Re: BUG: sleeping function called from invalid context at kernel/rwsem.c:131 XFS? (was: Re: linux-next: Tree for October 17)

["Alexander Beregalov" <a.beregalov@xxxxxxxxx>]

2008/10/18 Arjan van de Ven <arjan@xxxxxxxxxxxxx>:
> On Fri, 17 Oct 2008 16:37:10 -0400
> Christoph Hellwig <hch@xxxxxxxxxxxxx> wrote:
>
>> I just ran the xfs testsuite over linux-next on qemu (i386), and I
>> can't find anything at all.  Really strange.  Let's see if this still
>> there with mondays linux-next, and if yes can you just try the xfs
>> patch from the splitout linux-next patches and see if that alone
>> causes it?

Hi Christoph

I have the same result with next-1020 and today's xfs-2.6.git/master
(
commit bfd2bd10da76378dc4afd87d7d204a1d3d70b347
Author: David Chinner <david@xxxxxxxxxxxxx>
Date:   Fri Oct 17 15:36:23 2008 +1000
    Inode: Allow external list initialisation
)

Should I start bisecting?
>>
>> In fact that might be useful for todays linux-next, too.
>> --
>
> also if you enable lockdep, it tracks where irq's got turned off (with
> a stacktrace I think) so it would provide some good clues.

Yes, lockdep was enabled, but there is no more information than I
already posted.