On 12/4/23 14:18, Marcelo Leitner wrote: > On Mon, Dec 04, 2023 at 01:57:50PM +0100, Ilya Maximets wrote: >> On 12/4/23 13:40, Marcelo Leitner wrote: >>> On Sat, Dec 02, 2023 at 12:06:59AM +0100, Ilya Maximets wrote: >>> ... >>>> Adding a test case that demonstrates a scenario where the issue >>>> occurs - bridging of two tunnels. >>> >>> I get the fix and it LGTM. What I don't get is the test case. >>> Considering that the attr was getting simply ignored, wouldn't the >>> test case always succeed? That is, I don't see how ignoring tp_src >>> would cause the test to fail. Can you please elaborate? >> >> Ignoring the tp_src is causing tc code to warn about kernel >> acknowledgement mismatch. Testsuite doesn't tolerate warnings, >> so the test fails on checking the logs: > > Nice. > > ... >> We do not check if the offloading is actually happening, but we >> check if there is any miscommunication between ovs-vswitchd and >> the kernel (TC). >> >> Does that make sense? > > Yes, thanks. > > Reviewed-by: Marcelo Ricardo Leitner <mleit...@redhat.com> >
Thanks, everyone! Applied and backported down to 2.17. Best regards, Ilya Maximets. _______________________________________________ dev mailing list d...@openvswitch.org https://mail.openvswitch.org/mailman/listinfo/ovs-dev
