Date: Wed, 22 Apr 2009 18:30:32 +0200
From: Jens Axboe <jens.axboe@oracle.com>
To: Steve Wise <swise@opengridcomputing.com>
Cc: balbir@linux.vnet.ibm.com, Andrew Morton <akpm@linux-foundation.org>,
       "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
       Wolfram Strepp <wstrepp@gmx.de>
Subject: Re: [BUG] rbtree bug with mmotm 2009-04-14-17-24
Message-ID: <20090422163032.GV4593@kernel.dk>
References: <20090421184223.GP19637@balbir.in.ibm.com> <49EE42CC.7070002@opengridcomputing.com> <20090422131703.GO4593@kernel.dk> <49EF2882.1020306@opengridcomputing.com> <49EF293F.4030504@opengridcomputing.com> <49EF2C32.4030409@opengridcomputing.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <49EF2C32.4030409@opengridcomputing.com>
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1101
Lines: 34

On Wed, Apr 22 2009, Steve Wise wrote:
>
>>>
>>> No there are a few patches applied that are heading upstream, but one 
>>> is in the NFS RDMA server which isn't loaded yet and the rest are in  
>>> iw_cxgb3 (iwarp driver) which also hasn't loaded at the time we  
>>> crash. NOTE:  Out of 4 power cycles, one booted up ok, 3 hit the 
>>> crash.
>>>
>>>
>>>
>>
>> By the way, this one looks different from the last one I saw.  So its  
>> not consistently crashing in the same spot.  The one I hit yesterday  
>> (which I don't have the OOPs dump for was in rb_erase().
>>
>> Steve.
>
> I'll start bisecting to isolate this.

Don't bother, I see what the bug is now. It's caused by commit
a36e71f996e25d6213f57951f7ae1874086ec57e and it's due to ->ioprio being
changed without the prio rb root being adjusted.

I'll provide a fix shortly.

-- 
Jens Axboe

--
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/