> + br0 > + + > + vrf-h1 | vrf-h2 > + + +---+----+ + > + | | | | > + 192.0.2.1/24 + + + + 192.0.2.2/24 > + swp1 swp2 swp3 swp4 > + + + + + > + | | | | > + +--------+ +--------+ > +
> +setup_prepare > +setup_wait > + > +ping_test $h1 192.0.2.2 > +ping6_test $h1 2001:db8:1::2 How about putting 192.0.2.3/24 on br0, and pinging to/from it. Otherwise you don't test the slow path. I would also suggest a much simpler first test. > + > + > + vrf-h1 vrf-h2 > + + + > + | | > + 192.0.2.1/24 + + 192.0.2.2/24 192.0.3.1/24 + + > 192.0.3.2/24 > + swp1 swp2 swp3 swp4 > + + + + + > + | | | | > + +--------+ +--------+ Ping from 192.0.2.1 to 192.0.2.2 and test there are 0 packets seen on swp3 and swp4. I've seen DSA drivers get this wrong before, forgetting to ensure ports are properly separated by default. Andrew
