On Tue, 22 Apr 2008 15:40:34 +0200, Ingo Molnar <mingo@elte.hu> wrote:

quoted text
> > I got the following warning while running sched-devel/latest.
> > WARNING: at mm/slub.c:2443 kmem_cache_destroy+0x168/0x1a9()

quoted text
> >  [<ffffffff80293c5d>] kmem_cache_destroy+0x168/0x1a9
> >  [<ffffffff803974d8>] mon_text_release+0x89/0xae
> >  [<ffffffff80299254>] __fput+0xb9/0x161
> >  [<ffffffff8029957b>] fput+0x14/0x16
> >  [<ffffffff80296892>] filp_close+0x66/0x71
> >  [<ffffffff80239eea>] put_files_struct+0x77/0xcb
> >  [<ffffffff80239f72>] __exit_files+0x34/0x39

quoted text
> Cc:-ing USB folks as the kmem_cache_destroy() comes from 
> drivers/usb/mon/mon_text.c.

I looked at this whole day today, but found nothing.

The code analysis for usbmon shows nothing. Anyone wants to have
a look?

Reproduction does not work either. I tried various loads, loops
of opening/closing while pushing events from USB, nothing.

BTW, I tried to create memory pressure with mem=300MB, result:

top: page allocation failure. order:4, mode:0xc0d0
Pid: 2337, comm: top Not tainted 2.6.25-ub #33

Call Trace:
 [<ffffffff8027450d>] __alloc_pages+0x32f/0x352
 [<ffffffff80272252>] rmqueue_bulk+0x8a/0x9b
 [<ffffffff80273a3d>] __get_free_pages+0xe/0x4d
 [<ffffffff802ddd14>] show_stat+0x25/0x4f1
 [<ffffffff80252f6f>] __lock_acquire+0xd95/0xda4
 [<ffffffff80252f6f>] __lock_acquire+0xd95/0xda4
 [<ffffffff802b349c>] seq_read+0x37/0x2a7
 [<ffffffff80251a2a>] mark_held_locks+0x57/0x73
 [<ffffffff802b349c>] seq_read+0x37/0x2a7
 [<ffffffff80470fff>] mutex_lock_nested+0x23a/0x256
 [<ffffffff80251bd4>] trace_hardirqs_on+0xf9/0x123
 [<ffffffff8047100c>] mutex_lock_nested+0x247/0x256
 [<ffffffff80470fff>] mutex_lock_nested+0x23a/0x256
 [<ffffffff8047100c>] mutex_lock_nested+0x247/0x256
 [<ffffffff802b3579>] seq_read+0x114/0x2a7
 [<ffffffff802b3465>] seq_read+0x0/0x2a7
 [<ffffffff802d74a7>] proc_reg_read+0x80/0x9b

The error is not just advisory, the application (top) dies because
/proc/stat cannot be read. I'm surprised that you people managed
to get the warning at 2443, because things start unravel way earlier
for me.

-- Pete
--
unsubscribe notice
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/