Jason Wang 5bc3be4a0b vhost_net: flush batched heads before trying to busy polling ...

[ Upstream commit f5a4941aa6 ]

After commit e2b3b35eb9 ("vhost_net: batch used ring update in rx"),
we tend to batch updating used heads. But it doesn't flush batched
heads before trying to do busy polling, this will cause vhost to wait
for guest TX which waits for the used RX. Fixing by flush batched
heads before busy loop.

1 byte TCP_RR performance recovers from 13107.83 to 50402.65.

Fixes: e2b3b35eb9 ("vhost_net: batch used ring update in rx")
Signed-off-by: Jason Wang <jasowang@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

2018-06-11 22:48:19 +02:00

..

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

License cleanup: add SPDX GPL-2.0 license identifier to files with no license

2017-11-02 11:10:55 +01:00

net.c

vhost_net: flush batched heads before trying to busy polling

2018-06-11 22:48:19 +02:00

scsi.c

vhost: remove unused lock check flag in vhost_dev_cleanup()

2018-02-01 16:26:44 +02:00

test.c

vhost: remove unused lock check flag in vhost_dev_cleanup()

2018-02-01 16:26:44 +02:00

test.h

License cleanup: add SPDX GPL-2.0 license identifier to files with no license

2017-11-02 11:10:55 +01:00

vhost.c

vhost: synchronize IOTLB message with dev cleanup

2018-06-11 22:48:17 +02:00

vhost.h

Merge tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost

2018-02-08 10:41:00 -08:00

vringh.c

vringh: kill off ACCESS_ONCE()

2016-12-16 00:13:36 +02:00

vsock.c

vhost: remove unused lock check flag in vhost_dev_cleanup()

2018-02-01 16:26:44 +02:00