Re: BUG: soft lockup - is this XFS problem?

[Guus Sliepen]

On Thu, Jul 14, 2011 at 08:03:09PM +0200, Peter Klotz wrote:

> On 07/14/2011 01:23 PM, Guus Sliepen wrote:
> 
> >I'm having a problem with a system having an XFS filesystem on RAID locking up
> >fairly consistently when writing large amounts of data to it, with several
> >kernels, including 2.6.38.2 and 2.6.39.3, on both AMD and Intel multi-core
> >processors. The kernel always logs this several times:
> >
> >BUG: soft lockup - CPU#2 stuck for 67s! [kswapd0:33]
[...]
> This Bugzilla entry documents the XFS bug from 2009 in detail
> including links:
> 
> http://oss.sgi.com/bugzilla/show_bug.cgi?id=805

Aha, I did not look at that before.

> The problem was finally solved by a patch proposed by Linus. This is
> the reason the original patch developed by Nick never made it into
> the kernel.
> 
> My tests back then showed that both patches fixed the problem.
> 
> It seems you have found a test case where just Nick's patch helps.

Yes. I agree with Linus that the root cause should be fixed, not the symptoms.
I don't have time to dive in the kernel code myself, but I do have several
nearly identical machines where I can test things on. I will be happy to test
out patches and/or different kernel versions or kernel configurations, and I
can provide dmesg output and perhaps other information if necessary.

-- 
Met vriendelijke groet / with kind regards,
Guus Sliepen <Guus.Sliepen@xxxxxxxxxxx>

Attachment: signature.asc
Description: Digital signature