1 files changed, 192 insertions, 99 deletions

diff --git a/drivers/infiniband/hw/mlx5/qp.c b/drivers/infiniband/hw/mlx5/qp.c
index 379328b2598f..957f3a52589b 100644
--- a/drivers/infiniband/hw/mlx5/qp.c
+++ b/drivers/infiniband/hw/mlx5/qp.c
@@ -129,14 +129,10 @@ static int is_sqp(enum ib_qp_type qp_type)
  *
  * Return: zero on success, or an error code.
  */
-static int mlx5_ib_read_user_wqe_common(struct ib_umem *umem,
-					void *buffer,
-					u32 buflen,
-					int wqe_index,
-					int wq_offset,
-					int wq_wqe_cnt,
-					int wq_wqe_shift,
-					int bcnt,
+static int mlx5_ib_read_user_wqe_common(struct ib_umem *umem, void *buffer,
+					size_t buflen, int wqe_index,
+					int wq_offset, int wq_wqe_cnt,
+					int wq_wqe_shift, int bcnt,
 					size_t *bytes_copied)
 {
 	size_t offset = wq_offset + ((wqe_index % wq_wqe_cnt) << wq_wqe_shift);
@@ -160,11 +156,43 @@ static int mlx5_ib_read_user_wqe_common(struct ib_umem *umem,
 	return 0;
 }
 
-int mlx5_ib_read_user_wqe_sq(struct mlx5_ib_qp *qp,
-			     int wqe_index,
-			     void *buffer,
-			     int buflen,
-			     size_t *bc)
+static int mlx5_ib_read_kernel_wqe_sq(struct mlx5_ib_qp *qp, int wqe_index,
+				      void *buffer, size_t buflen, size_t *bc)
+{
+	struct mlx5_wqe_ctrl_seg *ctrl;
+	size_t bytes_copied = 0;
+	size_t wqe_length;
+	void *p;
+	int ds;
+
+	wqe_index = wqe_index & qp->sq.fbc.sz_m1;
+
+	/* read the control segment first */
+	p = mlx5_frag_buf_get_wqe(&qp->sq.fbc, wqe_index);
+	ctrl = p;
+	ds = be32_to_cpu(ctrl->qpn_ds) & MLX5_WQE_CTRL_DS_MASK;
+	wqe_length = ds * MLX5_WQE_DS_UNITS;
+
+	/* read rest of WQE if it spreads over more than one stride */
+	while (bytes_copied < wqe_length) {
+		size_t copy_length =
+			min_t(size_t, buflen - bytes_copied, MLX5_SEND_WQE_BB);
+
+		if (!copy_length)
+			break;
+
+		memcpy(buffer + bytes_copied, p, copy_length);
+		bytes_copied += copy_length;
+
+		wqe_index = (wqe_index + 1) & qp->sq.fbc.sz_m1;
+		p = mlx5_frag_buf_get_wqe(&qp->sq.fbc, wqe_index);
+	}
+	*bc = bytes_copied;
+	return 0;
+}
+
+static int mlx5_ib_read_user_wqe_sq(struct mlx5_ib_qp *qp, int wqe_index,
+				    void *buffer, size_t buflen, size_t *bc)
 {
 	struct mlx5_ib_qp_base *base = &qp->trans_qp.base;
 	struct ib_umem *umem = base->ubuffer.umem;
@@ -176,18 +204,10 @@ int mlx5_ib_read_user_wqe_sq(struct mlx5_ib_qp *qp,
 	int ret;
 	int ds;
 
-	if (buflen < sizeof(*ctrl))
-		return -EINVAL;
-
 	/* at first read as much as possible */
-	ret = mlx5_ib_read_user_wqe_common(umem,
-					   buffer,
-					   buflen,
-					   wqe_index,
-					   wq->offset,
-					   wq->wqe_cnt,
-					   wq->wqe_shift,
-					   buflen,
+	ret = mlx5_ib_read_user_wqe_common(umem, buffer, buflen, wqe_index,
+					   wq->offset, wq->wqe_cnt,
+					   wq->wqe_shift, buflen,
 					   &bytes_copied);
 	if (ret)
 		return ret;
@@ -210,13 +230,9 @@ int mlx5_ib_read_user_wqe_sq(struct mlx5_ib_qp *qp,
 	 * so read the remaining bytes starting
 	 * from  wqe_index 0
 	 */
-	ret = mlx5_ib_read_user_wqe_common(umem,
-					   buffer + bytes_copied,
-					   buflen - bytes_copied,
-					   0,
-					   wq->offset,
-					   wq->wqe_cnt,
-					   wq->wqe_shift,
+	ret = mlx5_ib_read_user_wqe_common(umem, buffer + bytes_copied,
+					   buflen - bytes_copied, 0, wq->offset,
+					   wq->wqe_cnt, wq->wqe_shift,
 					   wqe_length - bytes_copied,
 					   &bytes_copied2);
 
@@ -226,11 +242,24 @@ int mlx5_ib_read_user_wqe_sq(struct mlx5_ib_qp *qp,
 	return 0;
 }
 
-int mlx5_ib_read_user_wqe_rq(struct mlx5_ib_qp *qp,
-			     int wqe_index,
-			     void *buffer,
-			     int buflen,
-			     size_t *bc)
+int mlx5_ib_read_wqe_sq(struct mlx5_ib_qp *qp, int wqe_index, void *buffer,
+			size_t buflen, size_t *bc)
+{
+	struct mlx5_ib_qp_base *base = &qp->trans_qp.base;
+	struct ib_umem *umem = base->ubuffer.umem;
+
+	if (buflen < sizeof(struct mlx5_wqe_ctrl_seg))
+		return -EINVAL;
+
+	if (!umem)
+		return mlx5_ib_read_kernel_wqe_sq(qp, wqe_index, buffer,
+						  buflen, bc);
+
+	return mlx5_ib_read_user_wqe_sq(qp, wqe_index, buffer, buflen, bc);
+}
+
+static int mlx5_ib_read_user_wqe_rq(struct mlx5_ib_qp *qp, int wqe_index,
+				    void *buffer, size_t buflen, size_t *bc)
 {
 	struct mlx5_ib_qp_base *base = &qp->trans_qp.base;
 	struct ib_umem *umem = base->ubuffer.umem;
@@ -238,14 +267,9 @@ int mlx5_ib_read_user_wqe_rq(struct mlx5_ib_qp *qp,
 	size_t bytes_copied;
 	int ret;
 
-	ret = mlx5_ib_read_user_wqe_common(umem,
-					   buffer,
-					   buflen,
-					   wqe_index,
-					   wq->offset,
-					   wq->wqe_cnt,
-					   wq->wqe_shift,
-					   buflen,
+	ret = mlx5_ib_read_user_wqe_common(umem, buffer, buflen, wqe_index,
+					   wq->offset, wq->wqe_cnt,
+					   wq->wqe_shift, buflen,
 					   &bytes_copied);
 
 	if (ret)
@@ -254,25 +278,33 @@ int mlx5_ib_read_user_wqe_rq(struct mlx5_ib_qp *qp,
 	return 0;
 }
 
-int mlx5_ib_read_user_wqe_srq(struct mlx5_ib_srq *srq,
-			      int wqe_index,
-			      void *buffer,
-			      int buflen,
-			      size_t *bc)
+int mlx5_ib_read_wqe_rq(struct mlx5_ib_qp *qp, int wqe_index, void *buffer,
+			size_t buflen, size_t *bc)
+{
+	struct mlx5_ib_qp_base *base = &qp->trans_qp.base;
+	struct ib_umem *umem = base->ubuffer.umem;
+	struct mlx5_ib_wq *wq = &qp->rq;
+	size_t wqe_size = 1 << wq->wqe_shift;
+
+	if (buflen < wqe_size)
+		return -EINVAL;
+
+	if (!umem)
+		return -EOPNOTSUPP;
+
+	return mlx5_ib_read_user_wqe_rq(qp, wqe_index, buffer, buflen, bc);
+}
+
+static int mlx5_ib_read_user_wqe_srq(struct mlx5_ib_srq *srq, int wqe_index,
+				     void *buffer, size_t buflen, size_t *bc)
 {
 	struct ib_umem *umem = srq->umem;
 	size_t bytes_copied;
 	int ret;
 
-	ret = mlx5_ib_read_user_wqe_common(umem,
-					   buffer,
-					   buflen,
-					   wqe_index,
-					   0,
-					   srq->msrq.max,
-					   srq->msrq.wqe_shift,
-					   buflen,
-					   &bytes_copied);
+	ret = mlx5_ib_read_user_wqe_common(umem, buffer, buflen, wqe_index, 0,
+					   srq->msrq.max, srq->msrq.wqe_shift,
+					   buflen, &bytes_copied);
 
 	if (ret)
 		return ret;
@@ -280,6 +312,21 @@ int mlx5_ib_read_user_wqe_srq(struct mlx5_ib_srq *srq,
 	return 0;
 }
 
+int mlx5_ib_read_wqe_srq(struct mlx5_ib_srq *srq, int wqe_index, void *buffer,
+			 size_t buflen, size_t *bc)
+{
+	struct ib_umem *umem = srq->umem;
+	size_t wqe_size = 1 << srq->msrq.wqe_shift;
+
+	if (buflen < wqe_size)
+		return -EINVAL;
+
+	if (!umem)
+		return -EOPNOTSUPP;
+
+	return mlx5_ib_read_user_wqe_srq(srq, wqe_index, buffer, buflen, bc);
+}
+
 static void mlx5_ib_qp_event(struct mlx5_core_qp *qp, int type)
 {
 	struct ib_qp *ibqp = &to_mibqp(qp)->ibqp;
@@ -749,7 +796,7 @@ static int mlx5_ib_umem_get(struct mlx5_ib_dev *dev, struct ib_udata *udata,
 {
 	int err;
 
-	*umem = ib_umem_get(udata, addr, size, 0, 0);
+	*umem = ib_umem_get(&dev->ib_dev, addr, size, 0);
 	if (IS_ERR(*umem)) {
 		mlx5_ib_dbg(dev, "umem_get failed\n");
 		return PTR_ERR(*umem);
@@ -806,7 +853,7 @@ static int create_user_rq(struct mlx5_ib_dev *dev, struct ib_pd *pd,
 	if (!ucmd->buf_addr)
 		return -EINVAL;
 
-	rwq->umem = ib_umem_get(udata, ucmd->buf_addr, rwq->buf_size, 0, 0);
+	rwq->umem = ib_umem_get(&dev->ib_dev, ucmd->buf_addr, rwq->buf_size, 0);
 	if (IS_ERR(rwq->umem)) {
 		mlx5_ib_dbg(dev, "umem_get failed\n");
 		err = PTR_ERR(rwq->umem);
@@ -1041,11 +1088,14 @@ static int create_kernel_qp(struct mlx5_ib_dev *dev,
 					IB_QP_CREATE_BLOCK_MULTICAST_LOOPBACK |
 					IB_QP_CREATE_IPOIB_UD_LSO |
 					IB_QP_CREATE_NETIF_QP |
-					mlx5_ib_create_qp_sqpn_qp1()))
+					MLX5_IB_QP_CREATE_SQPN_QP1 |
+					MLX5_IB_QP_CREATE_WC_TEST))
 		return -EINVAL;
 
 	if (init_attr->qp_type == MLX5_IB_QPT_REG_UMR)
 		qp->bf.bfreg = &dev->fp_bfreg;
+	else if (init_attr->create_flags & MLX5_IB_QP_CREATE_WC_TEST)
+		qp->bf.bfreg = &dev->wc_bfreg;
 	else
 		qp->bf.bfreg = &dev->bfreg;
 
@@ -1104,7 +1154,7 @@ static int create_kernel_qp(struct mlx5_ib_dev *dev,
 	MLX5_SET(qpc, qpc, fre, 1);
 	MLX5_SET(qpc, qpc, rlky, 1);
 
-	if (init_attr->create_flags & mlx5_ib_create_qp_sqpn_qp1()) {
+	if (init_attr->create_flags & MLX5_IB_QP_CREATE_SQPN_QP1) {
 		MLX5_SET(qpc, qpc, deth_sqpn, 1);
 		qp->flags |= MLX5_IB_QP_SQPN_QP1;
 	}
@@ -1868,7 +1918,7 @@ static void configure_requester_scat_cqe(struct mlx5_ib_dev *dev,
 {
 	enum ib_qp_type qpt = init_attr->qp_type;
 	int scqe_sz;
-	bool allow_scat_cqe = 0;
+	bool allow_scat_cqe = false;
 
 	if (qpt == IB_QPT_UC || qpt == IB_QPT_UD)
 		return;
@@ -2140,7 +2190,7 @@ static int create_qp_common(struct mlx5_ib_dev *dev, struct ib_pd *pd,
 				return -EINVAL;
 			}
 			if (init_attr->create_flags &
-			    mlx5_ib_create_qp_sqpn_qp1()) {
+			    MLX5_IB_QP_CREATE_SQPN_QP1) {
 				mlx5_ib_dbg(dev, "user-space is not allowed to create UD QPs spoofing as QP1\n");
 				return -EINVAL;
 			}
@@ -3249,10 +3299,12 @@ static int modify_raw_packet_qp_sq(
 	}
 
 	/* Only remove the old rate after new rate was set */
-	if ((old_rl.rate &&
-	     !mlx5_rl_are_equal(&old_rl, &new_rl)) ||
-	    (new_state != MLX5_SQC_STATE_RDY))
+	if ((old_rl.rate && !mlx5_rl_are_equal(&old_rl, &new_rl)) ||
+	    (new_state != MLX5_SQC_STATE_RDY)) {
 		mlx5_rl_remove_rate(dev, &old_rl);
+		if (new_state != MLX5_SQC_STATE_RDY)
+			memset(&new_rl, 0, sizeof(new_rl));
+	}
 
 	ibqp->rl = new_rl;
 	sq->state = new_state;
@@ -3386,19 +3438,13 @@ static int __mlx5_ib_qp_set_counter(struct ib_qp *qp,
 	struct mlx5_ib_dev *dev = to_mdev(qp->device);
 	struct mlx5_ib_qp *mqp = to_mqp(qp);
 	struct mlx5_qp_context context = {};
-	struct mlx5_ib_port *mibport = NULL;
 	struct mlx5_ib_qp_base *base;
 	u32 set_id;
 
-	if (!MLX5_CAP_GEN(dev->mdev, rts2rts_qp_counters_set_id))
-		return 0;
-
-	if (counter) {
+	if (counter)
 		set_id = counter->id;
-	} else {
-		mibport = &dev->port[mqp->port - 1];
-		set_id = mibport->cnts.set_id;
-	}
+	else
+		set_id = mlx5_ib_get_counters_id(dev, mqp->port - 1);
 
 	base = &mqp->trans_qp.base;
 	context.qp_counter_set_usr_page &= cpu_to_be32(0xffffff);
@@ -3459,7 +3505,6 @@ static int __mlx5_ib_modify_qp(struct ib_qp *ibqp,
 	struct mlx5_ib_cq *send_cq, *recv_cq;
 	struct mlx5_qp_context *context;
 	struct mlx5_ib_pd *pd;
-	struct mlx5_ib_port *mibport = NULL;
 	enum mlx5_qp_state mlx5_cur, mlx5_new;
 	enum mlx5_qp_optpar optpar;
 	u32 set_id = 0;
@@ -3624,11 +3669,10 @@ static int __mlx5_ib_modify_qp(struct ib_qp *ibqp,
 		if (qp->flags & MLX5_IB_QP_UNDERLAY)
 			port_num = 0;
 
-		mibport = &dev->port[port_num];
 		if (ibqp->counter)
 			set_id = ibqp->counter->id;
 		else
-			set_id = mibport->cnts.set_id;
+			set_id = mlx5_ib_get_counters_id(dev, port_num);
 		context->qp_counter_set_usr_page |=
 			cpu_to_be32(set_id << 24);
 	}
@@ -3817,6 +3861,8 @@ static int mlx5_ib_modify_dct(struct ib_qp *ibqp, struct ib_qp_attr *attr,
 
 	dctc = MLX5_ADDR_OF(create_dct_in, qp->dct.in, dct_context_entry);
 	if (cur_state == IB_QPS_RESET && new_state == IB_QPS_INIT) {
+		u16 set_id;
+
 		required |= IB_QP_ACCESS_FLAGS | IB_QP_PKEY_INDEX | IB_QP_PORT;
 		if (!is_valid_mask(attr_mask, required, 0))
 			return -EINVAL;
@@ -3843,7 +3889,9 @@ static int mlx5_ib_modify_dct(struct ib_qp *ibqp, struct ib_qp_attr *attr,
 		}
 		MLX5_SET(dctc, dctc, pkey_index, attr->pkey_index);
 		MLX5_SET(dctc, dctc, port, attr->port_num);
-		MLX5_SET(dctc, dctc, counter_set_id, dev->port[attr->port_num - 1].cnts.set_id);
+
+		set_id = mlx5_ib_get_counters_id(dev, attr->port_num - 1);
+		MLX5_SET(dctc, dctc, counter_set_id, set_id);
 
 	} else if (cur_state == IB_QPS_INIT && new_state == IB_QPS_RTR) {
 		struct mlx5_ib_modify_qp_resp resp = {};
@@ -4162,7 +4210,7 @@ static u64 get_xlt_octo(u64 bytes)
 	       MLX5_IB_UMR_OCTOWORD;
 }
 
-static __be64 frwr_mkey_mask(void)
+static __be64 frwr_mkey_mask(bool atomic)
 {
 	u64 result;
 
@@ -4175,10 +4223,12 @@ static __be64 frwr_mkey_mask(void)
 		MLX5_MKEY_MASK_LW		|
 		MLX5_MKEY_MASK_RR		|
 		MLX5_MKEY_MASK_RW		|
-		MLX5_MKEY_MASK_A		|
 		MLX5_MKEY_MASK_SMALL_FENCE	|
 		MLX5_MKEY_MASK_FREE;
 
+	if (atomic)
+		result |= MLX5_MKEY_MASK_A;
+
 	return cpu_to_be64(result);
 }
 
@@ -4204,7 +4254,7 @@ static __be64 sig_mkey_mask(void)
 }
 
 static void set_reg_umr_seg(struct mlx5_wqe_umr_ctrl_seg *umr,
-			    struct mlx5_ib_mr *mr, u8 flags)
+			    struct mlx5_ib_mr *mr, u8 flags, bool atomic)
 {
 	int size = (mr->ndescs + mr->meta_ndescs) * mr->desc_size;
 
@@ -4212,7 +4262,7 @@ static void set_reg_umr_seg(struct mlx5_wqe_umr_ctrl_seg *umr,
 
 	umr->flags = flags;
 	umr->xlt_octowords = cpu_to_be16(get_xlt_octo(size));
-	umr->mkey_mask = frwr_mkey_mask();
+	umr->mkey_mask = frwr_mkey_mask(atomic);
 }
 
 static void set_linv_umr_seg(struct mlx5_wqe_umr_ctrl_seg *umr)
@@ -4811,10 +4861,22 @@ static int set_reg_wr(struct mlx5_ib_qp *qp,
 {
 	struct mlx5_ib_mr *mr = to_mmr(wr->mr);
 	struct mlx5_ib_pd *pd = to_mpd(qp->ibqp.pd);
+	struct mlx5_ib_dev *dev = to_mdev(pd->ibpd.device);
 	int mr_list_size = (mr->ndescs + mr->meta_ndescs) * mr->desc_size;
 	bool umr_inline = mr_list_size <= MLX5_IB_SQ_UMR_INLINE_THRESHOLD;
+	bool atomic = wr->access & IB_ACCESS_REMOTE_ATOMIC;
 	u8 flags = 0;
 
+	if (!mlx5_ib_can_use_umr(dev, atomic, wr->access)) {
+		mlx5_ib_warn(to_mdev(qp->ibqp.device),
+			     "Fast update of %s for MR is disabled\n",
+			     (MLX5_CAP_GEN(dev->mdev,
+					   umr_modify_entity_size_disabled)) ?
+				     "entity size" :
+				     "atomic access");
+		return -EINVAL;
+	}
+
 	if (unlikely(wr->wr.send_flags & IB_SEND_INLINE)) {
 		mlx5_ib_warn(to_mdev(qp->ibqp.device),
 			     "Invalid IB_SEND_INLINE send flag\n");
@@ -4826,7 +4888,7 @@ static int set_reg_wr(struct mlx5_ib_qp *qp,
 	if (umr_inline)
 		flags |= MLX5_UMR_INLINE;
 
-	set_reg_umr_seg(*seg, mr, flags);
+	set_reg_umr_seg(*seg, mr, flags, atomic);
 	*seg += sizeof(struct mlx5_wqe_umr_ctrl_seg);
 	*size += sizeof(struct mlx5_wqe_umr_ctrl_seg) / 16;
 	handle_post_send_edge(&qp->sq, seg, *size, cur_edge);
@@ -5315,7 +5377,6 @@ out:
 		 * we hit doorbell */
 		wmb();
 
-		/* currently we support only regular doorbells */
 		mlx5_write64((__be32 *)ctrl, bf->bfreg->map + bf->offset);
 		/* Make sure doorbells don't leak out of SQ spinlock
 		 * and reach the HCA out of order.
@@ -5810,7 +5871,7 @@ int mlx5_ib_query_qp(struct ib_qp *ibqp, struct ib_qp_attr *qp_attr,
 	if (qp->flags & MLX5_IB_QP_MANAGED_RECV)
 		qp_init_attr->create_flags |= IB_QP_CREATE_MANAGED_RECV;
 	if (qp->flags & MLX5_IB_QP_SQPN_QP1)
-		qp_init_attr->create_flags |= mlx5_ib_create_qp_sqpn_qp1();
+		qp_init_attr->create_flags |= MLX5_IB_QP_CREATE_SQPN_QP1;
 
 	qp_init_attr->sq_sig_type = qp->sq_signal_bits & MLX5_WQE_CTRL_CQ_UPDATE ?
 		IB_SIGNAL_ALL_WR : IB_SIGNAL_REQ_WR;
@@ -5942,12 +6003,21 @@ static int  create_rq(struct mlx5_ib_rwq *rwq, struct ib_pd *pd,
 	}
 	MLX5_SET(wq, wq, log_wq_stride, rwq->log_rq_stride);
 	if (rwq->create_flags & MLX5_IB_WQ_FLAGS_STRIDING_RQ) {
+		/*
+		 * In Firmware number of strides in each WQE is:
+		 *   "512 * 2^single_wqe_log_num_of_strides"
+		 * Values 3 to 8 are accepted as 10 to 15, 9 to 18 are
+		 * accepted as 0 to 9
+		 */
+		static const u8 fw_map[] = { 10, 11, 12, 13, 14, 15, 0, 1,
+					     2,  3,  4,  5,  6,  7,  8, 9 };
 		MLX5_SET(wq, wq, two_byte_shift_en, rwq->two_byte_shift_en);
 		MLX5_SET(wq, wq, log_wqe_stride_size,
 			 rwq->single_stride_log_num_of_bytes -
 			 MLX5_MIN_SINGLE_STRIDE_LOG_NUM_BYTES);
-		MLX5_SET(wq, wq, log_wqe_num_of_strides, rwq->log_num_strides -
-			 MLX5_MIN_SINGLE_WQE_LOG_NUM_STRIDES);
+		MLX5_SET(wq, wq, log_wqe_num_of_strides,
+			 fw_map[rwq->log_num_strides -
+				MLX5_EXT_MIN_SINGLE_WQE_LOG_NUM_STRIDES]);
 	}
 	MLX5_SET(wq, wq, log_wq_sz, rwq->log_rq_size);
 	MLX5_SET(wq, wq, pd, to_mpd(pd)->pdn);
@@ -6022,6 +6092,19 @@ static int set_user_rq_size(struct mlx5_ib_dev *dev,
 	return 0;
 }
 
+static bool log_of_strides_valid(struct mlx5_ib_dev *dev, u32 log_num_strides)
+{
+	if ((log_num_strides > MLX5_MAX_SINGLE_WQE_LOG_NUM_STRIDES) ||
+	    (log_num_strides < MLX5_EXT_MIN_SINGLE_WQE_LOG_NUM_STRIDES))
+		return false;
+
+	if (!MLX5_CAP_GEN(dev->mdev, ext_stride_num_range) &&
+	    (log_num_strides < MLX5_MIN_SINGLE_WQE_LOG_NUM_STRIDES))
+		return false;
+
+	return true;
+}
+
 static int prepare_user_rq(struct ib_pd *pd,
 			   struct ib_wq_init_attr *init_attr,
 			   struct ib_udata *udata,
@@ -6069,14 +6152,16 @@ static int prepare_user_rq(struct ib_pd *pd,
 				    MLX5_MAX_SINGLE_STRIDE_LOG_NUM_BYTES);
 			return -EINVAL;
 		}
-		if ((ucmd.single_wqe_log_num_of_strides >
-		    MLX5_MAX_SINGLE_WQE_LOG_NUM_STRIDES) ||
-		     (ucmd.single_wqe_log_num_of_strides <
-			MLX5_MIN_SINGLE_WQE_LOG_NUM_STRIDES)) {
-			mlx5_ib_dbg(dev, "Invalid log num strides (%u. Range is %u - %u)\n",
-				    ucmd.single_wqe_log_num_of_strides,
-				    MLX5_MIN_SINGLE_WQE_LOG_NUM_STRIDES,
-				    MLX5_MAX_SINGLE_WQE_LOG_NUM_STRIDES);
+		if (!log_of_strides_valid(dev,
+					  ucmd.single_wqe_log_num_of_strides)) {
+			mlx5_ib_dbg(
+				dev,
+				"Invalid log num strides (%u. Range is %u - %u)\n",
+				ucmd.single_wqe_log_num_of_strides,
+				MLX5_CAP_GEN(dev->mdev, ext_stride_num_range) ?
+					MLX5_EXT_MIN_SINGLE_WQE_LOG_NUM_STRIDES :
+					MLX5_MIN_SINGLE_WQE_LOG_NUM_STRIDES,
+				MLX5_MAX_SINGLE_WQE_LOG_NUM_STRIDES);
 			return -EINVAL;
 		}
 		rwq->single_stride_log_num_of_bytes =
@@ -6331,11 +6416,13 @@ int mlx5_ib_modify_wq(struct ib_wq *wq, struct ib_wq_attr *wq_attr,
 	}
 
 	if (curr_wq_state == IB_WQS_RESET && wq_state == IB_WQS_RDY) {
+		u16 set_id;
+
+		set_id = mlx5_ib_get_counters_id(dev, 0);
 		if (MLX5_CAP_GEN(dev->mdev, modify_rq_counter_set_id)) {
 			MLX5_SET64(modify_rq_in, in, modify_bitmask,
 				   MLX5_MODIFY_RQ_IN_MODIFY_BITMASK_RQ_COUNTER_SET_ID);
-			MLX5_SET(rqc, rqc, counter_set_id,
-				 dev->port->cnts.set_id);
+			MLX5_SET(rqc, rqc, counter_set_id, set_id);
 		} else
 			dev_info_once(
 				&dev->ib_dev.dev,
@@ -6486,6 +6573,7 @@ void mlx5_ib_drain_rq(struct ib_qp *qp)
  */
 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;
 
@@ -6495,6 +6583,11 @@ int mlx5_ib_qp_set_counter(struct ib_qp *qp, struct rdma_counter *counter)
 		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)