diff options
author | David S. Miller <davem@davemloft.net> | 2016-02-21 22:05:50 -0500 |
---|---|---|
committer | David S. Miller <davem@davemloft.net> | 2016-02-21 22:05:50 -0500 |
commit | 8b393f833346e0a5f0cb4269271d47be8fecd372 (patch) | |
tree | 889175291bb173b738b2a3341c96d895c7c0c991 /Documentation/numastat.txt | |
parent | 417b7ca44427ab8ddaa65eece094edf5e1404586 (diff) | |
parent | 6ceb31ca5f65acff299dbc3da5854d54e147b7d8 (diff) | |
download | op-kernel-dev-8b393f833346e0a5f0cb4269271d47be8fecd372.zip op-kernel-dev-8b393f833346e0a5f0cb4269271d47be8fecd372.tar.gz |
Merge branch 'geneve-vxlan-outer-checksum'
Alexander Duyck says:
====================
GENEVE/VXLAN: Enable outer Tx checksum by default
This patch series makes it so that we enable the outer Tx checksum for IPv4
tunnels by default. This makes the behavior consistent with how we were
handling this for IPv6. In addition I have updated the internal flags for
these tunnels so that we use a ZERO_CSUM_TX flag for IPv4 which should
match up will with the ZERO_CSUM6_TX flag which was already in use for
IPv6.
For most network devices this should be a net gain in terms of performance
as having the outer header checksum present allows for devices to report
CHECKSUM_UNNECESSARY which we can then convert to CHECKSUM_COMPLETE in order
to determine if the inner header checksum is valid.
Below is some data I collected with ixgbe with an X540 that demonstrates
this. I located two PFs connected back to back in two different name
spaces and then setup a pair of tunnels on each, one with checksum enabled
and one without.
Recv Send Send Utilization
Socket Socket Message Elapsed Send
Size Size Size Time Throughput local
bytes bytes bytes secs. 10^6bits/s % S
noudpcsum:
87380 16384 16384 30.00 8898.67 12.80
udpcsum:
87380 16384 16384 30.00 9088.47 5.69
The one spot where this may cause a performance regression is if the
environment contains devices that can parse the inner headers and a device
supports NETIF_F_GSO_UDP_TUNNEL but not NETIF_F_GSO_UDP_TUNNEL_CSUM. In
the case of such a device we have to fall back to using GSO to segment the
tunnel instead of TSO and as a result we may take a performance hit as seen
below with i40e.
Recv Send Send Utilization
Socket Socket Message Elapsed Send
Size Size Size Time Throughput local
bytes bytes bytes secs. 10^6bits/s % S
noudpcsum:
87380 16384 16384 30.00 9085.21 3.32
udpcsum:
87380 16384 16384 30.00 9089.23 5.54
In addition it will be necessary to update iproute2 so that we don't
provide the checksum attribute unless specified. This way on older kernels
which don't have local checksum offload we will default to disabling the
outer checksum, and on newer kernels that have LCO we can default to
enabling it.
I also haven't investigated the effect this will have on OVS. However I
suspect the impact should be minimal as the worst case scenario should be
that Tx checksumming will become enabled by default which should be
consistent with the existing behavior for IPv6.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'Documentation/numastat.txt')
0 files changed, 0 insertions, 0 deletions