On Fri, Feb 03, 2023 at 08:36:12PM +0100, Ilya Maximets wrote: > On 2/3/23 17:48, Simon Horman wrote: > > On Thu, Feb 02, 2023 at 12:09:47PM +0100, Eelco Chaudron wrote: > >> Include and run the system-traffic.at tests as part of the system offload > >> testsuite. Exclude all the tests that will not run without any special > >> modifications. > >> > >> Signed-off-by: Eelco Chaudron <echau...@redhat.com> > >> Acked-by: Roi Dayan <r...@nvidia.com> > > > > ... > > > >> diff --git a/tests/ofproto-macros.at b/tests/ofproto-macros.at > >> index 676d55aa9..690331f35 100644 > >> --- a/tests/ofproto-macros.at > >> +++ b/tests/ofproto-macros.at > >> @@ -195,6 +195,7 @@ m4_define([_OVS_VSWITCHD_START], > >> /netdev_offload|INFO|netdev: Flow API Enabled/d > >> /probe tc:/d > >> /setting extended ack support failed/d > >> +/recirc_id sharing not supported/d > >> /tc: Using policy/d']]) > >> ]) > > > > Hi Eelco, > > > > I gather the above is to address a problem I reported with an older kernel, > > which indeed doesn't support recirc_id sharing not supported. > > > > I am, to be honest, not entirely sure of the wisdom in me, or anyone else, > > running this code in such an environment. But perhaps this particular > > problem is harmless enough to fix. > > > > However, I don't think the above fixes the problem. > > This is because the false-positive on the logs is triggered not when > > starting OVS, but rather when stopping it. > > > > The following, somewhat global, change did address the problem for me. > > A more local version might bean override of OVS_VSWITCHD_STOP for > > system-offloads-traffic. > > Another option might be to reduce the log level from ERR to something > more reasonable. There is really no reason to report unsupported > feature as an error. If we'll demote it down to INFO level or even DBG, > check_logs will stop complaining. > > What do you think?
Sounds good to me. _______________________________________________ dev mailing list d...@openvswitch.org https://mail.openvswitch.org/mailman/listinfo/ovs-dev