LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
* [linux-next] possible circular locking dependency for cryptsetup
@ 2008-11-07  8:32 Markus Dahms
  2008-11-07  8:47 ` KOSAKI Motohiro
  2008-11-07  9:37 ` Stephen Rothwell
  0 siblings, 2 replies; 4+ messages in thread
From: Markus Dahms @ 2008-11-07  8:32 UTC (permalink / raw)
  To: linux-next; +Cc: linux-kernel

Hi there,

I got the following warning for 2.6.28-rc3-next-20081106, it's quiet for
2.6.28-rc3. Besides the new warning everything seems to work fine.

As a side note, I wanted to file this at bugzilla.k.o, but there is no
"next" tree in "alternate trees" - intended?

FWIW: x86, 2xP3

Greetings,

	Markus

=======================================================
[ INFO: possible circular locking dependency detected ]
2.6.28-rc3-next-20081106-mad-1 #2
-------------------------------------------------------
cryptsetup/2554 is trying to acquire lock:
 (&cpu_hotplug.lock){--..}, at: [<c012cd32>] get_online_cpus+0x22/0x40

but task is already holding lock:
 (&mm->mmap_sem){----}, at: [<c018b98e>] sys_mlockall+0x5e/0xd0

which lock already depends on the new lock.


the existing dependency chain (in reverse order) is:

-> #3 (&mm->mmap_sem){----}:
       [<c0151560>] __lock_acquire+0xf30/0x17f0
       [<c0151e7e>] lock_acquire+0x5e/0x80
       [<c0187d2c>] might_fault+0x5c/0x80
       [<c0267bc6>] copy_to_user+0x36/0x120
       [<c01af074>] filldir64+0xa4/0xf0
       [<c01ead29>] sysfs_readdir+0x129/0x220
       [<c01af315>] vfs_readdir+0x95/0xb0
       [<c01af399>] sys_getdents64+0x69/0xc0
       [<c010346d>] sysenter_do_call+0x12/0x31
       [<ffffffff>] 0xffffffff

-> #2 (sysfs_mutex){--..}:
       [<c0151560>] __lock_acquire+0xf30/0x17f0
       [<c0151e7e>] lock_acquire+0x5e/0x80
       [<c03e5be4>] mutex_lock_nested+0xb4/0x2c0
       [<c01eb0ac>] sysfs_addrm_start+0x2c/0xb0
       [<c01eb6b0>] create_dir+0x40/0x90
       [<c01eb72b>] sysfs_create_dir+0x2b/0x50
       [<c026116f>] kobject_add_internal+0xcf/0x240
       [<c02613b1>] kobject_add_varg+0x31/0x50
       [<c0261807>] kobject_init_and_add+0x27/0x30
       [<c019e4b3>] sysfs_slab_add+0x83/0x1a0
       [<c019e61e>] sysfs_add_func+0x4e/0x70
       [<c013c6b5>] run_workqueue+0xe5/0x1f0
       [<c013c848>] worker_thread+0x88/0xf0
       [<c01402ec>] kthread+0x3c/0x70
       [<c01042ff>] kernel_thread_helper+0x7/0x18
       [<ffffffff>] 0xffffffff

-> #1 (slub_lock){----}:
       [<c0151560>] __lock_acquire+0xf30/0x17f0
       [<c0151e7e>] lock_acquire+0x5e/0x80
       [<c03e65d7>] down_read+0x37/0x50
       [<c03e40da>] slab_cpuup_callback+0x3a/0x192
       [<c01452cd>] notifier_call_chain+0x2d/0x70
       [<c01453c9>] __raw_notifier_call_chain+0x19/0x20
       [<c03e334e>] _cpu_up+0x5e/0x105
       [<c03e3459>] cpu_up+0x64/0x7b
       [<c055430f>] kernel_init+0xb6/0x173
       [<c01042ff>] kernel_thread_helper+0x7/0x18
       [<ffffffff>] 0xffffffff

-> #0 (&cpu_hotplug.lock){--..}:
       [<c0151691>] __lock_acquire+0x1061/0x17f0
       [<c0151e7e>] lock_acquire+0x5e/0x80
       [<c03e5be4>] mutex_lock_nested+0xb4/0x2c0
       [<c012cd32>] get_online_cpus+0x22/0x40
       [<c013d2c8>] schedule_on_each_cpu+0x38/0xe0
       [<c018115d>] lru_add_drain_all+0xd/0x10
       [<c018b486>] __mlock_vma_pages_range+0x36/0x1f0
       [<c018b74f>] mlock_fixup+0x10f/0x210
       [<c018b8db>] do_mlockall+0x8b/0xa0
       [<c018b9c1>] sys_mlockall+0x91/0xd0
       [<c010346d>] sysenter_do_call+0x12/0x31
       [<ffffffff>] 0xffffffff

other info that might help us debug this:

1 lock held by cryptsetup/2554:
 #0:  (&mm->mmap_sem){----}, at: [<c018b98e>] sys_mlockall+0x5e/0xd0

stack backtrace:
Pid: 2554, comm: cryptsetup Not tainted 2.6.28-rc3-next-20081106-mad-1 #2
Call Trace:
 [<c03e45d2>] ? printk+0x18/0x1e
 [<c0150213>] print_circular_bug_tail+0xc3/0xd0
 [<c014ff4b>] ? print_circular_bug_entry+0x4b/0x50
 [<c0151691>] __lock_acquire+0x1061/0x17f0
 [<c0151e7e>] lock_acquire+0x5e/0x80
 [<c012cd32>] ? get_online_cpus+0x22/0x40
 [<c03e5be4>] mutex_lock_nested+0xb4/0x2c0
 [<c012cd32>] ? get_online_cpus+0x22/0x40
 [<c012cd32>] ? get_online_cpus+0x22/0x40
 [<c019f391>] ? __percpu_alloc_mask+0xa1/0x100
 [<c012cd32>] get_online_cpus+0x22/0x40
 [<c013d2c8>] schedule_on_each_cpu+0x38/0xe0
 [<c0180ed0>] ? lru_add_drain_per_cpu+0x0/0x10
 [<c018115d>] lru_add_drain_all+0xd/0x10
 [<c018b486>] __mlock_vma_pages_range+0x36/0x1f0
 [<c022c109>] ? avc_has_perm+0x59/0x70
 [<c018b74f>] mlock_fixup+0x10f/0x210
 [<c018b8db>] do_mlockall+0x8b/0xa0
 [<c018b9c1>] sys_mlockall+0x91/0xd0
 [<c010346d>] sysenter_do_call+0x12/0x31

-- 
ubuntu is an ancient african word meaning "i can't install debian."
                                                         -- unknown

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [linux-next] possible circular locking dependency for cryptsetup
  2008-11-07  8:32 [linux-next] possible circular locking dependency for cryptsetup Markus Dahms
@ 2008-11-07  8:47 ` KOSAKI Motohiro
  2008-11-07  9:37 ` Stephen Rothwell
  1 sibling, 0 replies; 4+ messages in thread
From: KOSAKI Motohiro @ 2008-11-07  8:47 UTC (permalink / raw)
  To: Markus Dahms; +Cc: kosaki.motohiro, linux-next, linux-kernel

Hi

I already fixed this bug.
its patch queued in mm tree, now.
(file name is mm-remove-lru_add_drain_all-from-the-munlock-path)

I expect it merge to mainline soon.

Thanks!


> Hi there,
> 
> I got the following warning for 2.6.28-rc3-next-20081106, it's quiet for
> 2.6.28-rc3. Besides the new warning everything seems to work fine.
> 
> As a side note, I wanted to file this at bugzilla.k.o, but there is no
> "next" tree in "alternate trees" - intended?
> 
> FWIW: x86, 2xP3
> 
> Greetings,
> 
> 	Markus
> 
> =======================================================
> [ INFO: possible circular locking dependency detected ]
> 2.6.28-rc3-next-20081106-mad-1 #2
> -------------------------------------------------------
> cryptsetup/2554 is trying to acquire lock:
>  (&cpu_hotplug.lock){--..}, at: [<c012cd32>] get_online_cpus+0x22/0x40
> 
> but task is already holding lock:
>  (&mm->mmap_sem){----}, at: [<c018b98e>] sys_mlockall+0x5e/0xd0
> 
> which lock already depends on the new lock.
> 
> 
> the existing dependency chain (in reverse order) is:
> 
> -> #3 (&mm->mmap_sem){----}:
>        [<c0151560>] __lock_acquire+0xf30/0x17f0
>        [<c0151e7e>] lock_acquire+0x5e/0x80
>        [<c0187d2c>] might_fault+0x5c/0x80
>        [<c0267bc6>] copy_to_user+0x36/0x120
>        [<c01af074>] filldir64+0xa4/0xf0
>        [<c01ead29>] sysfs_readdir+0x129/0x220
>        [<c01af315>] vfs_readdir+0x95/0xb0
>        [<c01af399>] sys_getdents64+0x69/0xc0
>        [<c010346d>] sysenter_do_call+0x12/0x31
>        [<ffffffff>] 0xffffffff
> 
> -> #2 (sysfs_mutex){--..}:
>        [<c0151560>] __lock_acquire+0xf30/0x17f0
>        [<c0151e7e>] lock_acquire+0x5e/0x80
>        [<c03e5be4>] mutex_lock_nested+0xb4/0x2c0
>        [<c01eb0ac>] sysfs_addrm_start+0x2c/0xb0
>        [<c01eb6b0>] create_dir+0x40/0x90
>        [<c01eb72b>] sysfs_create_dir+0x2b/0x50
>        [<c026116f>] kobject_add_internal+0xcf/0x240
>        [<c02613b1>] kobject_add_varg+0x31/0x50
>        [<c0261807>] kobject_init_and_add+0x27/0x30
>        [<c019e4b3>] sysfs_slab_add+0x83/0x1a0
>        [<c019e61e>] sysfs_add_func+0x4e/0x70
>        [<c013c6b5>] run_workqueue+0xe5/0x1f0
>        [<c013c848>] worker_thread+0x88/0xf0
>        [<c01402ec>] kthread+0x3c/0x70
>        [<c01042ff>] kernel_thread_helper+0x7/0x18
>        [<ffffffff>] 0xffffffff
> 
> -> #1 (slub_lock){----}:
>        [<c0151560>] __lock_acquire+0xf30/0x17f0
>        [<c0151e7e>] lock_acquire+0x5e/0x80
>        [<c03e65d7>] down_read+0x37/0x50
>        [<c03e40da>] slab_cpuup_callback+0x3a/0x192
>        [<c01452cd>] notifier_call_chain+0x2d/0x70
>        [<c01453c9>] __raw_notifier_call_chain+0x19/0x20
>        [<c03e334e>] _cpu_up+0x5e/0x105
>        [<c03e3459>] cpu_up+0x64/0x7b
>        [<c055430f>] kernel_init+0xb6/0x173
>        [<c01042ff>] kernel_thread_helper+0x7/0x18
>        [<ffffffff>] 0xffffffff
> 
> -> #0 (&cpu_hotplug.lock){--..}:
>        [<c0151691>] __lock_acquire+0x1061/0x17f0
>        [<c0151e7e>] lock_acquire+0x5e/0x80
>        [<c03e5be4>] mutex_lock_nested+0xb4/0x2c0
>        [<c012cd32>] get_online_cpus+0x22/0x40
>        [<c013d2c8>] schedule_on_each_cpu+0x38/0xe0
>        [<c018115d>] lru_add_drain_all+0xd/0x10
>        [<c018b486>] __mlock_vma_pages_range+0x36/0x1f0
>        [<c018b74f>] mlock_fixup+0x10f/0x210
>        [<c018b8db>] do_mlockall+0x8b/0xa0
>        [<c018b9c1>] sys_mlockall+0x91/0xd0
>        [<c010346d>] sysenter_do_call+0x12/0x31
>        [<ffffffff>] 0xffffffff
> 
> other info that might help us debug this:
> 
> 1 lock held by cryptsetup/2554:
>  #0:  (&mm->mmap_sem){----}, at: [<c018b98e>] sys_mlockall+0x5e/0xd0
> 
> stack backtrace:
> Pid: 2554, comm: cryptsetup Not tainted 2.6.28-rc3-next-20081106-mad-1 #2
> Call Trace:
>  [<c03e45d2>] ? printk+0x18/0x1e
>  [<c0150213>] print_circular_bug_tail+0xc3/0xd0
>  [<c014ff4b>] ? print_circular_bug_entry+0x4b/0x50
>  [<c0151691>] __lock_acquire+0x1061/0x17f0
>  [<c0151e7e>] lock_acquire+0x5e/0x80
>  [<c012cd32>] ? get_online_cpus+0x22/0x40
>  [<c03e5be4>] mutex_lock_nested+0xb4/0x2c0
>  [<c012cd32>] ? get_online_cpus+0x22/0x40
>  [<c012cd32>] ? get_online_cpus+0x22/0x40
>  [<c019f391>] ? __percpu_alloc_mask+0xa1/0x100
>  [<c012cd32>] get_online_cpus+0x22/0x40
>  [<c013d2c8>] schedule_on_each_cpu+0x38/0xe0
>  [<c0180ed0>] ? lru_add_drain_per_cpu+0x0/0x10
>  [<c018115d>] lru_add_drain_all+0xd/0x10
>  [<c018b486>] __mlock_vma_pages_range+0x36/0x1f0
>  [<c022c109>] ? avc_has_perm+0x59/0x70
>  [<c018b74f>] mlock_fixup+0x10f/0x210
>  [<c018b8db>] do_mlockall+0x8b/0xa0
>  [<c018b9c1>] sys_mlockall+0x91/0xd0
>  [<c010346d>] sysenter_do_call+0x12/0x31
> 
> -- 
> ubuntu is an ancient african word meaning "i can't install debian."
>                                                          -- unknown
> --
> 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/




^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [linux-next] possible circular locking dependency for cryptsetup
  2008-11-07  8:32 [linux-next] possible circular locking dependency for cryptsetup Markus Dahms
  2008-11-07  8:47 ` KOSAKI Motohiro
@ 2008-11-07  9:37 ` Stephen Rothwell
  2008-11-07 12:40   ` Herbert Xu
  1 sibling, 1 reply; 4+ messages in thread
From: Stephen Rothwell @ 2008-11-07  9:37 UTC (permalink / raw)
  To: Markus Dahms; +Cc: linux-next, linux-kernel, Herbert Xu

[-- Attachment #1: Type: text/plain, Size: 4755 bytes --]

[Just adding a cc]

On Fri, 7 Nov 2008 09:32:00 +0100 Markus Dahms <mad@automagically.de> wrote:
>
> I got the following warning for 2.6.28-rc3-next-20081106, it's quiet for
> 2.6.28-rc3. Besides the new warning everything seems to work fine.
> 
> As a side note, I wanted to file this at bugzilla.k.o, but there is no
> "next" tree in "alternate trees" - intended?
> 
> FWIW: x86, 2xP3
> 
> Greetings,
> 
> 	Markus
> 
> =======================================================
> [ INFO: possible circular locking dependency detected ]
> 2.6.28-rc3-next-20081106-mad-1 #2
> -------------------------------------------------------
> cryptsetup/2554 is trying to acquire lock:
>  (&cpu_hotplug.lock){--..}, at: [<c012cd32>] get_online_cpus+0x22/0x40
> 
> but task is already holding lock:
>  (&mm->mmap_sem){----}, at: [<c018b98e>] sys_mlockall+0x5e/0xd0
> 
> which lock already depends on the new lock.
> 
> 
> the existing dependency chain (in reverse order) is:
> 
> -> #3 (&mm->mmap_sem){----}:
>        [<c0151560>] __lock_acquire+0xf30/0x17f0
>        [<c0151e7e>] lock_acquire+0x5e/0x80
>        [<c0187d2c>] might_fault+0x5c/0x80
>        [<c0267bc6>] copy_to_user+0x36/0x120
>        [<c01af074>] filldir64+0xa4/0xf0
>        [<c01ead29>] sysfs_readdir+0x129/0x220
>        [<c01af315>] vfs_readdir+0x95/0xb0
>        [<c01af399>] sys_getdents64+0x69/0xc0
>        [<c010346d>] sysenter_do_call+0x12/0x31
>        [<ffffffff>] 0xffffffff
> 
> -> #2 (sysfs_mutex){--..}:
>        [<c0151560>] __lock_acquire+0xf30/0x17f0
>        [<c0151e7e>] lock_acquire+0x5e/0x80
>        [<c03e5be4>] mutex_lock_nested+0xb4/0x2c0
>        [<c01eb0ac>] sysfs_addrm_start+0x2c/0xb0
>        [<c01eb6b0>] create_dir+0x40/0x90
>        [<c01eb72b>] sysfs_create_dir+0x2b/0x50
>        [<c026116f>] kobject_add_internal+0xcf/0x240
>        [<c02613b1>] kobject_add_varg+0x31/0x50
>        [<c0261807>] kobject_init_and_add+0x27/0x30
>        [<c019e4b3>] sysfs_slab_add+0x83/0x1a0
>        [<c019e61e>] sysfs_add_func+0x4e/0x70
>        [<c013c6b5>] run_workqueue+0xe5/0x1f0
>        [<c013c848>] worker_thread+0x88/0xf0
>        [<c01402ec>] kthread+0x3c/0x70
>        [<c01042ff>] kernel_thread_helper+0x7/0x18
>        [<ffffffff>] 0xffffffff
> 
> -> #1 (slub_lock){----}:
>        [<c0151560>] __lock_acquire+0xf30/0x17f0
>        [<c0151e7e>] lock_acquire+0x5e/0x80
>        [<c03e65d7>] down_read+0x37/0x50
>        [<c03e40da>] slab_cpuup_callback+0x3a/0x192
>        [<c01452cd>] notifier_call_chain+0x2d/0x70
>        [<c01453c9>] __raw_notifier_call_chain+0x19/0x20
>        [<c03e334e>] _cpu_up+0x5e/0x105
>        [<c03e3459>] cpu_up+0x64/0x7b
>        [<c055430f>] kernel_init+0xb6/0x173
>        [<c01042ff>] kernel_thread_helper+0x7/0x18
>        [<ffffffff>] 0xffffffff
> 
> -> #0 (&cpu_hotplug.lock){--..}:
>        [<c0151691>] __lock_acquire+0x1061/0x17f0
>        [<c0151e7e>] lock_acquire+0x5e/0x80
>        [<c03e5be4>] mutex_lock_nested+0xb4/0x2c0
>        [<c012cd32>] get_online_cpus+0x22/0x40
>        [<c013d2c8>] schedule_on_each_cpu+0x38/0xe0
>        [<c018115d>] lru_add_drain_all+0xd/0x10
>        [<c018b486>] __mlock_vma_pages_range+0x36/0x1f0
>        [<c018b74f>] mlock_fixup+0x10f/0x210
>        [<c018b8db>] do_mlockall+0x8b/0xa0
>        [<c018b9c1>] sys_mlockall+0x91/0xd0
>        [<c010346d>] sysenter_do_call+0x12/0x31
>        [<ffffffff>] 0xffffffff
> 
> other info that might help us debug this:
> 
> 1 lock held by cryptsetup/2554:
>  #0:  (&mm->mmap_sem){----}, at: [<c018b98e>] sys_mlockall+0x5e/0xd0
> 
> stack backtrace:
> Pid: 2554, comm: cryptsetup Not tainted 2.6.28-rc3-next-20081106-mad-1 #2
> Call Trace:
>  [<c03e45d2>] ? printk+0x18/0x1e
>  [<c0150213>] print_circular_bug_tail+0xc3/0xd0
>  [<c014ff4b>] ? print_circular_bug_entry+0x4b/0x50
>  [<c0151691>] __lock_acquire+0x1061/0x17f0
>  [<c0151e7e>] lock_acquire+0x5e/0x80
>  [<c012cd32>] ? get_online_cpus+0x22/0x40
>  [<c03e5be4>] mutex_lock_nested+0xb4/0x2c0
>  [<c012cd32>] ? get_online_cpus+0x22/0x40
>  [<c012cd32>] ? get_online_cpus+0x22/0x40
>  [<c019f391>] ? __percpu_alloc_mask+0xa1/0x100
>  [<c012cd32>] get_online_cpus+0x22/0x40
>  [<c013d2c8>] schedule_on_each_cpu+0x38/0xe0
>  [<c0180ed0>] ? lru_add_drain_per_cpu+0x0/0x10
>  [<c018115d>] lru_add_drain_all+0xd/0x10
>  [<c018b486>] __mlock_vma_pages_range+0x36/0x1f0
>  [<c022c109>] ? avc_has_perm+0x59/0x70
>  [<c018b74f>] mlock_fixup+0x10f/0x210
>  [<c018b8db>] do_mlockall+0x8b/0xa0
>  [<c018b9c1>] sys_mlockall+0x91/0xd0
>  [<c010346d>] sysenter_do_call+0x12/0x31

-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au
http://www.canb.auug.org.au/~sfr/

[-- Attachment #2: Type: application/pgp-signature, Size: 197 bytes --]

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [linux-next] possible circular locking dependency for cryptsetup
  2008-11-07  9:37 ` Stephen Rothwell
@ 2008-11-07 12:40   ` Herbert Xu
  0 siblings, 0 replies; 4+ messages in thread
From: Herbert Xu @ 2008-11-07 12:40 UTC (permalink / raw)
  To: Stephen Rothwell; +Cc: Markus Dahms, linux-next, linux-kernel

On Fri, Nov 07, 2008 at 08:37:50PM +1100, Stephen Rothwell wrote:
>
> > =======================================================
> > [ INFO: possible circular locking dependency detected ]
> > 2.6.28-rc3-next-20081106-mad-1 #2
> > -------------------------------------------------------
> > cryptsetup/2554 is trying to acquire lock:
> >  (&cpu_hotplug.lock){--..}, at: [<c012cd32>] get_online_cpus+0x22/0x40
> > 
> > but task is already holding lock:
> >  (&mm->mmap_sem){----}, at: [<c018b98e>] sys_mlockall+0x5e/0xd0

I don't see anything in the back-trace that's crypto related
or even dm related.  Therefore I suspect the problem is in core
sysfs, mm or CPU hot plugging.

Cheers,
-- 
Visit Openswan at http://www.openswan.org/
Email: Herbert Xu ~{PmV>HI~} <herbert@gondor.apana.org.au>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt

^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2008-11-07 12:40 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2008-11-07  8:32 [linux-next] possible circular locking dependency for cryptsetup Markus Dahms
2008-11-07  8:47 ` KOSAKI Motohiro
2008-11-07  9:37 ` Stephen Rothwell
2008-11-07 12:40   ` Herbert Xu

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).