Re: bnx2/5709: Strange interrupts spread

["Michael Chan" <mchan@broadcom.com>]

On Mon, 2010-07-19 at 12:47 -0700, Eric Dumazet wrote:
> Le lundi 19 juillet 2010 à 11:47 -0700, Michael Chan a écrit :
> > On Mon, 2010-07-19 at 08:55 -0700, Christophe Ngo Van Duc wrote:
> > > So i've been able to do some test today:
> > > If I put the 2 interface in a bridge with no IP adress, the interrupts
> > > are on 1 CPU
> > > If I put the 2 interface in a bridge with IP adress, the interrupts
> > > are still on 1 CPU
> > > If I put the 2 interface outside the bridge with IP address,
> > > everything works fine the interrupts get spread on the CPU
> > > 
> > > So the conclusion seems to be that when the bnx2 is put into
> > > promiscuous mode by the bridge, the RSS hash stop to work even if
> > > traffic is IP in nature.
> > 
> > I did a quick test with bridging and saw no problem with RSS.  I did see
> > this though:
> > 
> > br0 received packet on queue 4, but number of RX queues is 1
> > 
> > Looks like it is a warning message from RPS.
> > 
> 
> Christophe uses an old kernel, not RPS enabled ;)
> 
> 

Right, I'm reporting a related problem on a newer kernel with RPS
enabled.  And the fact that it is receiving packets on queue 4 shows
that RSS is working together with bridging.

I'll try to use an older kernel to see what happens later today.