Re: 2.6.4-rc1-mm[12] - dm_any_congested issues

[Miquel van Smoorenburg]

In article <200403051754.i25Hsjg7015052@xxxxxxxxxxxxxxxxxxxxxxx>,
 <Valdis.Kletnieks@xxxxxx> wrote:
>On Tue, 02 Mar 2004 20:15:36 PST, Andrew Morton <akpm@xxxxxxxx>  said:
>
>(Added in -rc1-mm1 which I didn't try, problem noticed in -rc2-mm2)
>
>> queue-congestion-dm-implementation.patch
>>   Implement queue congestion callout for device mapper
>
>This is causing the following trace every second or 2 on my laptop:
>
>Mar  4 17:47:26 turing-police kernel: Debug: sleeping function called
>from invalid context at include/linux/rwsem.h:43
>Of course backing it out makes the messages go away, since dm_any_congested()
>was added by that patch.  This patch just not ready for prime time, or
>am I (as usual)
>managing to trip over some silly corner case due to odd configuration?

It's a bug. The new queue congestion function cannot sleep, yet
it calls parts of DM that can sleep (indirectly). This needs to
be fixed in DM (I think).
See https://www.redhat.com/archives/linux-lvm/2004-March/msg00033.html

Mike.

-
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/