Re: [x86,SMP,patch] smp-2.3.18, please test.

Ingo Molnar (mingo@chiara.csoma.elte.hu)
Tue, 21 Sep 1999 09:41:24 +0200 (CEST)


On Mon, 20 Sep 1999, Pete Wyckoff wrote:

> > > I was really looking forward to getting some info from the NMI watchdog,
> > > but it stays silent. Any guesses where to look next on my hard-lockup
> > > problem?
> >
> > could you kill syslogd and klogd and run 'dmesg -n 8' - this makes it sure
> > all kernel messages go to the console immediately without buffering.
>
> Ensured no loggers were running, killed off everything else useless,
> and hit dmesg -n 8. Same silent result. Bad hardware?

yes, the kind of workload you are generating hits the hardware really
hard.

_such_ types of lockups (if they are software) can only be debugged with
IKD's print-EIP component. See the IKD patch how to use that, it has a
learning curve, but since you can reproduce this easily it could debug the
lockup.

-- mingo

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