[dpdk-stable] [PATCH] net/mlx5: fix Rx buffer replenishment threshold

[Yongseok Koh]

The threshold of buffer replenishment for vectorized Rx burst is a constant
value (64). If the size of Rx queue is comparatively small, device could
run out of buffers. For example, if the size of Rx queue is 128, buffers
are replenished only twice per a wraparound. This can cause jitter in
receiving packets and the jitter can cause unnecessary retransmission for
TCP connections.

Fixes: 6cb559d67b83 ("net/mlx5: add vectorized Rx/Tx burst for x86")
Fixes: 570acdb1da8a ("net/mlx5: add vectorized Rx/Tx burst for ARM")
Cc: stable at dpdk.org

Signed-off-by: Yongseok Koh <yskoh at mellanox.com>
Acked-by: Shahaf Shuler <shahafs at mellanox.com>
---
 drivers/net/mlx5/mlx5_defs.h          | 5 +++--
 drivers/net/mlx5/mlx5_rxtx_vec.h      | 4 ++--
 drivers/net/mlx5/mlx5_rxtx_vec_neon.h | 2 +-
 drivers/net/mlx5/mlx5_rxtx_vec_sse.h  | 2 +-
 4 files changed, 7 insertions(+), 6 deletions(-)

diff --git a/drivers/net/mlx5/mlx5_defs.h b/drivers/net/mlx5/mlx5_defs.h
index 51124cdcc..b52f7dcbe 100644
--- a/drivers/net/mlx5/mlx5_defs.h
+++ b/drivers/net/mlx5/mlx5_defs.h
@@ -64,10 +64,11 @@
 #define MLX5_VPMD_MIN_TXQS 4
 
 /* Threshold of buffer replenishment for vectorized Rx. */
-#define MLX5_VPMD_RXQ_RPLNSH_THRESH   64U
+#define MLX5_VPMD_RXQ_RPLNSH_THRESH(n) \
+	(RTE_MIN(MLX5_VPMD_RX_MAX_BURST, (unsigned int)(n) >> 2))
 
 /* Maximum size of burst for vectorized Rx. */
-#define MLX5_VPMD_RX_MAX_BURST        MLX5_VPMD_RXQ_RPLNSH_THRESH
+#define MLX5_VPMD_RX_MAX_BURST 64U
 
 /*
  * Maximum size of burst for vectorized Tx. This is related to the maximum size
diff --git a/drivers/net/mlx5/mlx5_rxtx_vec.h b/drivers/net/mlx5/mlx5_rxtx_vec.h
index 598dc7519..fb884f92f 100644
--- a/drivers/net/mlx5/mlx5_rxtx_vec.h
+++ b/drivers/net/mlx5/mlx5_rxtx_vec.h
@@ -91,9 +91,9 @@ mlx5_rx_replenish_bulk_mbuf(struct mlx5_rxq_data *rxq, uint16_t n)
 		&((volatile struct mlx5_wqe_data_seg *)rxq->wqes)[elts_idx];
 	unsigned int i;
 
-	assert(n >= MLX5_VPMD_RXQ_RPLNSH_THRESH);
+	assert(n >= MLX5_VPMD_RXQ_RPLNSH_THRESH(q_n));
 	assert(n <= (uint16_t)(q_n - (rxq->rq_ci - rxq->rq_pi)));
-	assert(MLX5_VPMD_RXQ_RPLNSH_THRESH > MLX5_VPMD_DESCS_PER_LOOP);
+	assert(MLX5_VPMD_RXQ_RPLNSH_THRESH(q_n) > MLX5_VPMD_DESCS_PER_LOOP);
 	/* Not to cross queue end. */
 	n = RTE_MIN(n - MLX5_VPMD_DESCS_PER_LOOP, q_n - elts_idx);
 	if (rte_mempool_get_bulk(rxq->mp, (void *)elts, n) < 0) {
diff --git a/drivers/net/mlx5/mlx5_rxtx_vec_neon.h b/drivers/net/mlx5/mlx5_rxtx_vec_neon.h
index 71a5eaf23..3ebf34823 100644
--- a/drivers/net/mlx5/mlx5_rxtx_vec_neon.h
+++ b/drivers/net/mlx5/mlx5_rxtx_vec_neon.h
@@ -739,7 +739,7 @@ rxq_burst_v(struct mlx5_rxq_data *rxq, struct rte_mbuf **pkts, uint16_t pkts_n,
 	 *   N - (rq_ci - rq_pi) := # of buffers consumed (to be replenished).
 	 */
 	repl_n = q_n - (rxq->rq_ci - rxq->rq_pi);
-	if (repl_n >= MLX5_VPMD_RXQ_RPLNSH_THRESH)
+	if (repl_n >= MLX5_VPMD_RXQ_RPLNSH_THRESH(q_n))
 		mlx5_rx_replenish_bulk_mbuf(rxq, repl_n);
 	/* See if there're unreturned mbufs from compressed CQE. */
 	rcvd_pkt = rxq->cq_ci - rxq->rq_pi;
diff --git a/drivers/net/mlx5/mlx5_rxtx_vec_sse.h b/drivers/net/mlx5/mlx5_rxtx_vec_sse.h
index 3e985d61d..f0d888831 100644
--- a/drivers/net/mlx5/mlx5_rxtx_vec_sse.h
+++ b/drivers/net/mlx5/mlx5_rxtx_vec_sse.h
@@ -724,7 +724,7 @@ rxq_burst_v(struct mlx5_rxq_data *rxq, struct rte_mbuf **pkts, uint16_t pkts_n,
 	 *   N - (rq_ci - rq_pi) := # of buffers consumed (to be replenished).
 	 */
 	repl_n = q_n - (rxq->rq_ci - rxq->rq_pi);
-	if (repl_n >= MLX5_VPMD_RXQ_RPLNSH_THRESH)
+	if (repl_n >= MLX5_VPMD_RXQ_RPLNSH_THRESH(q_n))
 		mlx5_rx_replenish_bulk_mbuf(rxq, repl_n);
 	/* See if there're unreturned mbufs from compressed CQE. */
 	rcvd_pkt = rxq->cq_ci - rxq->rq_pi;
-- 
2.11.0