linux-ext4.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* ext4: possible unsafe locking scenario
@ 2014-11-26 14:59 Sergey Senozhatsky
  2014-11-26 15:28 ` Theodore Ts'o
  0 siblings, 1 reply; 5+ messages in thread
From: Sergey Senozhatsky @ 2014-11-26 14:59 UTC (permalink / raw)
  To: Theodore Ts'o; +Cc: Andreas Dilger, linux-ext4, linux-kernel

Hello,

locked spew with the current -next.

[  427.152842] =============================================
[  427.152843] [ INFO: possible recursive locking detected ]
[  427.152847] 3.18.0-rc6-next-20141126-dbg-00001-gb80a67f-dirty #402 Not tainted
[  427.152848] ---------------------------------------------
[  427.152850] gcc/10495 is trying to acquire lock:
[  427.152851]  (&ei->i_data_sem){++++..}, at: [<ffffffff8119022f>] ext4_truncate+0x1cc/0x2bc
[  427.152862] 
but task is already holding lock:
[  427.152864]  (&ei->i_data_sem){++++..}, at: [<ffffffff8119022f>] ext4_truncate+0x1cc/0x2bc
[  427.152868] 
other info that might help us debug this:
[  427.152870]  Possible unsafe locking scenario:

[  427.152872]        CPU0
[  427.152873]        ----
[  427.152874]   lock(&ei->i_data_sem);
[  427.152875]   lock(&ei->i_data_sem);
[  427.152877] 
 *** DEADLOCK ***

[  427.152879]  May be due to missing lock nesting notation

[  427.152881] 5 locks held by gcc/10495:
[  427.152883]  #0:  (sb_writers#7){.+.+.+}, at: [<ffffffff81101e62>] mnt_want_write+0x1f/0x46
[  427.152891]  #1:  (sb_internal){.+.+..}, at: [<ffffffff8119119c>] ext4_evict_inode+0x1cb/0x3d6
[  427.152897]  #2:  (jbd2_handle){+.+...}, at: [<ffffffff811c4099>] start_this_handle+0x4e0/0x566
[  427.152902]  #3:  (&ei->i_data_sem){++++..}, at: [<ffffffff8119022f>] ext4_truncate+0x1cc/0x2bc
[  427.152907]  #4:  (sb_internal){.+.+..}, at: [<ffffffff8119119c>] ext4_evict_inode+0x1cb/0x3d6
[  427.152911] 
stack backtrace:
[  427.152915] CPU: 3 PID: 10495 Comm: gcc Not tainted 3.18.0-rc6-next-20141126-dbg-00001-gb80a67f-dirty #402
[  427.152917] Hardware name: Acer             Aspire 5741G    /Aspire 5741G    , BIOS V1.20 02/08/2011
[  427.152919]  ffffffff820f6e40 ffff880061a13908 ffffffff813cd7ff 0000000000000006
[  427.152922]  ffffffff820f6e40 ffff880061a139d8 ffffffff8106c27d ffff880152eb9c30
[  427.152925]  ffff880152eb9c30 ffff8800377a2000 ffff88014a6e47d8 0000000000000003
[  427.152928] Call Trace:
[  427.152934]  [<ffffffff813cd7ff>] dump_stack+0x4c/0x65
[  427.152938]  [<ffffffff8106c27d>] __lock_acquire+0x7b5/0x1a17
[  427.152941]  [<ffffffff8118eff7>] ? ext4_mark_iloc_dirty+0x70c/0x7ab
[  427.152944]  [<ffffffff8106db62>] lock_acquire+0x4f/0x6e
[  427.152946]  [<ffffffff8119022f>] ? ext4_truncate+0x1cc/0x2bc
[  427.152951]  [<ffffffff81058acf>] ? __might_sleep+0xac/0xb4
[  427.152955]  [<ffffffff813d1695>] down_write+0x3d/0x6d
[  427.152958]  [<ffffffff8119022f>] ? ext4_truncate+0x1cc/0x2bc
[  427.152961]  [<ffffffff8119022f>] ext4_truncate+0x1cc/0x2bc
[  427.152963]  [<ffffffff811912a2>] ext4_evict_inode+0x2d1/0x3d6
[  427.152967]  [<ffffffff810fd871>] evict+0x102/0x1ad
[  427.152969]  [<ffffffff810fe490>] ? iput+0x120/0x131
[  427.152971]  [<ffffffff810fe498>] iput+0x128/0x131
[  427.152974]  [<ffffffff8118ef21>] ext4_mark_iloc_dirty+0x636/0x7ab
[  427.152977]  [<ffffffff8118f1a3>] ext4_mark_inode_dirty+0x8a/0x1a2
[  427.152982]  [<ffffffff811a484a>] __ext4_ext_dirty+0x6d/0x79
[  427.152985]  [<ffffffff811a731f>] ext4_ext_remove_space+0x858/0xcdc
[  427.152988]  [<ffffffff811a8cc7>] ext4_ext_truncate+0x9a/0xbe
[  427.152991]  [<ffffffff81190250>] ext4_truncate+0x1ed/0x2bc
[  427.152994]  [<ffffffff811912a2>] ext4_evict_inode+0x2d1/0x3d6
[  427.152996]  [<ffffffff810fd871>] evict+0x102/0x1ad
[  427.152998]  [<ffffffff8106b837>] ? trace_hardirqs_on_caller+0x180/0x19c
[  427.153001]  [<ffffffff810fe490>] ? iput+0x120/0x131
[  427.153003]  [<ffffffff810fe498>] iput+0x128/0x131
[  427.153007]  [<ffffffff810f4449>] do_unlinkat+0x141/0x213
[  427.153013]  [<ffffffff8100da89>] ? syscall_trace_enter_phase1+0x112/0x122
[  427.153018]  [<ffffffff812166db>] ? trace_hardirqs_on_thunk+0x3a/0x3f
[  427.153021]  [<ffffffff810f4b84>] SyS_unlink+0x11/0x13
[  427.153025]  [<ffffffff813d34d6>] system_call_fastpath+0x16/0x1b

	-ss

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: ext4: possible unsafe locking scenario
  2014-11-26 14:59 ext4: possible unsafe locking scenario Sergey Senozhatsky
@ 2014-11-26 15:28 ` Theodore Ts'o
  2014-11-27 12:32   ` Sergey Senozhatsky
  0 siblings, 1 reply; 5+ messages in thread
From: Theodore Ts'o @ 2014-11-26 15:28 UTC (permalink / raw)
  To: Sergey Senozhatsky; +Cc: Andreas Dilger, linux-ext4, linux-kernel

On Wed, Nov 26, 2014 at 11:59:51PM +0900, Sergey Senozhatsky wrote:
> Hello,
> 
> locked spew with the current -next.

Hmm.... what was the system doing at that time?  Is it something that
reliably reproduces, or is this a one-off?

Thanks,

						- Ted

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: ext4: possible unsafe locking scenario
  2014-11-26 15:28 ` Theodore Ts'o
@ 2014-11-27 12:32   ` Sergey Senozhatsky
  2014-11-27 15:38     ` Theodore Ts'o
  0 siblings, 1 reply; 5+ messages in thread
From: Sergey Senozhatsky @ 2014-11-27 12:32 UTC (permalink / raw)
  To: Theodore Ts'o, Sergey Senozhatsky, Andreas Dilger, linux-ext4,
	linux-kernel

On (11/26/14 10:28), Theodore Ts'o wrote:
> Date: Wed, 26 Nov 2014 10:28:21 -0500
> From: Theodore Ts'o <tytso@mit.edu>
> To: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
> Cc: Andreas Dilger <adilger.kernel@dilger.ca>, linux-ext4@vger.kernel.org,
>  linux-kernel@vger.kernel.org
> Subject: Re: ext4: possible unsafe locking scenario
> User-Agent: Mutt/1.5.23 (2014-03-12)
> 
> On Wed, Nov 26, 2014 at 11:59:51PM +0900, Sergey Senozhatsky wrote:
> > Hello,
> > 
> > locked spew with the current -next.
> 
> Hmm.... what was the system doing at that time?  Is it something that
> reliably reproduces, or is this a one-off?
> 
> Thanks,
> 
> 						- Ted

sorry for long response, can reply only in my spare time.


yet another one.
in both cases: compiling glibc with gcc (-j2).


[ 1127.189118] =============================================
[ 1127.189119] [ INFO: possible recursive locking detected ]
[ 1127.189122] 3.18.0-rc6-next-20141126-dbg-00001-gb80a67f-dirty #402 Not tainted
[ 1127.189123] ---------------------------------------------
[ 1127.189125] rm/27750 is trying to acquire lock:
[ 1127.189127]  (&ei->i_data_sem){++++..}, at: [<ffffffff8119022f>] ext4_truncate+0x1cc/0x2bc
[ 1127.189137] 
but task is already holding lock:
[ 1127.189139]  (&ei->i_data_sem){++++..}, at: [<ffffffff8119022f>] ext4_truncate+0x1cc/0x2bc
[ 1127.189144] 
other info that might help us debug this:
[ 1127.189145]  Possible unsafe locking scenario:

[ 1127.189147]        CPU0
[ 1127.189148]        ----
[ 1127.189149]   lock(&ei->i_data_sem);
[ 1127.189151]   lock(&ei->i_data_sem);
[ 1127.189153] 
 *** DEADLOCK ***

[ 1127.189155]  May be due to missing lock nesting notation

[ 1127.189157] 5 locks held by rm/27750:
[ 1127.189158]  #0:  (sb_writers#7){.+.+.+}, at: [<ffffffff81101e62>] mnt_want_write+0x1f/0x46
[ 1127.189166]  #1:  (sb_internal){.+.+..}, at: [<ffffffff8119119c>] ext4_evict_inode+0x1cb/0x3d6
[ 1127.189171]  #2:  (jbd2_handle){+.+...}, at: [<ffffffff811c4099>] start_this_handle+0x4e0/0x566
[ 1127.189177]  #3:  (&ei->i_data_sem){++++..}, at: [<ffffffff8119022f>] ext4_truncate+0x1cc/0x2bc
[ 1127.189181]  #4:  (sb_internal){.+.+..}, at: [<ffffffff8119119c>] ext4_evict_inode+0x1cb/0x3d6
[ 1127.189186] 
stack backtrace:
[ 1127.189189] CPU: 0 PID: 27750 Comm: rm Not tainted 3.18.0-rc6-next-20141126-dbg-00001-gb80a67f-dirty #402
[ 1127.189191] Hardware name: Acer             Aspire 5741G    /Aspire 5741G    , BIOS V1.20 02/08/2011
[ 1127.189193]  ffffffff820f71a0 ffff8801045d7908 ffffffff813cd7ff 0000000000000006
[ 1127.189196]  ffffffff820f71a0 ffff8801045d79d8 ffffffff8106c27d ffff880111aafdd0
[ 1127.189200]  ffff880111aafdd0 ffff88015255e800 ffff8801514b0798 0000000000000003
[ 1127.189203] Call Trace:
[ 1127.189208]  [<ffffffff813cd7ff>] dump_stack+0x4c/0x65
[ 1127.189212]  [<ffffffff8106c27d>] __lock_acquire+0x7b5/0x1a17
[ 1127.189216]  [<ffffffff810fe317>] ? find_active_inode_nowait+0x91/0xa8
[ 1127.189219]  [<ffffffff8118eff7>] ? ext4_mark_iloc_dirty+0x70c/0x7ab
[ 1127.189221]  [<ffffffff8106db62>] lock_acquire+0x4f/0x6e
[ 1127.189225]  [<ffffffff8119022f>] ? ext4_truncate+0x1cc/0x2bc
[ 1127.189231]  [<ffffffff81058acf>] ? __might_sleep+0xac/0xb4
[ 1127.189236]  [<ffffffff813d1695>] down_write+0x3d/0x6d
[ 1127.189240]  [<ffffffff8119022f>] ? ext4_truncate+0x1cc/0x2bc
[ 1127.189243]  [<ffffffff8119022f>] ext4_truncate+0x1cc/0x2bc
[ 1127.189247]  [<ffffffff811912a2>] ext4_evict_inode+0x2d1/0x3d6
[ 1127.189250]  [<ffffffff810fd871>] evict+0x102/0x1ad
[ 1127.189253]  [<ffffffff810fe490>] ? iput+0x120/0x131
[ 1127.189257]  [<ffffffff810fe498>] iput+0x128/0x131
[ 1127.189261]  [<ffffffff8118ef21>] ext4_mark_iloc_dirty+0x636/0x7ab
[ 1127.189265]  [<ffffffff8118f1a3>] ext4_mark_inode_dirty+0x8a/0x1a2
[ 1127.189270]  [<ffffffff811a484a>] __ext4_ext_dirty+0x6d/0x79
[ 1127.189274]  [<ffffffff811a731f>] ext4_ext_remove_space+0x858/0xcdc
[ 1127.189278]  [<ffffffff811a8cc7>] ext4_ext_truncate+0x9a/0xbe
[ 1127.189282]  [<ffffffff81190250>] ext4_truncate+0x1ed/0x2bc
[ 1127.189286]  [<ffffffff811912a2>] ext4_evict_inode+0x2d1/0x3d6
[ 1127.189289]  [<ffffffff810fd871>] evict+0x102/0x1ad
[ 1127.189293]  [<ffffffff8106b837>] ? trace_hardirqs_on_caller+0x180/0x19c
[ 1127.189296]  [<ffffffff810fe490>] ? iput+0x120/0x131
[ 1127.189299]  [<ffffffff810fe498>] iput+0x128/0x131
[ 1127.189304]  [<ffffffff810f4449>] do_unlinkat+0x141/0x213
[ 1127.189311]  [<ffffffff8100da89>] ? syscall_trace_enter_phase1+0x112/0x122
[ 1127.189317]  [<ffffffff812166db>] ? trace_hardirqs_on_thunk+0x3a/0x3f
[ 1127.189322]  [<ffffffff810f4b71>] SyS_unlinkat+0x24/0x26
[ 1127.189328]  [<ffffffff813d34d6>] system_call_fastpath+0x16/0x1b


	-ss

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: ext4: possible unsafe locking scenario
  2014-11-27 12:32   ` Sergey Senozhatsky
@ 2014-11-27 15:38     ` Theodore Ts'o
  2014-11-28 11:31       ` Sergey Senozhatsky
  0 siblings, 1 reply; 5+ messages in thread
From: Theodore Ts'o @ 2014-11-27 15:38 UTC (permalink / raw)
  To: Sergey Senozhatsky; +Cc: Andreas Dilger, linux-ext4, linux-kernel

Ah, OK, I know what's going in; this should be fixed soon.  Thanks for
reporting it, and my apologies for the inconvenience.

    	       	      	    	     	    - Ted

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: ext4: possible unsafe locking scenario
  2014-11-27 15:38     ` Theodore Ts'o
@ 2014-11-28 11:31       ` Sergey Senozhatsky
  0 siblings, 0 replies; 5+ messages in thread
From: Sergey Senozhatsky @ 2014-11-28 11:31 UTC (permalink / raw)
  To: Theodore Ts'o, Sergey Senozhatsky, Andreas Dilger, linux-ext4,
	linux-kernel

On (11/27/14 10:38), Theodore Ts'o wrote:
> Ah, OK, I know what's going in; this should be fixed soon.  Thanks for
> reporting it, and my apologies for the inconvenience.
> 

no problem, sir. thank you.

	-ss

^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2014-11-28 11:31 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2014-11-26 14:59 ext4: possible unsafe locking scenario Sergey Senozhatsky
2014-11-26 15:28 ` Theodore Ts'o
2014-11-27 12:32   ` Sergey Senozhatsky
2014-11-27 15:38     ` Theodore Ts'o
2014-11-28 11:31       ` Sergey Senozhatsky

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).