From 80fe326ab8f5ac6c41786fa534a220c6e3f5beaf Mon Sep 17 00:00:00 2001 From: Saeed Mahameed Date: Sat, 25 Mar 2017 00:52:03 +0300 Subject: [PATCH] net/mlx5e: Use dma_rmb rather than rmb in CQE fetch routine Use dma_rmb in mlx5e_get_cqe rather than aggressive rmb (at least on some architectures), this should help improve the performance on such CPU archs where dma_rmb is optimized. Performance improvement: System: Intel(R) Xeon(R) CPU E5-2620 v3 @ 2.40GHz Test case Baseline Now improvement --------------------------------------------------------------- TX packets (24 threads) 45Mpps 50Mpps 11% TC stack Drop (1 core) 3.45Mpps 3.6Mpps 5% XDP Drop (1 core) 14Mpps 16.9Mpps 20% XDP TX (1 core) 10.4Mpps 12Mpps 15% Signed-off-by: Saeed Mahameed Reviewed-by: Tariq Toukan Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c index e5c12a7..d8cda2f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c @@ -44,7 +44,7 @@ struct mlx5_cqe64 *mlx5e_get_cqe(struct mlx5e_cq *cq) return NULL; /* ensure cqe content is read after cqe ownership bit */ - rmb(); + dma_rmb(); return cqe; } -- 2.7.4