[dpdk-dev] [PATCH] net/mlx5: fix queue wrap-around of multi-packet send

[Yongseok Koh]

When the WQ is wrapped around, it wrongly checks the condition when
resetting the pointer. It should be compared against the end of the queue,
not the beginning of the queue. And this isn't even needed when the length
of the copying data crosses the boundary.

Fixes: e5291c280c08 ("net/mlx5: use work queue buffer as a raw buffer")

CC: stable at dpdk.org
Signed-off-by: Yongseok Koh <yskoh at mellanox.com>
---
 drivers/net/mlx5/mlx5_rxtx.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/drivers/net/mlx5/mlx5_rxtx.c b/drivers/net/mlx5/mlx5_rxtx.c
index 2ae949295..f54db5b85 100644
--- a/drivers/net/mlx5/mlx5_rxtx.c
+++ b/drivers/net/mlx5/mlx5_rxtx.c
@@ -1037,11 +1037,13 @@ mlx5_tx_burst_mpw_inline(void *dpdk_txq, struct rte_mbuf **pkts,
 				rte_memcpy((void *)(uintptr_t)mpw.data.raw,
 					   (void *)addr,
 					   length);
-				mpw.data.raw += length;
+
+				if (length == max)
+					mpw.data.raw =
+						(volatile void *)txq->wqes;
+				else
+					mpw.data.raw += length;
 			}
-			if ((uintptr_t)mpw.data.raw ==
-			    (uintptr_t)tx_mlx5_wqe(txq, 1 << txq->wqe_n))
-				mpw.data.raw = (volatile void *)txq->wqes;
 			++mpw.pkts_n;
 			++j;
 			if (mpw.pkts_n == MLX5_MPW_DSEG_MAX) {
-- 
2.11.0