Re: 2.4.23 SMP: kernel BUG at mmap.c

[Chris Stromsoe]

On Mon, 12 Jan 2004, Marcelo Tosatti wrote:
> On Mon, 12 Jan 2004, Chris Stromsoe wrote:

> > Any hints about a cause or where to start looking for a problem?
>
> Hi Chris,
>
> Both BUGs seem to be related to some kind of race (the first one is a
> check for mm->map_count !=0 on exit_mmap and the second is a BH_Lock
> check at submit_bh()).
>
> First try to not use the md0 (use md1 or md2 directly then reconstruct
> the mirror later).
>
> If that fails, recompile your kernel without SMP.
>
> It might me a JFS problem.

I tried the suggestions and was able to force the error to occur.  I took
the box out of service and ran memtest86 for a week and got a bunch of
errors.  The problem appears to have been caused by faulty hardware.

I have the exact same configuration running in a box with duplicate
hardware and am not having any problems.


-Chris
-
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/