[panic] netmap(4) and if_lagg(4)

[Vincenzo Maffione]

Is lagg0 the only interface attached to vale0?
Is lagg0 aggregating a VLAN interface?

You can try this trivial patch

diff --git a/sys/dev/netmap/netmap_generic.c
b/sys/dev/netmap/netmap_generic.c
index f148b228..46a3c2c6 100644
--- a/sys/dev/netmap/netmap_generic.c
+++ b/sys/dev/netmap/netmap_generic.c
@@ -950,6 +950,10 @@ generic_rx_handler(struct ifnet *ifp, struct mbuf *m)
        u_int work_done;
        u_int r = MBUF_RXQ(m); /* receive ring number */

+       if (!NM_NA_VALID(ifp)) {
+               return 0;
+       }
+
        if (r >= na->num_rx_rings) {
                r = r % na->num_rx_rings;
        }



2017-05-26 9:21 GMT+02:00 Harry Schmalzbauer <freebsd at omnilan.de>:

> Bezüglich Vincenzo Maffione's Nachricht vom 26.05.2017 09:14 (localtime):
> > Hi,
> >   Your stack trace report this:
> >
> > #7 0xffffffff8069dc50 at vlan_input+0x1f0
> >
> > which means VLANs are involved, in some way. Is that the correct trace?
>
> The trace is from the pnaic after doing 'vale-ctl -a vale0:lagg0' (while
> lagg0 can have various names, but I'm not using a vlan clone).
>
> Might be that existing, but to my understanding uninvolved vlan clones
> interfere here...
> The lagg0 does have vlan clones (lots of) defined.
> Unfortunately I can't take them out of the game for testing...
>
> Does that picture match the trace?
>
> thanks,
>
> -harry
>
>
>


-- 
Vincenzo Maffione