Re: [PATCH net-next 5/5] selftests/net: add offload checking test for virtual interface

[Hangbin Liu <liuhangbin@gmail.com>]

On Sun, Aug 31, 2025 at 06:52:01PM +0300, Ido Schimmel wrote:
> On Fri, Aug 29, 2025 at 09:54:30AM +0000, Hangbin Liu wrote:
> > make sure the virtual interface offload setting is correct after
> > changing lower devices.
> > 
> > Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
> > ---
> >  tools/testing/selftests/net/config          |   2 +
> >  tools/testing/selftests/net/vdev_offload.sh | 174 ++++++++++++++++++++
> >  2 files changed, 176 insertions(+)
> >  create mode 100755 tools/testing/selftests/net/vdev_offload.sh
> 
> Need to add to the Makefile

Sigh, I already added it to Makefile in my local branch, but forgot to
re-format-patch. Thanks for catching it.

> > +	# Common features
> > +	# NETIF_F_SG: tx-scatter-gather
> > +	ip netns exec "$ns" ethtool -K "$dev" tx-scatter-gather "$state" &> /dev/null
> 
> Why the redirection here? I don't see it in other places

When I tested local, I got message like

Actual changes:
tx-scatter-gather: on
tx-generic-segmentation: on [not requested]
tx-tcp-segmentation: on [not requested]
tx-tcp-ecn-segmentation: on [not requested]
tx-tcp-mangleid-segmentation: on [not requested]
tx-tcp6-segmentation: on [not requested]

So I redirected the log.

> > +do_test()
> > +{
> > +	local dev=$1
> 
> IMO, it makes more sense to put "RET=0" in the same function that calls
> log_test() (like you have it in check_xfrm()), so I would put it here...
> 
> > +	set_offload veth0 "on"
> > +	set_offload veth1 "on"
> > +	check_offload "$dev" "true"
> > +	log_test "$dev" "enable offload"
> > +
> 
> ... and here (instead of in check_offload())

OK, I will

Thanks
Hangbin