IB/mlx5: Packet packing enhancement for RAW QP

Enable RAW QP to be able to configure burst control by modify_qp. By
using burst control with rate limiting, user can achieve best
performance and accuracy. The burst control information is passed by
user through udata.

This patch also reports burst control capability for mlx5 related
hardwares, burst control is only marked as supported when both
packet_pacing_burst_bound and packet_pacing_typical_size are
supported.

Signed-off-by: Bodong Wang <bodong@mellanox.com>
Reviewed-by: Daniel Jurgens <danielj@mellanox.com>
Reviewed-by: Yishai Hadas <yishaih@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>

4 files changed, 98 insertions, 21 deletions

diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c
index 3408bede0ee5..d06aae9aa600 100644
--- a/drivers/infiniband/hw/mlx5/main.c
+++ b/drivers/infiniband/hw/mlx5/main.c
@@ -989,6 +989,10 @@ static int mlx5_ib_query_device(struct ib_device *ibdev,
 				MLX5_CAP_QOS(mdev, packet_pacing_min_rate);
 			resp.packet_pacing_caps.supported_qpts |=
 				1 << IB_QPT_RAW_PACKET;
+			if (MLX5_CAP_QOS(mdev, packet_pacing_burst_bound) &&
+			    MLX5_CAP_QOS(mdev, packet_pacing_typical_size))
+				resp.packet_pacing_caps.cap_flags |=
+					MLX5_IB_PP_SUPPORT_BURST;
 		}
 		resp.response_length += sizeof(resp.packet_pacing_caps);
 	}
diff --git a/drivers/infiniband/hw/mlx5/mlx5_ib.h b/drivers/infiniband/hw/mlx5/mlx5_ib.h
index f9ba1ea94f0f..aeea74357cbe 100644
--- a/drivers/infiniband/hw/mlx5/mlx5_ib.h
+++ b/drivers/infiniband/hw/mlx5/mlx5_ib.h
@@ -403,7 +403,7 @@ struct mlx5_ib_qp {
 	struct list_head	qps_list;
 	struct list_head	cq_recv_list;
 	struct list_head	cq_send_list;
-	u32			rate_limit;
+	struct mlx5_rate_limit	rl;
 	u32                     underlay_qpn;
 	bool			tunnel_offload_en;
 	/* storage for qp sub type when core qp type is IB_QPT_DRIVER */
diff --git a/drivers/infiniband/hw/mlx5/qp.c b/drivers/infiniband/hw/mlx5/qp.c
index 6c7b4c2bfaa4..2fb3d9a400d3 100644
--- a/drivers/infiniband/hw/mlx5/qp.c
+++ b/drivers/infiniband/hw/mlx5/qp.c
@@ -86,7 +86,9 @@ struct mlx5_modify_raw_qp_param {
 	u16 operation;
 
 	u32 set_mask; /* raw_qp_set_mask_map */
-	u32 rate_limit;
+
+	struct mlx5_rate_limit rl;
+
 	u8 rq_q_ctr_id;
 };
 
@@ -2774,8 +2776,9 @@ static int modify_raw_packet_qp_sq(struct mlx5_core_dev *dev,
 				   const struct mlx5_modify_raw_qp_param *raw_qp_param)
 {
 	struct mlx5_ib_qp *ibqp = sq->base.container_mibqp;
-	u32 old_rate = ibqp->rate_limit;
-	u32 new_rate = old_rate;
+	struct mlx5_rate_limit old_rl = ibqp->rl;
+	struct mlx5_rate_limit new_rl = old_rl;
+	bool new_rate_added = false;
 	u16 rl_index = 0;
 	void *in;
 	void *sqc;
@@ -2797,39 +2800,43 @@ static int modify_raw_packet_qp_sq(struct mlx5_core_dev *dev,
 			pr_warn("%s: Rate limit can only be changed when SQ is moving to RDY\n",
 				__func__);
 		else
-			new_rate = raw_qp_param->rate_limit;
+			new_rl = raw_qp_param->rl;
 	}
 
-	if (old_rate != new_rate) {
-		if (new_rate) {
-			err = mlx5_rl_add_rate(dev, new_rate, &rl_index);
+	if (!mlx5_rl_are_equal(&old_rl, &new_rl)) {
+		if (new_rl.rate) {
+			err = mlx5_rl_add_rate(dev, &rl_index, &new_rl);
 			if (err) {
-				pr_err("Failed configuring rate %u: %d\n",
-				       new_rate, err);
+				pr_err("Failed configuring rate limit(err %d): \
+				       rate %u, max_burst_sz %u, typical_pkt_sz %u\n",
+				       err, new_rl.rate, new_rl.max_burst_sz,
+				       new_rl.typical_pkt_sz);
+
 				goto out;
 			}
+			new_rate_added = true;
 		}
 
 		MLX5_SET64(modify_sq_in, in, modify_bitmask, 1);
+		/* index 0 means no limit */
 		MLX5_SET(sqc, sqc, packet_pacing_rate_limit_index, rl_index);
 	}
 
 	err = mlx5_core_modify_sq(dev, sq->base.mqp.qpn, in, inlen);
 	if (err) {
 		/* Remove new rate from table if failed */
-		if (new_rate &&
-		    old_rate != new_rate)
-			mlx5_rl_remove_rate(dev, new_rate);
+		if (new_rate_added)
+			mlx5_rl_remove_rate(dev, &new_rl);
 		goto out;
 	}
 
 	/* Only remove the old rate after new rate was set */
-	if ((old_rate &&
-	    (old_rate != new_rate)) ||
+	if ((old_rl.rate &&
+	     !mlx5_rl_are_equal(&old_rl, &new_rl)) ||
 	    (new_state != MLX5_SQC_STATE_RDY))
-		mlx5_rl_remove_rate(dev, old_rate);
+		mlx5_rl_remove_rate(dev, &old_rl);
 
-	ibqp->rate_limit = new_rate;
+	ibqp->rl = new_rl;
 	sq->state = new_state;
 
 out:
@@ -2906,7 +2913,8 @@ static int modify_raw_packet_qp(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp,
 
 static int __mlx5_ib_modify_qp(struct ib_qp *ibqp,
 			       const struct ib_qp_attr *attr, int attr_mask,
-			       enum ib_qp_state cur_state, enum ib_qp_state new_state)
+			       enum ib_qp_state cur_state, enum ib_qp_state new_state,
+			       const struct mlx5_ib_modify_qp *ucmd)
 {
 	static const u16 optab[MLX5_QP_NUM_STATE][MLX5_QP_NUM_STATE] = {
 		[MLX5_QP_STATE_RST] = {
@@ -3144,7 +3152,30 @@ static int __mlx5_ib_modify_qp(struct ib_qp *ibqp,
 		}
 
 		if (attr_mask & IB_QP_RATE_LIMIT) {
-			raw_qp_param.rate_limit = attr->rate_limit;
+			raw_qp_param.rl.rate = attr->rate_limit;
+
+			if (ucmd->burst_info.max_burst_sz) {
+				if (attr->rate_limit &&
+				    MLX5_CAP_QOS(dev->mdev, packet_pacing_burst_bound)) {
+					raw_qp_param.rl.max_burst_sz =
+						ucmd->burst_info.max_burst_sz;
+				} else {
+					err = -EINVAL;
+					goto out;
+				}
+			}
+
+			if (ucmd->burst_info.typical_pkt_sz) {
+				if (attr->rate_limit &&
+				    MLX5_CAP_QOS(dev->mdev, packet_pacing_typical_size)) {
+					raw_qp_param.rl.typical_pkt_sz =
+						ucmd->burst_info.typical_pkt_sz;
+				} else {
+					err = -EINVAL;
+					goto out;
+				}
+			}
+
 			raw_qp_param.set_mask |= MLX5_RAW_QP_RATE_LIMIT;
 		}
 
@@ -3332,8 +3363,10 @@ int mlx5_ib_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr,
 {
 	struct mlx5_ib_dev *dev = to_mdev(ibqp->device);
 	struct mlx5_ib_qp *qp = to_mqp(ibqp);
+	struct mlx5_ib_modify_qp ucmd = {};
 	enum ib_qp_type qp_type;
 	enum ib_qp_state cur_state, new_state;
+	size_t required_cmd_sz;
 	int err = -EINVAL;
 	int port;
 	enum rdma_link_layer ll = IB_LINK_LAYER_UNSPECIFIED;
@@ -3341,6 +3374,28 @@ int mlx5_ib_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr,
 	if (ibqp->rwq_ind_tbl)
 		return -ENOSYS;
 
+	if (udata && udata->inlen) {
+		required_cmd_sz = offsetof(typeof(ucmd), reserved) +
+			sizeof(ucmd.reserved);
+		if (udata->inlen < required_cmd_sz)
+			return -EINVAL;
+
+		if (udata->inlen > sizeof(ucmd) &&
+		    !ib_is_udata_cleared(udata, sizeof(ucmd),
+					 udata->inlen - sizeof(ucmd)))
+			return -EOPNOTSUPP;
+
+		if (ib_copy_from_udata(&ucmd, udata,
+				       min(udata->inlen, sizeof(ucmd))))
+			return -EFAULT;
+
+		if (ucmd.comp_mask ||
+		    memchr_inv(&ucmd.reserved, 0, sizeof(ucmd.reserved)) ||
+		    memchr_inv(&ucmd.burst_info.reserved, 0,
+			       sizeof(ucmd.burst_info.reserved)))
+			return -EOPNOTSUPP;
+	}
+
 	if (unlikely(ibqp->qp_type == IB_QPT_GSI))
 		return mlx5_ib_gsi_modify_qp(ibqp, attr, attr_mask);
 
@@ -3421,7 +3476,8 @@ int mlx5_ib_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr,
 		goto out;
 	}
 
-	err = __mlx5_ib_modify_qp(ibqp, attr, attr_mask, cur_state, new_state);
+	err = __mlx5_ib_modify_qp(ibqp, attr, attr_mask, cur_state,
+				  new_state, &ucmd);
 
 out:
 	mutex_unlock(&qp->mutex);
diff --git a/include/uapi/rdma/mlx5-abi.h b/include/uapi/rdma/mlx5-abi.h
index 1111aa4e7c1e..d2e0d234704f 100644
--- a/include/uapi/rdma/mlx5-abi.h
+++ b/include/uapi/rdma/mlx5-abi.h
@@ -163,6 +163,10 @@ struct mlx5_ib_cqe_comp_caps {
 	__u32 supported_format; /* enum mlx5_ib_cqe_comp_res_format */
 };
 
+enum mlx5_ib_packet_pacing_cap_flags {
+	MLX5_IB_PP_SUPPORT_BURST	= 1 << 0,
+};
+
 struct mlx5_packet_pacing_caps {
 	__u32 qp_rate_limit_min;
 	__u32 qp_rate_limit_max; /* In kpbs */
@@ -172,7 +176,8 @@ struct mlx5_packet_pacing_caps {
 	 * supported_qpts |= 1 << IB_QPT_RAW_PACKET
 	 */
 	__u32 supported_qpts;
-	__u32 reserved;
+	__u8  cap_flags; /* enum mlx5_ib_packet_pacing_cap_flags */
+	__u8  reserved[3];
 };
 
 enum mlx5_ib_mpw_caps {
@@ -362,6 +367,18 @@ struct mlx5_ib_create_ah_resp {
 	__u8	reserved[6];
 };
 
+struct mlx5_ib_burst_info {
+	__u32       max_burst_sz;
+	__u16       typical_pkt_sz;
+	__u16       reserved;
+};
+
+struct mlx5_ib_modify_qp {
+	__u32			   comp_mask;
+	struct mlx5_ib_burst_info  burst_info;
+	__u32			   reserved;
+};
+
 struct mlx5_ib_modify_qp_resp {
 	__u32	response_length;
 	__u32	dctn;