Re: 2.6.15-mm1 - locks solid when starting KDE (EDAC errors)

From: Dave Jones
Date: Fri Jan 06 2006 - 22:53:56 EST


On Sat, Jan 07, 2006 at 03:51:48AM +0100, Jesper Juhl wrote:

> Ok, this is with a pristine 2.6.15-mm1 + Dave's oops-pausing-patch
> Captured by switching to tty1 just after logging in via kdm.
> A *lot* of info still scrolls down when the problem hits before Daves
> patch stops it at a BUG dump, it scrolls by too fast for me to see
> what it is, but I guess it must be warning/error messages other than
> Oops's or BUG()'s ???
>
> Anyway, here's the entire contents of my screen after Daves patch
> stops the output - again written down by hand and then typed in from
> my handwritten notes, so there may be typos, but I've tried to be
> accurate.
>
>
> 050: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> 060: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> 070: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> 080: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> 090: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> 0a0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> 0b0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

My first thought was 'slab poison', but you don't have slab debugging turned on.
Randy's followup to my patch: http://lkml.org/lkml/2006/1/5/15
had another trick that may be useful. It'll slow down printk's
to a point where you might be able to see what happened.

another trick may be to just add a for (;;) into the BUG macro.

> ------------{ cut here ]------------
> kernel BUG at include/linux/list.h:166!
> invalid opcode: 0000 [#1]
> PREEMPT SMP
> Last sysfs file: /class/vc/vcsa2/dev
> Modules linked in: snd_seq_oss snd_seq_midi_event snd_seq snd_pcm_oss
> snd_mixer_oss uhci_hcd usbcore snd_emu10k1 snd_rawmidi snd_ac97_codec
> snd_ac97_bus snd_pcm snd_seq_device snd_timer snd_page_alloc
> snd_util_mem snd_hwdep snd agpgart
> CPU: 0
> EIP: 0060:[<c01475f5>] Tainted: G B VLI

Hmm, you had hit bad_page(), so that may be a more useful place
to add a pause.

Dave

-
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/