* liburing's eventfd test reliably oopses on 5.6-94f2630b1897
@ 2020-02-02 16:56 Andres Freund
2020-02-02 19:07 ` Jens Axboe
0 siblings, 1 reply; 2+ messages in thread
From: Andres Freund @ 2020-02-02 16:56 UTC (permalink / raw)
To: Jens Axboe, io-uring, linux-fsdevel
Hi,
Updated to linus' current master (with just one perf build fix applied
on top) for reasons unrelated to uring, got the oops below when running
the uring tests. It's sufficient to just run the eventfd test.
[ 4085.054332] rcu: INFO: rcu_sched detected stalls on CPUs/tasks:
[ 4085.054336] rcu: 8-...0: (0 ticks this GP) idle=346/1/0x4000000000000000 softirq=81543/81543 fqs=14982
[ 4085.054371] rcu: 14-...0: (1 GPs behind) idle=46e/1/0x4000000000000000 softirq=78025/78025 fqs=14982
[ 4085.054373] (detected by 0, t=60002 jiffies, g=246209, q=21702)
[ 4085.054375] Sending NMI from CPU 0 to CPUs 8:
[ 4085.055377] NMI backtrace for cpu 8
[ 4085.055378] CPU: 8 PID: 14268 Comm: eventfd Not tainted 5.5.0-andres-06886-g050b83aa3110-dirty #20
[ 4085.055378] Hardware name: LENOVO 20QVS0FP00/20QVS0FP00, BIOS N2OET41W (1.28 ) 11/25/2019
[ 4085.055378] RIP: 0010:queued_spin_lock_slowpath+0x42/0x1a0
[ 4085.055379] Code: 41 f0 0f ba 2f 08 0f 92 c0 0f b6 c0 c1 e0 08 89 c2 8b 07 30 e4 09 d0 a9 00 01 ff ff 75 1b 85 c0 74 0e 8b 07 84 c0 74 08 f3 90 <8b> 07 84 c0 75 f8 b8 01 00 00 00 66 89 07 c3 f6 c4 01 75 04 c6 47
[ 4085.055380] RSP: 0018:ffffc900050abb98 EFLAGS: 00000002
[ 4085.055381] RAX: 00000000003c0101 RBX: ffff888804459080 RCX: 0000000000000000
[ 4085.055381] RDX: 0000000000000000 RSI: 0000000000000000 RDI: ffff888804459088
[ 4085.055381] RBP: ffff888804459088 R08: ffffffff8245dfa0 R09: 0000000000000000
[ 4085.055382] R10: ffff88880bf4dcd8 R11: ffff88881c7ab638 R12: 0000000000000046
[ 4085.055382] R13: ffff888804459088 R14: 0000000000000046 R15: ffff88880f41a000
[ 4085.055383] FS: 00007f2383091500(0000) GS:ffff88881c600000(0000) knlGS:0000000000000000
[ 4085.055383] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 4085.055383] CR2: 00007f2382f0cb90 CR3: 0000000803216002 CR4: 00000000003606e0
[ 4085.055384] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 4085.055384] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 4085.055384] Call Trace:
[ 4085.055385] _raw_spin_lock_irqsave+0x33/0x40
[ 4085.055385] eventfd_signal+0x1d/0x70
[ 4085.055385] io_poll_wake+0x1ec/0x360
[ 4085.055386] ? __switch_to_asm+0x40/0x70
[ 4085.055386] ? __switch_to_asm+0x34/0x70
[ 4085.055386] __wake_up_common+0x7a/0x140
[ 4085.055386] eventfd_signal+0x57/0x70
[ 4085.055387] io_issue_sqe+0x580/0xf80
[ 4085.055387] ? io_submit_sqes+0x425/0xb10
[ 4085.055387] ? io_import_iovec+0x220/0x220
[ 4085.055388] ? rw_copy_check_uvector+0x4b/0x100
[ 4085.055388] io_queue_sqe+0x2d2/0x790
[ 4085.055388] ? io_read_prep+0x89/0xc0
[ 4085.055388] ? _cond_resched+0x19/0x30
[ 4085.055389] io_submit_sqes+0x839/0xb10
[ 4085.055389] ? trace_hardirqs_on+0x2c/0xd0
[ 4085.055389] ? __io_uring_register+0x199/0xe00
[ 4085.055390] ? alloc_file_pseudo+0xa3/0x110
[ 4085.055390] __x64_sys_io_uring_enter+0x253/0x350
[ 4085.055390] do_syscall_64+0x50/0x140
[ 4085.055390] entry_SYSCALL_64_after_hwframe+0x49/0xbe
[ 4085.055391] RIP: 0033:0x7f2382fc50a9
[ 4085.055392] Code: 00 c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d b7 3d 0c 00 f7 d8 64 89 01 48
[ 4085.055392] RSP: 002b:00007ffd259b6f88 EFLAGS: 00000212 ORIG_RAX: 00000000000001aa
[ 4085.055393] RAX: ffffffffffffffda RBX: 00007ffd259b7060 RCX: 00007f2382fc50a9
[ 4085.055393] RDX: 0000000000000000 RSI: 0000000000000001 RDI: 0000000000000003
[ 4085.055393] RBP: 0000000000000003 R08: 0000000000000000 R09: 0000000000000008
[ 4085.055394] R10: 0000000000000000 R11: 0000000000000212 R12: 0000000000000004
[ 4085.055394] R13: 00007ffd259b6fc0 R14: 0000000000000000 R15: 0000000000000000
[ 4085.055396] Sending NMI from CPU 0 to CPUs 14:
[ 4085.056412] NMI backtrace for cpu 14
[ 4085.056412] CPU: 14 PID: 14270 Comm: io_wqe_worker-0 Not tainted 5.5.0-andres-06886-g050b83aa3110-dirty #20
[ 4085.056413] Hardware name: LENOVO 20QVS0FP00/20QVS0FP00, BIOS N2OET41W (1.28 ) 11/25/2019
[ 4085.056413] RIP: 0010:queued_spin_lock_slowpath+0x118/0x1a0
[ 4085.056414] Code: 74 f2 eb f6 41 83 c0 01 c1 e1 10 41 c1 e0 12 44 09 c1 89 c8 c1 e8 10 66 87 47 02 89 c6 c1 e6 10 85 f6 75 3c 31 f6 eb 02 f3 90 <8b> 07 66 85 c0 75 f7 41 89 c0 66 45 31 c0 41 39 c8 74 64 48 85 f6
[ 4085.056414] RSP: 0018:ffffc900050c3bd0 EFLAGS: 00000002
[ 4085.056415] RAX: 00000000003c0101 RBX: ffff888804459080 RCX: 00000000003c0000
[ 4085.056430] RDX: ffff88881c7ac200 RSI: 0000000000000000 RDI: ffff888804459088
[ 4085.056431] RBP: ffff888804459088 R08: 00000000003c0000 R09: 0000000000000000
[ 4085.056431] R10: 00007ffd259b6fd0 R11: 0000000000105931 R12: 00007ffd259b6fc8
[ 4085.056431] R13: ffff888814fbb800 R14: ffff88880bf4dc40 R15: 0000000000000008
[ 4085.056432] FS: 0000000000000000(0000) GS:ffff88881c780000(0000) knlGS:0000000000000000
[ 4085.056432] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 4085.056433] CR2: 00007f4e7499b000 CR3: 0000000803216002 CR4: 00000000003606e0
[ 4085.056433] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 4085.056433] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 4085.056434] Call Trace:
[ 4085.056434] eventfd_read+0x78/0x210
[ 4085.056434] ? wake_up_q+0xa0/0xa0
[ 4085.056434] loop_rw_iter.part.0+0x40/0x110
[ 4085.056435] io_read+0x228/0x280
[ 4085.056435] ? __switch_to_asm+0x40/0x70
[ 4085.056435] ? __switch_to_asm+0x34/0x70
[ 4085.056435] ? __switch_to_asm+0x40/0x70
[ 4085.056436] ? __switch_to_asm+0x34/0x70
[ 4085.056436] ? __switch_to_asm+0x40/0x70
[ 4085.056436] ? __switch_to_asm+0x34/0x70
[ 4085.056436] ? __switch_to_asm+0x40/0x70
[ 4085.056437] ? __switch_to_asm+0x34/0x70
[ 4085.056437] ? __switch_to_asm+0x40/0x70
[ 4085.056437] ? __switch_to_asm+0x34/0x70
[ 4085.056437] ? __switch_to_asm+0x40/0x70
[ 4085.056438] ? __switch_to_asm+0x34/0x70
[ 4085.056438] ? __switch_to_asm+0x40/0x70
[ 4085.056438] ? __switch_to_asm+0x34/0x70
[ 4085.056438] ? __switch_to_asm+0x40/0x70
[ 4085.056438] ? __switch_to_asm+0x34/0x70
[ 4085.056439] ? __switch_to_asm+0x40/0x70
[ 4085.056439] ? finish_task_switch+0x74/0x240
[ 4085.056439] io_issue_sqe+0x562/0xf80
[ 4085.056455] ? trace_hardirqs_off+0x27/0xd0
[ 4085.056455] ? switch_mm+0x41/0x50
[ 4085.056456] io_wq_submit_work+0x73/0x220
[ 4085.056456] io_worker_handle_work+0x1f1/0x4b0
[ 4085.056456] io_wqe_worker+0x27a/0x350
[ 4085.056456] kthread+0xfb/0x130
[ 4085.056457] ? io_wqe_enqueue+0xf0/0xf0
[ 4085.056457] ? kthread_park+0x90/0x90
[ 4085.056457] ret_from_fork+0x24/0x50
[ 4265.059294] rcu: INFO: rcu_sched detected stalls on CPUs/tasks:
[ 4265.059298] rcu: 8-...0: (0 ticks this GP) idle=346/1/0x4000000000000000 softirq=81543/81543 fqs=59754
[ 4265.059300] rcu: 14-...0: (1 GPs behind) idle=46e/1/0x4000000000000000 softirq=78025/78025 fqs=59754
[ 4265.059334] (detected by 0, t=240007 jiffies, g=246209, q=55199)
[ 4265.059337] Sending NMI from CPU 0 to CPUs 8:
[ 4265.060339] NMI backtrace for cpu 8
[ 4265.060339] CPU: 8 PID: 14268 Comm: eventfd Not tainted 5.5.0-andres-06886-g050b83aa3110-dirty #20
[ 4265.060340] Hardware name: LENOVO 20QVS0FP00/20QVS0FP00, BIOS N2OET41W (1.28 ) 11/25/2019
[ 4265.060340] RIP: 0010:queued_spin_lock_slowpath+0x42/0x1a0
[ 4265.060341] Code: 41 f0 0f ba 2f 08 0f 92 c0 0f b6 c0 c1 e0 08 89 c2 8b 07 30 e4 09 d0 a9 00 01 ff ff 75 1b 85 c0 74 0e 8b 07 84 c0 74 08 f3 90 <8b> 07 84 c0 75 f8 b8 01 00 00 00 66 89 07 c3 f6 c4 01 75 04 c6 47
[ 4265.060342] RSP: 0018:ffffc900050abb98 EFLAGS: 00000002
[ 4265.060343] RAX: 00000000003c0101 RBX: ffff888804459080 RCX: 0000000000000000
[ 4265.060343] RDX: 0000000000000000 RSI: 0000000000000000 RDI: ffff888804459088
[ 4265.060343] RBP: ffff888804459088 R08: ffffffff8245dfa0 R09: 0000000000000000
[ 4265.060344] R10: ffff88880bf4dcd8 R11: ffff88881c7ab638 R12: 0000000000000046
[ 4265.060344] R13: ffff888804459088 R14: 0000000000000046 R15: ffff88880f41a000
[ 4265.060345] FS: 00007f2383091500(0000) GS:ffff88881c600000(0000) knlGS:0000000000000000
[ 4265.060345] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 4265.060346] CR2: 00007f2382f0cb90 CR3: 0000000803216002 CR4: 00000000003606e0
[ 4265.060346] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 4265.060347] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 4265.060347] Call Trace:
[ 4265.060347] _raw_spin_lock_irqsave+0x33/0x40
[ 4265.060347] eventfd_signal+0x1d/0x70
[ 4265.060348] io_poll_wake+0x1ec/0x360
[ 4265.060348] ? __switch_to_asm+0x40/0x70
[ 4265.060348] ? __switch_to_asm+0x34/0x70
[ 4265.060349] __wake_up_common+0x7a/0x140
[ 4265.060349] eventfd_signal+0x57/0x70
[ 4265.060349] io_issue_sqe+0x580/0xf80
[ 4265.060350] ? io_submit_sqes+0x425/0xb10
[ 4265.060350] ? io_import_iovec+0x220/0x220
[ 4265.060350] ? rw_copy_check_uvector+0x4b/0x100
[ 4265.060351] io_queue_sqe+0x2d2/0x790
[ 4265.060351] ? io_read_prep+0x89/0xc0
[ 4265.060351] ? _cond_resched+0x19/0x30
[ 4265.060352] io_submit_sqes+0x839/0xb10
[ 4265.060352] ? trace_hardirqs_on+0x2c/0xd0
[ 4265.060352] ? __io_uring_register+0x199/0xe00
[ 4265.060353] ? alloc_file_pseudo+0xa3/0x110
[ 4265.060353] __x64_sys_io_uring_enter+0x253/0x350
[ 4265.060353] do_syscall_64+0x50/0x140
[ 4265.060354] entry_SYSCALL_64_after_hwframe+0x49/0xbe
[ 4265.060354] RIP: 0033:0x7f2382fc50a9
[ 4265.060355] Code: 00 c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d b7 3d 0c 00 f7 d8 64 89 01 48
[ 4265.060355] RSP: 002b:00007ffd259b6f88 EFLAGS: 00000212 ORIG_RAX: 00000000000001aa
[ 4265.060356] RAX: ffffffffffffffda RBX: 00007ffd259b7060 RCX: 00007f2382fc50a9
[ 4265.060357] RDX: 0000000000000000 RSI: 0000000000000001 RDI: 0000000000000003
[ 4265.060357] RBP: 0000000000000003 R08: 0000000000000000 R09: 0000000000000008
[ 4265.060358] R10: 0000000000000000 R11: 0000000000000212 R12: 0000000000000004
[ 4265.060358] R13: 00007ffd259b6fc0 R14: 0000000000000000 R15: 0000000000000000
[ 4265.060360] Sending NMI from CPU 0 to CPUs 14:
[ 4265.061378] NMI backtrace for cpu 14
[ 4265.061378] CPU: 14 PID: 14270 Comm: io_wqe_worker-0 Not tainted 5.5.0-andres-06886-g050b83aa3110-dirty #20
[ 4265.061379] Hardware name: LENOVO 20QVS0FP00/20QVS0FP00, BIOS N2OET41W (1.28 ) 11/25/2019
[ 4265.061379] RIP: 0010:queued_spin_lock_slowpath+0x118/0x1a0
[ 4265.061380] Code: 74 f2 eb f6 41 83 c0 01 c1 e1 10 41 c1 e0 12 44 09 c1 89 c8 c1 e8 10 66 87 47 02 89 c6 c1 e6 10 85 f6 75 3c 31 f6 eb 02 f3 90 <8b> 07 66 85 c0 75 f7 41 89 c0 66 45 31 c0 41 39 c8 74 64 48 85 f6
[ 4265.061380] RSP: 0018:ffffc900050c3bd0 EFLAGS: 00000002
[ 4265.061381] RAX: 00000000003c0101 RBX: ffff888804459080 RCX: 00000000003c0000
[ 4265.061381] RDX: ffff88881c7ac200 RSI: 0000000000000000 RDI: ffff888804459088
[ 4265.061382] RBP: ffff888804459088 R08: 00000000003c0000 R09: 0000000000000000
[ 4265.061382] R10: 00007ffd259b6fd0 R11: 0000000000105931 R12: 00007ffd259b6fc8
[ 4265.061383] R13: ffff888814fbb800 R14: ffff88880bf4dc40 R15: 0000000000000008
[ 4265.061383] FS: 0000000000000000(0000) GS:ffff88881c780000(0000) knlGS:0000000000000000
[ 4265.061383] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 4265.061384] CR2: 00007f4e7499b000 CR3: 0000000803216002 CR4: 00000000003606e0
[ 4265.061384] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 4265.061385] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 4265.061385] Call Trace:
[ 4265.061385] eventfd_read+0x78/0x210
[ 4265.061385] ? wake_up_q+0xa0/0xa0
[ 4265.061386] loop_rw_iter.part.0+0x40/0x110
[ 4265.061386] io_read+0x228/0x280
[ 4265.061386] ? __switch_to_asm+0x40/0x70
[ 4265.061387] ? __switch_to_asm+0x34/0x70
[ 4265.061387] ? __switch_to_asm+0x40/0x70
[ 4265.061387] ? __switch_to_asm+0x34/0x70
[ 4265.061388] ? __switch_to_asm+0x40/0x70
[ 4265.061388] ? __switch_to_asm+0x34/0x70
[ 4265.061388] ? __switch_to_asm+0x40/0x70
[ 4265.061388] ? __switch_to_asm+0x34/0x70
[ 4265.061389] ? __switch_to_asm+0x40/0x70
[ 4265.061389] ? __switch_to_asm+0x34/0x70
[ 4265.061389] ? __switch_to_asm+0x40/0x70
[ 4265.061389] ? __switch_to_asm+0x34/0x70
[ 4265.061390] ? __switch_to_asm+0x40/0x70
[ 4265.061390] ? __switch_to_asm+0x34/0x70
[ 4265.061390] ? __switch_to_asm+0x40/0x70
[ 4265.061391] ? __switch_to_asm+0x34/0x70
[ 4265.061391] ? __switch_to_asm+0x40/0x70
[ 4265.061391] ? finish_task_switch+0x74/0x240
[ 4265.061391] io_issue_sqe+0x562/0xf80
[ 4265.061392] ? trace_hardirqs_off+0x27/0xd0
[ 4265.061392] ? switch_mm+0x41/0x50
[ 4265.061392] io_wq_submit_work+0x73/0x220
[ 4265.061393] io_worker_handle_work+0x1f1/0x4b0
[ 4265.061412] io_wqe_worker+0x27a/0x350
[ 4265.061412] kthread+0xfb/0x130
[ 4265.061412] ? io_wqe_enqueue+0xf0/0xf0
[ 4265.061413] ? kthread_park+0x90/0x90
[ 4265.061413] ret_from_fork+0x24/0x50
Greetings,
Andres Freund
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: liburing's eventfd test reliably oopses on 5.6-94f2630b1897
2020-02-02 16:56 liburing's eventfd test reliably oopses on 5.6-94f2630b1897 Andres Freund
@ 2020-02-02 19:07 ` Jens Axboe
0 siblings, 0 replies; 2+ messages in thread
From: Jens Axboe @ 2020-02-02 19:07 UTC (permalink / raw)
To: Andres Freund, io-uring, linux-fsdevel
On 2/2/20 9:56 AM, Andres Freund wrote:
> Hi,
>
> Updated to linus' current master (with just one perf build fix applied
> on top) for reasons unrelated to uring, got the oops below when running
> the uring tests. It's sufficient to just run the eventfd test.
This is known, I have a pending fix. I did gate the test on 5.6-rc, but
that'll only truly work once I ship those fixes out this week.
--
Jens Axboe
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2020-02-02 19:07 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-02-02 16:56 liburing's eventfd test reliably oopses on 5.6-94f2630b1897 Andres Freund
2020-02-02 19:07 ` Jens Axboe
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).