Re: linux-next: Tree for July 11

On Sat, Jul 12, 2008 at 8:51 PM, Rafael J. Wysocki [off-list ref] wrote:

and the second one is the following:

WARNING: at /home/rafael/src/linux-next/include/linux/blkdev.h:447 blk_plug_device+0x9b/0xb0()
Modules linked in: snd_hda_intel ohci1394 ieee1394 snd_pcm rtc_cmos sr_mod rtc_core floppy snd_timer wmi button cdrom rtc_lib serio_raw sky2 snd_page_alloc snd_hwdep snd evdev joydev sg soundcore raid456 async_xor async_memcpy async_tx xor raid0 usbhid ff_memless ehci_hcd ohci_hcd sd_mod edd raid1 ext3 jbd fan pata_marvell pata_atiixp thermal processor
Pid: 2244, comm: kjournald Not tainted 2.6.26-rc9-next #44

Call Trace:
 [<ffffffff8023d37f>] warn_on_slowpath+0x5f/0x80
 [<ffffffff80220030>] ? hpet_unregister_irq_handler+0x0/0x30
 [<ffffffff8022a8ae>] ? kmemcheck_mark_initialized+0xe/0x10
 [<ffffffff802b97fb>] ? kmemcheck_slab_alloc+0x2b/0x50
 [<ffffffff802b8610>] ? kmem_cache_alloc+0xc0/0x140
 [<ffffffff80291921>] ? mempool_alloc_slab+0x11/0x20
 [<ffffffff80291a8b>] ? mempool_alloc+0x5b/0x140
 [<ffffffff803575db>] blk_plug_device+0x9b/0xb0
 [<ffffffff80453d9f>] bitmap_startwrite+0xbf/0x1b0
 [<ffffffff802e87e4>] ? bio_alloc_bioset+0x54/0xb0
 [<ffffffffa004eafa>] make_request+0x39a/0x810 [raid1]
 [<ffffffff80291a8b>] ? mempool_alloc+0x5b/0x140
 [<ffffffff80291a8b>] ? mempool_alloc+0x5b/0x140
 [<ffffffff8035682d>] generic_make_request+0x17d/0x2b0
 [<ffffffff803581dc>] submit_bio+0x6c/0xf0
 [<ffffffff802e3c60>] submit_bh+0xf0/0x130
 [<ffffffffa001cce0>] journal_commit_transaction+0xa40/0x1000 [jbd]
 [<ffffffff80248504>] ? try_to_del_timer_sync+0x44/0x90
 [<ffffffffa0020967>] kjournald+0xe7/0x250 [jbd]
 [<ffffffff80254350>] ? autoremove_wake_function+0x0/0x40
 [<ffffffffa0020880>] ? kjournald+0x0/0x250 [jbd]
 [<ffffffff80253efd>] kthread+0x4d/0x80
 [<ffffffff8020c6d9>] child_rip+0xa/0x11
 [<ffffffff80500f9b>] ? _spin_unlock_irq+0x2b/0x40
 [<ffffffff8020bd0f>] ? restore_args+0x0/0x30
 [<ffffffff80254052>] ? kthreadd+0x122/0x1a0
 [<ffffffff80253eb0>] ? kthread+0x0/0x80
 [<ffffffff8020c6cf>] ? child_rip+0x0/0x11

Hm, even though there's kmemcheck in the stacktrace, I think it's
unrelated. Probably it's just a call that returned and the addresses
were left on the stack. You can also see that the warning clearly
comes from blkdev.h, and since kmemcheck does *not* call into any
blkdev stuff, those stacktrace entries must be just left-overs, as is
also indicated by the unreliable mark, so although those are wrong in
certain circumstances, I don't think they are in this case.

The warning comes from here:

static inline int queue_flag_test_and_set(unsigned int flag,
                                          struct request_queue *q)
{
        WARN_ON_ONCE(!queue_is_locked(q));

I'll add Jens Axboe to Cc.

Thanks for the heads-up!


Vegard

-- 
"The animistic metaphor of the bug that maliciously sneaked in while
the programmer was not looking is intellectually dishonest as it
disguises that the error is the programmer's own creation."
	-- E. W. Dijkstra, EWD1036