check for device disassociation during ioctl
  IB/uverbs: Remove struct uverbs_root_spec and all supporting code
  IB/uverbs: Use uverbs_api to unmarshal ioctl commands
  IB/uverbs: Use uverbs_alloc for allocations
  IB/uverbs: Add a simple allocator to uverbs_attr_bundle
  ...

14 files changed, 2083 insertions, 320 deletions

diff --git a/drivers/infiniband/hw/mlx5/Makefile b/drivers/infiniband/hw/mlx5/Makefile
index d42b922bede8..b8e4b15e2674 100644
--- a/drivers/infiniband/hw/mlx5/Makefile
+++ b/drivers/infiniband/hw/mlx5/Makefile
@@ -3,3 +3,5 @@ obj-$(CONFIG_MLX5_INFINIBAND)	+= mlx5_ib.o
 mlx5_ib-y :=	main.o cq.o doorbell.o qp.o mem.o srq.o mr.o ah.o mad.o gsi.o ib_virt.o cmd.o cong.o
 mlx5_ib-$(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
diff --git a/drivers/infiniband/hw/mlx5/ah.c b/drivers/infiniband/hw/mlx5/ah.c
index e6bde32a83f3..ffd03bf1a71e 100644
--- a/drivers/infiniband/hw/mlx5/ah.c
+++ b/drivers/infiniband/hw/mlx5/ah.c
@@ -37,7 +37,6 @@ static struct ib_ah *create_ib_ah(struct mlx5_ib_dev *dev,
 				  struct rdma_ah_attr *ah_attr)
 {
 	enum ib_gid_type gid_type;
-	int err;
 
 	if (rdma_ah_get_ah_flags(ah_attr) & IB_AH_GRH) {
 		const struct ib_global_route *grh = rdma_ah_read_grh(ah_attr);
@@ -53,18 +52,12 @@ static struct ib_ah *create_ib_ah(struct mlx5_ib_dev *dev,
 	ah->av.stat_rate_sl = (rdma_ah_get_static_rate(ah_attr) << 4);
 
 	if (ah_attr->type == RDMA_AH_ATTR_TYPE_ROCE) {
-		err = mlx5_get_roce_gid_type(dev, ah_attr->port_num,
-					     ah_attr->grh.sgid_index,
-					     &gid_type);
-		if (err)
-			return ERR_PTR(err);
+		gid_type = ah_attr->grh.sgid_attr->gid_type;
 
 		memcpy(ah->av.rmac, ah_attr->roce.dmac,
 		       sizeof(ah_attr->roce.dmac));
 		ah->av.udp_sport =
-		mlx5_get_roce_udp_sport(dev,
-					rdma_ah_get_port_num(ah_attr),
-					rdma_ah_read_grh(ah_attr)->sgid_index);
+			mlx5_get_roce_udp_sport(dev, ah_attr->grh.sgid_attr);
 		ah->av.stat_rate_sl |= (rdma_ah_get_sl(ah_attr) & 0x7) << 1;
 		if (gid_type == IB_GID_TYPE_ROCE_UDP_ENCAP)
 #define MLX5_ECN_ENABLED BIT(1)
diff --git a/drivers/infiniband/hw/mlx5/cmd.c b/drivers/infiniband/hw/mlx5/cmd.c
index ccc0b5d06a7d..c84fef9a8a08 100644
--- a/drivers/infiniband/hw/mlx5/cmd.c
+++ b/drivers/infiniband/hw/mlx5/cmd.c
@@ -185,3 +185,15 @@ int mlx5_cmd_dealloc_memic(struct mlx5_memic *memic, u64 addr, u64 length)
 
 	return err;
 }
+
+int mlx5_cmd_query_ext_ppcnt_counters(struct mlx5_core_dev *dev, void *out)
+{
+	u32 in[MLX5_ST_SZ_DW(ppcnt_reg)] = {};
+	int sz = MLX5_ST_SZ_BYTES(ppcnt_reg);
+
+	MLX5_SET(ppcnt_reg, in, local_port, 1);
+
+	MLX5_SET(ppcnt_reg, in, grp, MLX5_ETHERNET_EXTENDED_COUNTERS_GROUP);
+	return  mlx5_core_access_reg(dev, in, sz, out, sz, MLX5_REG_PPCNT,
+				     0, 0);
+}
diff --git a/drivers/infiniband/hw/mlx5/cmd.h b/drivers/infiniband/hw/mlx5/cmd.h
index 98ea4648c655..88cbb1c41703 100644
--- a/drivers/infiniband/hw/mlx5/cmd.h
+++ b/drivers/infiniband/hw/mlx5/cmd.h
@@ -41,6 +41,7 @@ int mlx5_cmd_dump_fill_mkey(struct mlx5_core_dev *dev, u32 *mkey);
 int mlx5_cmd_null_mkey(struct mlx5_core_dev *dev, u32 *null_mkey);
 int mlx5_cmd_query_cong_params(struct mlx5_core_dev *dev, int cong_point,
 			       void *out, int out_size);
+int mlx5_cmd_query_ext_ppcnt_counters(struct mlx5_core_dev *dev, void *out);
 int mlx5_cmd_modify_cong_params(struct mlx5_core_dev *mdev,
 				void *in, int in_size);
 int mlx5_cmd_alloc_memic(struct mlx5_memic *memic, phys_addr_t *addr,
diff --git a/drivers/infiniband/hw/mlx5/cong.c b/drivers/infiniband/hw/mlx5/cong.c
index 985fa2637390..7e4e358a4fd8 100644
--- a/drivers/infiniband/hw/mlx5/cong.c
+++ b/drivers/infiniband/hw/mlx5/cong.c
@@ -359,9 +359,6 @@ static ssize_t get_param(struct file *filp, char __user *buf, size_t count,
 	int ret;
 	char lbuf[11];
 
-	if (*pos)
-		return 0;
-
 	ret = mlx5_ib_get_cc_params(param->dev, param->port_num, offset, &var);
 	if (ret)
 		return ret;
@@ -370,11 +367,7 @@ static ssize_t get_param(struct file *filp, char __user *buf, size_t count,
 	if (ret < 0)
 		return ret;
 
-	if (copy_to_user(buf, lbuf, ret))
-		return -EFAULT;
-
-	*pos += ret;
-	return ret;
+	return simple_read_from_buffer(buf, count, pos, lbuf, ret);
 }
 
 static const struct file_operations dbg_cc_fops = {
diff --git a/drivers/infiniband/hw/mlx5/cq.c b/drivers/infiniband/hw/mlx5/cq.c
index ad39d64b8108..088205d7f1a1 100644
--- a/drivers/infiniband/hw/mlx5/cq.c
+++ b/drivers/infiniband/hw/mlx5/cq.c
@@ -1184,7 +1184,7 @@ int mlx5_ib_modify_cq(struct ib_cq *cq, u16 cq_count, u16 cq_period)
 	int err;
 
 	if (!MLX5_CAP_GEN(dev->mdev, cq_moderation))
-		return -ENOSYS;
+		return -EOPNOTSUPP;
 
 	if (cq_period > MLX5_MAX_CQ_PERIOD)
 		return -EINVAL;
diff --git a/drivers/infiniband/hw/mlx5/devx.c b/drivers/infiniband/hw/mlx5/devx.c
new file mode 100644
index 000000000000..ac116d63e466
--- /dev/null
+++ b/drivers/infiniband/hw/mlx5/devx.c
@@ -0,0 +1,1119 @@
+// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB
+/*
+ * Copyright (c) 2018, Mellanox Technologies inc.  All rights reserved.
+ */
+
+#include <rdma/ib_user_verbs.h>
+#include <rdma/ib_verbs.h>
+#include <rdma/uverbs_types.h>
+#include <rdma/uverbs_ioctl.h>
+#include <rdma/mlx5_user_ioctl_cmds.h>
+#include <rdma/ib_umem.h>
+#include <linux/mlx5/driver.h>
+#include <linux/mlx5/fs.h>
+#include "mlx5_ib.h"
+
+#define UVERBS_MODULE_NAME mlx5_ib
+#include <rdma/uverbs_named_ioctl.h>
+
+#define MLX5_MAX_DESTROY_INBOX_SIZE_DW MLX5_ST_SZ_DW(delete_fte_in)
+struct devx_obj {
+	struct mlx5_core_dev	*mdev;
+	u32			obj_id;
+	u32			dinlen; /* destroy inbox length */
+	u32			dinbox[MLX5_MAX_DESTROY_INBOX_SIZE_DW];
+};
+
+struct devx_umem {
+	struct mlx5_core_dev		*mdev;
+	struct ib_umem			*umem;
+	u32				page_offset;
+	int				page_shift;
+	int				ncont;
+	u32				dinlen;
+	u32				dinbox[MLX5_ST_SZ_DW(general_obj_in_cmd_hdr)];
+};
+
+struct devx_umem_reg_cmd {
+	void				*in;
+	u32				inlen;
+	u32				out[MLX5_ST_SZ_DW(general_obj_out_cmd_hdr)];
+};
+
+static struct mlx5_ib_ucontext *devx_ufile2uctx(struct ib_uverbs_file *file)
+{
+	return to_mucontext(ib_uverbs_get_ucontext(file));
+}
+
+int mlx5_ib_devx_create(struct mlx5_ib_dev *dev, struct mlx5_ib_ucontext *context)
+{
+	u32 in[MLX5_ST_SZ_DW(create_uctx_in)] = {0};
+	u32 out[MLX5_ST_SZ_DW(general_obj_out_cmd_hdr)] = {0};
+	u64 general_obj_types;
+	void *hdr;
+	int err;
+
+	hdr = MLX5_ADDR_OF(create_uctx_in, in, hdr);
+
+	general_obj_types = MLX5_CAP_GEN_64(dev->mdev, general_obj_types);
+	if (!(general_obj_types & MLX5_GENERAL_OBJ_TYPES_CAP_UCTX) ||
+	    !(general_obj_types & MLX5_GENERAL_OBJ_TYPES_CAP_UMEM))
+		return -EINVAL;
+
+	if (!capable(CAP_NET_RAW))
+		return -EPERM;
+
+	MLX5_SET(general_obj_in_cmd_hdr, hdr, opcode, MLX5_CMD_OP_CREATE_GENERAL_OBJECT);
+	MLX5_SET(general_obj_in_cmd_hdr, hdr, obj_type, MLX5_OBJ_TYPE_UCTX);
+
+	err = mlx5_cmd_exec(dev->mdev, in, sizeof(in), out, sizeof(out));
+	if (err)
+		return err;
+
+	context->devx_uid = MLX5_GET(general_obj_out_cmd_hdr, out, obj_id);
+	return 0;
+}
+
+void mlx5_ib_devx_destroy(struct mlx5_ib_dev *dev,
+			  struct mlx5_ib_ucontext *context)
+{
+	u32 in[MLX5_ST_SZ_DW(general_obj_in_cmd_hdr)] = {0};
+	u32 out[MLX5_ST_SZ_DW(general_obj_out_cmd_hdr)] = {0};
+
+	MLX5_SET(general_obj_in_cmd_hdr, in, opcode, MLX5_CMD_OP_DESTROY_GENERAL_OBJECT);
+	MLX5_SET(general_obj_in_cmd_hdr, in, obj_type, MLX5_OBJ_TYPE_UCTX);
+	MLX5_SET(general_obj_in_cmd_hdr, in, obj_id, context->devx_uid);
+
+	mlx5_cmd_exec(dev->mdev, in, sizeof(in), out, sizeof(out));
+}
+
+bool mlx5_ib_devx_is_flow_dest(void *obj, int *dest_id, int *dest_type)
+{
+	struct devx_obj *devx_obj = obj;
+	u16 opcode = MLX5_GET(general_obj_in_cmd_hdr, devx_obj->dinbox, opcode);
+
+	switch (opcode) {
+	case MLX5_CMD_OP_DESTROY_TIR:
+		*dest_type = MLX5_FLOW_DESTINATION_TYPE_TIR;
+		*dest_id = MLX5_GET(general_obj_in_cmd_hdr, devx_obj->dinbox,
+				    obj_id);
+		return true;
+
+	case MLX5_CMD_OP_DESTROY_FLOW_TABLE:
+		*dest_type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
+		*dest_id = MLX5_GET(destroy_flow_table_in, devx_obj->dinbox,
+				    table_id);
+		return true;
+	default:
+		return false;
+	}
+}
+
+static int devx_is_valid_obj_id(struct devx_obj *obj, const void *in)
+{
+	u16 opcode = MLX5_GET(general_obj_in_cmd_hdr, in, opcode);
+	u32 obj_id;
+
+	switch (opcode) {
+	case MLX5_CMD_OP_MODIFY_GENERAL_OBJECT:
+	case MLX5_CMD_OP_QUERY_GENERAL_OBJECT:
+		obj_id = MLX5_GET(general_obj_in_cmd_hdr, in, obj_id);
+		break;
+	case MLX5_CMD_OP_QUERY_MKEY:
+		obj_id = MLX5_GET(query_mkey_in, in, mkey_index);
+		break;
+	case MLX5_CMD_OP_QUERY_CQ:
+		obj_id = MLX5_GET(query_cq_in, in, cqn);
+		break;
+	case MLX5_CMD_OP_MODIFY_CQ:
+		obj_id = MLX5_GET(modify_cq_in, in, cqn);
+		break;
+	case MLX5_CMD_OP_QUERY_SQ:
+		obj_id = MLX5_GET(query_sq_in, in, sqn);
+		break;
+	case MLX5_CMD_OP_MODIFY_SQ:
+		obj_id = MLX5_GET(modify_sq_in, in, sqn);
+		break;
+	case MLX5_CMD_OP_QUERY_RQ:
+		obj_id = MLX5_GET(query_rq_in, in, rqn);
+		break;
+	case MLX5_CMD_OP_MODIFY_RQ:
+		obj_id = MLX5_GET(modify_rq_in, in, rqn);
+		break;
+	case MLX5_CMD_OP_QUERY_RMP:
+		obj_id = MLX5_GET(query_rmp_in, in, rmpn);
+		break;
+	case MLX5_CMD_OP_MODIFY_RMP:
+		obj_id = MLX5_GET(modify_rmp_in, in, rmpn);
+		break;
+	case MLX5_CMD_OP_QUERY_RQT:
+		obj_id = MLX5_GET(query_rqt_in, in, rqtn);
+		break;
+	case MLX5_CMD_OP_MODIFY_RQT:
+		obj_id = MLX5_GET(modify_rqt_in, in, rqtn);
+		break;
+	case MLX5_CMD_OP_QUERY_TIR:
+		obj_id = MLX5_GET(query_tir_in, in, tirn);
+		break;
+	case MLX5_CMD_OP_MODIFY_TIR:
+		obj_id = MLX5_GET(modify_tir_in, in, tirn);
+		break;
+	case MLX5_CMD_OP_QUERY_TIS:
+		obj_id = MLX5_GET(query_tis_in, in, tisn);
+		break;
+	case MLX5_CMD_OP_MODIFY_TIS:
+		obj_id = MLX5_GET(modify_tis_in, in, tisn);
+		break;
+	case MLX5_CMD_OP_QUERY_FLOW_TABLE:
+		obj_id = MLX5_GET(query_flow_table_in, in, table_id);
+		break;
+	case MLX5_CMD_OP_MODIFY_FLOW_TABLE:
+		obj_id = MLX5_GET(modify_flow_table_in, in, table_id);
+		break;
+	case MLX5_CMD_OP_QUERY_FLOW_GROUP:
+		obj_id = MLX5_GET(query_flow_group_in, in, group_id);
+		break;
+	case MLX5_CMD_OP_QUERY_FLOW_TABLE_ENTRY:
+		obj_id = MLX5_GET(query_fte_in, in, flow_index);
+		break;
+	case MLX5_CMD_OP_SET_FLOW_TABLE_ENTRY:
+		obj_id = MLX5_GET(set_fte_in, in, flow_index);
+		break;
+	case MLX5_CMD_OP_QUERY_Q_COUNTER:
+		obj_id = MLX5_GET(query_q_counter_in, in, counter_set_id);
+		break;
+	case MLX5_CMD_OP_QUERY_FLOW_COUNTER:
+		obj_id = MLX5_GET(query_flow_counter_in, in, flow_counter_id);
+		break;
+	case MLX5_CMD_OP_QUERY_MODIFY_HEADER_CONTEXT:
+		obj_id = MLX5_GET(general_obj_in_cmd_hdr, in, obj_id);
+		break;
+	case MLX5_CMD_OP_QUERY_SCHEDULING_ELEMENT:
+		obj_id = MLX5_GET(query_scheduling_element_in, in,
+				  scheduling_element_id);
+		break;
+	case MLX5_CMD_OP_MODIFY_SCHEDULING_ELEMENT:
+		obj_id = MLX5_GET(modify_scheduling_element_in, in,
+				  scheduling_element_id);
+		break;
+	case MLX5_CMD_OP_ADD_VXLAN_UDP_DPORT:
+		obj_id = MLX5_GET(add_vxlan_udp_dport_in, in, vxlan_udp_port);
+		break;
+	case MLX5_CMD_OP_QUERY_L2_TABLE_ENTRY:
+		obj_id = MLX5_GET(query_l2_table_entry_in, in, table_index);
+		break;
+	case MLX5_CMD_OP_SET_L2_TABLE_ENTRY:
+		obj_id = MLX5_GET(set_l2_table_entry_in, in, table_index);
+		break;
+	case MLX5_CMD_OP_QUERY_QP:
+		obj_id = MLX5_GET(query_qp_in, in, qpn);
+		break;
+	case MLX5_CMD_OP_RST2INIT_QP:
+		obj_id = MLX5_GET(rst2init_qp_in, in, qpn);
+		break;
+	case MLX5_CMD_OP_INIT2RTR_QP:
+		obj_id = MLX5_GET(init2rtr_qp_in, in, qpn);
+		break;
+	case MLX5_CMD_OP_RTR2RTS_QP:
+		obj_id = MLX5_GET(rtr2rts_qp_in, in, qpn);
+		break;
+	case MLX5_CMD_OP_RTS2RTS_QP:
+		obj_id = MLX5_GET(rts2rts_qp_in, in, qpn);
+		break;
+	case MLX5_CMD_OP_SQERR2RTS_QP:
+		obj_id = MLX5_GET(sqerr2rts_qp_in, in, qpn);
+		break;
+	case MLX5_CMD_OP_2ERR_QP:
+		obj_id = MLX5_GET(qp_2err_in, in, qpn);
+		break;
+	case MLX5_CMD_OP_2RST_QP:
+		obj_id = MLX5_GET(qp_2rst_in, in, qpn);
+		break;
+	case MLX5_CMD_OP_QUERY_DCT:
+		obj_id = MLX5_GET(query_dct_in, in, dctn);
+		break;
+	case MLX5_CMD_OP_QUERY_XRQ:
+		obj_id = MLX5_GET(query_xrq_in, in, xrqn);
+		break;
+	case MLX5_CMD_OP_QUERY_XRC_SRQ:
+		obj_id = MLX5_GET(query_xrc_srq_in, in, xrc_srqn);
+		break;
+	case MLX5_CMD_OP_ARM_XRC_SRQ:
+		obj_id = MLX5_GET(arm_xrc_srq_in, in, xrc_srqn);
+		break;
+	case MLX5_CMD_OP_QUERY_SRQ:
+		obj_id = MLX5_GET(query_srq_in, in, srqn);
+		break;
+	case MLX5_CMD_OP_ARM_RQ:
+		obj_id = MLX5_GET(arm_rq_in, in, srq_number);
+		break;
+	case MLX5_CMD_OP_DRAIN_DCT:
+	case MLX5_CMD_OP_ARM_DCT_FOR_KEY_VIOLATION:
+		obj_id = MLX5_GET(drain_dct_in, in, dctn);
+		break;
+	case MLX5_CMD_OP_ARM_XRQ:
+		obj_id = MLX5_GET(arm_xrq_in, in, xrqn);
+		break;
+	default:
+		return false;
+	}
+
+	if (obj_id == obj->obj_id)
+		return true;
+
+	return false;
+}
+
+static bool devx_is_obj_create_cmd(const void *in)
+{
+	u16 opcode = MLX5_GET(general_obj_in_cmd_hdr, in, opcode);
+
+	switch (opcode) {
+	case MLX5_CMD_OP_CREATE_GENERAL_OBJECT:
+	case MLX5_CMD_OP_CREATE_MKEY:
+	case MLX5_CMD_OP_CREATE_CQ:
+	case MLX5_CMD_OP_ALLOC_PD:
+	case MLX5_CMD_OP_ALLOC_TRANSPORT_DOMAIN:
+	case MLX5_CMD_OP_CREATE_RMP:
+	case MLX5_CMD_OP_CREATE_SQ:
+	case MLX5_CMD_OP_CREATE_RQ:
+	case MLX5_CMD_OP_CREATE_RQT:
+	case MLX5_CMD_OP_CREATE_TIR:
+	case MLX5_CMD_OP_CREATE_TIS:
+	case MLX5_CMD_OP_ALLOC_Q_COUNTER:
+	case MLX5_CMD_OP_CREATE_FLOW_TABLE:
+	case MLX5_CMD_OP_CREATE_FLOW_GROUP:
+	case MLX5_CMD_OP_ALLOC_FLOW_COUNTER:
+	case MLX5_CMD_OP_ALLOC_ENCAP_HEADER:
+	case MLX5_CMD_OP_ALLOC_MODIFY_HEADER_CONTEXT:
+	case MLX5_CMD_OP_CREATE_SCHEDULING_ELEMENT:
+	case MLX5_CMD_OP_ADD_VXLAN_UDP_DPORT:
+	case MLX5_CMD_OP_SET_L2_TABLE_ENTRY:
+	case MLX5_CMD_OP_CREATE_QP:
+	case MLX5_CMD_OP_CREATE_SRQ:
+	case MLX5_CMD_OP_CREATE_XRC_SRQ:
+	case MLX5_CMD_OP_CREATE_DCT:
+	case MLX5_CMD_OP_CREATE_XRQ:
+	case MLX5_CMD_OP_ATTACH_TO_MCG:
+	case MLX5_CMD_OP_ALLOC_XRCD:
+		return true;
+	case MLX5_CMD_OP_SET_FLOW_TABLE_ENTRY:
+	{
+		u16 op_mod = MLX5_GET(set_fte_in, in, op_mod);
+		if (op_mod == 0)
+			return true;
+		return false;
+	}
+	default:
+		return false;
+	}
+}
+
+static bool devx_is_obj_modify_cmd(const void *in)
+{
+	u16 opcode = MLX5_GET(general_obj_in_cmd_hdr, in, opcode);
+
+	switch (opcode) {
+	case MLX5_CMD_OP_MODIFY_GENERAL_OBJECT:
+	case MLX5_CMD_OP_MODIFY_CQ:
+	case MLX5_CMD_OP_MODIFY_RMP:
+	case MLX5_CMD_OP_MODIFY_SQ:
+	case MLX5_CMD_OP_MODIFY_RQ:
+	case MLX5_CMD_OP_MODIFY_RQT:
+	case MLX5_CMD_OP_MODIFY_TIR:
+	case MLX5_CMD_OP_MODIFY_TIS:
+	case MLX5_CMD_OP_MODIFY_FLOW_TABLE:
+	case MLX5_CMD_OP_MODIFY_SCHEDULING_ELEMENT:
+	case MLX5_CMD_OP_ADD_VXLAN_UDP_DPORT:
+	case MLX5_CMD_OP_SET_L2_TABLE_ENTRY:
+	case MLX5_CMD_OP_RST2INIT_QP:
+	case MLX5_CMD_OP_INIT2RTR_QP:
+	case MLX5_CMD_OP_RTR2RTS_QP:
+	case MLX5_CMD_OP_RTS2RTS_QP:
+	case MLX5_CMD_OP_SQERR2RTS_QP:
+	case MLX5_CMD_OP_2ERR_QP:
+	case MLX5_CMD_OP_2RST_QP:
+	case MLX5_CMD_OP_ARM_XRC_SRQ:
+	case MLX5_CMD_OP_ARM_RQ:
+	case MLX5_CMD_OP_DRAIN_DCT:
+	case MLX5_CMD_OP_ARM_DCT_FOR_KEY_VIOLATION:
+	case MLX5_CMD_OP_ARM_XRQ:
+		return true;
+	case MLX5_CMD_OP_SET_FLOW_TABLE_ENTRY:
+	{
+		u16 op_mod = MLX5_GET(set_fte_in, in, op_mod);
+
+		if (op_mod == 1)
+			return true;
+		return false;
+	}
+	default:
+		return false;
+	}
+}
+
+static bool devx_is_obj_query_cmd(const void *in)
+{
+	u16 opcode = MLX5_GET(general_obj_in_cmd_hdr, in, opcode);
+
+	switch (opcode) {
+	case MLX5_CMD_OP_QUERY_GENERAL_OBJECT:
+	case MLX5_CMD_OP_QUERY_MKEY:
+	case MLX5_CMD_OP_QUERY_CQ:
+	case MLX5_CMD_OP_QUERY_RMP:
+	case MLX5_CMD_OP_QUERY_SQ:
+	case MLX5_CMD_OP_QUERY_RQ:
+	case MLX5_CMD_OP_QUERY_RQT:
+	case MLX5_CMD_OP_QUERY_TIR:
+	case MLX5_CMD_OP_QUERY_TIS:
+	case MLX5_CMD_OP_QUERY_Q_COUNTER:
+	case MLX5_CMD_OP_QUERY_FLOW_TABLE:
+	case MLX5_CMD_OP_QUERY_FLOW_GROUP:
+	case MLX5_CMD_OP_QUERY_FLOW_TABLE_ENTRY:
+	case MLX5_CMD_OP_QUERY_FLOW_COUNTER:
+	case MLX5_CMD_OP_QUERY_MODIFY_HEADER_CONTEXT:
+	case MLX5_CMD_OP_QUERY_SCHEDULING_ELEMENT:
+	case MLX5_CMD_OP_QUERY_L2_TABLE_ENTRY:
+	case MLX5_CMD_OP_QUERY_QP:
+	case MLX5_CMD_OP_QUERY_SRQ:
+	case MLX5_CMD_OP_QUERY_XRC_SRQ:
+	case MLX5_CMD_OP_QUERY_DCT:
+	case MLX5_CMD_OP_QUERY_XRQ:
+		return true;
+	default:
+		return false;
+	}
+}
+
+static bool devx_is_general_cmd(void *in)
+{
+	u16 opcode = MLX5_GET(general_obj_in_cmd_hdr, in, opcode);
+
+	switch (opcode) {
+	case MLX5_CMD_OP_QUERY_HCA_CAP:
+	case MLX5_CMD_OP_QUERY_VPORT_STATE:
+	case MLX5_CMD_OP_QUERY_ADAPTER:
+	case MLX5_CMD_OP_QUERY_ISSI:
+	case MLX5_CMD_OP_QUERY_NIC_VPORT_CONTEXT:
+	case MLX5_CMD_OP_QUERY_ROCE_ADDRESS:
+	case MLX5_CMD_OP_QUERY_VNIC_ENV:
+	case MLX5_CMD_OP_QUERY_VPORT_COUNTER:
+	case MLX5_CMD_OP_GET_DROPPED_PACKET_LOG:
+	case MLX5_CMD_OP_NOP:
+	case MLX5_CMD_OP_QUERY_CONG_STATUS:
+	case MLX5_CMD_OP_QUERY_CONG_PARAMS:
+	case MLX5_CMD_OP_QUERY_CONG_STATISTICS:
+		return true;
+	default:
+		return false;
+	}
+}
+
+static int UVERBS_HANDLER(MLX5_IB_METHOD_DEVX_QUERY_EQN)(
+	struct ib_uverbs_file *file, struct uverbs_attr_bundle *attrs)
+{
+	struct mlx5_ib_ucontext *c;
+	struct mlx5_ib_dev *dev;
+	int user_vector;
+	int dev_eqn;
+	unsigned int irqn;
+	int err;
+
+	if (uverbs_copy_from(&user_vector, attrs,
+			     MLX5_IB_ATTR_DEVX_QUERY_EQN_USER_VEC))
+		return -EFAULT;
+
+	c = devx_ufile2uctx(file);
+	if (IS_ERR(c))
+		return PTR_ERR(c);
+	dev = to_mdev(c->ibucontext.device);
+
+	err = mlx5_vector2eqn(dev->mdev, user_vector, &dev_eqn, &irqn);
+	if (err < 0)
+		return err;
+
+	if (uverbs_copy_to(attrs, MLX5_IB_ATTR_DEVX_QUERY_EQN_DEV_EQN,
+			   &dev_eqn, sizeof(dev_eqn)))
+		return -EFAULT;
+
+	return 0;
+}
+
+/*
+ *Security note:
+ * The hardware protection mechanism works like this: Each device object that
+ * is subject to UAR doorbells (QP/SQ/CQ) gets a UAR ID (called uar_page in
+ * the device specification manual) upon its creation. Then upon doorbell,
+ * hardware fetches the object context for which the doorbell was rang, and
+ * validates that the UAR through which the DB was rang matches the UAR ID
+ * of the object.
+ * If no match the doorbell is silently ignored by the hardware. Of course,
+ * the user cannot ring a doorbell on a UAR that was not mapped to it.
+ * Now in devx, as the devx kernel does not manipulate the QP/SQ/CQ command
+ * mailboxes (except tagging them with UID), we expose to the user its UAR
+ * ID, so it can embed it in these objects in the expected specification
+ * format. So the only thing the user can do is hurt itself by creating a
+ * QP/SQ/CQ with a UAR ID other than his, and then in this case other users
+ * may ring a doorbell on its objects.
+ * The consequence of that will be that another user can schedule a QP/SQ
+ * of the buggy user for execution (just insert it to the hardware schedule
+ * queue or arm its CQ for event generation), no further harm is expected.
+ */
+static int UVERBS_HANDLER(MLX5_IB_METHOD_DEVX_QUERY_UAR)(
+	struct ib_uverbs_file *file, struct uverbs_attr_bundle *attrs)
+{
+	struct mlx5_ib_ucontext *c;
+	struct mlx5_ib_dev *dev;
+	u32 user_idx;
+	s32 dev_idx;
+
+	c = devx_ufile2uctx(file);
+	if (IS_ERR(c))
+		return PTR_ERR(c);
+	dev = to_mdev(c->ibucontext.device);
+
+	if (uverbs_copy_from(&user_idx, attrs,
+			     MLX5_IB_ATTR_DEVX_QUERY_UAR_USER_IDX))
+		return -EFAULT;
+
+	dev_idx = bfregn_to_uar_index(dev, &c->bfregi, user_idx, true);
+	if (dev_idx < 0)
+		return dev_idx;
+
+	if (uverbs_copy_to(attrs, MLX5_IB_ATTR_DEVX_QUERY_UAR_DEV_IDX,
+			   &dev_idx, sizeof(dev_idx)))
+		return -EFAULT;
+
+	return 0;
+}
+
+static int UVERBS_HANDLER(MLX5_IB_METHOD_DEVX_OTHER)(
+	struct ib_uverbs_file *file, struct uverbs_attr_bundle *attrs)
+{
+	struct mlx5_ib_ucontext *c;
+	struct mlx5_ib_dev *dev;
+	void *cmd_in = uverbs_attr_get_alloced_ptr(
+		attrs, MLX5_IB_ATTR_DEVX_OTHER_CMD_IN);
+	int cmd_out_len = uverbs_attr_get_len(attrs,
+					MLX5_IB_ATTR_DEVX_OTHER_CMD_OUT);
+	void *cmd_out;
+	int err;
+
+	c = devx_ufile2uctx(file);
+	if (IS_ERR(c))
+		return PTR_ERR(c);
+	dev = to_mdev(c->ibucontext.device);
+
+	if (!c->devx_uid)
+		return -EPERM;
+
+	/* Only white list of some general HCA commands are allowed for this method. */
+	if (!devx_is_general_cmd(cmd_in))
+		return -EINVAL;
+
+	cmd_out = uverbs_zalloc(attrs, cmd_out_len);
+	if (IS_ERR(cmd_out))
+		return PTR_ERR(cmd_out);
+
+	MLX5_SET(general_obj_in_cmd_hdr, cmd_in, uid, c->devx_uid);
+	err = mlx5_cmd_exec(dev->mdev, cmd_in,
+			    uverbs_attr_get_len(attrs, MLX5_IB_ATTR_DEVX_OTHER_CMD_IN),
+			    cmd_out, cmd_out_len);
+	if (err)
+		return err;
+
+	return uverbs_copy_to(attrs, MLX5_IB_ATTR_DEVX_OTHER_CMD_OUT, cmd_out,
+			      cmd_out_len);
+}
+
+static void devx_obj_build_destroy_cmd(void *in, void *out, void *din,
+				       u32 *dinlen,
+				       u32 *obj_id)
+{
+	u16 obj_type = MLX5_GET(general_obj_in_cmd_hdr, in, obj_type);
+	u16 uid = MLX5_GET(general_obj_in_cmd_hdr, in, uid);
+
+	*obj_id = MLX5_GET(general_obj_out_cmd_hdr, out, obj_id);
+	*dinlen = MLX5_ST_SZ_BYTES(general_obj_in_cmd_hdr);
+
+	MLX5_SET(general_obj_in_cmd_hdr, din, obj_id, *obj_id);
+	MLX5_SET(general_obj_in_cmd_hdr, din, uid, uid);
+
+	switch (MLX5_GET(general_obj_in_cmd_hdr, in, opcode)) {
+	case MLX5_CMD_OP_CREATE_GENERAL_OBJECT:
+		MLX5_SET(general_obj_in_cmd_hdr, din, opcode, MLX5_CMD_OP_DESTROY_GENERAL_OBJECT);
+		MLX5_SET(general_obj_in_cmd_hdr, din, obj_type, obj_type);
+		break;
+
+	case MLX5_CMD_OP_CREATE_MKEY:
+		MLX5_SET(general_obj_in_cmd_hdr, din, opcode, MLX5_CMD_OP_DESTROY_MKEY);
+		break;
+	case MLX5_CMD_OP_CREATE_CQ:
+		MLX5_SET(general_obj_in_cmd_hdr, din, opcode, MLX5_CMD_OP_DESTROY_CQ);
+		break;
+	case MLX5_CMD_OP_ALLOC_PD:
+		MLX5_SET(general_obj_in_cmd_hdr, din, opcode, MLX5_CMD_OP_DEALLOC_PD);
+		break;
+	case MLX5_CMD_OP_ALLOC_TRANSPORT_DOMAIN:
+		MLX5_SET(general_obj_in_cmd_hdr, din, opcode,
+			 MLX5_CMD_OP_DEALLOC_TRANSPORT_DOMAIN);
+		break;
+	case MLX5_CMD_OP_CREATE_RMP:
+		MLX5_SET(general_obj_in_cmd_hdr, din, opcode, MLX5_CMD_OP_DESTROY_RMP);
+		break;
+	case MLX5_CMD_OP_CREATE_SQ:
+		MLX5_SET(general_obj_in_cmd_hdr, din, opcode, MLX5_CMD_OP_DESTROY_SQ);
+		break;
+	case MLX5_CMD_OP_CREATE_RQ:
+		MLX5_SET(general_obj_in_cmd_hdr, din, opcode, MLX5_CMD_OP_DESTROY_RQ);
+		break;
+	case MLX5_CMD_OP_CREATE_RQT:
+		MLX5_SET(general_obj_in_cmd_hdr, din, opcode, MLX5_CMD_OP_DESTROY_RQT);
+		break;
+	case MLX5_CMD_OP_CREATE_TIR:
+		MLX5_SET(general_obj_in_cmd_hdr, din, opcode, MLX5_CMD_OP_DESTROY_TIR);
+		break;
+	case MLX5_CMD_OP_CREATE_TIS:
+		MLX5_SET(general_obj_in_cmd_hdr, din, opcode, MLX5_CMD_OP_DESTROY_TIS);
+		break;
+	case MLX5_CMD_OP_ALLOC_Q_COUNTER:
+		MLX5_SET(general_obj_in_cmd_hdr, din, opcode,
+			 MLX5_CMD_OP_DEALLOC_Q_COUNTER);
+		break;
+	case MLX5_CMD_OP_CREATE_FLOW_TABLE:
+		*dinlen = MLX5_ST_SZ_BYTES(destroy_flow_table_in);
+		*obj_id = MLX5_GET(create_flow_table_out, out, table_id);
+		MLX5_SET(destroy_flow_table_in, din, other_vport,
+			 MLX5_GET(create_flow_table_in,  in, other_vport));
+		MLX5_SET(destroy_flow_table_in, din, vport_number,
+			 MLX5_GET(create_flow_table_in,  in, vport_number));
+		MLX5_SET(destroy_flow_table_in, din, table_type,
+			 MLX5_GET(create_flow_table_in,  in, table_type));
+		MLX5_SET(destroy_flow_table_in, din, table_id, *obj_id);
+		MLX5_SET(general_obj_in_cmd_hdr, din, opcode,
+			 MLX5_CMD_OP_DESTROY_FLOW_TABLE);
+		break;
+	case MLX5_CMD_OP_CREATE_FLOW_GROUP:
+		*dinlen = MLX5_ST_SZ_BYTES(destroy_flow_group_in);
+		*obj_id = MLX5_GET(create_flow_group_out, out, group_id);
+		MLX5_SET(destroy_flow_group_in, din, other_vport,
+			 MLX5_GET(create_flow_group_in, in, other_vport));
+		MLX5_SET(destroy_flow_group_in, din, vport_number,
+			 MLX5_GET(create_flow_group_in, in, vport_number));
+		MLX5_SET(destroy_flow_group_in, din, table_type,
+			 MLX5_GET(create_flow_group_in, in, table_type));
+		MLX5_SET(destroy_flow_group_in, din, table_id,
+			 MLX5_GET(create_flow_group_in, in, table_id));
+		MLX5_SET(destroy_flow_group_in, din, group_id, *obj_id);
+		MLX5_SET(general_obj_in_cmd_hdr, din, opcode,
+			 MLX5_CMD_OP_DESTROY_FLOW_GROUP);
+		break;
+	case MLX5_CMD_OP_SET_FLOW_TABLE_ENTRY:
+		*dinlen = MLX5_ST_SZ_BYTES(delete_fte_in);
+		*obj_id = MLX5_GET(set_fte_in, in, flow_index);
+		MLX5_SET(delete_fte_in, din, other_vport,
+			 MLX5_GET(set_fte_in,  in, other_vport));
+		MLX5_SET(delete_fte_in, din, vport_number,
+			 MLX5_GET(set_fte_in, in, vport_number));
+		MLX5_SET(delete_fte_in, din, table_type,
+			 MLX5_GET(set_fte_in, in, table_type));
+		MLX5_SET(delete_fte_in, din, table_id,
+			 MLX5_GET(set_fte_in, in, table_id));
+		MLX5_SET(delete_fte_in, din, flow_index, *obj_id);
+		MLX5_SET(general_obj_in_cmd_hdr, din, opcode,
+			 MLX5_CMD_OP_DELETE_FLOW_TABLE_ENTRY);
+		break;
+	case MLX5_CMD_OP_ALLOC_FLOW_COUNTER:
+		MLX5_SET(general_obj_in_cmd_hdr, din, opcode,
+			 MLX5_CMD_OP_DEALLOC_FLOW_COUNTER);
+		break;
+	case MLX5_CMD_OP_ALLOC_ENCAP_HEADER:
+		MLX5_SET(general_obj_in_cmd_hdr, din, opcode,
+			 MLX5_CMD_OP_DEALLOC_ENCAP_HEADER);
+		break;
+	case MLX5_CMD_OP_ALLOC_MODIFY_HEADER_CONTEXT:
+		MLX5_SET(general_obj_in_cmd_hdr, din, opcode,
+			 MLX5_CMD_OP_DEALLOC_MODIFY_HEADER_CONTEXT);
+		break;
+	case MLX5_CMD_OP_CREATE_SCHEDULING_ELEMENT:
+		*dinlen = MLX5_ST_SZ_BYTES(destroy_scheduling_element_in);
+		*obj_id = MLX5_GET(create_scheduling_element_out, out,
+				   scheduling_element_id);
+		MLX5_SET(destroy_scheduling_element_in, din,
+			 scheduling_hierarchy,
+			 MLX5_GET(create_scheduling_element_in, in,
+				  scheduling_hierarchy));
+		MLX5_SET(destroy_scheduling_element_in, din,
+			 scheduling_element_id, *obj_id);
+		MLX5_SET(general_obj_in_cmd_hdr, din, opcode,
+			 MLX5_CMD_OP_DESTROY_SCHEDULING_ELEMENT);
+		break;
+	case MLX5_CMD_OP_ADD_VXLAN_UDP_DPORT:
+		*dinlen = MLX5_ST_SZ_BYTES(delete_vxlan_udp_dport_in);
+		*obj_id = MLX5_GET(add_vxlan_udp_dport_in, in, vxlan_udp_port);
+		MLX5_SET(delete_vxlan_udp_dport_in, din, vxlan_udp_port, *obj_id);
+		MLX5_SET(general_obj_in_cmd_hdr, din, opcode,
+			 MLX5_CMD_OP_DELETE_VXLAN_UDP_DPORT);
+		break;
+	case MLX5_CMD_OP_SET_L2_TABLE_ENTRY:
+		*dinlen = MLX5_ST_SZ_BYTES(delete_l2_table_entry_in);
+		*obj_id = MLX5_GET(set_l2_table_entry_in, in, table_index);
+		MLX5_SET(delete_l2_table_entry_in, din, table_index, *obj_id);
+		MLX5_SET(general_obj_in_cmd_hdr, din, opcode,
+			 MLX5_CMD_OP_DELETE_L2_TABLE_ENTRY);
+		break;
+	case MLX5_CMD_OP_CREATE_QP:
+		MLX5_SET(general_obj_in_cmd_hdr, din, opcode, MLX5_CMD_OP_DESTROY_QP);
+		break;
+	case MLX5_CMD_OP_CREATE_SRQ:
+		MLX5_SET(general_obj_in_cmd_hdr, din, opcode, MLX5_CMD_OP_DESTROY_SRQ);
+		break;
+	case MLX5_CMD_OP_CREATE_XRC_SRQ:
+		MLX5_SET(general_obj_in_cmd_hdr, din, opcode,
+			 MLX5_CMD_OP_DESTROY_XRC_SRQ);
+		break;
+	case MLX5_CMD_OP_CREATE_DCT:
+		MLX5_SET(general_obj_in_cmd_hdr, din, opcode, MLX5_CMD_OP_DESTROY_DCT);
+		break;
+	case MLX5_CMD_OP_CREATE_XRQ:
+		MLX5_SET(general_obj_in_cmd_hdr, din, opcode, MLX5_CMD_OP_DESTROY_XRQ);
+		break;
+	case MLX5_CMD_OP_ATTACH_TO_MCG:
+		*dinlen = MLX5_ST_SZ_BYTES(detach_from_mcg_in);
+		MLX5_SET(detach_from_mcg_in, din, qpn,
+			 MLX5_GET(attach_to_mcg_in, in, qpn));
+		memcpy(MLX5_ADDR_OF(detach_from_mcg_in, din, multicast_gid),
+		       MLX5_ADDR_OF(attach_to_mcg_in, in, multicast_gid),
+		       MLX5_FLD_SZ_BYTES(attach_to_mcg_in, multicast_gid));
+		MLX5_SET(general_obj_in_cmd_hdr, din, opcode, MLX5_CMD_OP_DETACH_FROM_MCG);
+		break;
+	case MLX5_CMD_OP_ALLOC_XRCD:
+		MLX5_SET(general_obj_in_cmd_hdr, din, opcode, MLX5_CMD_OP_DEALLOC_XRCD);
+		break;
+	default:
+		/* The entry must match to one of the devx_is_obj_create_cmd */
+		WARN_ON(true);
+		break;
+	}
+}
+
+static int devx_obj_cleanup(struct ib_uobject *uobject,
+			    enum rdma_remove_reason why)
+{
+	u32 out[MLX5_ST_SZ_DW(general_obj_out_cmd_hdr)];
+	struct devx_obj *obj = uobject->object;
+	int ret;
+
+	ret = mlx5_cmd_exec(obj->mdev, obj->dinbox, obj->dinlen, out, sizeof(out));
+	if (ib_is_destroy_retryable(ret, why, uobject))
+		return ret;
+
+	kfree(obj);
+	return ret;
+}
+
+static int UVERBS_HANDLER(MLX5_IB_METHOD_DEVX_OBJ_CREATE)(
+	struct ib_uverbs_file *file, struct uverbs_attr_bundle *attrs)
+{
+	void *cmd_in = uverbs_attr_get_alloced_ptr(attrs, MLX5_IB_ATTR_DEVX_OBJ_CREATE_CMD_IN);
+	int cmd_out_len =  uverbs_attr_get_len(attrs,
+					MLX5_IB_ATTR_DEVX_OBJ_CREATE_CMD_OUT);
+	void *cmd_out;
+	struct ib_uobject *uobj = uverbs_attr_get_uobject(
+		attrs, MLX5_IB_ATTR_DEVX_OBJ_CREATE_HANDLE);
+	struct mlx5_ib_ucontext *c = to_mucontext(uobj->context);
+	struct mlx5_ib_dev *dev = to_mdev(c->ibucontext.device);
+	struct devx_obj *obj;
+	int err;
+
+	if (!c->devx_uid)
+		return -EPERM;
+
+	if (!devx_is_obj_create_cmd(cmd_in))
+		return -EINVAL;
+
+	cmd_out = uverbs_zalloc(attrs, cmd_out_len);
+	if (IS_ERR(cmd_out))
+		return PTR_ERR(cmd_out);
+
+	obj = kzalloc(sizeof(struct devx_obj), GFP_KERNEL);
+	if (!obj)
+		return -ENOMEM;
+
+	MLX5_SET(general_obj_in_cmd_hdr, cmd_in, uid, c->devx_uid);
+	err = mlx5_cmd_exec(dev->mdev, cmd_in,
+			    uverbs_attr_get_len(attrs, MLX5_IB_ATTR_DEVX_OBJ_CREATE_CMD_IN),
+			    cmd_out, cmd_out_len);
+	if (err)
+		goto obj_free;
+
+	uobj->object = obj;
+	obj->mdev = dev->mdev;
+	devx_obj_build_destroy_cmd(cmd_in, cmd_out, obj->dinbox, &obj->dinlen, &obj->obj_id);
+	WARN_ON(obj->dinlen > MLX5_MAX_DESTROY_INBOX_SIZE_DW * sizeof(u32));
+
+	err = uverbs_copy_to(attrs, MLX5_IB_ATTR_DEVX_OBJ_CREATE_CMD_OUT, cmd_out, cmd_out_len);
+	if (err)
+		goto obj_free;
+
+	return 0;
+
+obj_free:
+	kfree(obj);
+	return err;
+}
+
+static int UVERBS_HANDLER(MLX5_IB_METHOD_DEVX_OBJ_MODIFY)(
+	struct ib_uverbs_file *file, struct uverbs_attr_bundle *attrs)
+{
+	void *cmd_in = uverbs_attr_get_alloced_ptr(attrs, MLX5_IB_ATTR_DEVX_OBJ_MODIFY_CMD_IN);
+	int cmd_out_len = uverbs_attr_get_len(attrs,
+					MLX5_IB_ATTR_DEVX_OBJ_MODIFY_CMD_OUT);
+	struct ib_uobject *uobj = uverbs_attr_get_uobject(attrs,
+							  MLX5_IB_ATTR_DEVX_OBJ_MODIFY_HANDLE);
+	struct mlx5_ib_ucontext *c = to_mucontext(uobj->context);
+	struct devx_obj *obj = uobj->object;
+	void *cmd_out;
+	int err;
+
+	if (!c->devx_uid)
+		return -EPERM;
+
+	if (!devx_is_obj_modify_cmd(cmd_in))
+		return -EINVAL;
+
+	if (!devx_is_valid_obj_id(obj, cmd_in))
+		return -EINVAL;
+
+	cmd_out = uverbs_zalloc(attrs, cmd_out_len);
+	if (IS_ERR(cmd_out))
+		return PTR_ERR(cmd_out);
+
+	MLX5_SET(general_obj_in_cmd_hdr, cmd_in, uid, c->devx_uid);
+	err = mlx5_cmd_exec(obj->mdev, cmd_in,
+			    uverbs_attr_get_len(attrs, MLX5_IB_ATTR_DEVX_OBJ_MODIFY_CMD_IN),
+			    cmd_out, cmd_out_len);
+	if (err)
+		return err;
+
+	return uverbs_copy_to(attrs, MLX5_IB_ATTR_DEVX_OBJ_MODIFY_CMD_OUT,
+			      cmd_out, cmd_out_len);
+}
+
+static int UVERBS_HANDLER(MLX5_IB_METHOD_DEVX_OBJ_QUERY)(
+	struct ib_uverbs_file *file, struct uverbs_attr_bundle *attrs)
+{
+	void *cmd_in = uverbs_attr_get_alloced_ptr(attrs, MLX5_IB_ATTR_DEVX_OBJ_QUERY_CMD_IN);
+	int cmd_out_len = uverbs_attr_get_len(attrs,
+					      MLX5_IB_ATTR_DEVX_OBJ_QUERY_CMD_OUT);
+	struct ib_uobject *uobj = uverbs_attr_get_uobject(attrs,
+							  MLX5_IB_ATTR_DEVX_OBJ_QUERY_HANDLE);
+	struct mlx5_ib_ucontext *c = to_mucontext(uobj->context);
+	struct devx_obj *obj = uobj->object;
+	void *cmd_out;
+	int err;
+
+	if (!c->devx_uid)
+		return -EPERM;
+
+	if (!devx_is_obj_query_cmd(cmd_in))
+		return -EINVAL;
+
+	if (!devx_is_valid_obj_id(obj, cmd_in))
+		return -EINVAL;
+
+	cmd_out = uverbs_zalloc(attrs, cmd_out_len);
+	if (IS_ERR(cmd_out))
+		return PTR_ERR(cmd_out);
+
+	MLX5_SET(general_obj_in_cmd_hdr, cmd_in, uid, c->devx_uid);
+	err = mlx5_cmd_exec(obj->mdev, cmd_in,
+			    uverbs_attr_get_len(attrs, MLX5_IB_ATTR_DEVX_OBJ_QUERY_CMD_IN),
+			    cmd_out, cmd_out_len);
+	if (err)
+		return err;
+
+	return uverbs_copy_to(attrs, MLX5_IB_ATTR_DEVX_OBJ_QUERY_CMD_OUT,
+			      cmd_out, cmd_out_len);
+}
+
+static int devx_umem_get(struct mlx5_ib_dev *dev, struct ib_ucontext *ucontext,
+			 struct uverbs_attr_bundle *attrs,
+			 struct devx_umem *obj)
+{
+	u64 addr;
+	size_t size;
+	u32 access;
+	int npages;
+	int err;
+	u32 page_mask;
+
+	if (uverbs_copy_from(&addr, attrs, MLX5_IB_ATTR_DEVX_UMEM_REG_ADDR) ||
+	    uverbs_copy_from(&size, attrs, MLX5_IB_ATTR_DEVX_UMEM_REG_LEN))
+		return -EFAULT;
+
+	err = uverbs_get_flags32(&access, attrs,
+				 MLX5_IB_ATTR_DEVX_UMEM_REG_ACCESS,
+				 IB_ACCESS_SUPPORTED);
+	if (err)
+		return err;
+
+	err = ib_check_mr_access(access);
+	if (err)
+		return err;
+
+	obj->umem = ib_umem_get(ucontext, addr, size, access, 0);
+	if (IS_ERR(obj->umem))
+		return PTR_ERR(obj->umem);
+
+	mlx5_ib_cont_pages(obj->umem, obj->umem->address,
+			   MLX5_MKEY_PAGE_SHIFT_MASK, &npages,
+			   &obj->page_shift, &obj->ncont, NULL);
+
+	if (!npages) {
+		ib_umem_release(obj->umem);
+		return -EINVAL;
+	}
+
+	page_mask = (1 << obj->page_shift) - 1;
+	obj->page_offset = obj->umem->address & page_mask;
+
+	return 0;
+}
+
+static int devx_umem_reg_cmd_alloc(struct uverbs_attr_bundle *attrs,
+				   struct devx_umem *obj,
+				   struct devx_umem_reg_cmd *cmd)
+{
+	cmd->inlen = MLX5_ST_SZ_BYTES(create_umem_in) +
+		    (MLX5_ST_SZ_BYTES(mtt) * obj->ncont);
+	cmd->in = uverbs_zalloc(attrs, cmd->inlen);
+	return PTR_ERR_OR_ZERO(cmd->in);
+}
+
+static void devx_umem_reg_cmd_build(struct mlx5_ib_dev *dev,
+				    struct devx_umem *obj,
+				    struct devx_umem_reg_cmd *cmd)
+{
+	void *umem;
+	__be64 *mtt;
+
+	umem = MLX5_ADDR_OF(create_umem_in, cmd->in, umem);
+	mtt = (__be64 *)MLX5_ADDR_OF(umem, umem, mtt);
+
+	MLX5_SET(general_obj_in_cmd_hdr, cmd->in, opcode, MLX5_CMD_OP_CREATE_GENERAL_OBJECT);
+	MLX5_SET(general_obj_in_cmd_hdr, cmd->in, obj_type, MLX5_OBJ_TYPE_UMEM);
+	MLX5_SET64(umem, umem, num_of_mtt, obj->ncont);
+	MLX5_SET(umem, umem, log_page_size, obj->page_shift -
+					    MLX5_ADAPTER_PAGE_SHIFT);
+	MLX5_SET(umem, umem, page_offset, obj->page_offset);
+	mlx5_ib_populate_pas(dev, obj->umem, obj->page_shift, mtt,
+			     (obj->umem->writable ? MLX5_IB_MTT_WRITE : 0) |
+			     MLX5_IB_MTT_READ);
+}
+
+static int UVERBS_HANDLER(MLX5_IB_METHOD_DEVX_UMEM_REG)(
+	struct ib_uverbs_file *file, struct uverbs_attr_bundle *attrs)
+{
+	struct devx_umem_reg_cmd cmd;
+	struct devx_umem *obj;
+	struct ib_uobject *uobj = uverbs_attr_get_uobject(
+		attrs, MLX5_IB_ATTR_DEVX_UMEM_REG_HANDLE);
+	u32 obj_id;
+	struct mlx5_ib_ucontext *c = to_mucontext(uobj->context);
+	struct mlx5_ib_dev *dev = to_mdev(c->ibucontext.device);
+	int err;
+
+	if (!c->devx_uid)
+		return -EPERM;
+
+	obj = kzalloc(sizeof(struct devx_umem), GFP_KERNEL);
+	if (!obj)
+		return -ENOMEM;
+
+	err = devx_umem_get(dev, &c->ibucontext, attrs, obj);
+	if (err)
+		goto err_obj_free;
+
+	err = devx_umem_reg_cmd_alloc(attrs, obj, &cmd);
+	if (err)
+		goto err_umem_release;
+
+	devx_umem_reg_cmd_build(dev, obj, &cmd);
+
+	MLX5_SET(general_obj_in_cmd_hdr, cmd.in, uid, c->devx_uid);
+	err = mlx5_cmd_exec(dev->mdev, cmd.in, cmd.inlen, cmd.out,
+			    sizeof(cmd.out));
+	if (err)
+		goto err_umem_release;
+
+	obj->mdev = dev->mdev;
+	uobj->object = obj;
+	devx_obj_build_destroy_cmd(cmd.in, cmd.out, obj->dinbox, &obj->dinlen, &obj_id);
+	err = uverbs_copy_to(attrs, MLX5_IB_ATTR_DEVX_UMEM_REG_OUT_ID, &obj_id, sizeof(obj_id));
+	if (err)
+		goto err_umem_destroy;
+
+	return 0;
+
+err_umem_destroy:
+	mlx5_cmd_exec(obj->mdev, obj->dinbox, obj->dinlen, cmd.out, sizeof(cmd.out));
+err_umem_release:
+	ib_umem_release(obj->umem);
+err_obj_free:
+	kfree(obj);
+	return err;
+}
+
+static int devx_umem_cleanup(struct ib_uobject *uobject,
+			     enum rdma_remove_reason why)
+{
+	struct devx_umem *obj = uobject->object;
+	u32 out[MLX5_ST_SZ_DW(general_obj_out_cmd_hdr)];
+	int err;
+
+	err = mlx5_cmd_exec(obj->mdev, obj->dinbox, obj->dinlen, out, sizeof(out));
+	if (ib_is_destroy_retryable(err, why, uobject))
+		return err;
+
+	ib_umem_release(obj->umem);
+	kfree(obj);
+	return 0;
+}
+
+DECLARE_UVERBS_NAMED_METHOD(
+	MLX5_IB_METHOD_DEVX_UMEM_REG,
+	UVERBS_ATTR_IDR(MLX5_IB_ATTR_DEVX_UMEM_REG_HANDLE,
+			MLX5_IB_OBJECT_DEVX_UMEM,
+			UVERBS_ACCESS_NEW,
+			UA_MANDATORY),
+	UVERBS_ATTR_PTR_IN(MLX5_IB_ATTR_DEVX_UMEM_REG_ADDR,
+			   UVERBS_ATTR_TYPE(u64),
+			   UA_MANDATORY),
+	UVERBS_ATTR_PTR_IN(MLX5_IB_ATTR_DEVX_UMEM_REG_LEN,
+			   UVERBS_ATTR_TYPE(u64),
+			   UA_MANDATORY),
+	UVERBS_ATTR_FLAGS_IN(MLX5_IB_ATTR_DEVX_UMEM_REG_ACCESS,
+			     enum ib_access_flags),
+	UVERBS_ATTR_PTR_OUT(MLX5_IB_ATTR_DEVX_UMEM_REG_OUT_ID,
+			    UVERBS_ATTR_TYPE(u32),
+			    UA_MANDATORY));
+
+DECLARE_UVERBS_NAMED_METHOD_DESTROY(
+	MLX5_IB_METHOD_DEVX_UMEM_DEREG,
+	UVERBS_ATTR_IDR(MLX5_IB_ATTR_DEVX_UMEM_DEREG_HANDLE,
+			MLX5_IB_OBJECT_DEVX_UMEM,
+			UVERBS_ACCESS_DESTROY,
+			UA_MANDATORY));
+
+DECLARE_UVERBS_NAMED_METHOD(
+	MLX5_IB_METHOD_DEVX_QUERY_EQN,
+	UVERBS_ATTR_PTR_IN(MLX5_IB_ATTR_DEVX_QUERY_EQN_USER_VEC,
+			   UVERBS_ATTR_TYPE(u32),
+			   UA_MANDATORY),
+	UVERBS_ATTR_PTR_OUT(MLX5_IB_ATTR_DEVX_QUERY_EQN_DEV_EQN,
+			    UVERBS_ATTR_TYPE(u32),
+			    UA_MANDATORY));
+
+DECLARE_UVERBS_NAMED_METHOD(
+	MLX5_IB_METHOD_DEVX_QUERY_UAR,
+	UVERBS_ATTR_PTR_IN(MLX5_IB_ATTR_DEVX_QUERY_UAR_USER_IDX,
+			   UVERBS_ATTR_TYPE(u32),
+			   UA_MANDATORY),
+	UVERBS_ATTR_PTR_OUT(MLX5_IB_ATTR_DEVX_QUERY_UAR_DEV_IDX,
+			    UVERBS_ATTR_TYPE(u32),
+			    UA_MANDATORY));
+
+DECLARE_UVERBS_NAMED_METHOD(
+	MLX5_IB_METHOD_DEVX_OTHER,
+	UVERBS_ATTR_PTR_IN(
+		MLX5_IB_ATTR_DEVX_OTHER_CMD_IN,
+		UVERBS_ATTR_MIN_SIZE(MLX5_ST_SZ_BYTES(general_obj_in_cmd_hdr)),
+		UA_MANDATORY,
+		UA_ALLOC_AND_COPY),
+	UVERBS_ATTR_PTR_OUT(
+		MLX5_IB_ATTR_DEVX_OTHER_CMD_OUT,
+		UVERBS_ATTR_MIN_SIZE(MLX5_ST_SZ_BYTES(general_obj_out_cmd_hdr)),
+		UA_MANDATORY));
+
+DECLARE_UVERBS_NAMED_METHOD(
+	MLX5_IB_METHOD_DEVX_OBJ_CREATE,
+	UVERBS_ATTR_IDR(MLX5_IB_ATTR_DEVX_OBJ_CREATE_HANDLE,
+			MLX5_IB_OBJECT_DEVX_OBJ,
+			UVERBS_ACCESS_NEW,
+			UA_MANDATORY),
+	UVERBS_ATTR_PTR_IN(
+		MLX5_IB_ATTR_DEVX_OBJ_CREATE_CMD_IN,
+		UVERBS_ATTR_MIN_SIZE(MLX5_ST_SZ_BYTES(general_obj_in_cmd_hdr)),
+		UA_MANDATORY,
+		UA_ALLOC_AND_COPY),
+	UVERBS_ATTR_PTR_OUT(
+		MLX5_IB_ATTR_DEVX_OBJ_CREATE_CMD_OUT,
+		UVERBS_ATTR_MIN_SIZE(MLX5_ST_SZ_BYTES(general_obj_out_cmd_hdr)),
+		UA_MANDATORY));
+
+DECLARE_UVERBS_NAMED_METHOD_DESTROY(
+	MLX5_IB_METHOD_DEVX_OBJ_DESTROY,
+	UVERBS_ATTR_IDR(MLX5_IB_ATTR_DEVX_OBJ_DESTROY_HANDLE,
+			MLX5_IB_OBJECT_DEVX_OBJ,
+			UVERBS_ACCESS_DESTROY,
+			UA_MANDATORY));
+
+DECLARE_UVERBS_NAMED_METHOD(
+	MLX5_IB_METHOD_DEVX_OBJ_MODIFY,
+	UVERBS_ATTR_IDR(MLX5_IB_ATTR_DEVX_OBJ_MODIFY_HANDLE,
+			MLX5_IB_OBJECT_DEVX_OBJ,
+			UVERBS_ACCESS_WRITE,
+			UA_MANDATORY),
+	UVERBS_ATTR_PTR_IN(
+		MLX5_IB_ATTR_DEVX_OBJ_MODIFY_CMD_IN,
+		UVERBS_ATTR_MIN_SIZE(MLX5_ST_SZ_BYTES(general_obj_in_cmd_hdr)),
+		UA_MANDATORY,
+		UA_ALLOC_AND_COPY),
+	UVERBS_ATTR_PTR_OUT(
+		MLX5_IB_ATTR_DEVX_OBJ_MODIFY_CMD_OUT,
+		UVERBS_ATTR_MIN_SIZE(MLX5_ST_SZ_BYTES(general_obj_out_cmd_hdr)),
+		UA_MANDATORY));
+
+DECLARE_UVERBS_NAMED_METHOD(
+	MLX5_IB_METHOD_DEVX_OBJ_QUERY,
+	UVERBS_ATTR_IDR(MLX5_IB_ATTR_DEVX_OBJ_QUERY_HANDLE,
+			MLX5_IB_OBJECT_DEVX_OBJ,
+			UVERBS_ACCESS_READ,
+			UA_MANDATORY),
+	UVERBS_ATTR_PTR_IN(
+		MLX5_IB_ATTR_DEVX_OBJ_QUERY_CMD_IN,
+		UVERBS_ATTR_MIN_SIZE(MLX5_ST_SZ_BYTES(general_obj_in_cmd_hdr)),
+		UA_MANDATORY,
+		UA_ALLOC_AND_COPY),
+	UVERBS_ATTR_PTR_OUT(
+		MLX5_IB_ATTR_DEVX_OBJ_QUERY_CMD_OUT,
+		UVERBS_ATTR_MIN_SIZE(MLX5_ST_SZ_BYTES(general_obj_out_cmd_hdr)),
+		UA_MANDATORY));
+
+DECLARE_UVERBS_GLOBAL_METHODS(MLX5_IB_OBJECT_DEVX,
+			      &UVERBS_METHOD(MLX5_IB_METHOD_DEVX_OTHER),
+			      &UVERBS_METHOD(MLX5_IB_METHOD_DEVX_QUERY_UAR),
+			      &UVERBS_METHOD(MLX5_IB_METHOD_DEVX_QUERY_EQN));
+
+DECLARE_UVERBS_NAMED_OBJECT(MLX5_IB_OBJECT_DEVX_OBJ,
+			    UVERBS_TYPE_ALLOC_IDR(devx_obj_cleanup),
+			    &UVERBS_METHOD(MLX5_IB_METHOD_DEVX_OBJ_CREATE),
+			    &UVERBS_METHOD(MLX5_IB_METHOD_DEVX_OBJ_DESTROY),
+			    &UVERBS_METHOD(MLX5_IB_METHOD_DEVX_OBJ_MODIFY),
+			    &UVERBS_METHOD(MLX5_IB_METHOD_DEVX_OBJ_QUERY));
+
+DECLARE_UVERBS_NAMED_OBJECT(MLX5_IB_OBJECT_DEVX_UMEM,
+			    UVERBS_TYPE_ALLOC_IDR(devx_umem_cleanup),
+			    &UVERBS_METHOD(MLX5_IB_METHOD_DEVX_UMEM_REG),
+			    &UVERBS_METHOD(MLX5_IB_METHOD_DEVX_UMEM_DEREG));
+
+DECLARE_UVERBS_OBJECT_TREE(devx_objects,
+			   &UVERBS_OBJECT(MLX5_IB_OBJECT_DEVX),
+			   &UVERBS_OBJECT(MLX5_IB_OBJECT_DEVX_OBJ),
+			   &UVERBS_OBJECT(MLX5_IB_OBJECT_DEVX_UMEM));
+
+const struct uverbs_object_tree_def *mlx5_ib_get_devx_tree(void)
+{
+	return &devx_objects;
+}
diff --git a/drivers/infiniband/hw/mlx5/flow.c b/drivers/infiniband/hw/mlx5/flow.c
new file mode 100644
index 000000000000..1a29f47f836e
--- /dev/null
+++ b/drivers/infiniband/hw/mlx5/flow.c
@@ -0,0 +1,252 @@
+// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB
+/*
+ * Copyright (c) 2018, Mellanox Technologies inc.  All rights reserved.
+ */
+
+#include <rdma/ib_user_verbs.h>
+#include <rdma/ib_verbs.h>
+#include <rdma/uverbs_types.h>
+#include <rdma/uverbs_ioctl.h>
+#include <rdma/mlx5_user_ioctl_cmds.h>
+#include <rdma/ib_umem.h>
+#include <linux/mlx5/driver.h>
+#include <linux/mlx5/fs.h>
+#include "mlx5_ib.h"
+
+#define UVERBS_MODULE_NAME mlx5_ib
+#include <rdma/uverbs_named_ioctl.h>
+
+static const struct uverbs_attr_spec mlx5_ib_flow_type[] = {
+	[MLX5_IB_FLOW_TYPE_NORMAL] = {
+		.type = UVERBS_ATTR_TYPE_PTR_IN,
+		.u.ptr = {
+			.len = sizeof(u16), /* data is priority */
+			.min_len = sizeof(u16),
+		}
+	},
+	[MLX5_IB_FLOW_TYPE_SNIFFER] = {
+		.type = UVERBS_ATTR_TYPE_PTR_IN,
+		UVERBS_ATTR_NO_DATA(),
+	},
+	[MLX5_IB_FLOW_TYPE_ALL_DEFAULT] = {
+		.type = UVERBS_ATTR_TYPE_PTR_IN,
+		UVERBS_ATTR_NO_DATA(),
+	},
+	[MLX5_IB_FLOW_TYPE_MC_DEFAULT] = {
+		.type = UVERBS_ATTR_TYPE_PTR_IN,
+		UVERBS_ATTR_NO_DATA(),
+	},
+};
+
+static int UVERBS_HANDLER(MLX5_IB_METHOD_CREATE_FLOW)(
+	struct ib_uverbs_file *file, struct uverbs_attr_bundle *attrs)
+{
+	struct mlx5_ib_flow_handler *flow_handler;
+	struct mlx5_ib_flow_matcher *fs_matcher;
+	void *devx_obj;
+	int dest_id, dest_type;
+	void *cmd_in;
+	int inlen;
+	bool dest_devx, dest_qp;
+	struct ib_qp *qp = NULL;
+	struct ib_uobject *uobj =
+		uverbs_attr_get_uobject(attrs, MLX5_IB_ATTR_CREATE_FLOW_HANDLE);
+	struct mlx5_ib_dev *dev = to_mdev(uobj->context->device);
+
+	if (!capable(CAP_NET_RAW))
+		return -EPERM;
+
+	dest_devx =
+		uverbs_attr_is_valid(attrs, MLX5_IB_ATTR_CREATE_FLOW_DEST_DEVX);
+	dest_qp = uverbs_attr_is_valid(attrs,
+				       MLX5_IB_ATTR_CREATE_FLOW_DEST_QP);
+
+	if ((dest_devx && dest_qp) || (!dest_devx && !dest_qp))
+		return -EINVAL;
+
+	if (dest_devx) {
+		devx_obj = uverbs_attr_get_obj(
+			attrs, MLX5_IB_ATTR_CREATE_FLOW_DEST_DEVX);
+		if (IS_ERR(devx_obj))
+			return PTR_ERR(devx_obj);
+
+		/* Verify that the given DEVX object is a flow
+		 * steering destination.
+		 */
+		if (!mlx5_ib_devx_is_flow_dest(devx_obj, &dest_id, &dest_type))
+			return -EINVAL;
+	} else {
+		struct mlx5_ib_qp *mqp;
+
+		qp = uverbs_attr_get_obj(attrs,
+					 MLX5_IB_ATTR_CREATE_FLOW_DEST_QP);
+		if (IS_ERR(qp))
+			return PTR_ERR(qp);
+
+		if (qp->qp_type != IB_QPT_RAW_PACKET)
+			return -EINVAL;
+
+		mqp = to_mqp(qp);
+		if (mqp->flags & MLX5_IB_QP_RSS)
+			dest_id = mqp->rss_qp.tirn;
+		else
+			dest_id = mqp->raw_packet_qp.rq.tirn;
+		dest_type = MLX5_FLOW_DESTINATION_TYPE_TIR;
+	}
+
+	if (dev->rep)
+		return -ENOTSUPP;
+
+	cmd_in = uverbs_attr_get_alloced_ptr(
+		attrs, MLX5_IB_ATTR_CREATE_FLOW_MATCH_VALUE);
+	inlen = uverbs_attr_get_len(attrs,
+				    MLX5_IB_ATTR_CREATE_FLOW_MATCH_VALUE);
+	fs_matcher = uverbs_attr_get_obj(attrs,
+					 MLX5_IB_ATTR_CREATE_FLOW_MATCHER);
+	flow_handler = mlx5_ib_raw_fs_rule_add(dev, fs_matcher, cmd_in, inlen,
+					       dest_id, dest_type);
+	if (IS_ERR(flow_handler))
+		return PTR_ERR(flow_handler);
+
+	ib_set_flow(uobj, &flow_handler->ibflow, qp, &dev->ib_dev);
+
+	return 0;
+}
+
+static int flow_matcher_cleanup(struct ib_uobject *uobject,
+				enum rdma_remove_reason why)
+{
+	struct mlx5_ib_flow_matcher *obj = uobject->object;
+	int ret;
+
+	ret = ib_destroy_usecnt(&obj->usecnt, why, uobject);
+	if (ret)
+		return ret;
+
+	kfree(obj);
+	return 0;
+}
+
+static int UVERBS_HANDLER(MLX5_IB_METHOD_FLOW_MATCHER_CREATE)(
+	struct ib_uverbs_file *file, struct uverbs_attr_bundle *attrs)
+{
+	struct ib_uobject *uobj = uverbs_attr_get_uobject(
+		attrs, MLX5_IB_ATTR_FLOW_MATCHER_CREATE_HANDLE);
+	struct mlx5_ib_dev *dev = to_mdev(uobj->context->device);
+	struct mlx5_ib_flow_matcher *obj;
+	int err;
+
+	obj = kzalloc(sizeof(struct mlx5_ib_flow_matcher), GFP_KERNEL);
+	if (!obj)
+		return -ENOMEM;
+
+	obj->mask_len = uverbs_attr_get_len(
+		attrs, MLX5_IB_ATTR_FLOW_MATCHER_MATCH_MASK);
+	err = uverbs_copy_from(&obj->matcher_mask,
+			       attrs,
+			       MLX5_IB_ATTR_FLOW_MATCHER_MATCH_MASK);
+	if (err)
+		goto end;
+
+	obj->flow_type = uverbs_attr_get_enum_id(
+		attrs, MLX5_IB_ATTR_FLOW_MATCHER_FLOW_TYPE);
+
+	if (obj->flow_type == MLX5_IB_FLOW_TYPE_NORMAL) {
+		err = uverbs_copy_from(&obj->priority,
+				       attrs,
+				       MLX5_IB_ATTR_FLOW_MATCHER_FLOW_TYPE);
+		if (err)
+			goto end;
+	}
+
+	err = uverbs_copy_from(&obj->match_criteria_enable,
+			       attrs,
+			       MLX5_IB_ATTR_FLOW_MATCHER_MATCH_CRITERIA);
+	if (err)
+		goto end;
+
+	uobj->object = obj;
+	obj->mdev = dev->mdev;
+	atomic_set(&obj->usecnt, 0);
+	return 0;
+
+end:
+	kfree(obj);
+	return err;
+}
+
+DECLARE_UVERBS_NAMED_METHOD(
+	MLX5_IB_METHOD_CREATE_FLOW,
+	UVERBS_ATTR_IDR(MLX5_IB_ATTR_CREATE_FLOW_HANDLE,
+			UVERBS_OBJECT_FLOW,
+			UVERBS_ACCESS_NEW,
+			UA_MANDATORY),
+	UVERBS_ATTR_PTR_IN(
+		MLX5_IB_ATTR_CREATE_FLOW_MATCH_VALUE,
+		UVERBS_ATTR_SIZE(1, sizeof(struct mlx5_ib_match_params)),
+		UA_MANDATORY,
+		UA_ALLOC_AND_COPY),
+	UVERBS_ATTR_IDR(MLX5_IB_ATTR_CREATE_FLOW_MATCHER,
+			MLX5_IB_OBJECT_FLOW_MATCHER,
+			UVERBS_ACCESS_READ,
+			UA_MANDATORY),
+	UVERBS_ATTR_IDR(MLX5_IB_ATTR_CREATE_FLOW_DEST_QP,
+			UVERBS_OBJECT_QP,
+			UVERBS_ACCESS_READ),
+	UVERBS_ATTR_IDR(MLX5_IB_ATTR_CREATE_FLOW_DEST_DEVX,
+			MLX5_IB_OBJECT_DEVX_OBJ,
+			UVERBS_ACCESS_READ));
+
+DECLARE_UVERBS_NAMED_METHOD_DESTROY(
+	MLX5_IB_METHOD_DESTROY_FLOW,
+	UVERBS_ATTR_IDR(MLX5_IB_ATTR_CREATE_FLOW_HANDLE,
+			UVERBS_OBJECT_FLOW,
+			UVERBS_ACCESS_DESTROY,
+			UA_MANDATORY));
+
+ADD_UVERBS_METHODS(mlx5_ib_fs,
+		   UVERBS_OBJECT_FLOW,
+		   &UVERBS_METHOD(MLX5_IB_METHOD_CREATE_FLOW),
+		   &UVERBS_METHOD(MLX5_IB_METHOD_DESTROY_FLOW));
+
+DECLARE_UVERBS_NAMED_METHOD(
+	MLX5_IB_METHOD_FLOW_MATCHER_CREATE,
+	UVERBS_ATTR_IDR(MLX5_IB_ATTR_FLOW_MATCHER_CREATE_HANDLE,
+			MLX5_IB_OBJECT_FLOW_MATCHER,
+			UVERBS_ACCESS_NEW,
+			UA_MANDATORY),
+	UVERBS_ATTR_PTR_IN(
+		MLX5_IB_ATTR_FLOW_MATCHER_MATCH_MASK,
+		UVERBS_ATTR_SIZE(1, sizeof(struct mlx5_ib_match_params)),
+		UA_MANDATORY),
+	UVERBS_ATTR_ENUM_IN(MLX5_IB_ATTR_FLOW_MATCHER_FLOW_TYPE,
+			    mlx5_ib_flow_type,
+			    UA_MANDATORY),
+	UVERBS_ATTR_PTR_IN(MLX5_IB_ATTR_FLOW_MATCHER_MATCH_CRITERIA,
+			   UVERBS_ATTR_TYPE(u8),
+			   UA_MANDATORY));
+
+DECLARE_UVERBS_NAMED_METHOD_DESTROY(
+	MLX5_IB_METHOD_FLOW_MATCHER_DESTROY,
+	UVERBS_ATTR_IDR(MLX5_IB_ATTR_FLOW_MATCHER_DESTROY_HANDLE,
+			MLX5_IB_OBJECT_FLOW_MATCHER,
+			UVERBS_ACCESS_DESTROY,
+			UA_MANDATORY));
+
+DECLARE_UVERBS_NAMED_OBJECT(MLX5_IB_OBJECT_FLOW_MATCHER,
+			    UVERBS_TYPE_ALLOC_IDR(flow_matcher_cleanup),
+			    &UVERBS_METHOD(MLX5_IB_METHOD_FLOW_MATCHER_CREATE),
+			    &UVERBS_METHOD(MLX5_IB_METHOD_FLOW_MATCHER_DESTROY));
+
+DECLARE_UVERBS_OBJECT_TREE(flow_objects,
+			   &UVERBS_OBJECT(MLX5_IB_OBJECT_FLOW_MATCHER));
+
+int mlx5_ib_get_flow_trees(const struct uverbs_object_tree_def **root)
+{
+	int i = 0;
+
+	root[i++] = &flow_objects;
+	root[i++] = &mlx5_ib_fs;
+
+	return i;
+}
diff --git a/drivers/infiniband/hw/mlx5/gsi.c b/drivers/infiniband/hw/mlx5/gsi.c
index 79e6309460dc..4950df3f71b6 100644
--- a/drivers/infiniband/hw/mlx5/gsi.c
+++ b/drivers/infiniband/hw/mlx5/gsi.c
@@ -477,8 +477,8 @@ static struct ib_qp *get_tx_qp(struct mlx5_ib_gsi_qp *gsi, struct ib_ud_wr *wr)
 	return gsi->tx_qps[qp_index];
 }
 
-int mlx5_ib_gsi_post_send(struct ib_qp *qp, struct ib_send_wr *wr,
-			  struct ib_send_wr **bad_wr)
+int mlx5_ib_gsi_post_send(struct ib_qp *qp, const struct ib_send_wr *wr,
+			  const struct ib_send_wr **bad_wr)
 {
 	struct mlx5_ib_gsi_qp *gsi = gsi_qp(qp);
 	struct ib_qp *tx_qp;
@@ -522,8 +522,8 @@ err:
 	return ret;
 }
 
-int mlx5_ib_gsi_post_recv(struct ib_qp *qp, struct ib_recv_wr *wr,
-			  struct ib_recv_wr **bad_wr)
+int mlx5_ib_gsi_post_recv(struct ib_qp *qp, const struct ib_recv_wr *wr,
+			  const struct ib_recv_wr **bad_wr)
 {
 	struct mlx5_ib_gsi_qp *gsi = gsi_qp(qp);
 
diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c
index b3ba9a222550..c414f3809e5c 100644
--- a/drivers/infiniband/hw/mlx5/main.c
+++ b/drivers/infiniband/hw/mlx5/main.c
@@ -419,8 +419,8 @@ static int mlx5_query_port_roce(struct ib_device *device, u8 port_num,
 	translate_eth_proto_oper(eth_prot_oper, &props->active_speed,
 				 &props->active_width);
 
-	props->port_cap_flags  |= IB_PORT_CM_SUP;
-	props->port_cap_flags  |= IB_PORT_IP_BASED_GIDS;
+	props->port_cap_flags |= IB_PORT_CM_SUP;
+	props->ip_gids = true;
 
 	props->gid_tbl_len      = MLX5_CAP_ROCE(dev->mdev,
 						roce_address_table_size);
@@ -510,12 +510,11 @@ static int set_roce_addr(struct mlx5_ib_dev *dev, u8 port_num,
 				      vlan_id, port_num);
 }
 
-static int mlx5_ib_add_gid(const union ib_gid *gid,
-			   const struct ib_gid_attr *attr,
+static int mlx5_ib_add_gid(const struct ib_gid_attr *attr,
 			   __always_unused void **context)
 {
 	return set_roce_addr(to_mdev(attr->device), attr->port_num,
-			     attr->index, gid, attr);
+			     attr->index, &attr->gid, attr);
 }
 
 static int mlx5_ib_del_gid(const struct ib_gid_attr *attr,
@@ -525,41 +524,15 @@ static int mlx5_ib_del_gid(const struct ib_gid_attr *attr,
 			     attr->index, NULL, NULL);
 }
 
-__be16 mlx5_get_roce_udp_sport(struct mlx5_ib_dev *dev, u8 port_num,
-			       int index)
+__be16 mlx5_get_roce_udp_sport(struct mlx5_ib_dev *dev,
+			       const struct ib_gid_attr *attr)
 {
-	struct ib_gid_attr attr;
-	union ib_gid gid;
-
-	if (ib_get_cached_gid(&dev->ib_dev, port_num, index, &gid, &attr))
-		return 0;
-
-	dev_put(attr.ndev);
-
-	if (attr.gid_type != IB_GID_TYPE_ROCE_UDP_ENCAP)
+	if (attr->gid_type != IB_GID_TYPE_ROCE_UDP_ENCAP)
 		return 0;
 
 	return cpu_to_be16(MLX5_CAP_ROCE(dev->mdev, r_roce_min_src_udp_port));
 }
 
-int mlx5_get_roce_gid_type(struct mlx5_ib_dev *dev, u8 port_num,
-			   int index, enum ib_gid_type *gid_type)
-{
-	struct ib_gid_attr attr;
-	union ib_gid gid;
-	int ret;
-
-	ret = ib_get_cached_gid(&dev->ib_dev, port_num, index, &gid, &attr);
-	if (ret)
-		return ret;
-
-	dev_put(attr.ndev);
-
-	*gid_type = attr.gid_type;
-
-	return 0;
-}
-
 static int mlx5_use_mad_ifc(struct mlx5_ib_dev *dev)
 {
 	if (MLX5_CAP_GEN(dev->mdev, port_type) == MLX5_CAP_PORT_TYPE_IB)
@@ -915,7 +888,8 @@ static int mlx5_ib_query_device(struct ib_device *ibdev,
 	max_sq_sg = (max_sq_desc - sizeof(struct mlx5_wqe_ctrl_seg) -
 		     sizeof(struct mlx5_wqe_raddr_seg)) /
 		sizeof(struct mlx5_wqe_data_seg);
-	props->max_sge = min(max_rq_sg, max_sq_sg);
+	props->max_send_sge = max_sq_sg;
+	props->max_recv_sge = max_rq_sg;
 	props->max_sge_rd	   = MLX5_MAX_SGE_RD;
 	props->max_cq		   = 1 << MLX5_CAP_GEN(mdev, log_max_cq);
 	props->max_cqe = (1 << MLX5_CAP_GEN(mdev, log_max_cq_sz)) - 1;
@@ -1246,7 +1220,6 @@ static int mlx5_query_hca_port(struct ib_device *ibdev, u8 port,
 	props->qkey_viol_cntr	= rep->qkey_violation_counter;
 	props->subnet_timeout	= rep->subnet_timeout;
 	props->init_type_reply	= rep->init_type_reply;
-	props->grh_required	= rep->grh_required;
 
 	err = mlx5_query_port_link_width_oper(mdev, &ib_link_width_oper, port);
 	if (err)
@@ -1585,31 +1558,26 @@ error:
 	return err;
 }
 
-static int deallocate_uars(struct mlx5_ib_dev *dev, struct mlx5_ib_ucontext *context)
+static void deallocate_uars(struct mlx5_ib_dev *dev,
+			    struct mlx5_ib_ucontext *context)
 {
 	struct mlx5_bfreg_info *bfregi;
-	int err;
 	int i;
 
 	bfregi = &context->bfregi;
-	for (i = 0; i < bfregi->num_sys_pages; i++) {
+	for (i = 0; i < bfregi->num_sys_pages; i++)
 		if (i < bfregi->num_static_sys_pages ||
-		    bfregi->sys_pages[i] != MLX5_IB_INVALID_UAR_INDEX) {
-			err = mlx5_cmd_free_uar(dev->mdev, bfregi->sys_pages[i]);
-			if (err) {
-				mlx5_ib_warn(dev, "failed to free uar %d, err=%d\n", i, err);
-				return err;
-			}
-		}
-	}
-
-	return 0;
+		    bfregi->sys_pages[i] != MLX5_IB_INVALID_UAR_INDEX)
+			mlx5_cmd_free_uar(dev->mdev, bfregi->sys_pages[i]);
 }
 
 static int mlx5_ib_alloc_transport_domain(struct mlx5_ib_dev *dev, u32 *tdn)
 {
 	int err;
 
+	if (!MLX5_CAP_GEN(dev->mdev, log_max_transport_domain))
+		return 0;
+
 	err = mlx5_core_alloc_transport_domain(dev->mdev, tdn);
 	if (err)
 		return err;
@@ -1631,6 +1599,9 @@ static int mlx5_ib_alloc_transport_domain(struct mlx5_ib_dev *dev, u32 *tdn)
 
 static void mlx5_ib_dealloc_transport_domain(struct mlx5_ib_dev *dev, u32 tdn)
 {
+	if (!MLX5_CAP_GEN(dev->mdev, log_max_transport_domain))
+		return;
+
 	mlx5_core_dealloc_transport_domain(dev->mdev, tdn);
 
 	if ((MLX5_CAP_GEN(dev->mdev, port_type) != MLX5_CAP_PORT_TYPE_ETH) ||
@@ -1660,6 +1631,7 @@ static struct ib_ucontext *mlx5_ib_alloc_ucontext(struct ib_device *ibdev,
 	int err;
 	size_t min_req_v2 = offsetof(struct mlx5_ib_alloc_ucontext_req_v2,
 				     max_cqe_version);
+	u32 dump_fill_mkey;
 	bool lib_uar_4k;
 
 	if (!dev->ib_active)
@@ -1676,8 +1648,8 @@ static struct ib_ucontext *mlx5_ib_alloc_ucontext(struct ib_device *ibdev,
 	if (err)
 		return ERR_PTR(err);
 
-	if (req.flags)
-		return ERR_PTR(-EINVAL);
+	if (req.flags & ~MLX5_IB_ALLOC_UCTX_DEVX)
+		return ERR_PTR(-EOPNOTSUPP);
 
 	if (req.comp_mask || req.reserved0 || req.reserved1 || req.reserved2)
 		return ERR_PTR(-EOPNOTSUPP);
@@ -1755,10 +1727,26 @@ static struct ib_ucontext *mlx5_ib_alloc_ucontext(struct ib_device *ibdev,
 	context->ibucontext.invalidate_range = &mlx5_ib_invalidate_range;
 #endif
 
-	if (MLX5_CAP_GEN(dev->mdev, log_max_transport_domain)) {
-		err = mlx5_ib_alloc_transport_domain(dev, &context->tdn);
+	err = mlx5_ib_alloc_transport_domain(dev, &context->tdn);
+	if (err)
+		goto out_uars;
+
+	if (req.flags & MLX5_IB_ALLOC_UCTX_DEVX) {
+		/* Block DEVX on Infiniband as of SELinux */
+		if (mlx5_ib_port_link_layer(ibdev, 1) != IB_LINK_LAYER_ETHERNET) {
+			err = -EPERM;
+			goto out_td;
+		}
+
+		err = mlx5_ib_devx_create(dev, context);
 		if (err)
-			goto out_uars;
+			goto out_td;
+	}
+
+	if (MLX5_CAP_GEN(dev->mdev, dump_fill_mkey)) {
+		err = mlx5_cmd_dump_fill_mkey(dev->mdev, &dump_fill_mkey);
+		if (err)
+			goto out_mdev;
 	}
 
 	INIT_LIST_HEAD(&context->vma_private_list);
@@ -1819,9 +1807,18 @@ static struct ib_ucontext *mlx5_ib_alloc_ucontext(struct ib_device *ibdev,
 		resp.response_length += sizeof(resp.num_dyn_bfregs);
 	}
 
+	if (field_avail(typeof(resp), dump_fill_mkey, udata->outlen)) {
+		if (MLX5_CAP_GEN(dev->mdev, dump_fill_mkey)) {
+			resp.dump_fill_mkey = dump_fill_mkey;
+			resp.comp_mask |=
+				MLX5_IB_ALLOC_UCONTEXT_RESP_MASK_DUMP_FILL_MKEY;
+		}
+		resp.response_length += sizeof(resp.dump_fill_mkey);
+	}
+
 	err = ib_copy_to_udata(udata, &resp, resp.response_length);
 	if (err)
-		goto out_td;
+		goto out_mdev;
 
 	bfregi->ver = ver;
 	bfregi->num_low_latency_bfregs = req.num_low_latency_bfregs;
@@ -1831,9 +1828,11 @@ static struct ib_ucontext *mlx5_ib_alloc_ucontext(struct ib_device *ibdev,
 
 	return &context->ibucontext;
 
+out_mdev:
+	if (req.flags & MLX5_IB_ALLOC_UCTX_DEVX)
+		mlx5_ib_devx_destroy(dev, context);
 out_td:
-	if (MLX5_CAP_GEN(dev->mdev, log_max_transport_domain))
-		mlx5_ib_dealloc_transport_domain(dev, context->tdn);
+	mlx5_ib_dealloc_transport_domain(dev, context->tdn);
 
 out_uars:
 	deallocate_uars(dev, context);
@@ -1856,9 +1855,11 @@ static int mlx5_ib_dealloc_ucontext(struct ib_ucontext *ibcontext)
 	struct mlx5_ib_dev *dev = to_mdev(ibcontext->device);
 	struct mlx5_bfreg_info *bfregi;
 
+	if (context->devx_uid)
+		mlx5_ib_devx_destroy(dev, context);
+
 	bfregi = &context->bfregi;
-	if (MLX5_CAP_GEN(dev->mdev, log_max_transport_domain))
-		mlx5_ib_dealloc_transport_domain(dev, context->tdn);
+	mlx5_ib_dealloc_transport_domain(dev, context->tdn);
 
 	deallocate_uars(dev, context);
 	kfree(bfregi->sys_pages);
@@ -2040,7 +2041,7 @@ static int uar_mmap(struct mlx5_ib_dev *dev, enum mlx5_ib_mmap_cmd cmd,
 	struct mlx5_bfreg_info *bfregi = &context->bfregi;
 	int err;
 	unsigned long idx;
-	phys_addr_t pfn, pa;
+	phys_addr_t pfn;
 	pgprot_t prot;
 	u32 bfreg_dyn_idx = 0;
 	u32 uar_index;
@@ -2131,8 +2132,6 @@ static int uar_mmap(struct mlx5_ib_dev *dev, enum mlx5_ib_mmap_cmd cmd,
 		goto err;
 	}
 
-	pa = pfn << PAGE_SHIFT;
-
 	err = mlx5_ib_set_vma_data(vma, context);
 	if (err)
 		goto err;
@@ -2699,7 +2698,7 @@ static int parse_flow_attr(struct mlx5_core_dev *mdev, u32 *match_c,
 			 IPPROTO_GRE);
 
 		MLX5_SET(fte_match_set_misc, misc_params_c, gre_protocol,
-			 0xffff);
+			 ntohs(ib_spec->gre.mask.protocol));
 		MLX5_SET(fte_match_set_misc, misc_params_v, gre_protocol,
 			 ntohs(ib_spec->gre.val.protocol));
 
@@ -2979,11 +2978,11 @@ static void counters_clear_description(struct ib_counters *counters)
 
 static int mlx5_ib_destroy_flow(struct ib_flow *flow_id)
 {
-	struct mlx5_ib_dev *dev = to_mdev(flow_id->qp->device);
 	struct mlx5_ib_flow_handler *handler = container_of(flow_id,
 							  struct mlx5_ib_flow_handler,
 							  ibflow);
 	struct mlx5_ib_flow_handler *iter, *tmp;
+	struct mlx5_ib_dev *dev = handler->dev;
 
 	mutex_lock(&dev->flow_db->lock);
 
@@ -3001,6 +3000,8 @@ static int mlx5_ib_destroy_flow(struct ib_flow *flow_id)
 		counters_clear_description(handler->ibcounters);
 
 	mutex_unlock(&dev->flow_db->lock);
+	if (handler->flow_matcher)
+		atomic_dec(&handler->flow_matcher->usecnt);
 	kfree(handler);
 
 	return 0;
@@ -3021,6 +3022,26 @@ enum flow_table_type {
 
 #define MLX5_FS_MAX_TYPES	 6
 #define MLX5_FS_MAX_ENTRIES	 BIT(16)
+
+static struct mlx5_ib_flow_prio *_get_prio(struct mlx5_flow_namespace *ns,
+					   struct mlx5_ib_flow_prio *prio,
+					   int priority,
+					   int num_entries, int num_groups)
+{
+	struct mlx5_flow_table *ft;
+
+	ft = mlx5_create_auto_grouped_flow_table(ns, priority,
+						 num_entries,
+						 num_groups,
+						 0, 0);
+	if (IS_ERR(ft))
+		return ERR_CAST(ft);
+
+	prio->flow_table = ft;
+	prio->refcount = 0;
+	return prio;
+}
+
 static struct mlx5_ib_flow_prio *get_flow_table(struct mlx5_ib_dev *dev,
 						struct ib_flow_attr *flow_attr,
 						enum flow_table_type ft_type)
@@ -3033,7 +3054,6 @@ static struct mlx5_ib_flow_prio *get_flow_table(struct mlx5_ib_dev *dev,
 	int num_entries;
 	int num_groups;
 	int priority;
-	int err = 0;
 
 	max_table_size = BIT(MLX5_CAP_FLOWTABLE_NIC_RX(dev->mdev,
 						       log_max_ft_size));
@@ -3083,21 +3103,10 @@ static struct mlx5_ib_flow_prio *get_flow_table(struct mlx5_ib_dev *dev,
 		return ERR_PTR(-ENOMEM);
 
 	ft = prio->flow_table;
-	if (!ft) {
-		ft = mlx5_create_auto_grouped_flow_table(ns, priority,
-							 num_entries,
-							 num_groups,
-							 0, 0);
-
-		if (!IS_ERR(ft)) {
-			prio->refcount = 0;
-			prio->flow_table = ft;
-		} else {
-			err = PTR_ERR(ft);
-		}
-	}
+	if (!ft)
+		return _get_prio(ns, prio, priority, num_entries, num_groups);
 
-	return err ? ERR_PTR(err) : prio;
+	return prio;
 }
 
 static void set_underlay_qp(struct mlx5_ib_dev *dev,
@@ -3356,6 +3365,7 @@ static struct mlx5_ib_flow_handler *_create_flow_rule(struct mlx5_ib_dev *dev,
 
 	ft_prio->refcount++;
 	handler->prio = ft_prio;
+	handler->dev = dev;
 
 	ft_prio->flow_table = ft;
 free:
@@ -3648,6 +3658,189 @@ free_ucmd:
 	return ERR_PTR(err);
 }
 
+static struct mlx5_ib_flow_prio *_get_flow_table(struct mlx5_ib_dev *dev,
+						 int priority, bool mcast)
+{
+	int max_table_size;
+	struct mlx5_flow_namespace *ns = NULL;
+	struct mlx5_ib_flow_prio *prio;
+
+	max_table_size = BIT(MLX5_CAP_FLOWTABLE_NIC_RX(dev->mdev,
+			     log_max_ft_size));
+	if (max_table_size < MLX5_FS_MAX_ENTRIES)
+		return ERR_PTR(-ENOMEM);
+
+	if (mcast)
+		priority = MLX5_IB_FLOW_MCAST_PRIO;
+	else
+		priority = ib_prio_to_core_prio(priority, false);
+
+	ns = mlx5_get_flow_namespace(dev->mdev, MLX5_FLOW_NAMESPACE_BYPASS);
+	if (!ns)
+		return ERR_PTR(-ENOTSUPP);
+
+	prio = &dev->flow_db->prios[priority];
+
+	if (prio->flow_table)
+		return prio;
+
+	return _get_prio(ns, prio, priority, MLX5_FS_MAX_ENTRIES,
+			 MLX5_FS_MAX_TYPES);
+}
+
+static struct mlx5_ib_flow_handler *
+_create_raw_flow_rule(struct mlx5_ib_dev *dev,
+		      struct mlx5_ib_flow_prio *ft_prio,
+		      struct mlx5_flow_destination *dst,
+		      struct mlx5_ib_flow_matcher  *fs_matcher,
+		      void *cmd_in, int inlen)
+{
+	struct mlx5_ib_flow_handler *handler;
+	struct mlx5_flow_act flow_act = {.flow_tag = MLX5_FS_DEFAULT_FLOW_TAG};
+	struct mlx5_flow_spec *spec;
+	struct mlx5_flow_table *ft = ft_prio->flow_table;
+	int err = 0;
+
+	spec = kvzalloc(sizeof(*spec), GFP_KERNEL);
+	handler = kzalloc(sizeof(*handler), GFP_KERNEL);
+	if (!handler || !spec) {
+		err = -ENOMEM;
+		goto free;
+	}
+
+	INIT_LIST_HEAD(&handler->list);
+
+	memcpy(spec->match_value, cmd_in, inlen);
+	memcpy(spec->match_criteria, fs_matcher->matcher_mask.match_params,
+	       fs_matcher->mask_len);
+	spec->match_criteria_enable = fs_matcher->match_criteria_enable;
+
+	flow_act.action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;
+	handler->rule = mlx5_add_flow_rules(ft, spec,
+					    &flow_act, dst, 1);
+
+	if (IS_ERR(handler->rule)) {
+		err = PTR_ERR(handler->rule);
+		goto free;
+	}
+
+	ft_prio->refcount++;
+	handler->prio = ft_prio;
+	handler->dev = dev;
+	ft_prio->flow_table = ft;
+
+free:
+	if (err)
+		kfree(handler);
+	kvfree(spec);
+	return err ? ERR_PTR(err) : handler;
+}
+
+static bool raw_fs_is_multicast(struct mlx5_ib_flow_matcher *fs_matcher,
+				void *match_v)
+{
+	void *match_c;
+	void *match_v_set_lyr_2_4, *match_c_set_lyr_2_4;
+	void *dmac, *dmac_mask;
+	void *ipv4, *ipv4_mask;
+
+	if (!(fs_matcher->match_criteria_enable &
+	      (1 << MATCH_CRITERIA_ENABLE_OUTER_BIT)))
+		return false;
+
+	match_c = fs_matcher->matcher_mask.match_params;
+	match_v_set_lyr_2_4 = MLX5_ADDR_OF(fte_match_param, match_v,
+					   outer_headers);
+	match_c_set_lyr_2_4 = MLX5_ADDR_OF(fte_match_param, match_c,
+					   outer_headers);
+
+	dmac = MLX5_ADDR_OF(fte_match_set_lyr_2_4, match_v_set_lyr_2_4,
+			    dmac_47_16);
+	dmac_mask = MLX5_ADDR_OF(fte_match_set_lyr_2_4, match_c_set_lyr_2_4,
+				 dmac_47_16);
+
+	if (is_multicast_ether_addr(dmac) &&
+	    is_multicast_ether_addr(dmac_mask))
+		return true;
+
+	ipv4 = MLX5_ADDR_OF(fte_match_set_lyr_2_4, match_v_set_lyr_2_4,
+			    dst_ipv4_dst_ipv6.ipv4_layout.ipv4);
+
+	ipv4_mask = MLX5_ADDR_OF(fte_match_set_lyr_2_4, match_c_set_lyr_2_4,
+				 dst_ipv4_dst_ipv6.ipv4_layout.ipv4);
+
+	if (ipv4_is_multicast(*(__be32 *)(ipv4)) &&
+	    ipv4_is_multicast(*(__be32 *)(ipv4_mask)))
+		return true;
+
+	return false;
+}
+
+struct mlx5_ib_flow_handler *
+mlx5_ib_raw_fs_rule_add(struct mlx5_ib_dev *dev,
+			struct mlx5_ib_flow_matcher *fs_matcher,
+			void *cmd_in, int inlen, int dest_id,
+			int dest_type)
+{
+	struct mlx5_flow_destination *dst;
+	struct mlx5_ib_flow_prio *ft_prio;
+	int priority = fs_matcher->priority;
+	struct mlx5_ib_flow_handler *handler;
+	bool mcast;
+	int err;
+
+	if (fs_matcher->flow_type != MLX5_IB_FLOW_TYPE_NORMAL)
+		return ERR_PTR(-EOPNOTSUPP);
+
+	if (fs_matcher->priority > MLX5_IB_FLOW_LAST_PRIO)
+		return ERR_PTR(-ENOMEM);
+
+	dst = kzalloc(sizeof(*dst), GFP_KERNEL);
+	if (!dst)
+		return ERR_PTR(-ENOMEM);
+
+	mcast = raw_fs_is_multicast(fs_matcher, cmd_in);
+	mutex_lock(&dev->flow_db->lock);
+
+	ft_prio = _get_flow_table(dev, priority, mcast);
+	if (IS_ERR(ft_prio)) {
+		err = PTR_ERR(ft_prio);
+		goto unlock;
+	}
+
+	if (dest_type == MLX5_FLOW_DESTINATION_TYPE_TIR) {
+		dst->type = dest_type;
+		dst->tir_num = dest_id;
+	} else {
+		dst->type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE_NUM;
+		dst->ft_num = dest_id;
+	}
+
+	handler = _create_raw_flow_rule(dev, ft_prio, dst, fs_matcher, cmd_in,
+					inlen);
+
+	if (IS_ERR(handler)) {
+		err = PTR_ERR(handler);
+		goto destroy_ft;
+	}
+
+	mutex_unlock(&dev->flow_db->lock);
+	atomic_inc(&fs_matcher->usecnt);
+	handler->flow_matcher = fs_matcher;
+
+	kfree(dst);
+
+	return handler;
+
+destroy_ft:
+	put_flow_table(dev, ft_prio, false);
+unlock:
+	mutex_unlock(&dev->flow_db->lock);
+	kfree(dst);
+
+	return ERR_PTR(err);
+}
+
 static u32 mlx5_ib_flow_action_flags_to_accel_xfrm_flags(u32 mlx5_flags)
 {
 	u32 flags = 0;
@@ -3672,12 +3865,11 @@ mlx5_ib_create_flow_action_esp(struct ib_device *device,
 	u64 flags;
 	int err = 0;
 
-	if (IS_UVERBS_COPY_ERR(uverbs_copy_from(&action_flags, attrs,
-						MLX5_IB_ATTR_CREATE_FLOW_ACTION_FLAGS)))
-		return ERR_PTR(-EFAULT);
-
-	if (action_flags >= (MLX5_FLOW_ACTION_ESP_CREATE_LAST_SUPPORTED << 1))
-		return ERR_PTR(-EOPNOTSUPP);
+	err = uverbs_get_flags64(
+		&action_flags, attrs, MLX5_IB_ATTR_CREATE_FLOW_ACTION_FLAGS,
+		((MLX5_FLOW_ACTION_ESP_CREATE_LAST_SUPPORTED << 1) - 1));
+	if (err)
+		return ERR_PTR(err);
 
 	flags = mlx5_ib_flow_action_flags_to_accel_xfrm_flags(action_flags);
 
@@ -4466,7 +4658,8 @@ static void destroy_dev_resources(struct mlx5_ib_resources *devr)
 		cancel_work_sync(&devr->ports[port].pkey_change_work);
 }
 
-static u32 get_core_cap_flags(struct ib_device *ibdev)
+static u32 get_core_cap_flags(struct ib_device *ibdev,
+			      struct mlx5_hca_vport_context *rep)
 {
 	struct mlx5_ib_dev *dev = to_mdev(ibdev);
 	enum rdma_link_layer ll = mlx5_ib_port_link_layer(ibdev, 1);
@@ -4475,11 +4668,14 @@ static u32 get_core_cap_flags(struct ib_device *ibdev)
 	bool raw_support = !mlx5_core_mp_enabled(dev->mdev);
 	u32 ret = 0;
 
+	if (rep->grh_required)
+		ret |= RDMA_CORE_CAP_IB_GRH_REQUIRED;
+
 	if (ll == IB_LINK_LAYER_INFINIBAND)
-		return RDMA_CORE_PORT_IBA_IB;
+		return ret | RDMA_CORE_PORT_IBA_IB;
 
 	if (raw_support)
-		ret = RDMA_CORE_PORT_RAW_PACKET;
+		ret |= RDMA_CORE_PORT_RAW_PACKET;
 
 	if (!(l3_type_cap & MLX5_ROCE_L3_TYPE_IPV4_CAP))
 		return ret;
@@ -4502,17 +4698,23 @@ static int mlx5_port_immutable(struct ib_device *ibdev, u8 port_num,
 	struct ib_port_attr attr;
 	struct mlx5_ib_dev *dev = to_mdev(ibdev);
 	enum rdma_link_layer ll = mlx5_ib_port_link_layer(ibdev, port_num);
+	struct mlx5_hca_vport_context rep = {0};
 	int err;
 
-	immutable->core_cap_flags = get_core_cap_flags(ibdev);
-
 	err = ib_query_port(ibdev, port_num, &attr);
 	if (err)
 		return err;
 
+	if (ll == IB_LINK_LAYER_INFINIBAND) {
+		err = mlx5_query_hca_vport_context(dev->mdev, 0, port_num, 0,
+						   &rep);
+		if (err)
+			return err;
+	}
+
 	immutable->pkey_tbl_len = attr.pkey_tbl_len;
 	immutable->gid_tbl_len = attr.gid_tbl_len;
-	immutable->core_cap_flags = get_core_cap_flags(ibdev);
+	immutable->core_cap_flags = get_core_cap_flags(ibdev, &rep);
 	if ((ll == IB_LINK_LAYER_INFINIBAND) || MLX5_CAP_GEN(dev->mdev, roce))
 		immutable->max_mad_size = IB_MGMT_MAD_SIZE;
 
@@ -4610,7 +4812,7 @@ static void mlx5_remove_netdev_notifier(struct mlx5_ib_dev *dev, u8 port_num)
 	}
 }
 
-static int mlx5_enable_eth(struct mlx5_ib_dev *dev, u8 port_num)
+static int mlx5_enable_eth(struct mlx5_ib_dev *dev)
 {
 	int err;
 
@@ -4689,12 +4891,21 @@ static const struct mlx5_ib_counter extended_err_cnts[] = {
 	INIT_Q_COUNTER(req_cqe_flush_error),
 };
 
+#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),
+};
+
 static void mlx5_ib_dealloc_counters(struct mlx5_ib_dev *dev)
 {
 	int i;
 
 	for (i = 0; i < dev->num_ports; i++) {
-		if (dev->port[i].cnts.set_id)
+		if (dev->port[i].cnts.set_id_valid)
 			mlx5_core_dealloc_q_counter(dev->mdev,
 						    dev->port[i].cnts.set_id);
 		kfree(dev->port[i].cnts.names);
@@ -4724,7 +4935,10 @@ static int __mlx5_ib_alloc_counters(struct mlx5_ib_dev *dev,
 		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);
+	}
 	cnts->names = kcalloc(num_counters, sizeof(cnts->names), GFP_KERNEL);
 	if (!cnts->names)
 		return -ENOMEM;
@@ -4781,6 +4995,13 @@ static void mlx5_ib_fill_counters(struct mlx5_ib_dev *dev,
 			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++) {
+			names[j] = ext_ppcnt_cnts[i].name;
+			offsets[j] = ext_ppcnt_cnts[i].offset;
+		}
+	}
 }
 
 static int mlx5_ib_alloc_counters(struct mlx5_ib_dev *dev)
@@ -4826,7 +5047,8 @@ static struct rdma_hw_stats *mlx5_ib_alloc_hw_stats(struct ib_device *ibdev,
 
 	return rdma_alloc_hw_stats_struct(port->cnts.names,
 					  port->cnts.num_q_counters +
-					  port->cnts.num_cong_counters,
+					  port->cnts.num_cong_counters +
+					  port->cnts.num_ext_ppcnt_counters,
 					  RDMA_HW_STATS_DEFAULT_LIFESPAN);
 }
 
@@ -4859,6 +5081,34 @@ free:
 	return ret;
 }
 
+static int mlx5_ib_query_ext_ppcnt_counters(struct mlx5_ib_dev *dev,
+					  struct mlx5_ib_port *port,
+					  struct rdma_hw_stats *stats)
+{
+	int offset = port->cnts.num_q_counters + port->cnts.num_cong_counters;
+	int sz = MLX5_ST_SZ_BYTES(ppcnt_reg);
+	int ret, i;
+	void *out;
+
+	out = kvzalloc(sz, GFP_KERNEL);
+	if (!out)
+		return -ENOMEM;
+
+	ret = mlx5_cmd_query_ext_ppcnt_counters(dev->mdev, out);
+	if (ret)
+		goto free;
+
+	for (i = 0; i < port->cnts.num_ext_ppcnt_counters; i++) {
+		stats->value[i + offset] =
+			be64_to_cpup((__be64 *)(out +
+				    port->cnts.offsets[i + offset]));
+	}
+
+free:
+	kvfree(out);
+	return ret;
+}
+
 static int mlx5_ib_get_hw_stats(struct ib_device *ibdev,
 				struct rdma_hw_stats *stats,
 				u8 port_num, int index)
@@ -4872,13 +5122,21 @@ static int mlx5_ib_get_hw_stats(struct ib_device *ibdev,
 	if (!stats)
 		return -EINVAL;
 
-	num_counters = port->cnts.num_q_counters + port->cnts.num_cong_counters;
+	num_counters = port->cnts.num_q_counters +
+		       port->cnts.num_cong_counters +
+		       port->cnts.num_ext_ppcnt_counters;
 
 	/* q_counters are per IB device, query the master mdev */
 	ret = mlx5_ib_query_q_counters(dev->mdev, port, stats);
 	if (ret)
 		return ret;
 
+	if (MLX5_CAP_PCAM_FEATURE(dev->mdev, rx_icrc_encapsulated_counter)) {
+		ret =  mlx5_ib_query_ext_ppcnt_counters(dev, port, stats);
+		if (ret)
+			return ret;
+	}
+
 	if (MLX5_CAP_GEN(dev->mdev, cc_query_allowed)) {
 		mdev = mlx5_ib_get_native_port_mdev(dev, port_num,
 						    &mdev_port_num);
@@ -4905,11 +5163,6 @@ done:
 	return num_counters;
 }
 
-static void mlx5_ib_free_rdma_netdev(struct net_device *netdev)
-{
-	return mlx5_rdma_netdev_free(netdev);
-}
-
 static struct net_device*
 mlx5_ib_alloc_rdma_netdev(struct ib_device *hca,
 			  u8 port_num,
@@ -4919,17 +5172,12 @@ mlx5_ib_alloc_rdma_netdev(struct ib_device *hca,
 			  void (*setup)(struct net_device *))
 {
 	struct net_device *netdev;
-	struct rdma_netdev *rn;
 
 	if (type != RDMA_NETDEV_IPOIB)
 		return ERR_PTR(-EOPNOTSUPP);
 
 	netdev = mlx5_rdma_netdev_alloc(to_mdev(hca)->mdev, hca,
 					name, setup);
-	if (likely(!IS_ERR_OR_NULL(netdev))) {
-		rn = netdev_priv(netdev);
-		rn->free_rdma_netdev = mlx5_ib_free_rdma_netdev;
-	}
 	return netdev;
 }
 
@@ -5127,8 +5375,8 @@ static bool mlx5_ib_bind_slave_port(struct mlx5_ib_dev *ibdev,
 
 	spin_lock(&ibdev->port[port_num].mp.mpi_lock);
 	if (ibdev->port[port_num].mp.mpi) {
-		mlx5_ib_warn(ibdev, "port %d already affiliated.\n",
-			     port_num + 1);
+		mlx5_ib_dbg(ibdev, "port %d already affiliated.\n",
+			    port_num + 1);
 		spin_unlock(&ibdev->port[port_num].mp.mpi_lock);
 		return false;
 	}
@@ -5263,45 +5511,47 @@ static void mlx5_ib_cleanup_multiport_master(struct mlx5_ib_dev *dev)
 	mlx5_nic_vport_disable_roce(dev->mdev);
 }
 
-ADD_UVERBS_ATTRIBUTES_SIMPLE(mlx5_ib_dm, UVERBS_OBJECT_DM,
-			     UVERBS_METHOD_DM_ALLOC,
-			     &UVERBS_ATTR_PTR_OUT(MLX5_IB_ATTR_ALLOC_DM_RESP_START_OFFSET,
-						  UVERBS_ATTR_TYPE(u64),
-						  UA_FLAGS(UVERBS_ATTR_SPEC_F_MANDATORY)),
-			     &UVERBS_ATTR_PTR_OUT(MLX5_IB_ATTR_ALLOC_DM_RESP_PAGE_INDEX,
-						  UVERBS_ATTR_TYPE(u16),
-						  UA_FLAGS(UVERBS_ATTR_SPEC_F_MANDATORY)));
-
-ADD_UVERBS_ATTRIBUTES_SIMPLE(mlx5_ib_flow_action, UVERBS_OBJECT_FLOW_ACTION,
-			     UVERBS_METHOD_FLOW_ACTION_ESP_CREATE,
-			     &UVERBS_ATTR_PTR_IN(MLX5_IB_ATTR_CREATE_FLOW_ACTION_FLAGS,
-						 UVERBS_ATTR_TYPE(u64),
-						 UA_FLAGS(UVERBS_ATTR_SPEC_F_MANDATORY)));
+ADD_UVERBS_ATTRIBUTES_SIMPLE(
+	mlx5_ib_dm,
+	UVERBS_OBJECT_DM,
+	UVERBS_METHOD_DM_ALLOC,
+	UVERBS_ATTR_PTR_OUT(MLX5_IB_ATTR_ALLOC_DM_RESP_START_OFFSET,
+			    UVERBS_ATTR_TYPE(u64),
+			    UA_MANDATORY),
+	UVERBS_ATTR_PTR_OUT(MLX5_IB_ATTR_ALLOC_DM_RESP_PAGE_INDEX,
+			    UVERBS_ATTR_TYPE(u16),
+			    UA_MANDATORY));
+
+ADD_UVERBS_ATTRIBUTES_SIMPLE(
+	mlx5_ib_flow_action,
+	UVERBS_OBJECT_FLOW_ACTION,
+	UVERBS_METHOD_FLOW_ACTION_ESP_CREATE,
+	UVERBS_ATTR_FLAGS_IN(MLX5_IB_ATTR_CREATE_FLOW_ACTION_FLAGS,
+			     enum mlx5_ib_uapi_flow_action_flags));
 
-#define NUM_TREES	2
 static int populate_specs_root(struct mlx5_ib_dev *dev)
 {
-	const struct uverbs_object_tree_def *default_root[NUM_TREES + 1] = {
-		uverbs_default_get_objects()};
-	size_t num_trees = 1;
+	const struct uverbs_object_tree_def **trees = dev->driver_trees;
+	size_t num_trees = 0;
 
-	if (mlx5_accel_ipsec_device_caps(dev->mdev) & MLX5_ACCEL_IPSEC_CAP_DEVICE &&
-	    !WARN_ON(num_trees >= ARRAY_SIZE(default_root)))
-		default_root[num_trees++] = &mlx5_ib_flow_action;
+	if (mlx5_accel_ipsec_device_caps(dev->mdev) &
+	    MLX5_ACCEL_IPSEC_CAP_DEVICE)
+		trees[num_trees++] = &mlx5_ib_flow_action;
 
-	if (MLX5_CAP_DEV_MEM(dev->mdev, memic) &&
-	    !WARN_ON(num_trees >= ARRAY_SIZE(default_root)))
-		default_root[num_trees++] = &mlx5_ib_dm;
+	if (MLX5_CAP_DEV_MEM(dev->mdev, memic))
+		trees[num_trees++] = &mlx5_ib_dm;
 
-	dev->ib_dev.specs_root =
-		uverbs_alloc_spec_tree(num_trees, default_root);
+	if (MLX5_CAP_GEN_64(dev->mdev, general_obj_types) &
+	    MLX5_GENERAL_OBJ_TYPES_CAP_UCTX)
+		trees[num_trees++] = mlx5_ib_get_devx_tree();
 
-	return PTR_ERR_OR_ZERO(dev->ib_dev.specs_root);
-}
+	num_trees += mlx5_ib_get_flow_trees(trees + num_trees);
 
-static void depopulate_specs_root(struct mlx5_ib_dev *dev)
-{
-	uverbs_free_spec_tree(dev->ib_dev.specs_root);
+	WARN_ON(num_trees >= ARRAY_SIZE(dev->driver_trees));
+	trees[num_trees] = NULL;
+	dev->ib_dev.driver_specs = trees;
+
+	return 0;
 }
 
 static int mlx5_ib_read_counters(struct ib_counters *counters,
@@ -5552,6 +5802,8 @@ int mlx5_ib_stage_caps_init(struct mlx5_ib_dev *dev)
 	dev->ib_dev.modify_qp		= mlx5_ib_modify_qp;
 	dev->ib_dev.query_qp		= mlx5_ib_query_qp;
 	dev->ib_dev.destroy_qp		= mlx5_ib_destroy_qp;
+	dev->ib_dev.drain_sq		= mlx5_ib_drain_sq;
+	dev->ib_dev.drain_rq		= mlx5_ib_drain_rq;
 	dev->ib_dev.post_send		= mlx5_ib_post_send;
 	dev->ib_dev.post_recv		= mlx5_ib_post_recv;
 	dev->ib_dev.create_cq		= mlx5_ib_create_cq;
@@ -5649,9 +5901,9 @@ int mlx5_ib_stage_rep_non_default_cb(struct mlx5_ib_dev *dev)
 	return 0;
 }
 
-static int mlx5_ib_stage_common_roce_init(struct mlx5_ib_dev *dev,
-					  u8 port_num)
+static int mlx5_ib_stage_common_roce_init(struct mlx5_ib_dev *dev)
 {
+	u8 port_num;
 	int i;
 
 	for (i = 0; i < dev->num_ports; i++) {
@@ -5674,6 +5926,8 @@ static int mlx5_ib_stage_common_roce_init(struct mlx5_ib_dev *dev,
 			(1ull << IB_USER_VERBS_EX_CMD_CREATE_RWQ_IND_TBL) |
 			(1ull << IB_USER_VERBS_EX_CMD_DESTROY_RWQ_IND_TBL);
 
+	port_num = mlx5_core_native_port_num(dev->mdev) - 1;
+
 	return mlx5_add_netdev_notifier(dev, port_num);
 }
 
@@ -5690,14 +5944,12 @@ int mlx5_ib_stage_rep_roce_init(struct mlx5_ib_dev *dev)
 	enum rdma_link_layer ll;
 	int port_type_cap;
 	int err = 0;
-	u8 port_num;
 
-	port_num = mlx5_core_native_port_num(dev->mdev) - 1;
 	port_type_cap = MLX5_CAP_GEN(mdev, port_type);
 	ll = mlx5_port_type_cap_to_rdma_ll(port_type_cap);
 
 	if (ll == IB_LINK_LAYER_ETHERNET)
-		err = mlx5_ib_stage_common_roce_init(dev, port_num);
+		err = mlx5_ib_stage_common_roce_init(dev);
 
 	return err;
 }
@@ -5712,19 +5964,17 @@ static int mlx5_ib_stage_roce_init(struct mlx5_ib_dev *dev)
 	struct mlx5_core_dev *mdev = dev->mdev;
 	enum rdma_link_layer ll;
 	int port_type_cap;
-	u8 port_num;
 	int err;
 
-	port_num = mlx5_core_native_port_num(dev->mdev) - 1;
 	port_type_cap = MLX5_CAP_GEN(mdev, port_type);
 	ll = mlx5_port_type_cap_to_rdma_ll(port_type_cap);
 
 	if (ll == IB_LINK_LAYER_ETHERNET) {
-		err = mlx5_ib_stage_common_roce_init(dev, port_num);
+		err = mlx5_ib_stage_common_roce_init(dev);
 		if (err)
 			return err;
 
-		err = mlx5_enable_eth(dev, port_num);
+		err = mlx5_enable_eth(dev);
 		if (err)
 			goto cleanup;
 	}
@@ -5741,9 +5991,7 @@ static void mlx5_ib_stage_roce_cleanup(struct mlx5_ib_dev *dev)
 	struct mlx5_core_dev *mdev = dev->mdev;
 	enum rdma_link_layer ll;
 	int port_type_cap;
-	u8 port_num;
 
-	port_num = mlx5_core_native_port_num(dev->mdev) - 1;
 	port_type_cap = MLX5_CAP_GEN(mdev, port_type);
 	ll = mlx5_port_type_cap_to_rdma_ll(port_type_cap);
 
@@ -5842,11 +6090,6 @@ int mlx5_ib_stage_ib_reg_init(struct mlx5_ib_dev *dev)
 	return ib_register_device(&dev->ib_dev, NULL);
 }
 
-static void mlx5_ib_stage_depopulate_specs(struct mlx5_ib_dev *dev)
-{
-	depopulate_specs_root(dev);
-}
-
 void mlx5_ib_stage_pre_ib_reg_umr_cleanup(struct mlx5_ib_dev *dev)
 {
 	destroy_umrc_res(dev);
@@ -5915,8 +6158,6 @@ void __mlx5_ib_remove(struct mlx5_ib_dev *dev,
 	ib_dealloc_device((struct ib_device *)dev);
 }
 
-static void *mlx5_ib_add_slave_port(struct mlx5_core_dev *mdev, u8 port_num);
-
 void *__mlx5_ib_add(struct mlx5_ib_dev *dev,
 		    const struct mlx5_ib_profile *profile)
 {
@@ -5983,7 +6224,7 @@ static const struct mlx5_ib_profile pf_profile = {
 		     mlx5_ib_stage_pre_ib_reg_umr_cleanup),
 	STAGE_CREATE(MLX5_IB_STAGE_SPECS,
 		     mlx5_ib_stage_populate_specs,
-		     mlx5_ib_stage_depopulate_specs),
+		     NULL),
 	STAGE_CREATE(MLX5_IB_STAGE_IB_REG,
 		     mlx5_ib_stage_ib_reg_init,
 		     mlx5_ib_stage_ib_reg_cleanup),
@@ -6031,7 +6272,7 @@ static const struct mlx5_ib_profile nic_rep_profile = {
 		     mlx5_ib_stage_pre_ib_reg_umr_cleanup),
 	STAGE_CREATE(MLX5_IB_STAGE_SPECS,
 		     mlx5_ib_stage_populate_specs,
-		     mlx5_ib_stage_depopulate_specs),
+		     NULL),
 	STAGE_CREATE(MLX5_IB_STAGE_IB_REG,
 		     mlx5_ib_stage_ib_reg_init,
 		     mlx5_ib_stage_ib_reg_cleanup),
@@ -6046,7 +6287,7 @@ static const struct mlx5_ib_profile nic_rep_profile = {
 		     mlx5_ib_stage_rep_reg_cleanup),
 };
 
-static void *mlx5_ib_add_slave_port(struct mlx5_core_dev *mdev, u8 port_num)
+static void *mlx5_ib_add_slave_port(struct mlx5_core_dev *mdev)
 {
 	struct mlx5_ib_multiport_info *mpi;
 	struct mlx5_ib_dev *dev;
@@ -6080,8 +6321,6 @@ static void *mlx5_ib_add_slave_port(struct mlx5_core_dev *mdev, u8 port_num)
 	if (!bound) {
 		list_add_tail(&mpi->list, &mlx5_ib_unaffiliated_port_list);
 		dev_dbg(&mdev->pdev->dev, "no suitable IB device found to bind to, added to unaffiliated list.\n");
-	} else {
-		mlx5_ib_dbg(dev, "bound port %u\n", port_num + 1);
 	}
 	mutex_unlock(&mlx5_ib_multiport_mutex);
 
@@ -6099,11 +6338,8 @@ static void *mlx5_ib_add(struct mlx5_core_dev *mdev)
 	port_type_cap = MLX5_CAP_GEN(mdev, port_type);
 	ll = mlx5_port_type_cap_to_rdma_ll(port_type_cap);
 
-	if (mlx5_core_is_mp_slave(mdev) && ll == IB_LINK_LAYER_ETHERNET) {
-		u8 port_num = mlx5_core_native_port_num(mdev) - 1;
-
-		return mlx5_ib_add_slave_port(mdev, port_num);
-	}
+	if (mlx5_core_is_mp_slave(mdev) && ll == IB_LINK_LAYER_ETHERNET)
+		return mlx5_ib_add_slave_port(mdev);
 
 	dev = (struct mlx5_ib_dev *)ib_alloc_device(sizeof(*dev));
 	if (!dev)
diff --git a/drivers/infiniband/hw/mlx5/mlx5_ib.h b/drivers/infiniband/hw/mlx5/mlx5_ib.h
index d89c8fe626f6..320d4dfe8c2f 100644
--- a/drivers/infiniband/hw/mlx5/mlx5_ib.h
+++ b/drivers/infiniband/hw/mlx5/mlx5_ib.h
@@ -46,6 +46,7 @@
 #include <rdma/ib_user_verbs.h>
 #include <rdma/mlx5-abi.h>
 #include <rdma/uverbs_ioctl.h>
+#include <rdma/mlx5_user_ioctl_cmds.h>
 
 #define mlx5_ib_dbg(dev, format, arg...)				\
 pr_debug("%s:%s:%d:(pid %d): " format, (dev)->ib_dev.name, __func__,	\
@@ -78,12 +79,6 @@ enum {
 	MLX5_REQ_SCAT_DATA64_CQE	= 0x22,
 };
 
-enum mlx5_ib_latency_class {
-	MLX5_IB_LATENCY_CLASS_LOW,
-	MLX5_IB_LATENCY_CLASS_MEDIUM,
-	MLX5_IB_LATENCY_CLASS_HIGH,
-};
-
 enum mlx5_ib_mad_ifc_flags {
 	MLX5_MAD_IFC_IGNORE_MKEY	= 1,
 	MLX5_MAD_IFC_IGNORE_BKEY	= 2,
@@ -143,6 +138,7 @@ struct mlx5_ib_ucontext {
 
 	u64			lib_caps;
 	DECLARE_BITMAP(dm_pages, MLX5_MAX_MEMIC_PAGES);
+	u16			devx_uid;
 };
 
 static inline struct mlx5_ib_ucontext *to_mucontext(struct ib_ucontext *ibucontext)
@@ -176,6 +172,18 @@ struct mlx5_ib_flow_handler {
 	struct mlx5_ib_flow_prio	*prio;
 	struct mlx5_flow_handle		*rule;
 	struct ib_counters		*ibcounters;
+	struct mlx5_ib_dev		*dev;
+	struct mlx5_ib_flow_matcher	*flow_matcher;
+};
+
+struct mlx5_ib_flow_matcher {
+	struct mlx5_ib_match_params matcher_mask;
+	int			mask_len;
+	enum mlx5_ib_flow_type	flow_type;
+	u16			priority;
+	struct mlx5_core_dev	*mdev;
+	atomic_t		usecnt;
+	u8			match_criteria_enable;
 };
 
 struct mlx5_ib_flow_db {
@@ -461,7 +469,7 @@ struct mlx5_umr_wr {
 	u32				mkey;
 };
 
-static inline struct mlx5_umr_wr *umr_wr(struct ib_send_wr *wr)
+static inline const struct mlx5_umr_wr *umr_wr(const struct ib_send_wr *wr)
 {
 	return container_of(wr, struct mlx5_umr_wr, wr);
 }
@@ -665,6 +673,7 @@ struct mlx5_ib_counters {
 	size_t *offsets;
 	u32 num_q_counters;
 	u32 num_cong_counters;
+	u32 num_ext_ppcnt_counters;
 	u16 set_id;
 	bool set_id_valid;
 };
@@ -851,6 +860,7 @@ to_mcounters(struct ib_counters *ibcntrs)
 
 struct mlx5_ib_dev {
 	struct ib_device		ib_dev;
+	const struct uverbs_object_tree_def *driver_trees[6];
 	struct mlx5_core_dev		*mdev;
 	struct mlx5_roce		roce[MLX5_MAX_PORTS];
 	int				num_ports;
@@ -1004,8 +1014,8 @@ int mlx5_ib_modify_srq(struct ib_srq *ibsrq, struct ib_srq_attr *attr,
 		       enum ib_srq_attr_mask attr_mask, struct ib_udata *udata);
 int mlx5_ib_query_srq(struct ib_srq *ibsrq, struct ib_srq_attr *srq_attr);
 int mlx5_ib_destroy_srq(struct ib_srq *srq);
-int mlx5_ib_post_srq_recv(struct ib_srq *ibsrq, struct ib_recv_wr *wr,
-			  struct ib_recv_wr **bad_wr);
+int mlx5_ib_post_srq_recv(struct ib_srq *ibsrq, const struct ib_recv_wr *wr,
+			  const struct ib_recv_wr **bad_wr);
 struct ib_qp *mlx5_ib_create_qp(struct ib_pd *pd,
 				struct ib_qp_init_attr *init_attr,
 				struct ib_udata *udata);
@@ -1014,10 +1024,12 @@ int mlx5_ib_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr,
 int mlx5_ib_query_qp(struct ib_qp *ibqp, struct ib_qp_attr *qp_attr, int qp_attr_mask,
 		     struct ib_qp_init_attr *qp_init_attr);
 int mlx5_ib_destroy_qp(struct ib_qp *qp);
-int mlx5_ib_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
-		      struct ib_send_wr **bad_wr);
-int mlx5_ib_post_recv(struct ib_qp *ibqp, struct ib_recv_wr *wr,
-		      struct ib_recv_wr **bad_wr);
+void mlx5_ib_drain_sq(struct ib_qp *qp);
+void mlx5_ib_drain_rq(struct ib_qp *qp);
+int mlx5_ib_post_send(struct ib_qp *ibqp, const struct ib_send_wr *wr,
+		      const struct ib_send_wr **bad_wr);
+int mlx5_ib_post_recv(struct ib_qp *ibqp, const struct ib_recv_wr *wr,
+		      const struct ib_recv_wr **bad_wr);
 void *mlx5_get_send_wqe(struct mlx5_ib_qp *qp, int n);
 int mlx5_ib_read_user_wqe(struct mlx5_ib_qp *qp, int send, int wqe_index,
 			  void *buffer, u32 length,
@@ -1183,10 +1195,8 @@ int mlx5_ib_get_vf_stats(struct ib_device *device, int vf,
 int mlx5_ib_set_vf_guid(struct ib_device *device, int vf, u8 port,
 			u64 guid, int type);
 
-__be16 mlx5_get_roce_udp_sport(struct mlx5_ib_dev *dev, u8 port_num,
-			       int index);
-int mlx5_get_roce_gid_type(struct mlx5_ib_dev *dev, u8 port_num,
-			   int index, enum ib_gid_type *gid_type);
+__be16 mlx5_get_roce_udp_sport(struct mlx5_ib_dev *dev,
+			       const struct ib_gid_attr *attr);
 
 void mlx5_ib_cleanup_cong_debugfs(struct mlx5_ib_dev *dev, u8 port_num);
 int mlx5_ib_init_cong_debugfs(struct mlx5_ib_dev *dev, u8 port_num);
@@ -1200,10 +1210,10 @@ int mlx5_ib_gsi_modify_qp(struct ib_qp *qp, struct ib_qp_attr *attr,
 int mlx5_ib_gsi_query_qp(struct ib_qp *qp, struct ib_qp_attr *qp_attr,
 			 int qp_attr_mask,
 			 struct ib_qp_init_attr *qp_init_attr);
-int mlx5_ib_gsi_post_send(struct ib_qp *qp, struct ib_send_wr *wr,
-			  struct ib_send_wr **bad_wr);
-int mlx5_ib_gsi_post_recv(struct ib_qp *qp, struct ib_recv_wr *wr,
-			  struct ib_recv_wr **bad_wr);
+int mlx5_ib_gsi_post_send(struct ib_qp *qp, const struct ib_send_wr *wr,
+			  const struct ib_send_wr **bad_wr);
+int mlx5_ib_gsi_post_recv(struct ib_qp *qp, const struct ib_recv_wr *wr,
+			  const struct ib_recv_wr **bad_wr);
 void mlx5_ib_gsi_pkey_change(struct mlx5_ib_gsi_qp *gsi);
 
 int mlx5_ib_generate_wc(struct ib_cq *ibcq, struct ib_wc *wc);
@@ -1217,6 +1227,36 @@ struct mlx5_core_dev *mlx5_ib_get_native_port_mdev(struct mlx5_ib_dev *dev,
 void mlx5_ib_put_native_port_mdev(struct mlx5_ib_dev *dev,
 				  u8 port_num);
 
+#if IS_ENABLED(CONFIG_INFINIBAND_USER_ACCESS)
+int mlx5_ib_devx_create(struct mlx5_ib_dev *dev,
+			struct mlx5_ib_ucontext *context);
+void mlx5_ib_devx_destroy(struct mlx5_ib_dev *dev,
+			  struct mlx5_ib_ucontext *context);
+const struct uverbs_object_tree_def *mlx5_ib_get_devx_tree(void);
+struct mlx5_ib_flow_handler *mlx5_ib_raw_fs_rule_add(
+	struct mlx5_ib_dev *dev, struct mlx5_ib_flow_matcher *fs_matcher,
+	void *cmd_in, int inlen, int dest_id, int dest_type);
+bool mlx5_ib_devx_is_flow_dest(void *obj, int *dest_id, int *dest_type);
+int mlx5_ib_get_flow_trees(const struct uverbs_object_tree_def **root);
+#else
+static inline int
+mlx5_ib_devx_create(struct mlx5_ib_dev *dev,
+		    struct mlx5_ib_ucontext *context) { return -EOPNOTSUPP; };
+static inline void mlx5_ib_devx_destroy(struct mlx5_ib_dev *dev,
+					struct mlx5_ib_ucontext *context) {}
+static inline const struct uverbs_object_tree_def *
+mlx5_ib_get_devx_tree(void) { return NULL; }
+static inline bool mlx5_ib_devx_is_flow_dest(void *obj, int *dest_id,
+					     int *dest_type)
+{
+	return false;
+}
+static inline int
+mlx5_ib_get_flow_trees(const struct uverbs_object_tree_def **root)
+{
+	return 0;
+}
+#endif
 static inline void init_query_mad(struct ib_smp *mad)
 {
 	mad->base_version  = 1;
@@ -1318,4 +1358,7 @@ static inline int get_num_static_uars(struct mlx5_ib_dev *dev,
 unsigned long mlx5_ib_get_xlt_emergency_page(void);
 void mlx5_ib_put_xlt_emergency_page(void);
 
+int bfregn_to_uar_index(struct mlx5_ib_dev *dev,
+			struct mlx5_bfreg_info *bfregi, u32 bfregn,
+			bool dyn_bfreg);
 #endif /* MLX5_IB_H */
diff --git a/drivers/infiniband/hw/mlx5/mr.c b/drivers/infiniband/hw/mlx5/mr.c
index 90a9c461cedc..9fb1d9cb9401 100644
--- a/drivers/infiniband/hw/mlx5/mr.c
+++ b/drivers/infiniband/hw/mlx5/mr.c
@@ -271,16 +271,16 @@ static ssize_t size_write(struct file *filp, const char __user *buf,
 {
 	struct mlx5_cache_ent *ent = filp->private_data;
 	struct mlx5_ib_dev *dev = ent->dev;
-	char lbuf[20];
+	char lbuf[20] = {0};
 	u32 var;
 	int err;
 	int c;
 
-	if (copy_from_user(lbuf, buf, sizeof(lbuf)))
+	count = min(count, sizeof(lbuf) - 1);
+	if (copy_from_user(lbuf, buf, count))
 		return -EFAULT;
 
 	c = order2idx(dev, ent->order);
-	lbuf[sizeof(lbuf) - 1] = 0;
 
 	if (sscanf(lbuf, "%u", &var) != 1)
 		return -EINVAL;
@@ -310,19 +310,11 @@ static ssize_t size_read(struct file *filp, char __user *buf, size_t count,
 	char lbuf[20];
 	int err;
 
-	if (*pos)
-		return 0;
-
 	err = snprintf(lbuf, sizeof(lbuf), "%d\n", ent->size);
 	if (err < 0)
 		return err;
 
-	if (copy_to_user(buf, lbuf, err))
-		return -EFAULT;
-
-	*pos += err;
-
-	return err;
+	return simple_read_from_buffer(buf, count, pos, lbuf, err);
 }
 
 static const struct file_operations size_fops = {
@@ -337,16 +329,16 @@ static ssize_t limit_write(struct file *filp, const char __user *buf,
 {
 	struct mlx5_cache_ent *ent = filp->private_data;
 	struct mlx5_ib_dev *dev = ent->dev;
-	char lbuf[20];
+	char lbuf[20] = {0};
 	u32 var;
 	int err;
 	int c;
 
-	if (copy_from_user(lbuf, buf, sizeof(lbuf)))
+	count = min(count, sizeof(lbuf) - 1);
+	if (copy_from_user(lbuf, buf, count))
 		return -EFAULT;
 
 	c = order2idx(dev, ent->order);
-	lbuf[sizeof(lbuf) - 1] = 0;
 
 	if (sscanf(lbuf, "%u", &var) != 1)
 		return -EINVAL;
@@ -372,19 +364,11 @@ static ssize_t limit_read(struct file *filp, char __user *buf, size_t count,
 	char lbuf[20];
 	int err;
 
-	if (*pos)
-		return 0;
-
 	err = snprintf(lbuf, sizeof(lbuf), "%d\n", ent->limit);
 	if (err < 0)
 		return err;
 
-	if (copy_to_user(buf, lbuf, err))
-		return -EFAULT;
-
-	*pos += err;
-
-	return err;
+	return simple_read_from_buffer(buf, count, pos, lbuf, err);
 }
 
 static const struct file_operations limit_fops = {
@@ -914,7 +898,7 @@ static int mlx5_ib_post_send_wait(struct mlx5_ib_dev *dev,
 				  struct mlx5_umr_wr *umrwr)
 {
 	struct umr_common *umrc = &dev->umrc;
-	struct ib_send_wr *bad;
+	const struct ib_send_wr *bad;
 	int err;
 	struct mlx5_ib_umr_context umr_context;
 
diff --git a/drivers/infiniband/hw/mlx5/qp.c b/drivers/infiniband/hw/mlx5/qp.c
index a4f1f638509f..6cba2a02d11b 100644
--- a/drivers/infiniband/hw/mlx5/qp.c
+++ b/drivers/infiniband/hw/mlx5/qp.c
@@ -563,32 +563,21 @@ static int alloc_med_class_bfreg(struct mlx5_ib_dev *dev,
 }
 
 static int alloc_bfreg(struct mlx5_ib_dev *dev,
-		       struct mlx5_bfreg_info *bfregi,
-		       enum mlx5_ib_latency_class lat)
+		       struct mlx5_bfreg_info *bfregi)
 {
-	int bfregn = -EINVAL;
+	int bfregn = -ENOMEM;
 
 	mutex_lock(&bfregi->lock);
-	switch (lat) {
-	case MLX5_IB_LATENCY_CLASS_LOW:
+	if (bfregi->ver >= 2) {
+		bfregn = alloc_high_class_bfreg(dev, bfregi);
+		if (bfregn < 0)
+			bfregn = alloc_med_class_bfreg(dev, bfregi);
+	}
+
+	if (bfregn < 0) {
 		BUILD_BUG_ON(NUM_NON_BLUE_FLAME_BFREGS != 1);
 		bfregn = 0;
 		bfregi->count[bfregn]++;
-		break;
-
-	case MLX5_IB_LATENCY_CLASS_MEDIUM:
-		if (bfregi->ver < 2)
-			bfregn = -ENOMEM;
-		else
-			bfregn = alloc_med_class_bfreg(dev, bfregi);
-		break;
-
-	case MLX5_IB_LATENCY_CLASS_HIGH:
-		if (bfregi->ver < 2)
-			bfregn = -ENOMEM;
-		else
-			bfregn = alloc_high_class_bfreg(dev, bfregi);
-		break;
 	}
 	mutex_unlock(&bfregi->lock);
 
@@ -641,13 +630,13 @@ static void mlx5_ib_lock_cqs(struct mlx5_ib_cq *send_cq,
 static void mlx5_ib_unlock_cqs(struct mlx5_ib_cq *send_cq,
 			       struct mlx5_ib_cq *recv_cq);
 
-static int bfregn_to_uar_index(struct mlx5_ib_dev *dev,
-			       struct mlx5_bfreg_info *bfregi, int bfregn,
-			       bool dyn_bfreg)
+int bfregn_to_uar_index(struct mlx5_ib_dev *dev,
+			struct mlx5_bfreg_info *bfregi, u32 bfregn,
+			bool dyn_bfreg)
 {
-	int bfregs_per_sys_page;
-	int index_of_sys_page;
-	int offset;
+	unsigned int bfregs_per_sys_page;
+	u32 index_of_sys_page;
+	u32 offset;
 
 	bfregs_per_sys_page = get_uars_per_sys_page(dev, bfregi->lib_uar_4k) *
 				MLX5_NON_FP_BFREGS_PER_UAR;
@@ -655,6 +644,10 @@ static int bfregn_to_uar_index(struct mlx5_ib_dev *dev,
 
 	if (dyn_bfreg) {
 		index_of_sys_page += bfregi->num_static_sys_pages;
+
+		if (index_of_sys_page >= bfregi->num_sys_pages)
+			return -EINVAL;
+
 		if (bfregn > bfregi->num_dyn_bfregs ||
 		    bfregi->sys_pages[index_of_sys_page] == MLX5_IB_INVALID_UAR_INDEX) {
 			mlx5_ib_dbg(dev, "Invalid dynamic uar index\n");
@@ -819,21 +812,9 @@ static int create_user_qp(struct mlx5_ib_dev *dev, struct ib_pd *pd,
 		bfregn = MLX5_CROSS_CHANNEL_BFREG;
 	}
 	else {
-		bfregn = alloc_bfreg(dev, &context->bfregi, MLX5_IB_LATENCY_CLASS_HIGH);
-		if (bfregn < 0) {
-			mlx5_ib_dbg(dev, "failed to allocate low latency BFREG\n");
-			mlx5_ib_dbg(dev, "reverting to medium latency\n");
-			bfregn = alloc_bfreg(dev, &context->bfregi, MLX5_IB_LATENCY_CLASS_MEDIUM);
-			if (bfregn < 0) {
-				mlx5_ib_dbg(dev, "failed to allocate medium latency BFREG\n");
-				mlx5_ib_dbg(dev, "reverting to high latency\n");
-				bfregn = alloc_bfreg(dev, &context->bfregi, MLX5_IB_LATENCY_CLASS_LOW);
-				if (bfregn < 0) {
-					mlx5_ib_warn(dev, "bfreg allocation failed\n");
-					return bfregn;
-				}
-			}
-		}
+		bfregn = alloc_bfreg(dev, &context->bfregi);
+		if (bfregn < 0)
+			return bfregn;
 	}
 
 	mlx5_ib_dbg(dev, "bfregn 0x%x, uar_index 0x%x\n", bfregn, uar_index);
@@ -1626,7 +1607,7 @@ static int create_qp_common(struct mlx5_ib_dev *dev, struct ib_pd *pd,
 	struct mlx5_ib_resources *devr = &dev->devr;
 	int inlen = MLX5_ST_SZ_BYTES(create_qp_in);
 	struct mlx5_core_dev *mdev = dev->mdev;
-	struct mlx5_ib_create_qp_resp resp;
+	struct mlx5_ib_create_qp_resp resp = {};
 	struct mlx5_ib_cq *send_cq;
 	struct mlx5_ib_cq *recv_cq;
 	unsigned long flags;
@@ -2555,18 +2536,16 @@ static int mlx5_set_path(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp,
 	if (ah->type == RDMA_AH_ATTR_TYPE_ROCE) {
 		if (!(ah_flags & IB_AH_GRH))
 			return -EINVAL;
-		err = mlx5_get_roce_gid_type(dev, port, grh->sgid_index,
-					     &gid_type);
-		if (err)
-			return err;
+
 		memcpy(path->rmac, ah->roce.dmac, sizeof(ah->roce.dmac));
 		if (qp->ibqp.qp_type == IB_QPT_RC ||
 		    qp->ibqp.qp_type == IB_QPT_UC ||
 		    qp->ibqp.qp_type == IB_QPT_XRC_INI ||
 		    qp->ibqp.qp_type == IB_QPT_XRC_TGT)
-			path->udp_sport = mlx5_get_roce_udp_sport(dev, port,
-								  grh->sgid_index);
+			path->udp_sport =
+				mlx5_get_roce_udp_sport(dev, ah->grh.sgid_attr);
 		path->dci_cfi_prio_sl = (sl & 0x7) << 4;
+		gid_type = ah->grh.sgid_attr->gid_type;
 		if (gid_type == IB_GID_TYPE_ROCE_UDP_ENCAP)
 			path->ecn_dscp = (grh->traffic_class >> 2) & 0x3f;
 	} else {
@@ -3529,7 +3508,7 @@ static __always_inline void set_raddr_seg(struct mlx5_wqe_raddr_seg *rseg,
 }
 
 static void *set_eth_seg(struct mlx5_wqe_eth_seg *eseg,
-			 struct ib_send_wr *wr, void *qend,
+			 const struct ib_send_wr *wr, void *qend,
 			 struct mlx5_ib_qp *qp, int *size)
 {
 	void *seg = eseg;
@@ -3582,7 +3561,7 @@ static void *set_eth_seg(struct mlx5_wqe_eth_seg *eseg,
 }
 
 static void set_datagram_seg(struct mlx5_wqe_datagram_seg *dseg,
-			     struct ib_send_wr *wr)
+			     const struct ib_send_wr *wr)
 {
 	memcpy(&dseg->av, &to_mah(ud_wr(wr)->ah)->av, sizeof(struct mlx5_av));
 	dseg->av.dqp_dct = cpu_to_be32(ud_wr(wr)->remote_qpn | MLX5_EXTENDED_UD_AV);
@@ -3730,9 +3709,9 @@ static int umr_check_mkey_mask(struct mlx5_ib_dev *dev, u64 mask)
 
 static int set_reg_umr_segment(struct mlx5_ib_dev *dev,
 			       struct mlx5_wqe_umr_ctrl_seg *umr,
-			       struct ib_send_wr *wr, int atomic)
+			       const struct ib_send_wr *wr, int atomic)
 {
-	struct mlx5_umr_wr *umrwr = umr_wr(wr);
+	const struct mlx5_umr_wr *umrwr = umr_wr(wr);
 
 	memset(umr, 0, sizeof(*umr));
 
@@ -3803,9 +3782,10 @@ static void set_linv_mkey_seg(struct mlx5_mkey_seg *seg)
 	seg->status = MLX5_MKEY_STATUS_FREE;
 }
 
-static void set_reg_mkey_segment(struct mlx5_mkey_seg *seg, struct ib_send_wr *wr)
+static void set_reg_mkey_segment(struct mlx5_mkey_seg *seg,
+				 const struct ib_send_wr *wr)
 {
-	struct mlx5_umr_wr *umrwr = umr_wr(wr);
+	const struct mlx5_umr_wr *umrwr = umr_wr(wr);
 
 	memset(seg, 0, sizeof(*seg));
 	if (wr->send_flags & MLX5_IB_SEND_UMR_DISABLE_MR)
@@ -3854,7 +3834,7 @@ static void set_reg_umr_inline_seg(void *seg, struct mlx5_ib_qp *qp,
 	seg += mr_list_size;
 }
 
-static __be32 send_ieth(struct ib_send_wr *wr)
+static __be32 send_ieth(const struct ib_send_wr *wr)
 {
 	switch (wr->opcode) {
 	case IB_WR_SEND_WITH_IMM:
@@ -3886,7 +3866,7 @@ static u8 wq_sig(void *wqe)
 	return calc_sig(wqe, (*((u8 *)wqe + 8) & 0x3f) << 4);
 }
 
-static int set_data_inl_seg(struct mlx5_ib_qp *qp, struct ib_send_wr *wr,
+static int set_data_inl_seg(struct mlx5_ib_qp *qp, const struct ib_send_wr *wr,
 			    void *wqe, int *sz)
 {
 	struct mlx5_wqe_inline_seg *seg;
@@ -4032,7 +4012,7 @@ static int mlx5_set_bsf(struct ib_mr *sig_mr,
 	return 0;
 }
 
-static int set_sig_data_segment(struct ib_sig_handover_wr *wr,
+static int set_sig_data_segment(const struct ib_sig_handover_wr *wr,
 				struct mlx5_ib_qp *qp, void **seg, int *size)
 {
 	struct ib_sig_attrs *sig_attrs = wr->sig_attrs;
@@ -4134,7 +4114,7 @@ static int set_sig_data_segment(struct ib_sig_handover_wr *wr,
 }
 
 static void set_sig_mkey_segment(struct mlx5_mkey_seg *seg,
-				 struct ib_sig_handover_wr *wr, u32 size,
+				 const struct ib_sig_handover_wr *wr, u32 size,
 				 u32 length, u32 pdn)
 {
 	struct ib_mr *sig_mr = wr->sig_mr;
@@ -4165,10 +4145,10 @@ static void set_sig_umr_segment(struct mlx5_wqe_umr_ctrl_seg *umr,
 }
 
 
-static int set_sig_umr_wr(struct ib_send_wr *send_wr, struct mlx5_ib_qp *qp,
-			  void **seg, int *size)
+static int set_sig_umr_wr(const struct ib_send_wr *send_wr,
+			  struct mlx5_ib_qp *qp, void **seg, int *size)
 {
-	struct ib_sig_handover_wr *wr = sig_handover_wr(send_wr);
+	const struct ib_sig_handover_wr *wr = sig_handover_wr(send_wr);
 	struct mlx5_ib_mr *sig_mr = to_mmr(wr->sig_mr);
 	u32 pdn = get_pd(qp)->pdn;
 	u32 xlt_size;
@@ -4243,7 +4223,7 @@ static int set_psv_wr(struct ib_sig_domain *domain,
 }
 
 static int set_reg_wr(struct mlx5_ib_qp *qp,
-		      struct ib_reg_wr *wr,
+		      const struct ib_reg_wr *wr,
 		      void **seg, int *size)
 {
 	struct mlx5_ib_mr *mr = to_mmr(wr->mr);
@@ -4314,10 +4294,10 @@ static void dump_wqe(struct mlx5_ib_qp *qp, int idx, int size_16)
 	}
 }
 
-static int begin_wqe(struct mlx5_ib_qp *qp, void **seg,
+static int __begin_wqe(struct mlx5_ib_qp *qp, void **seg,
 		     struct mlx5_wqe_ctrl_seg **ctrl,
-		     struct ib_send_wr *wr, unsigned *idx,
-		     int *size, int nreq)
+		     const struct ib_send_wr *wr, unsigned *idx,
+		     int *size, int nreq, bool send_signaled, bool solicited)
 {
 	if (unlikely(mlx5_wq_overflow(&qp->sq, nreq, qp->ibqp.send_cq)))
 		return -ENOMEM;
@@ -4328,10 +4308,8 @@ static int begin_wqe(struct mlx5_ib_qp *qp, void **seg,
 	*(uint32_t *)(*seg + 8) = 0;
 	(*ctrl)->imm = send_ieth(wr);
 	(*ctrl)->fm_ce_se = qp->sq_signal_bits |
-		(wr->send_flags & IB_SEND_SIGNALED ?
-		 MLX5_WQE_CTRL_CQ_UPDATE : 0) |
-		(wr->send_flags & IB_SEND_SOLICITED ?
-		 MLX5_WQE_CTRL_SOLICITED : 0);
+		(send_signaled ? MLX5_WQE_CTRL_CQ_UPDATE : 0) |
+		(solicited ? MLX5_WQE_CTRL_SOLICITED : 0);
 
 	*seg += sizeof(**ctrl);
 	*size = sizeof(**ctrl) / 16;
@@ -4339,6 +4317,16 @@ static int begin_wqe(struct mlx5_ib_qp *qp, void **seg,
 	return 0;
 }
 
+static int begin_wqe(struct mlx5_ib_qp *qp, void **seg,
+		     struct mlx5_wqe_ctrl_seg **ctrl,
+		     const struct ib_send_wr *wr, unsigned *idx,
+		     int *size, int nreq)
+{
+	return __begin_wqe(qp, seg, ctrl, wr, idx, size, nreq,
+			   wr->send_flags & IB_SEND_SIGNALED,
+			   wr->send_flags & IB_SEND_SOLICITED);
+}
+
 static void finish_wqe(struct mlx5_ib_qp *qp,
 		       struct mlx5_wqe_ctrl_seg *ctrl,
 		       u8 size, unsigned idx, u64 wr_id,
@@ -4360,9 +4348,8 @@ static void finish_wqe(struct mlx5_ib_qp *qp,
 	qp->sq.w_list[idx].next = qp->sq.cur_post;
 }
 
-
-int mlx5_ib_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
-		      struct ib_send_wr **bad_wr)
+static int _mlx5_ib_post_send(struct ib_qp *ibqp, const struct ib_send_wr *wr,
+			      const struct ib_send_wr **bad_wr, bool drain)
 {
 	struct mlx5_wqe_ctrl_seg *ctrl = NULL;  /* compiler warning */
 	struct mlx5_ib_dev *dev = to_mdev(ibqp->device);
@@ -4393,7 +4380,7 @@ int mlx5_ib_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
 
 	spin_lock_irqsave(&qp->sq.lock, flags);
 
-	if (mdev->state == MLX5_DEVICE_STATE_INTERNAL_ERROR) {
+	if (mdev->state == MLX5_DEVICE_STATE_INTERNAL_ERROR && !drain) {
 		err = -EIO;
 		*bad_wr = wr;
 		nreq = 0;
@@ -4498,10 +4485,8 @@ int mlx5_ib_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
 				 * SET_PSV WQEs are not signaled and solicited
 				 * on error
 				 */
-				wr->send_flags &= ~IB_SEND_SIGNALED;
-				wr->send_flags |= IB_SEND_SOLICITED;
-				err = begin_wqe(qp, &seg, &ctrl, wr,
-						&idx, &size, nreq);
+				err = __begin_wqe(qp, &seg, &ctrl, wr, &idx,
+						  &size, nreq, false, true);
 				if (err) {
 					mlx5_ib_warn(dev, "\n");
 					err = -ENOMEM;
@@ -4520,8 +4505,8 @@ int mlx5_ib_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
 
 				finish_wqe(qp, ctrl, size, idx, wr->wr_id, nreq,
 					   fence, MLX5_OPCODE_SET_PSV);
-				err = begin_wqe(qp, &seg, &ctrl, wr,
-						&idx, &size, nreq);
+				err = __begin_wqe(qp, &seg, &ctrl, wr, &idx,
+						  &size, nreq, false, true);
 				if (err) {
 					mlx5_ib_warn(dev, "\n");
 					err = -ENOMEM;
@@ -4690,13 +4675,19 @@ out:
 	return err;
 }
 
+int mlx5_ib_post_send(struct ib_qp *ibqp, const struct ib_send_wr *wr,
+		      const struct ib_send_wr **bad_wr)
+{
+	return _mlx5_ib_post_send(ibqp, wr, bad_wr, false);
+}
+
 static void set_sig_seg(struct mlx5_rwqe_sig *sig, int size)
 {
 	sig->signature = calc_sig(sig, size);
 }
 
-int mlx5_ib_post_recv(struct ib_qp *ibqp, struct ib_recv_wr *wr,
-		      struct ib_recv_wr **bad_wr)
+static int _mlx5_ib_post_recv(struct ib_qp *ibqp, const struct ib_recv_wr *wr,
+		      const struct ib_recv_wr **bad_wr, bool drain)
 {
 	struct mlx5_ib_qp *qp = to_mqp(ibqp);
 	struct mlx5_wqe_data_seg *scat;
@@ -4714,7 +4705,7 @@ int mlx5_ib_post_recv(struct ib_qp *ibqp, struct ib_recv_wr *wr,
 
 	spin_lock_irqsave(&qp->rq.lock, flags);
 
-	if (mdev->state == MLX5_DEVICE_STATE_INTERNAL_ERROR) {
+	if (mdev->state == MLX5_DEVICE_STATE_INTERNAL_ERROR && !drain) {
 		err = -EIO;
 		*bad_wr = wr;
 		nreq = 0;
@@ -4776,6 +4767,12 @@ out:
 	return err;
 }
 
+int mlx5_ib_post_recv(struct ib_qp *ibqp, const struct ib_recv_wr *wr,
+		      const struct ib_recv_wr **bad_wr)
+{
+	return _mlx5_ib_post_recv(ibqp, wr, bad_wr, false);
+}
+
 static inline enum ib_qp_state to_ib_qp_state(enum mlx5_qp_state mlx5_state)
 {
 	switch (mlx5_state) {
@@ -5365,7 +5362,9 @@ static int set_user_rq_size(struct mlx5_ib_dev *dev,
 
 	rwq->wqe_count = ucmd->rq_wqe_count;
 	rwq->wqe_shift = ucmd->rq_wqe_shift;
-	rwq->buf_size = (rwq->wqe_count << rwq->wqe_shift);
+	if (check_shl_overflow(rwq->wqe_count, rwq->wqe_shift, &rwq->buf_size))
+		return -EINVAL;
+
 	rwq->log_rq_stride = rwq->wqe_shift;
 	rwq->log_rq_size = ilog2(rwq->wqe_count);
 	return 0;
@@ -5697,3 +5696,132 @@ out:
 	kvfree(in);
 	return err;
 }
+
+struct mlx5_ib_drain_cqe {
+	struct ib_cqe cqe;
+	struct completion done;
+};
+
+static void mlx5_ib_drain_qp_done(struct ib_cq *cq, struct ib_wc *wc)
+{
+	struct mlx5_ib_drain_cqe *cqe = container_of(wc->wr_cqe,
+						     struct mlx5_ib_drain_cqe,
+						     cqe);
+
+	complete(&cqe->done);
+}
+
+/* This function returns only once the drained WR was completed */
+static void handle_drain_completion(struct ib_cq *cq,
+				    struct mlx5_ib_drain_cqe *sdrain,
+				    struct mlx5_ib_dev *dev)
+{
+	struct mlx5_core_dev *mdev = dev->mdev;
+
+	if (cq->poll_ctx == IB_POLL_DIRECT) {
+		while (wait_for_completion_timeout(&sdrain->done, HZ / 10) <= 0)
+			ib_process_cq_direct(cq, -1);
+		return;
+	}
+
+	if (mdev->state == MLX5_DEVICE_STATE_INTERNAL_ERROR) {
+		struct mlx5_ib_cq *mcq = to_mcq(cq);
+		bool triggered = false;
+		unsigned long flags;
+
+		spin_lock_irqsave(&dev->reset_flow_resource_lock, flags);
+		/* Make sure that the CQ handler won't run if wasn't run yet */
+		if (!mcq->mcq.reset_notify_added)
+			mcq->mcq.reset_notify_added = 1;
+		else
+			triggered = true;
+		spin_unlock_irqrestore(&dev->reset_flow_resource_lock, flags);
+
+		if (triggered) {
+			/* Wait for any scheduled/running task to be ended */
+			switch (cq->poll_ctx) {
+			case IB_POLL_SOFTIRQ:
+				irq_poll_disable(&cq->iop);
+				irq_poll_enable(&cq->iop);
+				break;
+			case IB_POLL_WORKQUEUE:
+				cancel_work_sync(&cq->work);
+				break;
+			default:
+				WARN_ON_ONCE(1);
+			}
+		}
+
+		/* Run the CQ handler - this makes sure that the drain WR will
+		 * be processed if wasn't processed yet.
+		 */
+		mcq->mcq.comp(&mcq->mcq);
+	}
+
+	wait_for_completion(&sdrain->done);
+}
+
+void mlx5_ib_drain_sq(struct ib_qp *qp)
+{
+	struct ib_cq *cq = qp->send_cq;
+	struct ib_qp_attr attr = { .qp_state = IB_QPS_ERR };
+	struct mlx5_ib_drain_cqe sdrain;
+	const struct ib_send_wr *bad_swr;
+	struct ib_rdma_wr swr = {
+		.wr = {
+			.next = NULL,
+			{ .wr_cqe	= &sdrain.cqe, },
+			.opcode	= IB_WR_RDMA_WRITE,
+		},
+	};
+	int ret;
+	struct mlx5_ib_dev *dev = to_mdev(qp->device);
+	struct mlx5_core_dev *mdev = dev->mdev;
+
+	ret = ib_modify_qp(qp, &attr, IB_QP_STATE);
+	if (ret && mdev->state != MLX5_DEVICE_STATE_INTERNAL_ERROR) {
+		WARN_ONCE(ret, "failed to drain send queue: %d\n", ret);
+		return;
+	}
+
+	sdrain.cqe.done = mlx5_ib_drain_qp_done;
+	init_completion(&sdrain.done);
+
+	ret = _mlx5_ib_post_send(qp, &swr.wr, &bad_swr, true);
+	if (ret) {
+		WARN_ONCE(ret, "failed to drain send queue: %d\n", ret);
+		return;
+	}
+
+	handle_drain_completion(cq, &sdrain, dev);
+}
+
+void mlx5_ib_drain_rq(struct ib_qp *qp)
+{
+	struct ib_cq *cq = qp->recv_cq;
+	struct ib_qp_attr attr = { .qp_state = IB_QPS_ERR };
+	struct mlx5_ib_drain_cqe rdrain;
+	struct ib_recv_wr rwr = {};
+	const struct ib_recv_wr *bad_rwr;
+	int ret;
+	struct mlx5_ib_dev *dev = to_mdev(qp->device);
+	struct mlx5_core_dev *mdev = dev->mdev;
+
+	ret = ib_modify_qp(qp, &attr, IB_QP_STATE);
+	if (ret && mdev->state != MLX5_DEVICE_STATE_INTERNAL_ERROR) {
+		WARN_ONCE(ret, "failed to drain recv queue: %d\n", ret);
+		return;
+	}
+
+	rwr.wr_cqe = &rdrain.cqe;
+	rdrain.cqe.done = mlx5_ib_drain_qp_done;
+	init_completion(&rdrain.done);
+
+	ret = _mlx5_ib_post_recv(qp, &rwr, &bad_rwr, true);
+	if (ret) {
+		WARN_ONCE(ret, "failed to drain recv queue: %d\n", ret);
+		return;
+	}
+
+	handle_drain_completion(cq, &rdrain, dev);
+}
diff --git a/drivers/infiniband/hw/mlx5/srq.c b/drivers/infiniband/hw/mlx5/srq.c
index f5de5adc9b1a..d359fecf7a5b 100644
--- a/drivers/infiniband/hw/mlx5/srq.c
+++ b/drivers/infiniband/hw/mlx5/srq.c
@@ -446,8 +446,8 @@ void mlx5_ib_free_srq_wqe(struct mlx5_ib_srq *srq, int wqe_index)
 	spin_unlock(&srq->lock);
 }
 
-int mlx5_ib_post_srq_recv(struct ib_srq *ibsrq, struct ib_recv_wr *wr,
-			  struct ib_recv_wr **bad_wr)
+int mlx5_ib_post_srq_recv(struct ib_srq *ibsrq, const struct ib_recv_wr *wr,
+			  const struct ib_recv_wr **bad_wr)
 {
 	struct mlx5_ib_srq *srq = to_msrq(ibsrq);
 	struct mlx5_wqe_srq_next_seg *next;