http://oss.sgi.com/bugzilla/show_bug.cgi?id=852
--- Comment #1 from tneumann@xxxxxxxxxxxxxxxxxxxxx 2009-10-06 21:32:35 CST ---
The same happens with a current 2.6.32 rc1 kernel. I can semi-reproducible
(i.e., most of the time) cause a deadlock with a certain workload. I include
the dmesg output below, it seems to deadlock with fallocate. Is there anything
else I should do to debug the problem?
uname -a
Linux infno5602 2.6.32-020632rc1-generic #020632rc1 SMP Thu Oct 1 01:45:15 UTC
2009 x86_64 GNU/Linux
dmesg excerpt
[36960.700132] INFO: task xfsconvertd/0:411 blocked for more than 120 seconds.
[36960.700138] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this
message.
[36960.700143] xfsconvertd/0 D 0000000000000000 0 411 2 0x00000000
[36960.700154] ffff88007b21d3e0 0000000000000046 ffff88007d4e8c40
ffff88007b21dfd8
[36960.700163] ffff88007adfdb40 0000000000015980 0000000000015980
ffff88007b21dfd8
[36960.700172] 0000000000015980 ffff88007b21dfd8 0000000000015980
ffff88007adfdf00
[36960.700180] Call Trace:
[36960.700197] [<ffffffff81526162>] io_schedule+0x42/0x60
[36960.700206] [<ffffffff810df6d5>] sync_page+0x35/0x50
[36960.700213] [<ffffffff815268e5>] __wait_on_bit+0x55/0x80
[36960.700220] [<ffffffff810df6a0>] ? sync_page+0x0/0x50
[36960.700227] [<ffffffff810df7f0>] wait_on_page_bit+0x70/0x80
[36960.700236] [<ffffffff8107afb0>] ? wake_bit_function+0x0/0x30
[36960.700245] [<ffffffff810ecce8>] shrink_page_list+0x3d8/0x550
[36960.700253] [<ffffffff8107b0de>] ? finish_wait+0x5e/0x80
[36960.700262] [<ffffffff810f85da>] ? congestion_wait+0x7a/0x90
[36960.700269] [<ffffffff8107af70>] ? autoremove_wake_function+0x0/0x40
[36960.700276] [<ffffffff810ed7e6>] shrink_inactive_list+0x6b6/0x700
[36960.700285] [<ffffffff81044250>] ? wake_affine+0x220/0x240
[36960.700292] [<ffffffff81048210>] ? __enqueue_entity+0x80/0x90
[36960.700299] [<ffffffff810ed881>] shrink_list+0x51/0xb0
[36960.700306] [<ffffffff810eddea>] shrink_zone+0x1ea/0x200
[36960.700314] [<ffffffff810ee823>] shrink_zones+0x63/0xf0
[36960.700321] [<ffffffff810ee920>] do_try_to_free_pages+0x70/0x280
[36960.700328] [<ffffffff810eec9c>] try_to_free_pages+0x9c/0xc0
[36960.700335] [<ffffffff810ed0e0>] ? isolate_pages_global+0x0/0x50
[36960.700342] [<ffffffff810e6342>] __alloc_pages_slowpath+0x232/0x520
[36960.700350] [<ffffffff810e6776>] __alloc_pages_nodemask+0x146/0x180
[36960.700358] [<ffffffff811143f7>] alloc_pages_current+0x87/0xd0
[36960.700366] [<ffffffff8111939c>] allocate_slab+0x11c/0x1b0
[36960.700372] [<ffffffff8111945b>] new_slab+0x2b/0x190
[36960.700379] [<ffffffff8111b641>] __slab_alloc+0x121/0x230
[36960.700427] [<ffffffffa009b57d>] ? kmem_zone_alloc+0x5d/0xd0 [xfs]
[36960.700457] [<ffffffffa009b57d>] ? kmem_zone_alloc+0x5d/0xd0 [xfs]
[36960.700465] [<ffffffff8111b980>] kmem_cache_alloc+0xf0/0x130
[36960.700495] [<ffffffffa009b57d>] kmem_zone_alloc+0x5d/0xd0 [xfs]
[36960.700524] [<ffffffffa009b609>] kmem_zone_zalloc+0x19/0x50 [xfs]
[36960.700555] [<ffffffffa009368f>] _xfs_trans_alloc+0x2f/0x70 [xfs]
[36960.700586] [<ffffffffa0093832>] xfs_trans_alloc+0x92/0xa0 [xfs]
[36960.700593] [<ffffffff810f7710>] ? __dec_zone_page_state+0x30/0x40
[36960.700624] [<ffffffffa0083691>] xfs_iomap_write_unwritten+0x71/0x200 [xfs]
[36960.700633] [<ffffffff810e1ad2>] ? mempool_free_slab+0x12/0x20
[36960.700639] [<ffffffff810e1b6a>] ? mempool_free+0x8a/0xa0
[36960.700669] [<ffffffffa009c3d0>] ? xfs_end_bio_unwritten+0x0/0x80 [xfs]
[36960.700698] [<ffffffffa009c435>] xfs_end_bio_unwritten+0x65/0x80 [xfs]
[36960.700707] [<ffffffff81075c47>] run_workqueue+0xb7/0x190
[36960.700713] [<ffffffff81076fa6>] worker_thread+0x96/0xf0
[36960.700720] [<ffffffff8107af70>] ? autoremove_wake_function+0x0/0x40
[36960.700727] [<ffffffff81076f10>] ? worker_thread+0x0/0xf0
[36960.700733] [<ffffffff8107ab8e>] kthread+0x8e/0xa0
[36960.700741] [<ffffffff81012f8a>] child_rip+0xa/0x20
[36960.700748] [<ffffffff8107ab00>] ? kthread+0x0/0xa0
[36960.700754] [<ffffffff81012f80>] ? child_rip+0x0/0x20
[36960.700793] INFO: task rdf3xload:3947 blocked for more than 120 seconds.
[36960.700797] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this
message.
[36960.700801] dataload D 0000000000000001 0 3947 2121 0x00000000
[36960.700810] ffff880074f37ce8 0000000000000082 0000000000000086
ffff880074f37fd8
[36960.700819] ffff88007b0316d0 0000000000015980 0000000000015980
ffff880074f37fd8
[36960.700827] 0000000000015980 ffff880074f37fd8 0000000000015980
ffff88007b031a90
[36960.700835] Call Trace:
[36960.700869] [<ffffffffa009c0e5>] xfs_ioend_wait+0x85/0xc0 [xfs]
[36960.700877] [<ffffffff8107af70>] ? autoremove_wake_function+0x0/0x40
[36960.700909] [<ffffffffa0097d1d>] xfs_setattr+0x85d/0xb20 [xfs]
[36960.700940] [<ffffffffa0094aa8>] ? xfs_trans_unlocked_item+0x38/0x50 [xfs]
[36960.700970] [<ffffffffa00a2ebd>] xfs_vn_fallocate+0xed/0x100 [xfs]
[36960.700978] [<ffffffff810334a9>] ? default_spin_lock_flags+0x9/0x10
[36960.700986] [<ffffffff81527ecf>] ? _spin_lock_irqsave+0x2f/0x50
[36960.700994] [<ffffffff8112556d>] do_fallocate+0xfd/0x110
[36960.701001] [<ffffffff811255c9>] sys_fallocate+0x49/0x70
[36960.701009] [<ffffffff81011f42>] system_call_fastpath+0x16/0x1b
--
Configure bugmail: http://oss.sgi.com/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.
|