diff options
| author | Daniel Jurgens <danielj@mellanox.com> | 2018-11-06 01:05:37 +0300 |
|---|---|---|
| committer | Saeed Mahameed <saeedm@mellanox.com> | 2018-12-12 01:52:20 +0300 |
| commit | 939de57d30344ce728b0de61be87984e75af420e (patch) | |
| tree | dc42835752a817f639508f23b45b1601fd644861 /include/linux/mlx5/cq.h | |
| parent | 8c4dc42bf6e4ffeda49cf5e26bfc991b548fc0aa (diff) | |
| download | linux-939de57d30344ce728b0de61be87984e75af420e.tar.xz | |
net/mlx5e: Use CQE padding for Ethernet CQs
Writing 64B CQEs to 128B cache lines results in a RMW operation. Padding
the CQEs to 128B if possible improves performance on 128B cache line
systems like PPC.
Testing on PPC showed up to a 24% improvement in small packet throughput
vs the default behavior, depending on the workload and system topology.
Signed-off-by: Daniel Jurgens <danielj@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Diffstat (limited to 'include/linux/mlx5/cq.h')
| -rw-r--r-- | include/linux/mlx5/cq.h | 10 |
1 files changed, 5 insertions, 5 deletions
diff --git a/include/linux/mlx5/cq.h b/include/linux/mlx5/cq.h index 28b757a64029..612c8c2f2466 100644 --- a/include/linux/mlx5/cq.h +++ b/include/linux/mlx5/cq.h @@ -125,9 +125,9 @@ struct mlx5_cq_modify_params { }; enum { - CQE_SIZE_64 = 0, - CQE_SIZE_128 = 1, - CQE_SIZE_128_PAD = 2, + CQE_STRIDE_64 = 0, + CQE_STRIDE_128 = 1, + CQE_STRIDE_128_PAD = 2, }; #define MLX5_MAX_CQ_PERIOD (BIT(__mlx5_bit_sz(cqc, cq_period)) - 1) @@ -135,8 +135,8 @@ enum { static inline int cqe_sz_to_mlx_sz(u8 size, int padding_128_en) { - return padding_128_en ? CQE_SIZE_128_PAD : - size == 64 ? CQE_SIZE_64 : CQE_SIZE_128; + return padding_128_en ? CQE_STRIDE_128_PAD : + size == 64 ? CQE_STRIDE_64 : CQE_STRIDE_128; } static inline void mlx5_cq_set_ci(struct mlx5_core_cq *cq) |