a non producible kernel panic

From: Alex Shi
Date: Sat Feb 04 2012 - 23:28:55 EST


When do performance testing for rc2 kernel on our NHM EX machine( 4
sockets * 8 cores), I find a kernel panic. but never reproduced it. and
didn't find some useful info after enable lockdep, rcu debug etc.

Any ideas on this?

INFO: task umount:4280 blocked for more than 120 seconds.
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
umount D ffff883025443198 0 4280 4235 0x00000080
ffff883027805d28 0000000000000086 ffff883025442de0 ffff883027805fd8
ffff883027805fd8 0000000000013100 ffff8810292eade0 ffff883025442de0
0000000000000018 7fffffffffffffff 7fffffffffffffff ffffffff81c180c8
Call Trace:
[<ffffffff81452caa>] schedule+0x55/0x57
[<ffffffff81451a93>] schedule_timeout+0x2f/0xd9
[<ffffffff8109dfe0>] ? rcu_panic+0x12/0x12
[<ffffffff81028259>] ? physflat_send_IPI_allbutself+0x12/0x14
[<ffffffff81024a33>] ? native_send_call_func_ipi+0x71/0x91
[<ffffffff81452b67>] wait_for_common+0x95/0xeb
[<ffffffff8105dfe7>] ? try_to_wake_up+0x1e0/0x1e0
[<ffffffff8109dfe0>] ? rcu_panic+0x12/0x12
[<ffffffff8109eb41>] ? call_rcu_bh+0x12/0x12
[<ffffffff81452c53>] wait_for_completion+0x18/0x1a
[<ffffffff8109ef36>] _rcu_barrier+0x96/0xa6
[<ffffffff8109ef5d>] rcu_barrier_sched+0x17/0x19
[<ffffffff8109ef68>] rcu_barrier+0x9/0xb
[<ffffffff8110bac0>] deactivate_locked_super+0x32/0x52
[<ffffffff8110c536>] deactivate_super+0x32/0x36
[<ffffffff81120691>] mntput_no_expire+0xde/0xe7
[<ffffffff811212f2>] sys_umount+0x2cd/0x2ff
[<ffffffff81112699>] ? path_put+0x1a/0x1e
[<ffffffff81459a79>] system_call_fastpath+0x16/0x1b
INFO: task umount:4280 blocked for more than 120 seconds.
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
umount D ffff883025443198 0 4280 4235 0x00000080
ffff883027805d28 0000000000000086 ffff883025442de0 ffff883027805fd8
ffff883027805fd8 0000000000013100 ffff8810292eade0 ffff883025442de0
0000000000000018 7fffffffffffffff 7fffffffffffffff ffffffff81c18
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/