diff options
author | David S. Miller <davem@davemloft.net> | 2014-07-23 22:12:32 (GMT) |
---|---|---|
committer | David S. Miller <davem@davemloft.net> | 2014-07-23 22:12:32 (GMT) |
commit | 92bab4e71522767bd9033fff60744b111b015fef (patch) | |
tree | 0b55a52980362bd053ef02615bb0b7630e86e02f /net/ipv4/raw.c | |
parent | 96b3bff4c94cebc97726dc7f019cfc7ba3ced2e1 (diff) | |
parent | 91815639d8804d1eee7ce2e1f7f60b36771db2c9 (diff) | |
download | linux-92bab4e71522767bd9033fff60744b111b015fef.tar.xz |
Merge branch 'virtio-net-busy-poll'
Jason Wang says:
====================
rx busy polling support for virtio-net
This series introduces the support for rx busy polling support. This
was useful for reducing the latency for a kvm guest. Instead of
introducing new states and spinlocks, this series re-uses NAPI state
to synchonrize between NAPI and busy polling. This grealy simplified
the codes and reduce the overheads of spinlocks for normal NAPI fast
path.
Test was done between a kvm guest and an external host. Two hosts were
connected through 40gb mlx4 cards. With both busy_poll and busy_read
are set to 50 in guest, 1 byte netperf tcp_rr shows 127% improvement:
transaction rate was increased from 8353.33 to 18966.87.
Changes from V2:
- Avoid introducing new states and spinlocks by reusuing the NAPI
state
- Fix the budget calculation in virtnet_poll()
- Drop patch 1/3 from V2 since it was useless
Changes from V1:
- split the patch info smaller ones
- add more details about test setup/configuration
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'net/ipv4/raw.c')
0 files changed, 0 insertions, 0 deletions