Dell Precision 530 + MPICH

Robert G. Brown rgb at phy.duke.edu
Tue Feb 19 11:45:18 PST 2002


On Tue, 19 Feb 2002, Mark Hahn wrote:

> > > > > Any ideas? I already tried the apm=power-off option at the boot
> > >
> > > should be "smp=power-off".
> > 
> > Do you mean apm=smp-power-off??? What the options that should be also compiled in the kernel to support it?
> 
> dang, I was misremembering.  after looking at the sources,
> on a modern kernel, it should be either "apm=power-off", "apm=power_off",
> or just use ACPI (which *does* work in SMP afaik).  for compiling,
> you just need the CONFIG_APM option - the other detailed 
> features are probably what don't work in SMP.

I think that there was a deadlock issue with APM and SMP -- maybe the
system had a hard time coming back out of a powerdown with the lock held
by the "right" processor?

In any event, I'm pretty sure that there are still APM issues even in
the 2.4 kernel.  We were getting fairly consistent crashes on dual AMD
tigers with the RH smp kernel (2.4.9, e.g.) and it turned out that
somehow they were getting installed with APM enabled.  We turned it off
and no more crashes.  I think.  Fingers crossed.

   rgb

-- 
Robert G. Brown	                       http://www.phy.duke.edu/~rgb/
Duke University Dept. of Physics, Box 90305
Durham, N.C. 27708-0305
Phone: 1-919-660-2567  Fax: 919-660-2525     email:rgb at phy.duke.edu






More information about the Beowulf mailing list