[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20091207164826.eb2bbc48.akpm@linux-foundation.org>
Date: Mon, 7 Dec 2009 16:48:26 -0800
From: Andrew Morton <akpm@...ux-foundation.org>
To: Richard Zidlicky <rz@...ux-m68k.org>
Cc: linux-kernel@...r.kernel.org, netdev@...r.kernel.org,
Takashi Iwai <tiwai@...e.de>
Subject: Re: 2.6.32 regression, hard lock
On Tue, 8 Dec 2009 01:23:53 +0100
Richard Zidlicky <rz@...ux-m68k.org> wrote:
>
> > > first visible symptom is that ppp over UMTS connection "stops working", connection
> > > did not die. Trying to restart the connection fails and the related processes hang.
> > >
> > > After that one after another everything stops working, "telinit 6" does not do its
> > > job and SysRQ sync-umount-reboot is logged in messages but has no visible effect.
> > >
> > > Attaching messages and .config.
> > >
> >
> > I uploaded your dmesg output to
> > http://userweb.kernel.org/~akpm/stuff/messages.txt
>
> >
> > I'm looking at those traces and am not able to develop a theory to
> > explain it :(
>
> thanks for looking at it.
>
> I did revert to 2.31.5 and to my big surprise found a similar thing happened
> so it is not a regression, at least not from 2.6.31.
This trace is different. It's a lock ranking warning and is surely
unrelated to the IRQs-off lockup which you earlier experienced.
> No idea why it did not occur anytime sooner while running earlier kernels, however
> the most recently plugged in device is an USB webcam with micro which I am suspecting
> to play a role in this message (posted more details about it earlier):
>
> Dec 6 10:31:39 localhost kernel: [ 162.087836] =======================================================
> Dec 6 10:31:39 localhost kernel: [ 162.087839] [ INFO: possible circular locking dependency detected ]
> Dec 6 10:31:39 localhost kernel: [ 162.087842] 2.6.32v0 #1
> Dec 6 10:31:39 localhost kernel: [ 162.087844] -------------------------------------------------------
> Dec 6 10:31:39 localhost kernel: [ 162.087846] pulseaudio/4506 is trying to acquire lock:
> Dec 6 10:31:39 localhost kernel: [ 162.087848] (sysfs_mutex){+.+.+.}, at: [<c050c99d>] sysfs_get_dirent+0x15/0x35
> Dec 6 10:31:39 localhost kernel: [ 162.087857]
> Dec 6 10:31:39 localhost kernel: [ 162.087858] but task is already holding lock:
> Dec 6 10:31:39 localhost kernel: [ 162.087860] (&pcm->open_mutex){+.+.+.}, at: [<f80f4b2a>] snd_pcm_release+0x55/0x9e [snd_pcm]
> Dec 6 10:31:39 localhost kernel: [ 162.087873]
> Dec 6 10:31:39 localhost kernel: [ 162.087874] which lock already depends on the new lock.
> Dec 6 10:31:39 localhost kernel: [ 162.087875]
> Dec 6 10:31:39 localhost kernel: [ 162.087876]
> Dec 6 10:31:39 localhost kernel: [ 162.087877] the existing dependency chain (in reverse order) is:
> Dec 6 10:31:39 localhost kernel: [ 162.087879]
> Dec 6 10:31:39 localhost kernel: [ 162.087880] -> #2 (&pcm->open_mutex){+.+.+.}:
> Dec 6 10:31:39 localhost kernel: [ 162.087884] [<c045a703>] __lock_acquire+0xa2a/0xbb5
>
> Dec 6 10:31:39 localhost kernel: [ 162.087890] [<c045a922>] lock_acquire+0x94/0xb1
> Dec 6 10:31:39 localhost kernel: [ 162.087893] [<c07324a6>] __mutex_lock_common+0x35/0x2dc
> Dec 6 10:31:39 localhost kernel: [ 162.087898] [<c07327eb>] mutex_lock_nested+0x30/0x38
> Dec 6 10:31:39 localhost kernel: [ 162.087901] [<f80f4b2a>] snd_pcm_release+0x55/0x9e [snd_pcm]
> Dec 6 10:31:39 localhost kernel: [ 162.087913] [<c04ca0b5>] __fput+0xf0/0x187
> Dec 6 10:31:39 localhost kernel: [ 162.087918] [<c04ca165>] fput+0x19/0x1b
> Dec 6 10:31:39 localhost kernel: [ 162.087921] [<c04b0d6b>] remove_vma+0x3e/0x5d
> Dec 6 10:31:39 localhost kernel: [ 162.087926] [<c04b19f0>] do_munmap+0x21c/0x237
> Dec 6 10:31:39 localhost kernel: [ 162.087929] [<c04b1a3b>] sys_munmap+0x30/0x3f
> Dec 6 10:31:39 localhost kernel: [ 162.087932] [<c040321d>] syscall_call+0x7/0xb
> Dec 6 10:31:39 localhost kernel: [ 162.087937]
> Dec 6 10:31:39 localhost kernel: [ 162.087938] -> #1 (&mm->mmap_sem){++++++}:
> Dec 6 10:31:39 localhost kernel: [ 162.087942] [<c045a703>] __lock_acquire+0xa2a/0xbb5
> Dec 6 10:31:39 localhost kernel: [ 162.087946] [<c045a922>] lock_acquire+0x94/0xb1
> Dec 6 10:31:39 localhost kernel: [ 162.087949] [<c04aba42>] might_fault+0x64/0x81
> Dec 6 10:31:39 localhost kernel: [ 162.087952] [<c05b34cc>] copy_to_user+0x2c/0xfc
> Dec 6 10:31:39 localhost kernel: [ 162.087957] [<c04d4879>] filldir+0x78/0xb7
> Dec 6 10:31:39 localhost kernel: [ 162.087961] [<c050c88d>] sysfs_readdir+0x117/0x14b
> Dec 6 10:31:39 localhost kernel: [ 162.087964] [<c04d49dd>] vfs_readdir+0x68/0x94
> Dec 6 10:31:39 localhost kernel: [ 162.087968] [<c04d4b0b>] sys_getdents+0x62/0xa1
> Dec 6 10:31:39 localhost kernel: [ 162.087971] [<c040321d>] syscall_call+0x7/0xb
> Dec 6 10:31:39 localhost kernel: [ 162.087975]
> Dec 6 10:31:39 localhost kernel: [ 162.087975] -> #0 (sysfs_mutex){+.+.+.}:
> Dec 6 10:31:39 localhost kernel: [ 162.087979] [<c045a610>] __lock_acquire+0x937/0xbb5
> Dec 6 10:31:39 localhost kernel: [ 162.087983] [<c045a922>] lock_acquire+0x94/0xb1
> Dec 6 10:31:39 localhost kernel: [ 162.087986] [<c07324a6>] __mutex_lock_common+0x35/0x2dc
> Dec 6 10:31:39 localhost kernel: [ 162.087990] [<c07327eb>] mutex_lock_nested+0x30/0x38
> Dec 6 10:31:39 localhost kernel: [ 162.087993] [<c050c99d>] sysfs_get_dirent+0x15/0x35
> Dec 6 10:31:39 localhost kernel: [ 162.087996] [<c050e03e>] sysfs_remove_group+0x1e/0xa3
> Dec 6 10:31:39 localhost kernel: [ 162.088000] [<c062bf50>] dpm_sysfs_remove+0x10/0x12
> Dec 6 10:31:39 localhost kernel: [ 162.088006] [<c0627108>] device_del+0x33/0x154
> Dec 6 10:31:39 localhost kernel: [ 162.088010] [<c0627251>] device_unregister+0x28/0x4b
> Dec 6 10:31:39 localhost kernel: [ 162.088014] [<c0678499>] usb_remove_ep_devs+0x15/0x1f
> Dec 6 10:31:39 localhost kernel: [ 162.088018] [<c0672ba2>] remove_intf_ep_devs+0x21/0x32
> Dec 6 10:31:39 localhost kernel: [ 162.088023] [<c0673b3a>] usb_set_interface+0x10c/0x19f
> Dec 6 10:31:39 localhost kernel: [ 162.088027] [<f832dce6>] snd_usb_capture_close+0x1e/0x37 [snd_usb_audio]
> Dec 6 10:31:39 localhost kernel: [ 162.088040] [<f80f4aac>] snd_pcm_release_substream+0x3d/0x66 [snd_pcm]
> Dec 6 10:31:39 localhost kernel: [ 162.088050] [<f80f4b31>] snd_pcm_release+0x5c/0x9e [snd_pcm]
> Dec 6 10:31:39 localhost kernel: [ 162.088058] [<c04ca0b5>] __fput+0xf0/0x187
> Dec 6 10:31:39 localhost kernel: [ 162.088062] [<c04ca165>] fput+0x19/0x1b
> Dec 6 10:31:39 localhost kernel: [ 162.088065] [<c04b0d6b>] remove_vma+0x3e/0x5d
> Dec 6 10:31:39 localhost kernel: [ 162.088069] [<c04b19f0>] do_munmap+0x21c/0x237
> Dec 6 10:31:39 localhost kernel: [ 162.088072] [<c04b1a3b>] sys_munmap+0x30/0x3f
> Dec 6 10:31:39 localhost kernel: [ 162.088076] [<c040321d>] syscall_call+0x7/0xb
> Dec 6 10:31:39 localhost kernel: [ 162.088079]
> Dec 6 10:31:39 localhost kernel: [ 162.088080] other info that might help us debug this:
> Dec 6 10:31:39 localhost kernel: [ 162.088080]
> Dec 6 10:31:39 localhost kernel: [ 162.088083] 2 locks held by pulseaudio/4506:
> Dec 6 10:31:39 localhost kernel: [ 162.088085] #0: (&mm->mmap_sem){++++++}, at: [<c04b1a2e>] sys_munmap+0x23/0x3f
> Dec 6 10:31:39 localhost kernel: [ 162.088090] #1: (&pcm->open_mutex){+.+.+.}, at: [<f80f4b2a>] snd_pcm_release+0x55/0x9e [snd_pcm]
> Dec 6 10:31:39 localhost kernel: [ 162.088104]
> Dec 6 10:31:39 localhost kernel: [ 162.088105] stack backtrace:
> Dec 6 10:31:39 localhost kernel: [ 162.088108] Pid: 4506, comm: pulseaudio Not tainted 2.6.32v0 #1
> Dec 6 10:31:39 localhost kernel: [ 162.088110] Call Trace:
> Dec 6 10:31:39 localhost kernel: [ 162.088113] [<c07312cc>] ? printk+0xf/0x13
> Dec 6 10:31:39 localhost kernel: [ 162.088117] [<c045999b>] print_circular_bug+0x91/0x9d
> Dec 6 10:31:39 localhost kernel: [ 162.088121] [<c045a610>] __lock_acquire+0x937/0xbb5
> Dec 6 10:31:39 localhost kernel: [ 162.088124] [<c04584f5>] ? save_trace+0x37/0xa3
> Dec 6 10:31:39 localhost kernel: [ 162.088128] [<c045a922>] lock_acquire+0x94/0xb1
> Dec 6 10:31:39 localhost kernel: [ 162.088131] [<c050c99d>] ? sysfs_get_dirent+0x15/0x35
> Dec 6 10:31:39 localhost kernel: [ 162.088135] [<c07324a6>] __mutex_lock_common+0x35/0x2dc
> Dec 6 10:31:39 localhost kernel: [ 162.088138] [<c050c99d>] ? sysfs_get_dirent+0x15/0x35
> Dec 6 10:31:39 localhost kernel: [ 162.088141] [<c0458f23>] ? mark_held_locks+0x43/0x5b
> Dec 6 10:31:39 localhost kernel: [ 162.088145] [<c04591bf>] ? trace_hardirqs_on+0xb/0xd
> Dec 6 10:31:39 localhost kernel: [ 162.088148] [<c07327eb>] mutex_lock_nested+0x30/0x38
> Dec 6 10:31:39 localhost kernel: [ 162.088151] [<c050c99d>] ? sysfs_get_dirent+0x15/0x35
> Dec 6 10:31:39 localhost kernel: [ 162.088154] [<c050c99d>] sysfs_get_dirent+0x15/0x35
> Dec 6 10:31:39 localhost kernel: [ 162.088157] [<c050e03e>] sysfs_remove_group+0x1e/0xa3
> Dec 6 10:31:39 localhost kernel: [ 162.088161] [<c062bf50>] dpm_sysfs_remove+0x10/0x12
> Dec 6 10:31:39 localhost kernel: [ 162.088164] [<c0627108>] device_del+0x33/0x154
> Dec 6 10:31:39 localhost kernel: [ 162.088168] [<c0627251>] device_unregister+0x28/0x4b
> Dec 6 10:31:39 localhost kernel: [ 162.088171] [<c0678499>] usb_remove_ep_devs+0x15/0x1f
> Dec 6 10:31:39 localhost kernel: [ 162.088174] [<c0672ba2>] remove_intf_ep_devs+0x21/0x32
> Dec 6 10:31:39 localhost kernel: [ 162.088178] [<c0673b3a>] usb_set_interface+0x10c/0x19f
> Dec 6 10:31:39 localhost kernel: [ 162.088190] [<f832dce6>] snd_usb_capture_close+0x1e/0x37 [snd_usb_audio]
> Dec 6 10:31:39 localhost kernel: [ 162.088199] [<f80f4aac>] snd_pcm_release_substream+0x3d/0x66 [snd_pcm]
> Dec 6 10:31:39 localhost kernel: [ 162.088208] [<f80f4b31>] snd_pcm_release+0x5c/0x9e [snd_pcm]
> Dec 6 10:31:39 localhost kernel: [ 162.088211] [<c04ca0b5>] __fput+0xf0/0x187
> Dec 6 10:31:39 localhost kernel: [ 162.088215] [<c04ca165>] fput+0x19/0x1b
> Dec 6 10:31:39 localhost kernel: [ 162.088218] [<c04b0d6b>] remove_vma+0x3e/0x5d
> Dec 6 10:31:39 localhost kernel: [ 162.088221] [<c04b19f0>] do_munmap+0x21c/0x237
> Dec 6 10:31:39 localhost kernel: [ 162.088225] [<c04b1a3b>] sys_munmap+0x30/0x3f
> Dec 6 10:31:39 localhost kernel: [ 162.088228] [<c040321d>] syscall_call+0x7/0xb
And fortunately I know who to cc on this one ;)
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Powered by blists - more mailing lists