* soft deadlock in blk_congestion_wait (2.6.7)
@ 2004-07-08 9:31 Miklos Szeredi
2004-07-08 13:23 ` Nuno Ferreira
0 siblings, 1 reply; 3+ messages in thread
From: Miklos Szeredi @ 2004-07-08 9:31 UTC (permalink / raw)
To: linux-kernel
While running LTP tests (more specifically the ftest* testcases) on
FUSE (http://sourceforge.net/projects/avf), I run into the following
problem: the process serving the FUSE filesystem requests does some
debugging write to an ext3 filesystem, and deadlocks in
blk_congestion_wait:
fusexmp D C043E760 0 2130 2125 (NOTLB)
c0c99c50 00200082 c09bb3f0 c043e760 00000664 c791f6b0 c043ec08 00000000
c0c99c64 c0c99c50 00200246 00000801 f70d3b39 00000664 c09bb598 0066663b
c0c99c64 c03ce968 c0c99c8c c033195e c0c99c64 0066663b c0115051 c0443f00
Call Trace:
[<c033195e>] schedule_timeout+0x5e/0xb0
[<c03318f1>] io_schedule_timeout+0x11/0x20
[<c02275be>] blk_congestion_wait+0x6e/0x90
[<c01369ce>] balance_dirty_pages+0xde/0x150
[<c01332a4>] generic_file_aio_write_nolock+0x4f4/0xc00
[<c0133ac1>] generic_file_aio_write+0x81/0xa0
[<c018609f>] ext3_file_write+0x3f/0xd0
[<c014eab7>] do_sync_write+0x87/0xc0
[<c014eb9a>] vfs_write+0xaa/0x130
[<c014ecbf>] sys_write+0x3f/0x60
[<c010423b>] syscall_call+0x7/0xb
Some other process performing a read will kick it out of this coma,
but if left alone it will wait there forever.
The FUSE request being served is a write, called from
aops->commit_write(). So my guess is that the kernel does not handle
this sort of recursive write too well.
Can someone explain what blk_congestion_wait() is waiting for, and why
does that event not happen?
I've seen a very similar report for CIFS here:
http://marc.theaimsgroup.com/?l=linux-kernel&m=108807439504273&w=2
So it's not even FUSE specific, though it could be a bug in the
filesystems themselves in both cases, or possibly a core kernel bug.
Thanks,
Miklos
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: soft deadlock in blk_congestion_wait (2.6.7)
2004-07-08 9:31 soft deadlock in blk_congestion_wait (2.6.7) Miklos Szeredi
@ 2004-07-08 13:23 ` Nuno Ferreira
2004-07-08 13:35 ` Miklos Szeredi
0 siblings, 1 reply; 3+ messages in thread
From: Nuno Ferreira @ 2004-07-08 13:23 UTC (permalink / raw)
To: Miklos Szeredi; +Cc: Linux Kernel
On Qui, 2004-07-08 at 11:31 +0200, Miklos Szeredi wrote:
> While running LTP tests (more specifically the ftest* testcases) on
> FUSE (http://sourceforge.net/projects/avf), I run into the following
> problem: the process serving the FUSE filesystem requests does some
> debugging write to an ext3 filesystem, and deadlocks in
> blk_congestion_wait:
[...]
> I've seen a very similar report for CIFS here:
>
> http://marc.theaimsgroup.com/?l=linux-kernel&m=108807439504273&w=2
>
I reported that bug, and at least for CIFS it was fixed using the patch
mentioned a few mails later on that thread, see
http://marc.theaimsgroup.com/?l=linux-kernel&m=108846032716689&w=2
I can no longer reproduce the problem, but it may be that the patch only
made it harder to trigger.
--
Nuno Ferreira
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: soft deadlock in blk_congestion_wait (2.6.7)
2004-07-08 13:23 ` Nuno Ferreira
@ 2004-07-08 13:35 ` Miklos Szeredi
0 siblings, 0 replies; 3+ messages in thread
From: Miklos Szeredi @ 2004-07-08 13:35 UTC (permalink / raw)
To: nuno.ferreira; +Cc: linux-kernel
> I reported that bug, and at least for CIFS it was fixed using the patch
> mentioned a few mails later on that thread, see
> http://marc.theaimsgroup.com/?l=linux-kernel&m=108846032716689&w=2
>
> I can no longer reproduce the problem, but it may be that the patch only
> made it harder to trigger.
Thanks, I should have looked the end of the thread. It could be that
this solution also applies to FUSE, though I don't yet understand
either the bug or the fix :).
Miklos
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2004-07-08 13:37 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2004-07-08 9:31 soft deadlock in blk_congestion_wait (2.6.7) Miklos Szeredi
2004-07-08 13:23 ` Nuno Ferreira
2004-07-08 13:35 ` Miklos Szeredi
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®