Re: Oops in 2.6.19.1

[Adrian Bunk]

On Sun, Dec 31, 2006 at 04:48:43PM +0000, Alistair John Strachan wrote:
> On Sunday 31 December 2006 16:28, Adrian Bunk wrote:
> > On Sat, Dec 30, 2006 at 06:29:15PM +0000, Alistair John Strachan wrote:
> > > On Saturday 30 December 2006 17:21, Chuck Ebbert wrote:
> > > > In-Reply-To: <200612301659.35982.s0348365@xxxxxxxxxxxx>
> > > >
> > > > On Sat, 30 Dec 2006 16:59:35 +0000, Alistair John Strachan wrote:
> > > > > I've eliminated 2.6.19.1 as the culprit, and also tried toggling
> > > > > "optimize for size", various debug options. 2.6.19 compiled with GCC
> > > > > 4.1.1 on an Via Nehemiah C3-2 seems to crash in pipe_poll reliably,
> > > > > within approximately 12 hours.
> > > >
> > > > Which CPU are you compiling for?  You should try different options.
> > >
> > > I should, I haven't thought of that. Currently it's compiling for
> > > CONFIG_MVIAC3_2, but I could try i686 for example.
> > >
> > > > Can you post disassembly of pipe_poll() for both the one that crashes
> > > > and the one that doesn't?  Use 'objdump -D -r fs/pipe.o' so we get the
> > > > relocation info and post just the one function from each for now.
> > >
> > > Sure, no problem:
> > >
> > > http://devzero.co.uk/~alistair/2.6.19-via-c3-pipe_poll/
> > >
> > > Both use identical configs, neither are optimised for size. The config is
> > > available from the same location.
> >
> > Can you try enabling as many debug options as possible?
> 
> Specifically what? I've already had:
> 
> CONFIG_DETECT_SOFTLOCKUP
> CONFIG_FRAME_POINTER
> CONFIG_UNWIND_INFO
> 
> Enabled. CONFIG_4KSTACKS is disabled. Are there any debugging features 
> actually pertinent to this bug?

No, that's only an "enable as much as possible and hope one helps" shot 
in the dark.

> Cheers,
> Alistair.

cu
Adrian

-- 

       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed

-
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/