All of lore.kernel.org
 help / color / mirror / Atom feed
* Bug report: Kernel IO lockup. Hard reboot required.
@ 2019-10-24  2:16 David Niklas
  0 siblings, 0 replies; only message in thread
From: David Niklas @ 2019-10-24  2:16 UTC (permalink / raw)
  To: linux-kernel

Ever since I started using Linux I've had problems with freezing.
Sometimes this has been the HW, sometimes its the FS being dirty.
This time I know, having checked, that the FS is clean, and having
verified that the disks are accessible, I managed to finally get a report
on this lockup before the entire system became unresponsive. When I say
unresponsive I mean the even the sysreq keys cease to work (Yes, its that
serious).
I've no idea what causes this lockup, or what I can do to trigger it
again, only that a lockup will happen sooner or latter.
The FS that the kernel was having trouble addressing was an MD array. Its
named /dev/md7p1. There is only one partition on it. Its formated
reiserfs. The version of the array is 1.2.
My kernel is using a custom config. I'm running Devuan (Debian) linux on
kernel 4.14.146. This is a desktop system running on an AMD 1090T
processor.
mdadm - v3.4 - 28th January 2016
GPT fdisk (gdisk) version 1.0.1
parted (GNU parted) 3.2
The system is not running any other special parts that I compiled that
might cause a crash (I use discord and the yuoa game engine now and
again, etc.). These are sand-boxed by apparmor.
I was not using the system when it locked-up. I happened to be looking at
it from a distance to see how it was progressing on the tasks it was
carrying out. It was downloading some stuff. I don't recall what exactly.
I noticed, because htop was running on my terminal, that the amount of IO
was extreme. I sat down and scrolled through the list of processes
waiting on IO and noticed that they were trying to access my MD array.
They were all stuck. I decided to sit down for a moment and tried to
switch to my desktop from my tty. That failed and the desktop couldn't
do any IO, which it wanted to for some reason (Its Icewm). I verified
using smartctl that all the disks were accessible. I walked away
and came back later to recored the dmesg output for our edification.

If you need more details just ask, I am subscribed.
Dmesg below: 

Thanks,
David

PS: The kernel was compiled in the normal way with gcc, make clean, make
prepare, make modules_prepare, make, make modules, make modules_install,
make install.

[    0.000000] Linux version 4.14.146-nopreempt-AMDGPU-dav9 (Snip host
name) (gcc version 6.3.0 20170516 (Debian 6.3.0-18+deb9u1)) #1 SMP Mon
Sep 30 00:49:45 EDT 2019 [    0.000000] Command line:
BOOT_IMAGE=Linux_AMDGPU2 ro root=824 root=/dev/sdc4 rootfstype=ext4
splash=verbose iommu=soft elevator=deadline ...<snip>... [190299.658191]
md: md7: data-check done. [193601.094230] usb 6-1: USB disconnect, device
number 5 [193601.617854] usb 6-1: new full-speed USB device number 6
using ohci-pci [193601.802333] usb 6-1: New USB device found,
idVendor=258a, idProduct=0012 [193601.802335] usb 6-1: New USB device
strings: Mfr=1, Product=2, SerialNumber=0 [193601.802337] usb 6-1:
Product: USBGamingMouse [193601.802337] usb 6-1: Manufacturer: SN TECH
[193601.809199] input: SN TECH      USBGamingMouse
as /devices/pci0000:00/0000:00:12.0/usb6/6-1/6-1:1.0/0003:258A:0012.0009/input/input24
[193601.864756] hid-generic 0003:258A:0012.0009: input,hidraw0: USB HID
v1.11 Mouse [SN TECH      USBGamingMouse] on usb-0000:00:12.0-1/input0
[193601.870103] input: SN TECH      USBGamingMouse
as /devices/pci0000:00/0000:00:12.0/usb6/6-1/6-1:1.1/0003:258A:0012.000A/input/input25
[193601.925117] hid-generic 0003:258A:0012.000A: input,hiddev0,hidraw1:
USB HID v1.11 Keyboard [SN TECH      USBGamingMouse] on
usb-0000:00:12.0-1/input1 [193602.008004] elogind[3860]: Watching system
buttons on /dev/input/event16 (SN TECH      USBGamingMouse)
[216672.407021] elogind[3860]: Removed session 9. [252149.611801] INFO:
task icewm:7918 blocked for more than 480 seconds. [252149.613376]
Not tainted 4.14.146-nopreempt-AMDGPU-dav9 #1 [252149.614771] "echo 0
> /proc/sys/kernel/hung_task_timeout_secs" disables this message.
> [252149.616255] icewm           D    0  7918   3491 0x00000080
> [252149.616258] Call Trace: [252149.617735]  ? __schedule+0x29e/0x6c0
> [252149.619206]  schedule+0x32/0x80 [252149.620641]
> schedule_preempt_disabled+0xa/0x10 [252149.622106]
> __mutex_lock.isra.1+0x26b/0x4e0
[252149.623566]  ? __wake_up_common_lock+0x89/0xc0
[252149.625025]  ? do_journal_begin_r+0xbe/0x390 [reiserfs]
[252149.626535]  do_journal_begin_r+0xbe/0x390 [reiserfs]
[252149.627993]  journal_begin+0x80/0x140 [reiserfs]
[252149.629402]  reiserfs_dirty_inode+0x3d/0xa0 [reiserfs]
[252149.630803]  ? unix_state_double_lock+0x60/0x60
[252149.632289]  __mark_inode_dirty+0x163/0x350
[252149.633757]  generic_update_time+0x79/0xc0
[252149.635156]  ? current_time+0x38/0x70
[252149.636572]  file_update_time+0xbe/0x110
[252149.637967]  __generic_file_write_iter+0x99/0x1b0
[252149.639362]  generic_file_write_iter+0xe2/0x1c0
[252149.640734]  __vfs_write+0x102/0x180
[252149.642170]  vfs_write+0xb0/0x190
[252149.643549]  SyS_write+0x5a/0xd0
[252149.644898]  do_syscall_64+0x6e/0x110
[252149.646261]  entry_SYSCALL_64_after_hwframe+0x3d/0xa2
[252149.647610] RIP: 0033:0x7f1be1ea4970
[252149.648964] RSP: 002b:00007fff27baa2c8 EFLAGS: 00000246 ORIG_RAX:
0000000000000001 [252149.650388] RAX: ffffffffffffffda RBX:
0000000000000007 RCX: 00007f1be1ea4970 [252149.651752] RDX:
0000000000000007 RSI: 00007fff27baa490 RDI: 0000000000000002
[252149.653136] RBP: 00007fff27baa490 R08: 00007f1be42ca880 R09:
0000000000000007 [252149.654487] R10: 0000000000000073 R11:
0000000000000246 R12: 0000000000000007 [252149.655849] R13:
0000000000000001 R14: 00007f1be2163520 R15: 0000000000000007
[252149.657205] INFO: task claws-mail:8342 blocked for more than 480
seconds. [252149.658625]       Not tainted 4.14.146-nopreempt-AMDGPU-dav9
#1 [252149.660037] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs"
disables this message. [252149.661399] claws-mail      D    0  8342
8293 0x00000080 [252149.661400] Call Trace: [252149.662761]  ?
__schedule+0x29e/0x6c0 [252149.664255]  schedule+0x32/0x80
[252149.665700]  schedule_preempt_disabled+0xa/0x10 [252149.667117]
__mutex_lock.isra.1+0x26b/0x4e0 [252149.668481]  ?
__wake_up_common_lock+0x89/0xc0 [252149.669905]  ?
do_journal_begin_r+0xbe/0x390 [reiserfs] [252149.671268]
do_journal_begin_r+0xbe/0x390 [reiserfs] [252149.672726]  ?
copyout+0x22/0x30 [252149.674081]  ? _copy_to_iter+0x8d/0x3f0
[252149.675501]  journal_begin+0x80/0x140 [reiserfs]
[252149.676846]  reiserfs_dirty_inode+0x3d/0xa0 [reiserfs]
[252149.678188]  ? release_sock+0x43/0x90
[252149.679534]  __mark_inode_dirty+0x163/0x350
[252149.680874]  generic_update_time+0x79/0xc0
[252149.682291]  ? current_time+0x38/0x70
[252149.683672]  file_update_time+0xbe/0x110
[252149.685042]  __generic_file_write_iter+0x99/0x1b0
[252149.686376]  generic_file_write_iter+0xe2/0x1c0
[252149.687700]  __vfs_write+0x102/0x180
[252149.689031]  vfs_write+0xb0/0x190
[252149.690342]  SyS_write+0x5a/0xd0
[252149.691767]  do_syscall_64+0x6e/0x110
[252149.693091]  entry_SYSCALL_64_after_hwframe+0x3d/0xa2
[252149.694372] RIP: 0033:0x7f457d7ef98d
[252149.695657] RSP: 002b:00007ffd3a42c8a0 EFLAGS: 00000293 ORIG_RAX:
0000000000000001 [252149.696960] RAX: ffffffffffffffda RBX:
000000000000002a RCX: 00007f457d7ef98d [252149.698268] RDX:
000000000000002a RSI: 0000000001fa3e30 RDI: 0000000000000009
[252149.699733] RBP: 0000000001fa3e30 R08: 0000000000b17560 R09:
00007f4582dc8a00 [252149.701036] R10: 000824f7579ca466 R11:
0000000000000293 R12: 000000000000002a [252149.702354] R13:
0000000000000001 R14: 0000000000b17480 R15: 000000000000002a
[252149.703667] INFO: task Cache2 I/O:8456 blocked for more than 480
seconds. [252149.704995]       Not tainted 4.14.146-nopreempt-AMDGPU-dav9
#1 [252149.706325] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs"
disables this message. [252149.707733] Cache2 I/O      D    0  8456
8293 0x00000080 [252149.707734] Call Trace: [252149.709091]  ?
__schedule+0x29e/0x6c0 [252149.710489]  schedule+0x32/0x80
[252149.711870]  schedule_preempt_disabled+0xa/0x10 [252149.713329]
__mutex_lock.isra.1+0x26b/0x4e0 [252149.714773]  ?
do_journal_begin_r+0xbe/0x390 [reiserfs] [252149.716198]
do_journal_begin_r+0xbe/0x390 [reiserfs] [252149.717545]
journal_begin+0x80/0x140 [reiserfs] [252149.718900]
reiserfs_rename+0x282/0x990 [reiserfs] [252149.720416]  ?
__wake_up_common_lock+0x89/0xc0 [252149.721812]  ? lookup_fast+0xc8/0x2c0
[252149.723203]  ? lookup_fast+0xc8/0x2c0
[252149.724662]  ? vfs_rename+0x6c9/0x900
[252149.725991]  vfs_rename+0x6c9/0x900
[252149.727314]  SyS_rename+0x3ce/0x3f0
[252149.728644]  do_syscall_64+0x6e/0x110
[252149.729967]  entry_SYSCALL_64_after_hwframe+0x3d/0xa2
[252149.731293] RIP: 0033:0x7f8927562d47
[252149.732671] RSP: 002b:00007f892872ec38 EFLAGS: 00000202 ORIG_RAX:
0000000000000052 [252149.734077] RAX: ffffffffffffffda RBX:
0000000000000000 RCX: 00007f8927562d47 [252149.735404] RDX:
0000000000000000 RSI: 00007f8733d47f08 RDI: 00007f87320d6b88
[252149.736767] RBP: 00007f872cb84c40 R08: 0000000000000000 R09:
0000000000000000 [252149.738066] R10: 0000000000000000 R11:
0000000000000202 R12: 00007f872cb84ce8 [252149.739380] R13:
00007f87322ada00 R14: 00007f892872ed50 R15: 00007f892872ec40
[252149.740773] INFO: task StreamT~ns #751:29976 blocked for more than
480 seconds. [252149.742093]       Not tainted
4.14.146-nopreempt-AMDGPU-dav9 #1 [252149.743475] "echo 0
> /proc/sys/kernel/hung_task_timeout_secs" disables this message.
> [252149.744805] StreamT~ns #751 D    0 29976   8293 0x00000080
> [252149.744806] Call Trace: [252149.746141]  ? __schedule+0x29e/0x6c0
[252149.747465]  schedule+0x32/0x80
[252149.748833]  schedule_preempt_disabled+0xa/0x10
[252149.750271]  __mutex_lock.isra.1+0x26b/0x4e0
[252149.751597]  ? do_journal_begin_r+0xbe/0x390 [reiserfs]
[252149.752918]  do_journal_begin_r+0xbe/0x390 [reiserfs]
[252149.754239]  ? reiserfs_lookup+0xb5/0x160 [reiserfs]
[252149.755573]  journal_begin+0x80/0x140 [reiserfs]
[252149.756959]  reiserfs_create+0xfc/0x210 [reiserfs]
[252149.758300]  path_openat+0x13e1/0x1490
[252149.759627]  ? futex_wake+0x91/0x170
[252149.760949]  do_filp_open+0x99/0x110
[252149.762309]  ? __check_object_size+0xfa/0x1a0
[252149.763688]  ? strncpy_from_user+0x48/0x130
[252149.765023]  ? __check_object_size+0xfa/0x1a0
[252149.766477]  ? __alloc_fd+0x3d/0x160
[252149.767800]  ? do_sys_open+0x12e/0x210
[252149.769126]  do_sys_open+0x12e/0x210
[252149.770432]  do_syscall_64+0x6e/0x110
[252149.771734]  entry_SYSCALL_64_after_hwframe+0x3d/0xa2
[252149.773084] RIP: 0033:0x7f892834e85d
[252149.774377] RSP: 002b:00007f87203aaa50 EFLAGS: 00000293 ORIG_RAX:
0000000000000002 [252149.775700] RAX: ffffffffffffffda RBX:
0000000000000241 RCX: 00007f892834e85d [252149.777059] RDX:
0000000000000180 RSI: 0000000000000241 RDI: 00007f8732bf4d08
[252149.778385] RBP: 00007f8732bf4d08 R08: 00007f89272290e0 R09:
00007f8732bf4d08 [252149.779759] R10: 00007f8732bfd0d0 R11:
0000000000000293 R12: 0000000000000180 [252149.781059] R13:
00007f87203aab70 R14: 00007f891d350f40 R15: 0000000000000001
[252149.782408] INFO: task CacheThread_Blo:8552 blocked for more than 480
seconds. [252149.783727]       Not tainted 4.14.146-nopreempt-AMDGPU-dav9
#1 [252149.785086] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs"
disables this message. [252149.786521] CacheThread_Blo D    0  8552
8293 0x00000080 [252149.786522] Call Trace: [252149.787842]  ?
__schedule+0x29e/0x6c0 [252149.789169]  schedule+0x32/0x80
[252149.790581]  schedule_preempt_disabled+0xa/0x10 [252149.791905]
__mutex_lock.isra.1+0x26b/0x4e0 [252149.793221]  ?
do_journal_begin_r+0xbe/0x390 [reiserfs] [252149.794544]
do_journal_begin_r+0xbe/0x390 [reiserfs] [252149.795932]  ?
__switch_to_asm+0x35/0x70 [252149.797304]  ? __switch_to_asm+0x41/0x70
[252149.798618]  ? __switch_to_asm+0x35/0x70
[252149.800025]  ? __switch_to_asm+0x41/0x70
[252149.801381]  ? __switch_to_asm+0x35/0x70
[252149.802674]  ? __switch_to_asm+0x41/0x70
[252149.803983]  ? __switch_to_asm+0x35/0x70
[252149.805257]  journal_begin+0x80/0x140 [reiserfs]
[252149.806659]  reiserfs_dirty_inode+0x3d/0xa0 [reiserfs]
[252149.807935]  ? __switch_to+0x1ee/0x3f0
[252149.809317]  ? __switch_to+0x1ee/0x3f0
[252149.810709]  __mark_inode_dirty+0x163/0x350
[252149.812074]  generic_update_time+0x79/0xc0
[252149.813416]  ? current_time+0x38/0x70
[252149.814831]  file_update_time+0xbe/0x110
[252149.816272]  __generic_file_write_iter+0x99/0x1b0
[252149.817596]  generic_file_write_iter+0xe2/0x1c0
[252149.818920]  __vfs_write+0x102/0x180
[252149.820324]  vfs_write+0xb0/0x190
[252149.821693]  SyS_pwrite64+0x90/0xb0
[252149.822998]  do_syscall_64+0x6e/0x110
[252149.824280]  entry_SYSCALL_64_after_hwframe+0x3d/0xa2
[252149.825630] RIP: 0033:0x7f45af7db983
[252149.826955] RSP: 002b:00007f4594c43620 EFLAGS: 00000293 ORIG_RAX:
0000000000000012 [252149.828200] RAX: ffffffffffffffda RBX:
00007f4594c43630 RCX: 00007f45af7db983 [252149.829558] RDX:
0000000000000128 RSI: 00000eb793a78e00 RDI: 0000000000000037
[252149.830889] RBP: 00007f4594c436b0 R08: 0000000000000008 R09:
00000eb793a78e90 [252149.832277] R10: 0000000000002000 R11:
0000000000000293 R12: 0000000000002000 [252149.833606] R13:
00000eb793a90c90 R14: 00000eb793a78e00 R15: 0000000000000128
[252149.834982] INFO: task Chrome_HistoryT:8874 blocked for more than 480
seconds. [252149.836374]       Not tainted 4.14.146-nopreempt-AMDGPU-dav9
#1 [252149.837737] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs"
disables this message. [252149.839077] Chrome_HistoryT D    0  8874
8293 0x00000080 [252149.839078] Call Trace: [252149.840406]  ?
__schedule+0x29e/0x6c0 [252149.841751]  schedule+0x32/0x80
[252149.843090]  schedule_preempt_disabled+0xa/0x10 [252149.844389]
__mutex_lock.isra.1+0x26b/0x4e0 [252149.845793]  ?
flush_commit_list+0x1fa/0x670 [reiserfs] [252149.847226]
flush_commit_list+0x1fa/0x670 [reiserfs] [252149.848600]
flush_commit_list+0x11f/0x670 [reiserfs] [252149.850009]
do_journal_end+0xc81/0xda0 [reiserfs] [252149.851306]
reiserfs_commit_for_inode+0x201/0x220 [reiserfs] [252149.852715]
reiserfs_sync_file+0x6b/0xe0 [reiserfs] [252149.854058]
do_fsync+0x38/0x60 [252149.855435]  SyS_fdatasync+0xf/0x20
[252149.856832]  do_syscall_64+0x6e/0x110
[252149.858155]  entry_SYSCALL_64_after_hwframe+0x3d/0xa2
[252149.859480] RIP: 0033:0x7f45a970f84d
[252149.860805] RSP: 002b:00007f45929e00e0 EFLAGS: 00000293 ORIG_RAX:
000000000000004b [252149.862149] RAX: ffffffffffffffda RBX:
00000eb795860e80 RCX: 00007f45a970f84d [252149.863511] RDX:
0000000000000008 RSI: 0000000000000002 RDI: 0000000000000257
[252149.864771] RBP: 00007f45929e0120 R08: 0000000000000000 R09:
0000000000001fdf [252149.866183] R10: 0000000000002400 R11:
0000000000000293 R12: 0000000000000000 [252149.867558] R13:
d763a120f905d5d9 R14: 0000000000000000 R15: 0000000000001000
[252149.868933] INFO: task ThreadPoolSingl:8875 blocked for more than 480
seconds. [252149.870329]       Not tainted 4.14.146-nopreempt-AMDGPU-dav9
#1 [252149.871652] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs"
disables this message. [252149.873063] ThreadPoolSingl D    0  8875
8293 0x00000080 [252149.873064] Call Trace: [252149.874335]  ?
__schedule+0x29e/0x6c0 [252149.875688]  schedule+0x32/0x80
[252149.877028]  schedule_preempt_disabled+0xa/0x10 [252149.878409]
__mutex_lock.isra.1+0x26b/0x4e0 [252149.879748]  ?
do_journal_begin_r+0xbe/0x390 [reiserfs] [252149.881139]
do_journal_begin_r+0xbe/0x390 [reiserfs] [252149.882572]  ?
reiserfs_lookup+0xb5/0x160 [reiserfs] [252149.883968]
journal_begin+0x80/0x140 [reiserfs] [252149.885272]
reiserfs_create+0xfc/0x210 [reiserfs] [252149.886663]
path_openat+0x13e1/0x1490 [252149.888017]  ? futex_wake+0x91/0x170
[252149.889372]  do_filp_open+0x99/0x110
[252149.890824]  ? __check_object_size+0xfa/0x1a0
[252149.892220]  ? __alloc_fd+0x3d/0x160
[252149.893644]  ? do_sys_open+0x12e/0x210
[252149.894914]  do_sys_open+0x12e/0x210
[252149.896294]  do_syscall_64+0x6e/0x110
[252149.897633]  entry_SYSCALL_64_after_hwframe+0x3d/0xa2
[252149.898986] RIP: 0033:0x7f45a970970d
[252149.900339] RSP: 002b:00007f459213f2c0 EFLAGS: 00000293 ORIG_RAX:
0000000000000002 [252149.901757] RAX: ffffffffffffffda RBX:
0000000000000000 RCX: 00007f45a970970d [252149.903135] RDX:
0000000000000180 RSI: 00000000000000c2 RDI: 00000eb7ae7f17e0
[252149.904463] RBP: 000000000003a2f8 R08: 0000000000000001 R09:
00000eb7ad9bbac0 [252149.905792] R10: 0000000000000000 R11:
0000000000000293 R12: 00000eb7ae7f181a [252149.907105] R13:
8421084210842109 R14: 00000000000000c2 R15: 00007f45a9797540
[252149.908413] INFO: task ThreadPoolForeg:28571 blocked for more than
480 seconds. [252149.909742]       Not tainted
4.14.146-nopreempt-AMDGPU-dav9 #1 [252149.911031] "echo 0
> /proc/sys/kernel/hung_task_timeout_secs" disables this message.
> [252149.912453] ThreadPoolForeg D    0 28571   8293 0x00000080
> [252149.912467] Call Trace: [252149.913820]  ? __schedule+0x29e/0x6c0
[252149.915128]  ? __switch_to_asm+0x35/0x70
[252149.916585]  schedule+0x32/0x80
[252149.917895]  schedule_preempt_disabled+0xa/0x10
[252149.919332]  __mutex_lock.isra.1+0x26b/0x4e0
[252149.920746]  ? __switch_to_asm+0x35/0x70
[252149.922180]  ? do_journal_begin_r+0xbe/0x390 [reiserfs]
[252149.923613]  do_journal_begin_r+0xbe/0x390 [reiserfs]
[252149.925010]  ? __remove_hrtimer+0x35/0x70
[252149.926474]  ? hrtimer_try_to_cancel+0xbd/0x110
[252149.927893]  journal_begin+0x80/0x140 [reiserfs]
[252149.929282]  reiserfs_dirty_inode+0x3d/0xa0 [reiserfs]
[252149.930623]  ? get_futex_key+0x4a0/0x4d0
[252149.932006]  __mark_inode_dirty+0x163/0x350
[252149.933434]  generic_update_time+0x79/0xc0
[252149.934769]  ? current_time+0x38/0x70
[252149.936158]  file_update_time+0xbe/0x110
[252149.937490]  __generic_file_write_iter+0x99/0x1b0
[252149.938936]  generic_file_write_iter+0xe2/0x1c0
[252149.940271]  __vfs_write+0x102/0x180
[252149.941618]  vfs_write+0xb0/0x190
[252149.943012]  SyS_write+0x5a/0xd0
[252149.944342]  do_syscall_64+0x6e/0x110
[252149.945726]  entry_SYSCALL_64_after_hwframe+0x3d/0xa2
[252149.947056] RIP: 0033:0x7f45af7db1cd
[252149.948402] RSP: 002b:00007f456a563af0 EFLAGS: 00000293 ORIG_RAX:
0000000000000001 [252149.949785] RAX: ffffffffffffffda RBX:
00007f456a563b08 RCX: 00007f45af7db1cd [252149.951154] RDX:
0000000000000007 RSI: 00007f456a563c18 RDI: 00000000000000f4
[252149.952596] RBP: 00007f456a563b80 R08: 0000000000000000 R09:
00005641a9f0b790 [252149.953937] R10: 0000cb689851cb28 R11:
0000000000000293 R12: 00000eb7aeb1d650 [252149.955335] R13:
00007f456a563c18 R14: 0000000000000007 R15: 0000000000000000
[252149.956727] INFO: task ThreadPoolForeg:27436 blocked for more than
480 seconds. [252149.958056]       Not tainted
4.14.146-nopreempt-AMDGPU-dav9 #1 [252149.959491] "echo 0
> /proc/sys/kernel/hung_task_timeout_secs" disables this message.
> [252149.960820] ThreadPoolForeg D    0 27436   8514 0x00000080
> [252149.960821] Call Trace: [252149.962253]  ? __schedule+0x29e/0x6c0
[252149.963600]  ? __switch_to_asm+0x41/0x70
[252149.964933]  schedule+0x32/0x80
[252149.966441]  schedule_preempt_disabled+0xa/0x10
[252149.967768]  __mutex_lock.isra.1+0x26b/0x4e0
[252149.969162]  ? __switch_to_asm+0x41/0x70
[252149.970492]  ? find_get_entry+0x21/0x180
[252149.971860]  ? flush_commit_list+0x1fa/0x670 [reiserfs]
[252149.973289]  flush_commit_list+0x1fa/0x670 [reiserfs]
[252149.974621]  get_list_bitmap+0x81/0xd0 [reiserfs]
[252149.976016]  do_journal_end+0xbce/0xda0 [reiserfs]
[252149.977414]  reiserfs_commit_for_inode+0x201/0x220 [reiserfs]
[252149.978813]  reiserfs_sync_file+0x6b/0xe0 [reiserfs]
[252149.980157]  do_fsync+0x38/0x60
[252149.981488]  SyS_fdatasync+0xf/0x20
[252149.982883]  do_syscall_64+0x6e/0x110
[252149.984191]  entry_SYSCALL_64_after_hwframe+0x3d/0xa2
[252149.985592] RIP: 0033:0x7f11f546884d
[252149.986901] RSP: 002b:00007f11e8f2a430 EFLAGS: 00000293 ORIG_RAX:
000000000000004b [252149.988234] RAX: ffffffffffffffda RBX:
00007f11e8f2a440 RCX: 00007f11f546884d [252149.989700] RDX:
00001c4f0a409e28 RSI: 0000000000000000 RDI: 0000000000000054
[252149.991098] RBP: 00007f11e8f2a490 R08: 0000000000000000 R09:
00001c4f0f5ea0c0 [252149.992484] R10: 00347354bacab6d5 R11:
0000000000000293 R12: 0000000000000000 [252149.993861] R13:
000000000002276a R14: 00007f11e8f2a528 R15: 00007f11e8f2a528
[252149.995216] INFO: task ThreadPoolForeg:29152 blocked for more than
480 seconds. [252149.996626]       Not tainted
4.14.146-nopreempt-AMDGPU-dav9 #1 [252149.997955] "echo 0
> /proc/sys/kernel/hung_task_timeout_secs" disables this message.
> [252149.999431] ThreadPoolForeg D    0 29152   8514 0x00000080
> [252149.999432] Call Trace: [252150.000867]  ? __schedule+0x29e/0x6c0
[252150.002282]  ? __switch_to_asm+0x41/0x70
[252150.003708]  schedule+0x32/0x80
[252150.005064]  schedule_preempt_disabled+0xa/0x10
[252150.006459]  __mutex_lock.isra.1+0x26b/0x4e0
[252150.007795]  ? __switch_to_asm+0x41/0x70
[252150.009192]  ? flush_commit_list+0x1fa/0x670 [reiserfs]
[252150.010638]  flush_commit_list+0x1fa/0x670 [reiserfs]
[252150.012050]  flush_commit_list+0x11f/0x670 [reiserfs]
[252150.013390]  do_journal_end+0xc81/0xda0 [reiserfs]
[252150.014734]  reiserfs_commit_for_inode+0x201/0x220 [reiserfs]
[252150.016196]  reiserfs_sync_file+0x6b/0xe0 [reiserfs]
[252150.017549]  do_fsync+0x38/0x60
[252150.018961]  SyS_fdatasync+0xf/0x20
[252150.020311]  do_syscall_64+0x6e/0x110
[252150.021670]  entry_SYSCALL_64_after_hwframe+0x3d/0xa2
[252150.023047] RIP: 0033:0x7f11f546884d
[252150.024376] RSP: 002b:00007f11e7726ea0 EFLAGS: 00000293 ORIG_RAX:
000000000000004b [252150.025788] RAX: ffffffffffffffda RBX:
00001c4f0af8fa00 RCX: 00007f11f546884d [252150.027150] RDX:
0000000000000008 RSI: 0000000000000002 RDI: 000000000000004f
[252150.028584] RBP: 00007f11e7726ee0 R08: 0000000000000001 R09:
0000000000001fdf [252150.029981] R10: 0000000000001000 R11:
0000000000000293 R12: 0000000000000000 [252150.031334] R13:
d763a120f905d5d9 R14: 0000000000000000 R15: 0000000000001000
[252207.504644] elogind[3860]: Removed session 5. [252212.001258]
elogind[3860]: Removed session 4. [252216.542880] elogind[3860]: Removed
session 6.

^ permalink raw reply	[flat|nested] only message in thread

only message in thread, other threads:[~2019-10-24  2:17 UTC | newest]

Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-10-24  2:16 Bug report: Kernel IO lockup. Hard reboot required David Niklas

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.