net/mlx5: remove excessive data prefetch
authorYongseok Koh <yskoh@mellanox.com>
Mon, 12 Mar 2018 17:05:45 +0000 (10:05 -0700)
committerFerruh Yigit <ferruh.yigit@intel.com>
Fri, 13 Apr 2018 22:40:21 +0000 (00:40 +0200)
commitf84411be9e307d2723052a8ae8ee80a1fd9f559b
tree8e4cad0be7ecb392c81cde7d4d1457f1f7b4bbb3
parent0915e287a6a7ed884a3de198f08926d8418366d4
net/mlx5: remove excessive data prefetch

In Enhanced Multi-Packet Send (eMPW), entire packet data is prefetched to
LLC if it isn't inlined. Even though this helps reducing jitter when HW
fetches data by DMA, this can thresh the LLC with evicting precious data.
And if the size of queue is large and there are many queues, this might not
be effective. Also, if application runs on a remote node from the PCIe
link, it may not be helpful and can even cause bad results.

Signed-off-by: Yongseok Koh <yskoh@mellanox.com>
Acked-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
drivers/net/mlx5/mlx5_rxtx.c