Re: 2.4.22-pre lockups (now decoded oops for pre10)

[Stephan von Krawczynski <skraw@ithnet.com>]

On Wed, 6 Aug 2003 11:09:20 +0200
Willy Tarreau <willy@w.ods.org> wrote:

> On Wed, Aug 06, 2003 at 09:41:50AM +0200, Stephan von Krawczynski wrote:
>  
> > Code;  c0144b14 <__remove_from_queues+14/30>
> > 00000000 <_EIP>:
> > Code;  c0144b14 <__remove_from_queues+14/30>   <=====
> >    0:   89 02                     mov    %eax,(%edx)   <=====
> > Code;  c0144b16 <__remove_from_queues+16/30>
> >    2:   c7 41 30 00 00 00 00      movl   $0x0,0x30(%ecx)
> > Code;  c0144b1d <__remove_from_queues+1d/30>
> >    9:   89 4c 24 04               mov    %ecx,0x4(%esp,1)
> > Code;  c0144b21 <__remove_from_queues+21/30>
> >    d:   e9 7a ff ff ff            jmp    ffffff8c <_EIP+0xffffff8c>
> > Code;  c0144b26 <__remove_from_queues+26/30>
> >   12:   8d 76 00                  lea    0x0(%esi),%esi
> 
> once again, it's *pprev=next which is is causing trouble, with pprev=6 this
> time (fs/buffer.c:523). There really seems to be something playing badly with
> this...
> 
> I find amazing that such widely used portions of code only trigger panics on
> your system ! either it's a rare combinations of several components/drivers,
> or a strange hardware problem, although I can't imagine which (cpu? bus
> locking?).

Hm, the hardware may not be that widespread. I guess not many people are really
using SMP, 64 bit PCI network, 3 GB RAM, 3ware RAID5 and serverworks board
altogether in one box. I can't fight the impression it has something to do with
locking issues. It doesn't look exactly like a hardware problem, you would not
expect crashes on the same type of code then.
The question is: what additional information is needed to find the underlying
problem?

Regards,
Stephan