Message ID | 20200515203619.1044803-3-aconole@redhat.com |
---|---|
State | Accepted |
Delegated to: | Ilya Maximets |
Headers | show |
Series | TC fixes | expand |
On 5/15/20 10:36 PM, Aaron Conole wrote: > It's possible that block_id could changes after the probe for block > support. Therefore, fetch the block_id again after the probe. > > Fixes: edc2055a2bf7 ("netdev-offload-tc: Flush rules on ingress block when init tc flow api") > Cc: Dmytro Linkin <dmitrolin@mellanox.com> > Co-authored-by: Marcelo Leitner <mleitner@redhat.com> > Acked-by: Roi Dayan <roid@mellanox.com> > Signed-off-by: Marcelo Leitner <mleitner@redhat.com> > Signed-off-by: Aaron Conole <aconole@redhat.com> > --- Thanks! Applied to master and branch-2.13. Best regards, Ilya Maximets.
diff --git a/lib/netdev-offload-tc.c b/lib/netdev-offload-tc.c index 875ebef719..e188e63e56 100644 --- a/lib/netdev-offload-tc.c +++ b/lib/netdev-offload-tc.c @@ -1931,6 +1931,8 @@ netdev_tc_init_flow_api(struct netdev *netdev) if (ovsthread_once_start(&block_once)) { probe_tc_block_support(ifindex); + /* Need to re-fetch block id as it depends on feature availability. */ + block_id = get_block_id_from_netdev(netdev); ovsthread_once_done(&block_once); }