Re: 2.1.125+: lockups, scheduler
Tom Eastep (eastep@loc1.tandem.com)
Mon, 23 Nov 1998 15:04:40 -0800
Jim Bray wrote:
>
> This may be an init bug of some sort, but it seems like other people
> are running into something like this.
>
> System (ide,pci,up,ppp) semi-locks occasionally on leaving X, and
> at other times. System is spinning in scheduler code. Running processes
> are xinit, bash, and two copies of init. Current alternates between
> them. No disk activity, plenty of free mem. After trying to kill them,
> signals for procs listed above are '1 00...4101'. Finally, with Sysrq
> killall procs, system hangs.
>
> Not reproducible. Occurs infrequently. Currently running 2.1.129. I
> am guessing that I started seeing this by about .126, maybe before.
>
> Behavior seems consistent with some sort of deadlock/race-condition
> between the procs in question.
This behavior sounds similar to what other people have experienced with recent
kernels. The symptoms are only observed on kernels compiled for UP (SMP=1
commented out).
The problem is actually in the fork() code in arch/i386/kernel/entry.S and Linus
recently posted a patch that corrects it. You can work around the problem by
uncommenting 'SMP = 1' in your Makefile.
Cheers,
-Tom
--
Tom Eastep
COMPAQ Computer Corporation
Enterprise Computing Group
Tandem Division
tom.eastep@compaq.com
-
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/