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

[Vincenzo Maffione]

No, the thing is that I misinterpreted your stack trace. The patch is ok
for a different bug. It seems that the problem are vlans more than lagg.
Which interface did you put in netmap mode, em or em.345?

Il 25 mag 2017 10:46 PM, "Harry Schmalzbauer" <freebsd at omnilan.de> ha
scritto:

 Bezüglich Harry Schmalzbauer's Nachricht vom 25.05.2017 18:09 (localtime):
> Bezüglich Vincenzo Maffione's Nachricht vom 25.05.2017 17:56 (localtime):
>> I see the bug is in FreeBSD 11. I attached the simple patch to fix it.
>> Can someone commit the patch to 11/stable?
>>
>> Harry: You should be able to workaround the bug by setting
>>
>> # sysctl dev.netmap.generic_rings=1
> I'll recompile with your patch, thanks a lot!

Hi, unfortunately I can't confirm it to be fixed.

The kgdb output is exactly the same:

(kgdb) list *0xffffffff80426714
0xffffffff80426714 is in generic_rx_handler
(/usr/local/share/deploy-tools/RELENG_11/src/sys/dev/netmap/
netmap_generic.c:628).
623             struct netmap_adapter *na = NA(ifp);
624             struct netmap_generic_adapter *gna = (struct
netmap_generic_adapter *)na;
625             u_int work_done;
626             u_int rr = MBUF_RXQ(m); // receive ring number
627
628             if (rr >= na->num_rx_rings) {
629                     rr = rr % na->num_rx_rings; // XXX expensive...
630             }
631
632             /* limit the size of the queue */
Current language:  auto; currently minimal

Will double-check tomorrow that the binary really includes your patch.
Ofcourse I checked briefly and revision and buildtime check affirms it.

best,

-harry