lkml.org 
[lkml]   [2013]   [May]   [31]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: 3.9.4+ watchdog overflow and migration lockup soon after boot (cpu_stopper_thread related?)
On 05/31/2013 02:40 PM, Ben Greear wrote:
> On 05/31/2013 12:22 PM, Ben Greear wrote:
>> While trying to verify that the kobject patch
>> (see "Please add to stable: module: don't unlink the module until we've removed all exposure." email)
>> fixed the problems I was seeing, I hit what I believe is a different problem. Much harder to reproduce
>> (maybe 15-20 reboots before I saw this..haven't seen it since).
>>
>> This was on a kernel with additional non-upstream patches, but it's not obviously related to anything
>> I've been hacking on. I'm trying to reproduce now on standard 3.9.4 plus the kobj patch,
>> but this seems to be quite rare, so not sure how long it will take.
>
> I reproduced this with only stock 3.9.4 and the kobj patch, so looks like it's
> a real bug. The info looks basically identical to what I posted previously.
>
> I have logs if someone wants to see it.
>
> It took around 20-30 reboots to hit the problem.

I added some printk debugging, and if I'm not mistaken, the
curstate and smdata->state is both '1' (STOPMACHINE_PREPARE)
in the stop_machine_cpu_stop method. Once you are in that state,
that method loops forever as far as I can tell.

Also, I was trying to manually rate-limit messages here based on
the jiffies timer...but it seems jiffies is not updating?

Any ideas?

Thanks,
Ben



>
> I'm thinking the cpu_stopper_thread is stuck in a loop..here are some of
> the stack points decoded:
>
> (gdb) l *(stop_machine_cpu_stop+0x68)
> 0xffffffff81103875 is in stop_machine_cpu_stop (/home/greearb/git/linux-2.6.linus/kernel/stop_machine.c:427).
> 422
> 423 /* Simple state machine */
> 424 do {
> 425 /* Chill out and ensure we re-read stopmachine_state. */
> 426 cpu_relax();
> 427 if (smdata->state != curstate) {
> 428 curstate = smdata->state;
> 429 switch (curstate) {
> 430 case STOPMACHINE_DISABLE_IRQ:
> 431 local_irq_disable();
> (gdb) l *(cpu_stopper_thread+0xbd)
> 0xffffffff81103574 is in cpu_stopper_thread (/home/greearb/git/linux-2.6.linus/kernel/stop_machine.c:286).
> 281
> 282 /* cpu stop callbacks are not allowed to sleep */
> 283 preempt_disable();
> 284
> 285 ret = fn(arg);
> 286 if (ret)
> 287 done->ret = ret;
> 288
> 289 /* restore preemption and check it's still balanced */
> 290 preempt_enable();
> (gdb)
>
>
> SysRq : Show backtrace of all active CPUs
> sending NMI to all CPUs:
> NMI backtrace for cpu 3
> CPU 3
> Pid: 29, comm: migration/3 Tainted: G WC 3.9.4+ #6 To be filled by O.E.M. To be filled by O.E.M./To be filled by O.E.M.
> RIP: 0010:[<ffffffff81103875>] [<ffffffff81103875>] stop_machine_cpu_stop+0x68/0xfb
> RSP: 0018:ffff8802223f3cc8 EFLAGS: 00000093
> RAX: 0000000000000002 RBX: ffff880219ac3e18 RCX: 0000000000000000
> RDX: ffff8802223ec2c0 RSI: 0000000000000040 RDI: ffffffff81103898
> RBP: ffff8802223f3d08 R08: 0000000000000000 R09: ffff8802223f3b98
> R10: ffffffff81cfc860 R11: 0000000000000002 R12: 0000000000000292
> R13: 0000000000000002 R14: ffff880219ac3e3c R15: 0000000000000000
> FS: 0000000000000000(0000) GS:ffff88022bd80000(0000) knlGS:0000000000000000
> CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b
> CR2: 0000003397a07cc0 CR3: 000000021b7ce000 CR4: 00000000000007e0
> DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
> Process migration/3 (pid: 29, threadinfo ffff8802223f2000, task ffff8802223ec2c0)
> Stack:
> ffff88022bd8e5b0 00ff88022bd8e610 ffff8802223f3ce8 ffff88022bd8e5b0
> ffff880219ac3d58 ffff8802223f2000 ffff8802223f2000 ffffffff8110380d
> ffff8802223f3de8 ffffffff81103574 ffff8802223f3fd8 ffff8802223f2010
> Call Trace:
> [<ffffffff8110380d>] ? stop_one_cpu_nowait+0x30/0x30
> [<ffffffff81103574>] cpu_stopper_thread+0xbd/0x176
> [<ffffffff815e8db7>] ? __schedule+0x5e7/0x62f
> [<ffffffff815ea223>] ? _raw_spin_unlock_irqrestore+0x47/0x77
> [<ffffffff810e7e6a>] ? trace_hardirqs_on_caller+0x123/0x15a
> [<ffffffff810e7eae>] ? trace_hardirqs_on+0xd/0xf
> [<ffffffff815ea245>] ? _raw_spin_unlock_irqrestore+0x69/0x77
> [<ffffffff810bdcb2>] smpboot_thread_fn+0x217/0x21d
> [<ffffffff810bda9b>] ? test_ti_thread_flag.clone.0+0x11/0x11
> [<ffffffff810b6a86>] kthread+0xc7/0xcf
> [<ffffffff810b69bf>] ? __init_kthread_worker+0x5b/0x5b
> [<ffffffff815f0cac>] ret_from_fork+0x7c/0xb0
> [<ffffffff810b69bf>] ? __init_kthread_worker+0x5b/0x5b
> Code: 40 00 00 00 e8 f5 ad 20 00 41 39 c5 0f 94 45 cf eb 0e 44 0f a3 28 45 19 ed 45 85 ed 0f 95 45 cf 4c 8d 73 24 45 31 ff 31 c0 f3 90 <44> 8b 6b 20 41 39 c5 74
> 44 41 83 fd 02 74 08 41 83 fd 03 75 1f
> NMI backtrace for cpu 1
> CPU 1
> Pid: 17, comm: migration/1 Tainted: G WC 3.9.4+ #6 To be filled by O.E.M. To be filled by O.E.M./To be filled by O.E.M.
> RIP: 0010:[<ffffffff81103875>] [<ffffffff81103875>] stop_machine_cpu_stop+0x68/0xfb
> RSP: 0018:ffff88022230fcc8 EFLAGS: 00000093
> RAX: 0000000000000002 RBX: ffff880219ac3e18 RCX: 0000000000000000
> RDX: ffff8802223142c0 RSI: 0000000000000040 RDI: ffffffff81103898
> RBP: ffff88022230fd08 R08: 0000000000000000 R09: ffff88022230fb98
> R10: ffff880219ac3cf8 R11: ffff8802223ec2c0 R12: 0000000000000292
> R13: 0000000000000002 R14: ffff880219ac3e3c R15: 0000000000000000
> FS: 0000000000000000(0000) GS:ffff88022bc80000(0000) knlGS:0000000000000000
> CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b
> CR2: 0000000000440b70 CR3: 000000021a03b000 CR4: 00000000000007e0
> DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
> Process migration/1 (pid: 17, threadinfo ffff88022230e000, task ffff8802223142c0)
> Stack:
> ffff88022bc8e5b0 00ff88022bc8e610 ffff88022230fce8 ffff88022bc8e5b0
> ffff880219ac3d58 ffff88022230e000 ffff88022230e000 ffffffff8110380d
> ffff88022230fde8 ffffffff81103574 ffff88022230ffd8 ffff88022230e010
> Call Trace:
> [<ffffffff8110380d>] ? stop_one_cpu_nowait+0x30/0x30
> [<ffffffff81103574>] cpu_stopper_thread+0xbd/0x176
> [<ffffffff815e8db7>] ? __schedule+0x5e7/0x62f
> [<ffffffff815ea223>] ? _raw_spin_unlock_irqrestore+0x47/0x77
> [<ffffffff810e7e6a>] ? trace_hardirqs_on_caller+0x123/0x15a
> [<ffffffff810e7eae>] ? trace_hardirqs_on+0xd/0xf
> [<ffffffff815ea245>] ? _raw_spin_unlock_irqrestore+0x69/0x77
> [<ffffffff810bdcb2>] smpboot_thread_fn+0x217/0x21d
> [<ffffffff810bda9b>] ? test_ti_thread_flag.clone.0+0x11/0x11
> [<ffffffff810b6a86>] kthread+0xc7/0xcf
> [<ffffffff810b69bf>] ? __init_kthread_worker+0x5b/0x5b
> [<ffffffff815f0cac>] ret_from_fork+0x7c/0xb0
> [<ffffffff810b69bf>] ? __init_kthread_worker+0x5b/0x5b
> Code: 40 00 00 00 e8 f5 ad 20 00 41 39 c5 0f 94 45 cf eb 0e 44 0f a3 28 45 19 ed 45 85 ed 0f 95 45 cf 4c 8d 73 24 45 31 ff 31 c0 f3 90 <44> 8b 6b 20 41 39 c5 74
> 44 41 83 fd 02 74 08 41 83 fd 03 75 1f
> NMI backtrace for cpu 2
> CPU 2
> Pid: 23, comm: migration/2 Tainted: G WC 3.9.4+ #6 To be filled by O.E.M. To be filled by O.E.M./To be filled by O.E.M.
> RIP: 0010:[<ffffffff81016549>] [<ffffffff81016549>] native_read_tsc+0x14/0x16
> RSP: 0000:ffff88022bd03b40 EFLAGS: 00000006
> RAX: 00000057e22024ac RBX: ffff8802223b4000 RCX: 00000000e22024ac
> RDX: 0000000000000057 RSI: 0000000000000002 RDI: 0000000000289789
> RBP: ffff88022bd03b40 R08: 0000000000000002 R09: 0000000000000000
> R10: ffff88022bd03a90 R11: 0000000000000006 R12: 000000000000006c
> R13: 0000000000000002 R14: 0000000000000086 R15: 0000000000000001
> FS: 0000000000000000(0000) GS:ffff88022bd00000(0000) knlGS:0000000000000000
> CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b
> CR2: 0000000000574fe0 CR3: 0000000214ca8000 CR4: 00000000000007e0
> DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
> Process migration/2 (pid: 23, threadinfo ffff8802223b4000, task ffff8802223bc2c0)
> Stack:
> ffff88022bd03ba0 ffffffff81309ea2 ffff88022bd03b70 0000000000000046
> ffff88022bd03b70 00289789810e5676 ffff88022bd03ba0 0000000000000001
> 000000000000006c 0000000000000007 0000000000000086 0000000000000001
> Call Trace:
> <IRQ>
> [<ffffffff81309ea2>] delay_tsc+0x40/0xee
> [<ffffffff81309e35>] __const_udelay+0x28/0x2a
> [<ffffffff8103053d>] arch_trigger_all_cpu_backtrace+0x66/0x7d
> [<ffffffff813c467a>] sysrq_handle_showallcpus+0xe/0x10
> [<ffffffff813c49ab>] __handle_sysrq+0xbf/0x15b
> [<ffffffff813c4d68>] handle_sysrq+0x2c/0x2e
> [<ffffffff813d8fcc>] serial8250_rx_chars+0x145/0x1c5
> [<ffffffff813d90be>] serial8250_handle_irq+0x72/0xa8
> [<ffffffff813d917f>] serial8250_default_handle_irq+0x23/0x28
> [<ffffffff813d7e14>] serial8250_interrupt+0x4d/0xc6
> [<ffffffff8110e760>] handle_irq_event_percpu+0x7a/0x1ea
> [<ffffffff8110e911>] handle_irq_event+0x41/0x61
> [<ffffffff8111113c>] handle_edge_irq+0xa6/0xcb
> [<ffffffff81011dd7>] handle_irq+0x24/0x2d
> [<ffffffff815f2a4d>] do_IRQ+0x4d/0xb4
> [<ffffffff815ea872>] common_interrupt+0x72/0x72
> [<ffffffff815ea937>] ? retint_restore_args+0x13/0x13
> [<ffffffff811146d2>] ? rcu_sched_qs+0x22/0x22
> [<ffffffff8109e4f1>] ? __do_softirq+0x185/0x25a
> [<ffffffff8109e68f>] irq_exit+0x5f/0xbc
> [<ffffffff815f2a9d>] do_IRQ+0x9d/0xb4
> [<ffffffff815ea872>] common_interrupt+0x72/0x72
> <EOI>
> [<ffffffff815ea937>] ? retint_restore_args+0x13/0x13
> [<ffffffff81103875>] ? stop_machine_cpu_stop+0x68/0xfb
> [<ffffffff81103853>] ? stop_machine_cpu_stop+0x46/0xfb
> [<ffffffff8110380d>] ? stop_one_cpu_nowait+0x30/0x30
> [<ffffffff81103574>] cpu_stopper_thread+0xbd/0x176
> [<ffffffff815e8db7>] ? __schedule+0x5e7/0x62f
> [<ffffffff815ea223>] ? _raw_spin_unlock_irqrestore+0x47/0x77
> [<ffffffff810e7e6a>] ? trace_hardirqs_on_caller+0x123/0x15a
> [<ffffffff810e7eae>] ? trace_hardirqs_on+0xd/0xf
> [<ffffffff815ea245>] ? _raw_spin_unlock_irqrestore+0x69/0x77
> [<ffffffff810bdcb2>] smpboot_thread_fn+0x217/0x21d
> [<ffffffff810bda9b>] ? test_ti_thread_flag.clone.0+0x11/0x11
> [<ffffffff810b6a86>] kthread+0xc7/0xcf
> [<ffffffff810b69bf>] ? __init_kthread_worker+0x5b/0x5b
> [<ffffffff815f0cac>] ret_from_fork+0x7c/0xb0
> [<ffffffff810b69bf>] ? __init_kthread_worker+0x5b/0x5b
> Code: 31 c9 48 c1 ef 0a 48 01 f8 c3 55 48 89 e5 e8 33 0a 00 00 66 90 c9 c3 55 48 89 e5 0f 31 89 c1 48 89 d0 48 c1 e0 20 89 c9 48 09 c8 <c9> c3 55 8b 05 12 c3 ad
> 00 48 89 e5 c9 c3 55 b8 ed ff ff ff 48
> NMI backtrace for cpu 0
> CPU 0
> Pid: 8, comm: migration/0 Tainted: G WC 3.9.4+ #6 To be filled by O.E.M. To be filled by O.E.M./To be filled by O.E.M.
> RIP: 0010:[<ffffffff81103879>] [<ffffffff81103879>] stop_machine_cpu_stop+0x6c/0xfb
> RSP: 0018:ffff8802222abcc8 EFLAGS: 00000093
> RAX: 0000000000000002 RBX: ffff880219ac3e18 RCX: 0000000000000000
> RDX: ffff8802222a42c0 RSI: 0000000000000040 RDI: ffffffff81103898
> RBP: ffff8802222abd08 R08: 0000000000000000 R09: ffff8802222abb98
> R10: ffffffff81a01ec8 R11: 0000000000000000 R12: 0000000000000292
> R13: 0000000000000002 R14: ffff880219ac3e3c R15: 0000000000000000
> FS: 0000000000000000(0000) GS:ffff88022bc00000(0000) knlGS:0000000000000000
> CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b
> CR2: 00007fff998bbff0 CR3: 0000000001a0c000 CR4: 00000000000007f0
> DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
> Process migration/0 (pid: 8, threadinfo ffff8802222aa000, task ffff8802222a42c0)
> Stack:
> ffff88022bc0e5b0 01ff88022bc0e610 ffff8802222abce8 ffff88022bc0e5b0
> ffff880219ac3d58 ffff8802222aa000 ffff8802222aa000 ffffffff8110380d
> ffff8802222abde8 ffffffff81103574 ffff8802222abfd8 ffff8802222aa010
> Call Trace:
> [<ffffffff8110380d>] ? stop_one_cpu_nowait+0x30/0x30
> [<ffffffff81103574>] cpu_stopper_thread+0xbd/0x176
> [<ffffffff815e8db7>] ? __schedule+0x5e7/0x62f
> [<ffffffff815ea223>] ? _raw_spin_unlock_irqrestore+0x47/0x77
> [<ffffffff810e7e6a>] ? trace_hardirqs_on_caller+0x123/0x15a
> [<ffffffff810e7eae>] ? trace_hardirqs_on+0xd/0xf
> [<ffffffff815ea245>] ? _raw_spin_unlock_irqrestore+0x69/0x77
> [<ffffffff810bdcb2>] smpboot_thread_fn+0x217/0x21d
> [<ffffffff810bda9b>] ? test_ti_thread_flag.clone.0+0x11/0x11
> [<ffffffff810b6a86>] kthread+0xc7/0xcf
> [<ffffffff810b69bf>] ? __init_kthread_worker+0x5b/0x5b
> [<ffffffff815f0cac>] ret_from_fork+0x7c/0xb0
> [<ffffffff810b69bf>] ? __init_kthread_worker+0x5b/0x5b
> Code: e8 f5 ad 20 00 41 39 c5 0f 94 45 cf eb 0e 44 0f a3 28 45 19 ed 45 85 ed 0f 95 45 cf 4c 8d 73 24 45 31 ff 31 c0 f3 90 44 8b 6b 20 <41> 39 c5 74 44 41 83 fd
> 02 74 08 41 83 fd 03 75 1f eb 0e fa 66
>
> Thanks,
> Ben
>


--
Ben Greear <greearb@candelatech.com>
Candela Technologies Inc http://www.candelatech.com



\
 
 \ /
  Last update: 2013-06-01 03:21    [W:0.099 / U:0.240 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site