Hi Ilya Thanks for those additional comments.
On Mon, Jun 29, 2026 at 7:21 PM Ilya Maximets <[email protected]> wrote: > On 6/29/26 2:40 PM, Xavier Simonart via dev wrote: > > Using ofport_request can make the test flaky, as OVS might have > > already used the requested ofport (typically when ovn created > > ovn-hv[0-9]-0 ports). In that case, it takes some time (between > > a few msec to multiple seconds) until the requested ofport is > > obtained. OVN handles this properly (flows are recalculated), but > > tests do not handle this properly (e.g. if it happens when comparing > > I-P and recompute flows). > > > > This patch removes all ofport-request, querying the obtained ofport > > when needed. > > > > Signed-off-by: Xavier Simonart <[email protected]> > > --- > > tests/ovn.at | 1055 ++++++++++++++++++-------------------------------- > > 1 file changed, 373 insertions(+), 682 deletions(-) > > > > diff --git a/tests/ovn.at b/tests/ovn.at > > index 5dda50044..6ae12acc7 100644 > > --- a/tests/ovn.at > > +++ b/tests/ovn.at > > @@ -2641,7 +2641,7 @@ for i in 1 2 3; do > > ovn_attach n1 br-phys 192.168.0.$i > > > > for j in 1 2 3; do > > - check ovs-vsctl add-port br-int vif$i$j -- set Interface > vif$i$j external-ids:iface-id=lp$i$j options:tx_pcap=hv$i/vif$i$j-tx.pcap > options:rxq_pcap=hv$i/vif$i$j-rx.pcap ofport-request=$i$j > > + check ovs-vsctl add-port br-int vif$i$j -- set Interface > vif$i$j external-ids:iface-id=lp$i$j options:tx_pcap=hv$i/vif$i$j-tx.pcap > options:rxq_pcap=hv$i/vif$i$j-rx.pcap > > check ovn-nbctl lsp-add lsw0 lp$i$j > > if test $j = 1; then > > check ovn-nbctl lsp-set-addresses lp$i$j > "f0:00:00:00:00:$i$j 192.168.0.$i$j" unknown > > @@ -2912,7 +2912,7 @@ for i in 1 2; do > > ovn_attach n1 br-phys 192.168.0.$i > > > > for j in 1 2; do > > - ovs-vsctl add-port br-int vif$i$j -- set Interface vif$i$j > external-ids:iface-id=lp$i$j options:tx_pcap=hv$i/vif$i$j-tx.pcap > options:rxq_pcap=hv$i/vif$i$j-rx.pcap ofport-request=$i$j > > + ovs-vsctl add-port br-int vif$i$j -- set Interface vif$i$j > external-ids:iface-id=lp$i$j options:tx_pcap=hv$i/vif$i$j-tx.pcap > options:rxq_pcap=hv$i/vif$i$j-rx.pcap > > check ovn-nbctl lsp-add lsw0 lp$i$j > > ip_addrs="192.168.0.$i$j" > > check ovn-nbctl lsp-set-addresses lp$i$j "f0:00:00:00:00:$i$j > $ip_addrs" > > @@ -3359,8 +3359,7 @@ for i in 1 2; do > > ovs-vsctl add-port br-int vif$i$j -- \ > > set Interface vif$i$j external-ids:iface-id=lp$i$j \ > > options:tx_pcap=hv$i/vif$i$j-tx.pcap \ > > - options:rxq_pcap=hv$i/vif$i$j-rx.pcap > \ > > - ofport-request=$i$j > > + options:rxq_pcap=hv$i/vif$i$j-rx.pcap > > > > I also wonder if all those 'ovs-vsctl add-port' commands should be > replaced with a macro, since we're touching all of them here anyway. > We could do that; a quick look shows around 1,500 fewer lines and more readable code in most cases. > So, the next time we don't have to. > I am not sure that, even with a macro, we would not have had to re-modify all lines (we would have, for example, OVS_ADD_PORT(br-int, vif$i$j, lp$i$j, ..., ofport-request=$i$j), and we still would have had to delete or modify all requested ofports. But it would have been easier. So I'll send a v2 with a new macro, and the changes mentioned in the previous mail (i.e. remove most ofport requests and use a prefix such as 1000 when requesting one) > > Best regards, Ilya Maximets. > Thanks Xavier _______________________________________________ dev mailing list [email protected] https://mail.openvswitch.org/mailman/listinfo/ovs-dev
