2.6.19-rc3-git7: BUG: mutex warning sysfs_dir_open
I don't know if the kernel is getting better at detecting its own bugs, or if 2.6.19 is shaping up to be the flakiest kernel in recent memory.
I found this in my syslog just now:
kernel: BUG: warning at kernel/mutex.c:132/__mutex_lock_common() kernel: [<c0104088>] dump_trace+0x1c8/0x200 kernel: [<c01040da>] show_trace_log_lvl+0x1a/0x30 kernel: [<c0104832>] show_trace+0x12/0x20 kernel: [<c0104999>] dump_stack+0x19/0x20 kernel: [<c035d8ef>] __mutex_lock_slowpath+0x20f/0x260 kernel: [<c035d961>] mutex_lock+0x21/0x30 kernel: [<c01a0286>] sysfs_dir_open+0x26/0x60 kernel: [<c0164925>] __dentry_open+0xb5/0x1e0 kernel: [<c0164af5>] nameidata_to_filp+0x35/0x40 kernel: [<c0164b49>] do_filp_open+0x49/0x50 kernel: [<c0164b97>] do_sys_open+0x47/0xe0 kernel: [<c0164c6c>] sys_open+0x1c/0x20 kernel: [<c0102f4d>] sysenter_past_esp+0x56/0x8d kernel: [<b7fc5410>] 0xb7fc5410 kernel: ======================= kernel: BUG: unable to handle kernel paging request at virtual address 6e726574 kernel: printing eip: kernel: c022a8fc kernel: *pde = 00000000 kernel: Oops: 0000 [#1] kernel: SMP kernel: Modules linked in: button battery ac dm_mod snd_hda_intel snd_hda_codec snd_pcm snd_timer snd soundcore snd_page_alloc r1000 ahci edd fan thermal processor kernel: CPU: 0 kernel: EIP: 0060:[<c022a8fc>] Not tainted VLI kernel: EFLAGS: 00010046 (2.6.19-rc3-git7-ml #9) kernel: EIP is at __list_add+0x1c/0x80 kernel: eax: 6e726574 ebx: dfe39120 ecx: dfe39120 edx: 6e726574 kernel: esi: 6e726574 edi: f7861ea4 ebp: f7861e88 esp: f7861e70 kernel: ds: 007b es: 007b ss: 0068 kernel: Process hald (pid: 3360, ti=f7860000 task=f7974af0 task.ti=f7860000) kernel: Stack: 00000000 00000002 00000001 dfe39100 00000246 f7974af0 f7861ec4 c035d76e kernel: 00000000 00000002 c035d961 c0362479 dfe39120 f7861ea4 f7861ea4 11111111 kernel: dfe39100 f7861ea4 dfe39100 f753ac00 dfe3a8a4 f7861ed0 c035d961 f78c0b34 kernel: Call Trace: kernel: [<c035d76e>] __mutex_lock_slowpath+0x8e/0x260 kernel: [<c035d961>] mutex_lock+0x21/0x30 kernel: [<c01a0286>] sysfs_dir_open+0x26/0x60 kernel: [<c0164925>] __dentry_open+0xb5/0x1e0 kernel: [<c0164af5>] nameidata_to_filp+0x35/0x40 kernel: [<c0164b49>] do_filp_open+0x49/0x50 kernel: [<c0164b97>] do_sys_open+0x47/0xe0 kernel: [<c0164c6c>] sys_open+0x1c/0x20 kernel: [<c0102f4d>] sysenter_past_esp+0x56/0x8d kernel: [<b7fc5410>] 0xb7fc5410 kernel: ======================= kernel: Code: 13 eb a7 8d b6 00 00 00 00 8d bf 00 00 00 00 55 89 e5 83 ec 18 89 5d f4 89 cb 89 75 f8 89 d6 89 7d fc 89 c7 8b 41 04 39 d0 75 1d <8b> 06 39 c3 75 35 89 7b 04 89 1f 8b 5d f4 89 77 04 89 3e 8b 75 kernel: EIP: [<c022a8fc>] __list_add+0x1c/0x80 SS:ESP 0068:f7861e70 kernel: <3>BUG: sleeping function called from invalid context at kernel/rwsem.c:20 kernel: in_atomic():0, irqs_disabled():1 kernel: [<c0104088>] dump_trace+0x1c8/0x200 kernel: [<c01040da>] show_trace_log_lvl+0x1a/0x30 kernel: [<c0104832>] show_trace+0x12/0x20 kernel: [<c0104999>] dump_stack+0x19/0x20 kernel: [<c01166f8>] __might_sleep+0xa8/0xb0 kernel: [<c0134468>] down_read+0x18/0x60 kernel: [<c0143869>] acct_collect+0x39/0x150 kernel: [<c0120715>] do_exit+0xf5/0x8a0 kernel: [<c0104725>] die+0x2f5/0x300 kernel: [<c0114c0f>] do_page_fault+0x1ff/0x5f0 kernel: [<c035f0a9>] error_code+0x39/0x40 kernel: [<c022a8fc>] __list_add+0x1c/0x80 kernel: [<c035d76e>] __mutex_lock_slowpath+0x8e/0x260 kernel: [<c035d961>] mutex_lock+0x21/0x30 kernel: [<c01a0286>] sysfs_dir_open+0x26/0x60 kernel: [<c0164925>] __dentry_open+0xb5/0x1e0 kernel: [<c0164af5>] nameidata_to_filp+0x35/0x40 kernel: [<c0164b49>] do_filp_open+0x49/0x50 kernel: [<c0164b97>] do_sys_open+0x47/0xe0 kernel: [<c0164c6c>] sys_open+0x1c/0x20 kernel: [<c0102f4d>] sysenter_past_esp+0x56/0x8d kernel: [<b7fc5410>] 0xb7fc5410 kernel: ======================= silvy:~#
Quoted 4 lines · Mark Lord
I found this in my syslog just now:
kernel: BUG: warning at kernel/mutex.c:132/__mutex_lock_common()
That's the "spin_lock_mutex()" debugging (kernel/mutex-debug.h), and it's one of either
DEBUG_LOCKS_WARN_ON(in_interrupt()); or DEBUG_LOCKS_WARN_ON(l->magic != l);
where the much more likely one is the "l->magic" one triggering, because your bug that follows is:
Quoted 1 line
kernel: BUG: unable to handle kernel paging request at virtual address
6e726574
64726574 is "tern", for whatever that's worth, and that EIP is in "__list_add()", which comes from list_add_tail(), and so it looks like the mutex "lock->lock_list" is also totally buggered (which would certainly fit with the lock "magic" field being crud too).
Quoted 4 lines
kernel: Call Trace: kernel: [<c035d76e>] __mutex_lock_slowpath+0x8e/0x260 kernel: [<c035d961>] mutex_lock+0x21/0x30 kernel: [<c01a0286>] sysfs_dir_open+0x26/0x60
This would be
mutex_lock(&dentry->d_inode->i_mutex);
in sysfs_dir_open(), so you would seem to have an inode that is overwritten by crud (or it could be a dentry with a bogus d_inode pointer, of course).
There's a ton of stuff with "tern" in it, mostly "Internal", so sadly even the nice string pattern doesn't seem to be all that helpful.
Is this somewhat reproducible?
HOWEVER, google does find a clue. We've had reports of something somewhat similar before: googling for "__mutex_lock_slowpath" and "sysfs_dir_open" shows for example
http://lkml.org/lkml/2006/8/18/72
where the thread ended up first blaming sound, but then deciding that maybe it was DRM-related. However, you don't seem to have any AGP or DRM support at all, so maybe it really _is_ a sound problem.
Mark - can you verify that you don't have any DRM support in your kernel?
Frederic, Takashi, did that bug-report ever get resolved?
Linus
Quoted 2 lines · Linus Torvalds
Is this somewhat reproducible?
Just the one occurance of it, with rc3-git7.
I've now moved on to -rc4.
Quoted 7 lines
HOWEVER, google does find a clue. We've had reports of something somewhat similar before: googling for "
__mutex_lock_slowpath" and "sysfs_dir_open" shows for examplehttp://lkml.org/lkml/2006/8/18/72where the thread ended up first blaming sound, but then deciding that maybe it was DRM-related. However, you don't seem to have any AGP or DRM support at all, so maybe it really _is_ a sound problem.
Mark - can you verify that you don't have any DRM support in your kernel?
# CONFIG_DRM is not set
(but I'm about to change that!).
Cheers
At Tue, 31 Oct 2006 09:53:35 -0800 (PST),
Quoted 50 lines · Linus Torvalds
Quoted 4 lines · Mark Lord
I found this in my syslog just now:
kernel: BUG: warning at kernel/mutex.c:132/__mutex_lock_common()
That's the "
spin_lock_mutex()" debugging (kernel/mutex-debug.h), and it's one of eitherDEBUG_LOCKS_WARN_ON(in_interrupt()); or DEBUG_LOCKS_WARN_ON(l->magic != l);where the much more likely one is the "
l->magic" one triggering, because your bug that follows is:Quoted 1 line
kernel: BUG: unable to handle kernel paging request at virtual address
6e72657464726574 is "tern", for whatever that's worth, and that EIP is in "
__list_add()", which comes fromlist_add_tail(), and so it looks like the mutex "lock->lock_list" is also totally buggered (which would certainly fit with the lock "magic" field being crud too).Quoted 4 lines
kernel: Call Trace: kernel: [<c035d76e>] __mutex_lock_slowpath+0x8e/0x260 kernel: [<c035d961>] mutex_lock+0x21/0x30 kernel: [<c01a0286>] sysfs_dir_open+0x26/0x60This would be
mutex_lock(&dentry->d_inode->i_mutex);in
sysfs_dir_open(), so you would seem to have an inode that is overwritten by crud (or it could be a dentry with a bogusd_inodepointer, of course).There's a ton of stuff with "tern" in it, mostly "Internal", so sadly even the nice string pattern doesn't seem to be all that helpful.
Is this somewhat reproducible?
HOWEVER, google does find a clue. We've had reports of something somewhat similar before: googling for "
__mutex_lock_slowpath" and "sysfs_dir_open" shows for examplehttp://lkml.org/lkml/2006/8/18/72
Quoted 7 lines
where the thread ended up first blaming sound, but then deciding that maybe it was DRM-related. However, you don't seem to have any AGP or DRM support at all, so maybe it really _is_ a sound problem.
Mark - can you verify that you don't have any DRM support in your kernel?
Frederic, Takashi, did that bug-report ever get resolved?
IIRC, in Frederik's case, it turned out to be irrelevant from the sound stuff since the bug occured even before loading ALSA modules.
Of course, it still _might_ be a sound problem :)
It'd be nice if we can see what sysfs entry triggers bugs...
Takashi