Re: [PATCH net-next 1/2] selftests/net: GRO coalesce test

[Coco Li <lixiaoyan@google.com>]

On Thu, Aug 5, 2021 at 4:16 AM Paolo Abeni <pabeni@redhat.com> wrote:
>
> Hello,
>
> On Thu, 2021-08-05 at 07:36 +0000, Coco Li wrote:
> > Implement a GRO testsuite that expects Linux kernel GRO behavior.
> > All tests pass with the kernel software GRO stack. Run against a device
> > with hardware GRO to verify that it matches the software stack.
> >
> > gro.c generates packets and sends them out through a packet socket. The
> > receiver in gro.c (run separately) receives the packets on a packet
> > socket, filters them by destination ports using BPF and checks the
> > packet geometry to see whether GRO was applied.
> >
> > gro.sh provides a wrapper to run the gro.c in NIC loopback mode.
> > It is not included in continuous testing because it modifies network
> > configuration around a physical NIC: gro.sh sets the NIC in loopback
> > mode, creates macvlan devices on the physical device in separate
> > namespaces, and sends traffic generated by gro.c between the two
> > namespaces to observe coalescing behavior.
>
> I like this idea a lot!
>
> Have you considered additionally run the same test of top of a veth
> pair, and have such tests always enabled, so we could have some
> coverage regardless of specific H/W available?
>
> To do the above you should disable TSO on the veth sender peer and
> enable GRO on the other end.

Thanks for the suggestion! To make sure I understand you correctly,
would this be another script that creates the veth pair separate from
the gro.sh wrapper?