Re: divide error in select_task_rq_fair()

[Bjorn Helgaas]

On Thu, Nov 04, 2010 at 03:37:54PM +0100, Eric Dumazet wrote:
> Le jeudi 04 novembre 2010 à 08:28 -0600, Bjorn Helgaas a écrit :
> > On Thu, Nov 04, 2010 at 06:19:52AM +0100, Eric Dumazet wrote:
> > > Le mercredi 03 novembre 2010 à 22:12 -0600, Bjorn Helgaas a écrit :
> > > > Hi,
> > > > 
> > > > With current upstream, I see the following crash at boot-time:
> > > > 
> > > >     Brought up 64 CPUs
> > > >     Total of 64 processors activated (289366.52 BogoMIPS).
> > > >     divide error: 0000 [#1] SMP 
> > > >     last sysfs file: 
> > > >     CPU 1 
> > > >     Modules linked in:
> > > > 
> > > >     Pid: 2, comm: kthreadd Not tainted 2.6.37-rc1-00027-gff8b16d #271 /ProLiant DL980 G7
> > > >     RIP: 0010:[<ffffffff81034645>]  [<ffffffff81034645>] select_task_rq_fair+0x62a/0x7a0
> > > > 
> > > > Complete dmesg below; let me know if you need more info.
> > > 
> > > Is the machine runs OK if you build a kernel with NR_CPUS=128 ?
> > 
> > Nope, it fails the same way with NR_CPUS=128.  Dmesg below.
> 
> Sorry, just try 256 or 512, it seems you have a pretty big machine ?

Is that going to help you debug the problem?  The solution is not going
to be something like "set NR_CPUS=x".  If NR_CPUS is too small, the
machine should still *boot*, even if we can't use all the CPUs in the
box.

Bjorn
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/