On 8/25/20 5:07 PM, Aaron Conole wrote:
> Ilya Maximets <i.maxim...@ovn.org> writes:
>
>> Packets in the benchmark must be treated as new packets, i.e. they
>> should not have conntrack metadata set. Current code will set up
>> 'pkt->md.conn' after the first run and all subsequent calls will hit
>> the 'fast' processing that is intended for recirculated packets making
>> a false impression that current conntrack implementation is lightning
>> fast.
>
> Oh no.
>
>> Before the change:
>> $ ./ovstest test-conntrack benchmark 4 33554432 32 1
>> conntrack: 1059 ms
>>
>> After (correct):
>> $ ./ovstest test-conntrack benchmark 4 33554432 32 1
>> conntrack: 92785 ms
>
> That's horrifying.
>
>> Fixes: 594570ea1cde ("conntrack: Optimize recirculations.")
>> Signed-off-by: Ilya Maximets <i.maxim...@ovn.org>
>> ---
>
> Sadly:
>
> Acked-by: Aaron Conole <acon...@redhat.com>
>
Thanks!
Applied to master and backported down to 2.13.
Best regards, Ilya Maximets.
_______________________________________________
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
