[ovs-dev] [PATCH v2 09/10] tc: Move tunnel_key unset action before output ports

Paul Blakey paulb at mellanox.com
Sun Dec 1 08:01:05 UTC 2019


On 11/27/2019 5:33 PM, Marcelo Ricardo Leitner wrote:
> On Wed, Nov 27, 2019 at 02:55:15PM +0200, Roi Dayan wrote: >> From: Paul Blakey <paulb at mellanox.com><mailto:paulb at mellanox.com> >> >> Since OvS datapath gets packets already decapsulated from tunnel >> devices, it doesn't explicitly decapsulate them. So in a >> recirculation setup, the tunnel matching continues in the >> recirculation as the tunnel metadata still exists on the SKB. >> >> Tunnel key unset action unsets this metadata. Some drivers might >> rely on this explicit tunnel key unset to know when to decapsulate >> the packet instead of the device type. So instead of removing it >> completly, we move it near the output actions. > > Hm, you said 'move', but the patch is only 'adding' stuff. > > Did you intend to: > > diff --git a/lib/tc.c b/lib/tc.c index 1541629a58e8..00d7f2f2239d > 100644 --- a/lib/tc.c +++ b/lib/tc.c @@ -2375,13 +2375,6 @@ > nl_msg_put_flower_acts(struct ofpbuf *request, struct tc_flower > *flower) { int error; > > - if (flower->tunnel) { - act_offset = > nl_msg_start_nested(request, act_index++); - > nl_msg_put_act_tunnel_key_release(request); - > nl_msg_put_act_flags(request); - > nl_msg_end_nested(request, act_offset); - } - action = > flower->actions; for (i = 0; i < flower->action_count; i++, action++) > { switch (action->type) { > > ? > >

Yes I did, thanks.

not sure how it got added back in :|




More information about the dev mailing list