Re: NULL pointer dereference panic in stable (2.6.33.2), amd64

[Eric Dumazet <eric.dumazet@gmail.com>]

Le lundi 12 avril 2010 à 02:04 +0300, Denys Fedorysychenko a écrit :
> > It is IMHO not safe, because route for this socket might have just
> > changed and we are transmitting an old packet (queued some milli seconds
> > before, when route was different).
> > 
> > We then memorize a queue_index that might be too big for the new device
> > of new selected route.
> > 
> > Next packet we want to transmit will take the cached value of
> > queue_index, correct for old device, maybe not correct for new device.
> Yes, it is possible, i have there RIP with 1k+ routes, changing non-stop.
> 
> > 
> > You could try to revert commit a4ee3ce3293dc931fab19beb472a8bde1295aebe
> > 
> > commit a4ee3ce3293dc931fab19beb472a8bde1295aebe
> I will try to revert it now. But to trigger bug probably i need 1-2 days.

I can try to reproduce bug here, with a multiqueue device, a non
multiqueue device, and changing routes while transmitting bulk of
packets, and of course some trafic shaping to slow down xmits.

Could you send :

ifconfig -a
lspci

Thanks