hi all,
using 2.6.31.1 with CONFIG_PREEMPT=y, i got the following call traces
from ext4 related code, when doing some io-heavy tasks:
[30477.126201] INFO: task gconfd-2:2296 blocked for more than 120 seconds.
[30477.126205] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[30477.126208] gconfd-2 D ffff8800280436e8 0 2296 1 0x00000000
[30477.126214] BUG: using smp_processor_id() in preemptible [00000000] code: khungtaskd/72
[30477.126221] caller is show_stack_log_lvl+0x32/0x170
[30477.126224] Pid: 72, comm: khungtaskd Tainted: P 2.6.31.1 #29
[30477.126227] Call Trace:
[30477.126234] [<ffffffff81266881>] debug_smp_processor_id+0xd1/0xf0
[30477.126238] [<ffffffff81014b62>] show_stack_log_lvl+0x32/0x170
[30477.126242] [<ffffffff814f57c0>] ? printk+0x3c/0x44
[30477.126246] [<ffffffff81015e17>] show_stack+0x17/0x20
[30477.126251] [<ffffffff81045b52>] sched_show_task+0x92/0x100
[30477.126256] [<ffffffff810ab66d>] watchdog+0x1fd/0x240
[30477.126260] [<ffffffff810ab470>] ? watchdog+0x0/0x240
[30477.126264] [<ffffffff81071216>] kthread+0xa6/0xb0
[30477.126268] [<ffffffff8101303a>] child_rip+0xa/0x20
[30477.126272] [<ffffffff81071170>] ? kthread+0x0/0xb0
[30477.126275] [<ffffffff81013030>] ? child_rip+0x0/0x20
[30477.126278] ffff8803331afc98 0000000000000086 0000000000000000 0000000000000000
[30477.126282] ffff8803331afc58 0000000000000000 ffff8803331afc58 00000001002ddc13
[30477.126287] 000000000000f6e8 00000000000147c0 ffff880331d3acb8 ffff880331d3acb8
[30477.126291] Call Trace:
[30477.126298] [<ffffffff811de775>] jbd2_log_wait_commit+0xb5/0x130
[30477.126303] [<ffffffff81071660>] ? autoremove_wake_function+0x0/0x40
[30477.126307] [<ffffffff81074b51>] ? ktime_get_ts+0x51/0x70
[30477.126311] [<ffffffff811d6300>] jbd2_journal_stop+0x1e0/0x2b0
[30477.126315] [<ffffffff811d6cbe>] ? jbd2_journal_start+0xae/0x100
[30477.126319] [<ffffffff811d6d32>] jbd2_journal_force_commit+0x22/0x30
[30477.126325] [<ffffffff811ae9f2>] ext4_force_commit+0x22/0x40
[30477.126329] [<ffffffff8119c004>] ext4_write_inode+0x34/0x50
[30477.126335] [<ffffffff8112b79c>] writeback_single_inode+0x2bc/0x420
[30477.126339] [<ffffffff8112b92e>] sync_inode+0x2e/0x50
[30477.126343] [<ffffffff81199220>] ext4_sync_file+0x100/0x1c0
[30477.126348] [<ffffffff810db128>] ? do_writepages+0x28/0x50
[30477.126352] [<ffffffff8112faf6>] vfs_fsync+0x86/0xf0
[30477.126356] [<ffffffff8112fb99>] do_fsync+0x39/0x60
[30477.126360] [<ffffffff8112fbeb>] sys_fsync+0xb/0x10
[30477.126365] [<ffffffff81011f42>] system_call_fastpath+0x16/0x1b
[30477.126374] INFO: task firefox:4898 blocked for more than 120 seconds.
[30477.126376] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[30477.126378] firefox D ffff8800280d46e8 0 4898 1 0x00000000
[30477.126383] BUG: using smp_processor_id() in preemptible [00000000] code: khungtaskd/72
[30477.126386] caller is show_stack_log_lvl+0x32/0x170
[30477.126389] Pid: 72, comm: khungtaskd Tainted: P 2.6.31.1 #29
[30477.126391] Call Trace:
[30477.126394] [<ffffffff81266881>] debug_smp_processor_id+0xd1/0xf0
[30477.126398] [<ffffffff81014b62>] show_stack_log_lvl+0x32/0x170
[30477.126401] [<ffffffff814f57c0>] ? printk+0x3c/0x44
[30477.126405] [<ffffffff81015e17>] show_stack+0x17/0x20
[30477.126408] [<ffffffff81045b52>] sched_show_task+0x92/0x100
[30477.126412] [<ffffffff810ab66d>] watchdog+0x1fd/0x240
[30477.126415] [<ffffffff810ab470>] ? watchdog+0x0/0x240
[30477.126419] [<ffffffff81071216>] kthread+0xa6/0xb0
[30477.126422] [<ffffffff8101303a>] child_rip+0xa/0x20
[30477.126426] [<ffffffff81071170>] ? kthread+0x0/0xb0
[30477.126429] [<ffffffff81013030>] ? child_rip+0x0/0x20
[30477.126431] ffff880309409ac8 0000000000000082 0000000000000000 0000000000000286
[30477.126435] ffff88012547fc78 0000000000000005 ffff8803309cf000 00000001002dc21f
[30477.126440] 000000000000f6e8 00000000000147c0 ffff88030954c438 ffff88030954c438
[30477.126444] Call Trace:
[30477.126448] [<ffffffff811d7255>] do_get_write_access+0x365/0x6d0
[30477.126453] [<ffffffff810716a0>] ? wake_bit_function+0x0/0x40
[30477.126456] [<ffffffff811e1437>] ? jbd2_journal_add_journal_head+0x117/0x220
[30477.126461] [<ffffffff811d77dc>] jbd2_journal_get_write_access+0x2c/0x50
[30477.126466] [<ffffffff811bf700>] __ext4_journal_get_write_access+0x30/0x70
[30477.126470] [<ffffffff811a3f31>] ext4_orphan_add+0xc1/0x1d0
[30477.126474] [<ffffffff811d6cbe>] ? jbd2_journal_start+0xae/0x100
[30477.126479] [<ffffffff814f8737>] ? _spin_unlock_irq+0x17/0x40
[30477.126483] [<ffffffff8119dd9a>] ext4_setattr+0x22a/0x360
[30477.126488] [<ffffffff811229c3>] notify_change+0x163/0x340
[30477.126493] [<ffffffff8110aae9>] do_truncate+0x69/0x90
[30477.126496] [<ffffffff81114edc>] ? get_write_access+0x3c/0x60
[30477.126500] [<ffffffff81115616>] may_open+0x1d6/0x200
[30477.126503] [<ffffffff811184e8>] do_filp_open+0x208/0xb50
[30477.126508] [<ffffffff814f81f6>] ? _spin_lock+0x16/0x40
[30477.126511] [<ffffffff811238f2>] ? alloc_fd+0xf2/0x140
[30477.126515] [<ffffffff81109b84>] do_sys_open+0x64/0x130
[30477.126518] [<ffffffff81109c7b>] sys_open+0x1b/0x20
[30477.126522] [<ffffffff81011f42>] system_call_fastpath+0x16/0x1b
[30477.126530] INFO: task thunderbird-bin:12190 blocked for more than 120 seconds.
[30477.126532] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[30477.126534] thunderbird-b D ffff8800280436e8 0 12190 12186 0x00000000
[30477.126539] BUG: using smp_processor_id() in preemptible [00000000] code: khungtaskd/72
[30477.126542] caller is show_stack_log_lvl+0x32/0x170
[30477.126545] Pid: 72, comm: khungtaskd Tainted: P 2.6.31.1 #29
[30477.126546] Call Trace:
[30477.126550] [<ffffffff81266881>] debug_smp_processor_id+0xd1/0xf0
[30477.126553] [<ffffffff81014b62>] show_stack_log_lvl+0x32/0x170
[30477.126556] [<ffffffff814f57c0>] ? printk+0x3c/0x44
[30477.126560] [<ffffffff81015e17>] show_stack+0x17/0x20
[30477.126563] [<ffffffff81045b52>] sched_show_task+0x92/0x100
[30477.126567] [<ffffffff810ab66d>] watchdog+0x1fd/0x240
[30477.126570] [<ffffffff810ab470>] ? watchdog+0x0/0x240
[30477.126574] [<ffffffff81071216>] kthread+0xa6/0xb0
[30477.126577] [<ffffffff8101303a>] child_rip+0xa/0x20
[30477.126581] [<ffffffff81071170>] ? kthread+0x0/0xb0
[30477.126584] [<ffffffff81013030>] ? child_rip+0x0/0x20
[30477.126586] ffff8803031c3a68 0000000000000082 0000000000000000 ffffea000b023168
[30477.126590] 00000000001800f2 0000000000000000 ffff880333713800 00000001002dba15
[30477.126594] 000000000000f6e8 00000000000147c0 ffff8803031c0938 ffff8803031c0938
[30477.126598] Call Trace:
[30477.126603] [<ffffffff811d7255>] do_get_write_access+0x365/0x6d0
[30477.126607] [<ffffffff810716a0>] ? wake_bit_function+0x0/0x40
[30477.126611] [<ffffffff811e1437>] ? jbd2_journal_add_journal_head+0x117/0x220
[30477.126615] [<ffffffff811d77dc>] jbd2_journal_get_write_access+0x2c/0x50
[30477.126619] [<ffffffff811bf700>] __ext4_journal_get_write_access+0x30/0x70
[30477.126624] [<ffffffff8119d7cb>] ext4_reserve_inode_write+0x7b/0xa0
[30477.126628] [<ffffffff8119d82c>] ext4_mark_inode_dirty+0x3c/0x1c0
[30477.126632] [<ffffffff810d7cec>] ? T.1235+0xac/0xe0
[30477.126636] [<ffffffff8119db22>] ext4_dirty_inode+0x62/0xb0
[30477.126640] [<ffffffff8112c346>] __mark_inode_dirty+0x36/0x130
[30477.126644] [<ffffffff81120ae4>] touch_atime+0xf4/0x150
[30477.126649] [<ffffffff810d4590>] generic_file_aio_read+0x2a0/0x620
[30477.126654] [<ffffffff8110b9f2>] do_sync_read+0xf2/0x130
[30477.126658] [<ffffffff810dcc69>] ? __lru_cache_add+0x79/0xd0
[30477.126662] [<ffffffff814f8630>] ? _spin_unlock+0x10/0x40
[30477.126665] [<ffffffff81071660>] ? autoremove_wake_function+0x0/0x40
[30477.126670] [<ffffffff8110fd4f>] ? cp_new_stat+0xdf/0xf0
[30477.126674] [<ffffffff8120d991>] ? security_file_permission+0x11/0x20
[30477.126678] [<ffffffff8110be55>] vfs_read+0xb5/0x1a0
[30477.126682] [<ffffffff8110cedd>] ? fget_light+0x9d/0xb0
[30477.126685] [<ffffffff8110c49c>] sys_read+0x4c/0x80
[30477.126689] [<ffffffff81011f42>] system_call_fastpath+0x16/0x1b
[30477.126693] INFO: task thunderbird-bin:12196 blocked for more than 120 seconds.
[30477.126695] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[30477.126697] thunderbird-b D ffff8800280436e8 0 12196 12186 0x00000000
[30477.126702] BUG: using smp_processor_id() in preemptible [00000000] code: khungtaskd/72
[30477.126705] caller is show_stack_log_lvl+0x32/0x170
[30477.126708] Pid: 72, comm: khungtaskd Tainted: P 2.6.31.1 #29
[30477.126710] Call Trace:
[30477.126713] [<ffffffff81266881>] debug_smp_processor_id+0xd1/0xf0
[30477.126716] [<ffffffff81014b62>] show_stack_log_lvl+0x32/0x170
[30477.126720] [<ffffffff814f57c0>] ? printk+0x3c/0x44
[30477.126723] [<ffffffff81015e17>] show_stack+0x17/0x20
[30477.126726] [<ffffffff81045b52>] sched_show_task+0x92/0x100
[30477.126730] [<ffffffff810ab66d>] watchdog+0x1fd/0x240
[30477.126734] [<ffffffff810ab470>] ? watchdog+0x0/0x240
[30477.126737] [<ffffffff81071216>] kthread+0xa6/0xb0
[30477.126740] [<ffffffff8101303a>] child_rip+0xa/0x20
[30477.126744] [<ffffffff81071170>] ? kthread+0x0/0xb0
[30477.126747] [<ffffffff81013030>] ? child_rip+0x0/0x20
[30477.126749] ffff8803031a1d68 0000000000000082 0000000000000000 ffff8803031a1e30
[30477.126753] 0000000000000000 0000000000000000 ffff8803031a1d78 00000001002dba15
[30477.126757] 000000000000f6e8 00000000000147c0 ffff8803031a4978 ffff8803031a4978
[30477.126761] Call Trace:
[30477.126766] [<ffffffff811de775>] jbd2_log_wait_commit+0xb5/0x130
[30477.126770] [<ffffffff81071660>] ? autoremove_wake_function+0x0/0x40
[30477.126774] [<ffffffff811d6300>] jbd2_journal_stop+0x1e0/0x2b0
[30477.126778] [<ffffffff811d6cbe>] ? jbd2_journal_start+0xae/0x100
[30477.126782] [<ffffffff811d6d32>] jbd2_journal_force_commit+0x22/0x30
[30477.126786] [<ffffffff811ae9f2>] ext4_force_commit+0x22/0x40
[30477.126790] [<ffffffff811991a3>] ext4_sync_file+0x83/0x1c0
[30477.126856] [<ffffffff810db13d>] ? do_writepages+0x3d/0x50
[30477.126860] [<ffffffff810d3103>] ? __filemap_fdatawrite_range+0x53/0x60
[30477.126865] [<ffffffff8112faf6>] vfs_fsync+0x86/0xf0
[30477.126869] [<ffffffff8112fb99>] do_fsync+0x39/0x60
[30477.126873] [<ffffffff8112fbce>] sys_fdatasync+0xe/0x20
[30477.126877] [<ffffffff81011f42>] system_call_fastpath+0x16/0x1b
[30477.126885] INFO: task java:14629 blocked for more than 120 seconds.
[30477.126887] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[30477.126889] java D ffff8800280436e8 0 14629 1 0x00000004
[30477.126894] BUG: using smp_processor_id() in preemptible [00000000] code: khungtaskd/72
[30477.126898] caller is show_stack_log_lvl+0x32/0x170
[30477.126900] Pid: 72, comm: khungtaskd Tainted: P 2.6.31.1 #29
[30477.126902] Call Trace:
[30477.126906] [<ffffffff81266881>] debug_smp_processor_id+0xd1/0xf0
[30477.126909] [<ffffffff81014b62>] show_stack_log_lvl+0x32/0x170
[30477.126912] [<ffffffff814f57c0>] ? printk+0x3c/0x44
[30477.126915] [<ffffffff81015e17>] show_stack+0x17/0x20
[30477.126919] [<ffffffff81045b52>] sched_show_task+0x92/0x100
[30477.126922] [<ffffffff810ab66d>] watchdog+0x1fd/0x240
[30477.126926] [<ffffffff810ab470>] ? watchdog+0x0/0x240
[30477.126929] [<ffffffff81071216>] kthread+0xa6/0xb0
[30477.126933] [<ffffffff8101303a>] child_rip+0xa/0x20
[30477.126937] [<ffffffff81071170>] ? kthread+0x0/0xb0
[30477.126940] [<ffffffff81013030>] ? child_rip+0x0/0x20
[30477.126942] ffff88026ec81c68 0000000000000086 0000000000000000 ffffffff811d7fb2
[30477.126946] ffff88026ec81cc8 0000000000000000 ffff88026ec81c18 00000001002dc64a
[30477.126950] 000000000000f6e8 00000000000147c0 ffff8801571a88b8 ffff8801571a88b8
[30477.126954] Call Trace:
[30477.126957] [<ffffffff811d7fb2>] ? jbd2_journal_dirty_metadata+0xb2/0x1e0
[30477.126962] [<ffffffff814f6f9b>] __mutex_lock_common+0x12b/0x270
[30477.126966] [<ffffffff8119d860>] ? ext4_mark_inode_dirty+0x70/0x1c0
[30477.126970] [<ffffffff814f70f4>] __mutex_lock_slowpath+0x14/0x20
[30477.126973] [<ffffffff814f6d8e>] mutex_lock+0x1e/0x40
[30477.126977] [<ffffffff811a3cf3>] ext4_orphan_del+0x53/0x1d0
[30477.126981] [<ffffffff814f81f6>] ? _spin_lock+0x16/0x40
[30477.126984] [<ffffffff814f8630>] ? _spin_unlock+0x10/0x40
[30477.126988] [<ffffffff8112c3a8>] ? __mark_inode_dirty+0x98/0x130
[30477.126992] [<ffffffff8119de9a>] ext4_setattr+0x32a/0x360
[30477.126996] [<ffffffff811229c3>] notify_change+0x163/0x340
[30477.126999] [<ffffffff81115770>] ? putname+0x30/0x50
[30477.127004] [<ffffffff81130087>] utimes_common+0xd7/0x1b0
[30477.127008] [<ffffffff811301f0>] do_utimes+0x90/0xf0
[30477.127012] [<ffffffff8113027b>] sys_futimesat+0x2b/0xb0
[30477.127016] [<ffffffff81130314>] sys_utimes+0x14/0x20
[30477.127020] [<ffffffff81011f42>] system_call_fastpath+0x16/0x1b
hth, tim
--
[email protected]
http://tim.klingt.org
You don't have to call it music if the term shocks you.
John Cage
Tim,
I have similar crash, but probably different cause-vector, posted on this list, fedora bugzilla, rt-users but no response so far from anyone. Did you find any work around? I tried looking at smp code for hints on problem it looks like in my case it could be trying to reschedule work on different CPU at time of crash. However, I admit that I could be way, way off on that cause. In my case, if I hammer my network interface with packets while another app is reading packets, my host locks up and clients are reporting dest unreachables.
I don't think I have the problem in non-rt kernel. Have you tried booting without smp support?
http://lkml.org/lkml/headers/2009/11/23/548
-Bob
> -----Original Message-----
> From: [email protected] [mailto:linux-kernel-
> [email protected]] On Behalf Of Tim Blechmann
> Sent: Monday, October 05, 2009 5:06 AM
> To: [email protected]; [email protected]
> Subject: ext4: smp_processor_id() in preemptible
>
> hi all,
>
> using 2.6.31.1 with CONFIG_PREEMPT=y, i got the following call traces
> from ext4 related code, when doing some io-heavy tasks:
>
> [30477.126201] INFO: task gconfd-2:2296 blocked for more than 120
> seconds.
> [30477.126205] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs"
> disables this message.
> [30477.126208] gconfd-2 D ffff8800280436e8 0 2296 1
> 0x00000000
> [30477.126214] BUG: using smp_processor_id() in preemptible [00000000]
> code: khungtaskd/72 [30477.126221] caller is
> show_stack_log_lvl+0x32/0x170
> [30477.126224] Pid: 72, comm: khungtaskd Tainted: P 2.6.31.1
> #29
> [30477.126227] Call Trace:
> [30477.126234] [<ffffffff81266881>] debug_smp_processor_id+0xd1/0xf0
> [30477.126238] [<ffffffff81014b62>] show_stack_log_lvl+0x32/0x170
> [30477.126242] [<ffffffff814f57c0>] ? printk+0x3c/0x44 [30477.126246]
> [<ffffffff81015e17>] show_stack+0x17/0x20 [30477.126251]
> [<ffffffff81045b52>] sched_show_task+0x92/0x100 [30477.126256]
> [<ffffffff810ab66d>] watchdog+0x1fd/0x240 [30477.126260]
> [<ffffffff810ab470>] ? watchdog+0x0/0x240 [30477.126264]
> [<ffffffff81071216>] kthread+0xa6/0xb0 [30477.126268]
> [<ffffffff8101303a>] child_rip+0xa/0x20 [30477.126272]
> [<ffffffff81071170>] ? kthread+0x0/0xb0 [30477.126275]
> [<ffffffff81013030>] ? child_rip+0x0/0x20 [30477.126278]
> I have similar crash, but probably different cause-vector, posted on this list, fedora bugzilla, rt-users but no response so far from anyone. Did you find any work around? I tried looking at smp code for hints on problem it looks like in my case it could be trying to reschedule work on different CPU at time of crash. However, I admit that I could be way, way off on that cause. In my case, if I hammer my network interface with packets while another app is reading packets, my host locks up and clients are reporting dest unreachables.
>
> I don't think I have the problem in non-rt kernel. Have you tried booting without smp support?
i haven't tried to reproduce it since then ... it occurred, when i was
doing some heavy io tasks ...
tim
--
[email protected]
http://tim.klingt.org
Nothing exists until or unless it is observed. An artist is making
something exist by observing it. And his hope for other people is that
they will also make it exist by observing it. I call it 'creative
observation.' Creative viewing.
William S. Burroughs
> -----Original Message-----
> From: Tim Blechmann [mailto:[email protected]]
> Sent: Friday, November 27, 2009 4:33 AM
> To: Leyendecker, Robert
> Cc: [email protected]; [email protected]
> Subject: Re: ext4: smp_processor_id() in preemptible
>
> > I have similar crash, but probably different cause-vector, posted on
> this list, fedora bugzilla, rt-users but no response so far from
> anyone. Did you find any work around? I tried looking at smp code for
> hints on problem it looks like in my case it could be trying to
> reschedule work on different CPU at time of crash. However, I admit
> that I could be way, way off on that cause. In my case, if I hammer my
> network interface with packets while another app is reading packets, my
> host locks up and clients are reporting dest unreachables.
> >
> > I don't think I have the problem in non-rt kernel. Have you tried
> booting without smp support?
>
> i haven't tried to reproduce it since then ... it occurred, when i was
> doing some heavy io tasks ...
>
> tim
>
> --
> [email protected]
> http://tim.klingt.org
>
> Nothing exists until or unless it is observed. An artist is making
> something exist by observing it. And his hope for other people is that
> they will also make it exist by observing it. I call it 'creative
> observation.' Creative viewing.
> William S. Burroughs
OK - thanks. I'm arriving here via heavy network io, relatively easy to repro in my case.
-Bob