From: Jesse Barnes <jbarnes@sgi.com>
To: linux-kernel@vger.kernel.org
Subject: Re: 2.6.5-rc1-mm2
Date: Wed, 31 Mar 2004 15:58:28 -0800
User-Agent: KMail/1.6.1
Cc: Andrew Morton <akpm@osdl.org>
References: <20040317201454.5b2e8a3c.akpm@osdl.org> <200403311515.01289.jbarnes@sgi.com> <20040331155626.3f196a18.akpm@osdl.org>
In-Reply-To: <20040331155626.3f196a18.akpm@osdl.org>
MIME-Version: 1.0
Content-Disposition: inline
Content-Type: text/plain;
  charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
Message-Id: <200403311558.28178.jbarnes@sgi.com>
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1285
Lines: 36

On Wednesday 31 March 2004 3:56 pm, Andrew Morton wrote:
> Jesse Barnes <jbarnes@sgi.com> wrote:
> > On Wednesday 31 March 2004 12:06 pm, Andrew Morton wrote:
> > > So are we to assume that this is the offending process?  That the
> > > periodic slab reaping code has screwed up?
> >
> > It looks like it.  Disabling the slab cache reaping function allows it to
> > boot again.
>
> I suspect that the reap timer is innocent and what we have is simply
> scribbled-on slab metadata.  Which means it could be anything at all.

That's what I thought too, I'm trying to track down exactly which slab is 
having problems now.

>
> One last thing: could you please stick a
>
> 	printk(KERN_EMERG "destroying slab %s\n", cachep->name);

I'm already booting up something similar...

> at the start of slab_destroy()?  That'll help narrow it down.
>
> Could you also punt me over the .config?  If I can make it happen, the
> binary search will find it.  But it probably won't happen here.

I'm using sn2_defconfig in arch/ia64/configs.

Thanks,
Jesse
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/