Update Linux to v5.4.148

Sourced from [1]

[1] https://cdn.kernel.org/pub/linux/kernel/v5.x/linux-5.4.148.tar.gz

Change-Id: Ib3d26c5ba9b022e2e03533005c4fed4d7c30b61b
Signed-off-by: Olivier Deprez <olivier.deprez@arm.com>
diff --git a/drivers/infiniband/hw/mlx5/cq.c b/drivers/infiniband/hw/mlx5/cq.c
index 45f48cd..23ce012 100644
--- a/drivers/infiniband/hw/mlx5/cq.c
+++ b/drivers/infiniband/hw/mlx5/cq.c
@@ -167,7 +167,7 @@
 {
 	enum rdma_link_layer ll = rdma_port_get_link_layer(qp->ibqp.device, 1);
 	struct mlx5_ib_dev *dev = to_mdev(qp->ibqp.device);
-	struct mlx5_ib_srq *srq;
+	struct mlx5_ib_srq *srq = NULL;
 	struct mlx5_ib_wq *wq;
 	u16 wqe_ctr;
 	u8  roce_packet_type;
@@ -179,7 +179,8 @@
 
 		if (qp->ibqp.xrcd) {
 			msrq = mlx5_cmd_get_srq(dev, be32_to_cpu(cqe->srqn));
-			srq = to_mibsrq(msrq);
+			if (msrq)
+				srq = to_mibsrq(msrq);
 		} else {
 			srq = to_msrq(qp->ibqp.srq);
 		}
@@ -330,6 +331,22 @@
 		dump_cqe(dev, cqe);
 }
 
+static void handle_atomics(struct mlx5_ib_qp *qp, struct mlx5_cqe64 *cqe64,
+			   u16 tail, u16 head)
+{
+	u16 idx;
+
+	do {
+		idx = tail & (qp->sq.wqe_cnt - 1);
+		if (idx == head)
+			break;
+
+		tail = qp->sq.w_list[idx].next;
+	} while (1);
+	tail = qp->sq.w_list[idx].next;
+	qp->sq.last_poll = tail;
+}
+
 static void free_cq_buf(struct mlx5_ib_dev *dev, struct mlx5_ib_cq_buf *buf)
 {
 	mlx5_frag_buf_free(dev->mdev, &buf->frag_buf);
@@ -368,7 +385,7 @@
 }
 
 static void sw_comp(struct mlx5_ib_qp *qp, int num_entries, struct ib_wc *wc,
-		    int *npolled, int is_send)
+		    int *npolled, bool is_send)
 {
 	struct mlx5_ib_wq *wq;
 	unsigned int cur;
@@ -383,10 +400,16 @@
 		return;
 
 	for (i = 0;  i < cur && np < num_entries; i++) {
-		wc->wr_id = wq->wrid[wq->tail & (wq->wqe_cnt - 1)];
+		unsigned int idx;
+
+		idx = (is_send) ? wq->last_poll : wq->tail;
+		idx &= (wq->wqe_cnt - 1);
+		wc->wr_id = wq->wrid[idx];
 		wc->status = IB_WC_WR_FLUSH_ERR;
 		wc->vendor_err = MLX5_CQE_SYNDROME_WR_FLUSH_ERR;
 		wq->tail++;
+		if (is_send)
+			wq->last_poll = wq->w_list[idx].next;
 		np++;
 		wc->qp = &qp->ibqp;
 		wc++;
@@ -476,6 +499,7 @@
 		wqe_ctr = be16_to_cpu(cqe64->wqe_counter);
 		idx = wqe_ctr & (wq->wqe_cnt - 1);
 		handle_good_req(wc, cqe64, wq, idx);
+		handle_atomics(*cur_qp, cqe64, wq->last_poll, idx);
 		wc->wr_id = wq->wrid[idx];
 		wq->tail = wq->wqe_head[idx] + 1;
 		wc->status = IB_WC_SUCCESS;
@@ -805,15 +829,14 @@
 	ib_umem_release(cq->buf.umem);
 }
 
-static void init_cq_frag_buf(struct mlx5_ib_cq *cq,
-			     struct mlx5_ib_cq_buf *buf)
+static void init_cq_frag_buf(struct mlx5_ib_cq_buf *buf)
 {
 	int i;
 	void *cqe;
 	struct mlx5_cqe64 *cqe64;
 
 	for (i = 0; i < buf->nent; i++) {
-		cqe = get_cqe(cq, i);
+		cqe = mlx5_frag_buf_get_wqe(&buf->fbc, i);
 		cqe64 = buf->cqe_size == 64 ? cqe : cqe + 64;
 		cqe64->op_own = MLX5_CQE_INVALID << 4;
 	}
@@ -839,7 +862,7 @@
 	if (err)
 		goto err_db;
 
-	init_cq_frag_buf(cq, &cq->buf);
+	init_cq_frag_buf(&cq->buf);
 
 	*inlen = MLX5_ST_SZ_BYTES(create_cq_in) +
 		 MLX5_FLD_SZ_BYTES(create_cq_in, pas[0]) *
@@ -1139,7 +1162,7 @@
 	if (err)
 		goto ex;
 
-	init_cq_frag_buf(cq, cq->resize_buf);
+	init_cq_frag_buf(cq->resize_buf);
 
 	return 0;
 
diff --git a/drivers/infiniband/hw/mlx5/devx.c b/drivers/infiniband/hw/mlx5/devx.c
index d609f46..664e0f3 100644
--- a/drivers/infiniband/hw/mlx5/devx.c
+++ b/drivers/infiniband/hw/mlx5/devx.c
@@ -489,6 +489,10 @@
 		obj_id = get_enc_obj_id(MLX5_CMD_OP_CREATE_QP,
 					MLX5_GET(rst2init_qp_in, in, qpn));
 		break;
+	case MLX5_CMD_OP_INIT2INIT_QP:
+		obj_id = get_enc_obj_id(MLX5_CMD_OP_CREATE_QP,
+					MLX5_GET(init2init_qp_in, in, qpn));
+		break;
 	case MLX5_CMD_OP_INIT2RTR_QP:
 		obj_id = get_enc_obj_id(MLX5_CMD_OP_CREATE_QP,
 					MLX5_GET(init2rtr_qp_in, in, qpn));
@@ -814,6 +818,7 @@
 	case MLX5_CMD_OP_SET_L2_TABLE_ENTRY:
 	case MLX5_CMD_OP_RST2INIT_QP:
 	case MLX5_CMD_OP_INIT2RTR_QP:
+	case MLX5_CMD_OP_INIT2INIT_QP:
 	case MLX5_CMD_OP_RTR2RTS_QP:
 	case MLX5_CMD_OP_RTS2RTS_QP:
 	case MLX5_CMD_OP_SQERR2RTS_QP:
@@ -1113,7 +1118,9 @@
 		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);
+		*obj_id = MLX5_GET(create_tir_out, out, tirn);
+		MLX5_SET(destroy_tir_in, din, opcode, MLX5_CMD_OP_DESTROY_TIR);
+		MLX5_SET(destroy_tir_in, din, tirn, *obj_id);
 		break;
 	case MLX5_CMD_OP_CREATE_TIS:
 		MLX5_SET(general_obj_in_cmd_hdr, din, opcode, MLX5_CMD_OP_DESTROY_TIS);
@@ -2015,8 +2022,10 @@
 
 		num_alloc_xa_entries++;
 		event_sub = kzalloc(sizeof(*event_sub), GFP_KERNEL);
-		if (!event_sub)
+		if (!event_sub) {
+			err = -ENOMEM;
 			goto err;
+		}
 
 		list_add_tail(&event_sub->event_list, &sub_list);
 		if (use_eventfd) {
diff --git a/drivers/infiniband/hw/mlx5/flow.c b/drivers/infiniband/hw/mlx5/flow.c
index b198ff1..fddefb2 100644
--- a/drivers/infiniband/hw/mlx5/flow.c
+++ b/drivers/infiniband/hw/mlx5/flow.c
@@ -13,6 +13,7 @@
 #include <rdma/ib_umem.h>
 #include <linux/mlx5/driver.h>
 #include <linux/mlx5/fs.h>
+#include <linux/mlx5/eswitch.h>
 #include "mlx5_ib.h"
 
 #define UVERBS_MODULE_NAME mlx5_ib
@@ -316,6 +317,13 @@
 	if (err)
 		goto end;
 
+	if (obj->ns_type == MLX5_FLOW_NAMESPACE_FDB &&
+	    mlx5_eswitch_mode(dev->mdev->priv.eswitch) !=
+			      MLX5_ESWITCH_OFFLOADS) {
+		err = -EINVAL;
+		goto end;
+	}
+
 	uobj->object = obj;
 	obj->mdev = dev->mdev;
 	atomic_set(&obj->usecnt, 0);
diff --git a/drivers/infiniband/hw/mlx5/gsi.c b/drivers/infiniband/hw/mlx5/gsi.c
index 4950df3..5c73c0a 100644
--- a/drivers/infiniband/hw/mlx5/gsi.c
+++ b/drivers/infiniband/hw/mlx5/gsi.c
@@ -507,8 +507,7 @@
 		ret = ib_post_send(tx_qp, &cur_wr.wr, bad_wr);
 		if (ret) {
 			/* Undo the effect of adding the outstanding wr */
-			gsi->outstanding_pi = (gsi->outstanding_pi - 1) %
-					      gsi->cap.max_send_wr;
+			gsi->outstanding_pi--;
 			goto err;
 		}
 		spin_unlock_irqrestore(&gsi->lock, flags);
diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c
index 8315394..9025086 100644
--- a/drivers/infiniband/hw/mlx5/main.c
+++ b/drivers/infiniband/hw/mlx5/main.c
@@ -517,7 +517,7 @@
 					   mdev_port_num);
 	if (err)
 		goto out;
-	ext = MLX5_CAP_PCAM_FEATURE(dev->mdev, ptys_extended_ethernet);
+	ext = !!MLX5_GET_ETH_PROTO(ptys_reg, out, true, eth_proto_capability);
 	eth_prot_oper = MLX5_GET_ETH_PROTO(ptys_reg, out, ext, eth_proto_oper);
 
 	props->active_width     = IB_WIDTH_4X;
@@ -829,6 +829,7 @@
 				struct ib_device_attr *props,
 				struct ib_udata *uhw)
 {
+	size_t uhw_outlen = (uhw) ? uhw->outlen : 0;
 	struct mlx5_ib_dev *dev = to_mdev(ibdev);
 	struct mlx5_core_dev *mdev = dev->mdev;
 	int err = -ENOMEM;
@@ -842,12 +843,12 @@
 	u64 max_tso;
 
 	resp_len = sizeof(resp.comp_mask) + sizeof(resp.response_length);
-	if (uhw->outlen && uhw->outlen < resp_len)
+	if (uhw_outlen && uhw_outlen < resp_len)
 		return -EINVAL;
 	else
 		resp.response_length = resp_len;
 
-	if (uhw->inlen && !ib_is_udata_cleared(uhw, 0, uhw->inlen))
+	if (uhw && uhw->inlen && !ib_is_udata_cleared(uhw, 0, uhw->inlen))
 		return -EINVAL;
 
 	memset(props, 0, sizeof(*props));
@@ -887,7 +888,9 @@
 		/* We support 'Gappy' memory registration too */
 		props->device_cap_flags |= IB_DEVICE_SG_GAPS_REG;
 	}
-	props->device_cap_flags |= IB_DEVICE_MEM_MGT_EXTENSIONS;
+	/* IB_WR_REG_MR always requires changing the entity size with UMR */
+	if (!MLX5_CAP_GEN(dev->mdev, umr_modify_entity_size_disabled))
+		props->device_cap_flags |= IB_DEVICE_MEM_MGT_EXTENSIONS;
 	if (MLX5_CAP_GEN(mdev, sho)) {
 		props->device_cap_flags |= IB_DEVICE_INTEGRITY_HANDOVER;
 		/* At this stage no support for signature handover */
@@ -911,7 +914,7 @@
 			props->raw_packet_caps |=
 				IB_RAW_PACKET_CAP_CVLAN_STRIPPING;
 
-		if (field_avail(typeof(resp), tso_caps, uhw->outlen)) {
+		if (field_avail(typeof(resp), tso_caps, uhw_outlen)) {
 			max_tso = MLX5_CAP_ETH(mdev, max_lso_cap);
 			if (max_tso) {
 				resp.tso_caps.max_tso = 1 << max_tso;
@@ -921,7 +924,7 @@
 			}
 		}
 
-		if (field_avail(typeof(resp), rss_caps, uhw->outlen)) {
+		if (field_avail(typeof(resp), rss_caps, uhw_outlen)) {
 			resp.rss_caps.rx_hash_function =
 						MLX5_RX_HASH_FUNC_TOEPLITZ;
 			resp.rss_caps.rx_hash_fields_mask =
@@ -941,9 +944,9 @@
 			resp.response_length += sizeof(resp.rss_caps);
 		}
 	} else {
-		if (field_avail(typeof(resp), tso_caps, uhw->outlen))
+		if (field_avail(typeof(resp), tso_caps, uhw_outlen))
 			resp.response_length += sizeof(resp.tso_caps);
-		if (field_avail(typeof(resp), rss_caps, uhw->outlen))
+		if (field_avail(typeof(resp), rss_caps, uhw_outlen))
 			resp.response_length += sizeof(resp.rss_caps);
 	}
 
@@ -1066,7 +1069,7 @@
 						MLX5_MAX_CQ_PERIOD;
 	}
 
-	if (field_avail(typeof(resp), cqe_comp_caps, uhw->outlen)) {
+	if (field_avail(typeof(resp), cqe_comp_caps, uhw_outlen)) {
 		resp.response_length += sizeof(resp.cqe_comp_caps);
 
 		if (MLX5_CAP_GEN(dev->mdev, cqe_compression)) {
@@ -1084,7 +1087,7 @@
 		}
 	}
 
-	if (field_avail(typeof(resp), packet_pacing_caps, uhw->outlen) &&
+	if (field_avail(typeof(resp), packet_pacing_caps, uhw_outlen) &&
 	    raw_support) {
 		if (MLX5_CAP_QOS(mdev, packet_pacing) &&
 		    MLX5_CAP_GEN(mdev, qos)) {
@@ -1103,7 +1106,7 @@
 	}
 
 	if (field_avail(typeof(resp), mlx5_ib_support_multi_pkt_send_wqes,
-			uhw->outlen)) {
+			uhw_outlen)) {
 		if (MLX5_CAP_ETH(mdev, multi_pkt_send_wqe))
 			resp.mlx5_ib_support_multi_pkt_send_wqes =
 				MLX5_IB_ALLOW_MPW;
@@ -1116,7 +1119,7 @@
 			sizeof(resp.mlx5_ib_support_multi_pkt_send_wqes);
 	}
 
-	if (field_avail(typeof(resp), flags, uhw->outlen)) {
+	if (field_avail(typeof(resp), flags, uhw_outlen)) {
 		resp.response_length += sizeof(resp.flags);
 
 		if (MLX5_CAP_GEN(mdev, cqe_compression_128))
@@ -1132,8 +1135,7 @@
 		resp.flags |= MLX5_IB_QUERY_DEV_RESP_FLAGS_SCAT2CQE_DCT;
 	}
 
-	if (field_avail(typeof(resp), sw_parsing_caps,
-			uhw->outlen)) {
+	if (field_avail(typeof(resp), sw_parsing_caps, uhw_outlen)) {
 		resp.response_length += sizeof(resp.sw_parsing_caps);
 		if (MLX5_CAP_ETH(mdev, swp)) {
 			resp.sw_parsing_caps.sw_parsing_offloads |=
@@ -1153,7 +1155,7 @@
 		}
 	}
 
-	if (field_avail(typeof(resp), striding_rq_caps, uhw->outlen) &&
+	if (field_avail(typeof(resp), striding_rq_caps, uhw_outlen) &&
 	    raw_support) {
 		resp.response_length += sizeof(resp.striding_rq_caps);
 		if (MLX5_CAP_GEN(mdev, striding_rq)) {
@@ -1170,8 +1172,7 @@
 		}
 	}
 
-	if (field_avail(typeof(resp), tunnel_offloads_caps,
-			uhw->outlen)) {
+	if (field_avail(typeof(resp), tunnel_offloads_caps, uhw_outlen)) {
 		resp.response_length += sizeof(resp.tunnel_offloads_caps);
 		if (MLX5_CAP_ETH(mdev, tunnel_stateless_vxlan))
 			resp.tunnel_offloads_caps |=
@@ -1182,17 +1183,15 @@
 		if (MLX5_CAP_ETH(mdev, tunnel_stateless_gre))
 			resp.tunnel_offloads_caps |=
 				MLX5_IB_TUNNELED_OFFLOADS_GRE;
-		if (MLX5_CAP_GEN(mdev, flex_parser_protocols) &
-		    MLX5_FLEX_PROTO_CW_MPLS_GRE)
+		if (MLX5_CAP_ETH(mdev, tunnel_stateless_mpls_over_gre))
 			resp.tunnel_offloads_caps |=
 				MLX5_IB_TUNNELED_OFFLOADS_MPLS_GRE;
-		if (MLX5_CAP_GEN(mdev, flex_parser_protocols) &
-		    MLX5_FLEX_PROTO_CW_MPLS_UDP)
+		if (MLX5_CAP_ETH(mdev, tunnel_stateless_mpls_over_udp))
 			resp.tunnel_offloads_caps |=
 				MLX5_IB_TUNNELED_OFFLOADS_MPLS_UDP;
 	}
 
-	if (uhw->outlen) {
+	if (uhw_outlen) {
 		err = ib_copy_to_udata(uhw, &resp, resp.response_length);
 
 		if (err)
@@ -3548,10 +3547,6 @@
 	}
 
 	INIT_LIST_HEAD(&handler->list);
-	if (dst) {
-		memcpy(&dest_arr[0], dst, sizeof(*dst));
-		dest_num++;
-	}
 
 	for (spec_index = 0; spec_index < flow_attr->num_of_specs; spec_index++) {
 		err = parse_flow_attr(dev->mdev, spec,
@@ -3564,6 +3559,11 @@
 		ib_flow += ((union ib_flow_spec *)ib_flow)->size;
 	}
 
+	if (dst && !(flow_act.action & MLX5_FLOW_CONTEXT_ACTION_DROP)) {
+		memcpy(&dest_arr[0], dst, sizeof(*dst));
+		dest_num++;
+	}
+
 	if (!flow_is_multicast_only(flow_attr))
 		set_underlay_qp(dev, spec, underlay_qpn);
 
@@ -3604,10 +3604,8 @@
 	}
 
 	if (flow_act.action & MLX5_FLOW_CONTEXT_ACTION_DROP) {
-		if (!(flow_act.action & MLX5_FLOW_CONTEXT_ACTION_COUNT)) {
+		if (!dest_num)
 			rule_dst = NULL;
-			dest_num = 0;
-		}
 	} else {
 		if (is_egress)
 			flow_act.action |= MLX5_FLOW_CONTEXT_ACTION_ALLOW;
@@ -4739,7 +4737,6 @@
 	struct ib_device_attr *dprops = NULL;
 	struct ib_port_attr *pprops = NULL;
 	int err = -ENOMEM;
-	struct ib_udata uhw = {.inlen = 0, .outlen = 0};
 
 	pprops = kzalloc(sizeof(*pprops), GFP_KERNEL);
 	if (!pprops)
@@ -4749,7 +4746,7 @@
 	if (!dprops)
 		goto out;
 
-	err = mlx5_ib_query_device(&dev->ib_dev, dprops, &uhw);
+	err = mlx5_ib_query_device(&dev->ib_dev, dprops, NULL);
 	if (err) {
 		mlx5_ib_warn(dev, "query_device failed %d\n", err);
 		goto out;
@@ -5641,9 +5638,10 @@
 	const struct mlx5_ib_counters *cnts =
 		get_counters(dev, counter->port - 1);
 
-	/* Q counters are in the beginning of all counters */
 	return rdma_alloc_hw_stats_struct(cnts->names,
-					  cnts->num_q_counters,
+					  cnts->num_q_counters +
+					  cnts->num_cong_counters +
+					  cnts->num_ext_ppcnt_counters,
 					  RDMA_HW_STATS_DEFAULT_LIFESPAN);
 }
 
@@ -5873,8 +5871,6 @@
 
 	port->mp.mpi = NULL;
 
-	list_add_tail(&mpi->list, &mlx5_ib_unaffiliated_port_list);
-
 	spin_unlock(&port->mp.mpi_lock);
 
 	err = mlx5_nic_vport_unaffiliate_multiport(mpi->mdev);
@@ -6027,6 +6023,8 @@
 				dev->port[i].mp.mpi = NULL;
 			} else {
 				mlx5_ib_dbg(dev, "unbinding port_num: %d\n", i + 1);
+				list_add_tail(&dev->port[i].mp.mpi->list,
+					      &mlx5_ib_unaffiliated_port_list);
 				mlx5_ib_unbind_slave_port(dev, dev->port[i].mp.mpi);
 			}
 		}
@@ -6175,7 +6173,7 @@
 
 	err = set_has_smi_cap(dev);
 	if (err)
-		return err;
+		goto err_mp;
 
 	if (!mlx5_core_mp_enabled(mdev)) {
 		for (i = 1; i <= dev->num_ports; i++) {
@@ -6215,8 +6213,7 @@
 
 err_mp:
 	mlx5_ib_cleanup_multiport_master(dev);
-
-	return -ENOMEM;
+	return err;
 }
 
 static int mlx5_ib_stage_flow_db_init(struct mlx5_ib_dev *dev)
@@ -6628,7 +6625,7 @@
 
 	err = mlx5_alloc_bfreg(dev->mdev, &dev->fp_bfreg, false, true);
 	if (err)
-		mlx5_free_bfreg(dev->mdev, &dev->fp_bfreg);
+		mlx5_free_bfreg(dev->mdev, &dev->bfreg);
 
 	return err;
 }
@@ -6882,6 +6879,7 @@
 
 		if (bound) {
 			rdma_roce_rescan_device(&dev->ib_dev);
+			mpi->ibdev->ib_active = true;
 			break;
 		}
 	}
diff --git a/drivers/infiniband/hw/mlx5/mlx5_ib.h b/drivers/infiniband/hw/mlx5/mlx5_ib.h
index 1a98ee2..a9ce46c 100644
--- a/drivers/infiniband/hw/mlx5/mlx5_ib.h
+++ b/drivers/infiniband/hw/mlx5/mlx5_ib.h
@@ -283,6 +283,7 @@
 	unsigned		head;
 	unsigned		tail;
 	u16			cur_post;
+	u16			last_poll;
 	void			*cur_edge;
 };
 
diff --git a/drivers/infiniband/hw/mlx5/mr.c b/drivers/infiniband/hw/mlx5/mr.c
index 7019c12..99d563d 100644
--- a/drivers/infiniband/hw/mlx5/mr.c
+++ b/drivers/infiniband/hw/mlx5/mr.c
@@ -428,7 +428,7 @@
 
 	if (entry < 0 || entry >= MAX_MR_CACHE_ENTRIES) {
 		mlx5_ib_err(dev, "cache entry %d is out of range\n", entry);
-		return NULL;
+		return ERR_PTR(-EINVAL);
 	}
 
 	ent = &cache->ent[entry];
diff --git a/drivers/infiniband/hw/mlx5/qp.c b/drivers/infiniband/hw/mlx5/qp.c
index 5fd071c..4540835 100644
--- a/drivers/infiniband/hw/mlx5/qp.c
+++ b/drivers/infiniband/hw/mlx5/qp.c
@@ -1463,6 +1463,8 @@
 	u16 uid = to_mpd(pd)->uid;
 	u32 out[MLX5_ST_SZ_DW(create_tir_out)] = {};
 
+	if (!qp->sq.wqe_cnt && !qp->rq.wqe_cnt)
+		return -EINVAL;
 	if (qp->sq.wqe_cnt) {
 		err = create_raw_packet_qp_tis(dev, qp, sq, tdn, pd);
 		if (err)
@@ -3391,9 +3393,6 @@
 	struct mlx5_ib_qp_base *base;
 	u32 set_id;
 
-	if (!MLX5_CAP_GEN(dev->mdev, rts2rts_qp_counters_set_id))
-		return 0;
-
 	if (counter)
 		set_id = counter->id;
 	else
@@ -3728,6 +3727,7 @@
 		qp->sq.cur_post = 0;
 		if (qp->sq.wqe_cnt)
 			qp->sq.cur_edge = get_sq_edge(&qp->sq, 0);
+		qp->sq.last_poll = 0;
 		qp->db.db[MLX5_RCV_DBR] = 0;
 		qp->db.db[MLX5_SND_DBR] = 0;
 	}
@@ -5498,7 +5498,9 @@
 	rdma_ah_set_path_bits(ah_attr, path->grh_mlid & 0x7f);
 	rdma_ah_set_static_rate(ah_attr,
 				path->static_rate ? path->static_rate - 5 : 0);
-	if (path->grh_mlid & (1 << 7)) {
+
+	if (path->grh_mlid & (1 << 7) ||
+	    ah_attr->type == RDMA_AH_ATTR_TYPE_ROCE) {
 		u32 tc_fl = be32_to_cpu(path->tclass_flowlabel);
 
 		rdma_ah_set_grh(ah_attr, NULL,
@@ -6134,6 +6136,10 @@
 	if (udata->outlen && udata->outlen < min_resp_len)
 		return ERR_PTR(-EINVAL);
 
+	if (!capable(CAP_SYS_RAWIO) &&
+	    init_attr->create_flags & IB_WQ_FLAGS_DELAY_DROP)
+		return ERR_PTR(-EPERM);
+
 	dev = to_mdev(pd->device);
 	switch (init_attr->wq_type) {
 	case IB_WQT_RQ:
@@ -6311,10 +6317,8 @@
 
 	rqc = MLX5_ADDR_OF(modify_rq_in, in, ctx);
 
-	curr_wq_state = (wq_attr_mask & IB_WQ_CUR_STATE) ?
-		wq_attr->curr_wq_state : wq->state;
-	wq_state = (wq_attr_mask & IB_WQ_STATE) ?
-		wq_attr->wq_state : curr_wq_state;
+	curr_wq_state = wq_attr->curr_wq_state;
+	wq_state = wq_attr->wq_state;
 	if (curr_wq_state == IB_WQS_ERR)
 		curr_wq_state = MLX5_RQC_STATE_ERR;
 	if (wq_state == IB_WQS_ERR)
@@ -6503,6 +6507,7 @@
  */
 int mlx5_ib_qp_set_counter(struct ib_qp *qp, struct rdma_counter *counter)
 {
+	struct mlx5_ib_dev *dev = to_mdev(qp->device);
 	struct mlx5_ib_qp *mqp = to_mqp(qp);
 	int err = 0;
 
@@ -6512,6 +6517,11 @@
 		goto out;
 	}
 
+	if (!MLX5_CAP_GEN(dev->mdev, rts2rts_qp_counters_set_id)) {
+		err = -EOPNOTSUPP;
+		goto out;
+	}
+
 	if (mqp->state == IB_QPS_RTS) {
 		err = __mlx5_ib_qp_set_counter(qp, counter);
 		if (!err)
diff --git a/drivers/infiniband/hw/mlx5/srq.c b/drivers/infiniband/hw/mlx5/srq.c
index 4e7fde8..c29c1f7 100644
--- a/drivers/infiniband/hw/mlx5/srq.c
+++ b/drivers/infiniband/hw/mlx5/srq.c
@@ -310,12 +310,18 @@
 	srq->msrq.event = mlx5_ib_srq_event;
 	srq->ibsrq.ext.xrc.srq_num = srq->msrq.srqn;
 
-	if (udata)
-		if (ib_copy_to_udata(udata, &srq->msrq.srqn, sizeof(__u32))) {
+	if (udata) {
+		struct mlx5_ib_create_srq_resp resp = {
+			.srqn = srq->msrq.srqn,
+		};
+
+		if (ib_copy_to_udata(udata, &resp, min(udata->outlen,
+				     sizeof(resp)))) {
 			mlx5_ib_dbg(dev, "copy to user failed\n");
 			err = -EFAULT;
 			goto err_core;
 		}
+	}
 
 	init_attr->attr.max_wr = srq->msrq.max - 1;
 
diff --git a/drivers/infiniband/hw/mlx5/srq_cmd.c b/drivers/infiniband/hw/mlx5/srq_cmd.c
index 8fc3630..0224231 100644
--- a/drivers/infiniband/hw/mlx5/srq_cmd.c
+++ b/drivers/infiniband/hw/mlx5/srq_cmd.c
@@ -83,11 +83,11 @@
 	struct mlx5_srq_table *table = &dev->srq_table;
 	struct mlx5_core_srq *srq;
 
-	xa_lock(&table->array);
+	xa_lock_irq(&table->array);
 	srq = xa_load(&table->array, srqn);
 	if (srq)
 		refcount_inc(&srq->common.refcount);
-	xa_unlock(&table->array);
+	xa_unlock_irq(&table->array);
 
 	return srq;
 }