Jason Wang c67df11f6e vhost_net: try batch dequing from skb array ... We used to dequeue one skb during recvmsg() from skb_array, this could be inefficient because of the bad cache utilization and spinlock touching for each packet. This patch tries to batch them by calling batch dequeuing helpers explicitly on the exported skb array and pass the skb back through msg_control for underlayer socket to finish the userspace copying. Batch dequeuing is also the requirement for more batching improvement on receive path. Tests were done by pktgen on tap with XDP1 in guest. Host is Intel(R) Xeon(R) CPU E5-2650 0 @ 2.00GHz. rx batch | pps 0 2.25Mpps 1 2.33Mpps (+3.56%) 4 2.33Mpps (+3.56%) 16 2.35Mpps (+4.44%) 64 2.42Mpps (+7.56%) <- Default rx batching 128 2.40Mpps (+6.67%) 256 2.38Mpps (+5.78%) Signed-off-by: Jason Wang <jasowang@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>		2017-05-18 10:07:42 -04:00
..
Kconfig	tap: tap as an independent module	2017-02-11 20:59:41 -05:00
Kconfig.vringh	vhost: split out vringh Kconfig	2016-08-02 16:54:28 +03:00
Makefile
net.c	vhost_net: try batch dequing from skb array	2017-05-18 10:07:42 -04:00
scsi.c	vhost/scsi: silence uninitialized variable warning	2017-01-19 23:46:30 +02:00
test.c	Merge 4.8-rc5 into char-misc-next	2016-09-05 08:04:07 +02:00
test.h
vhost.c	mm: support __GFP_REPEAT in kvmalloc_node for >32kB	2017-05-08 17:15:12 -07:00
vhost.h	vhost: introduce O(1) vq metadata cache	2017-03-02 01:35:06 +02:00
vringh.c	vringh: kill off ACCESS_ONCE()	2016-12-16 00:13:36 +02:00
vsock.c	mm: support __GFP_REPEAT in kvmalloc_node for >32kB	2017-05-08 17:15:12 -07:00