You are not logged in.

#1 2021-01-08 11:01:19

TheArc07
Member
Registered: 2021-01-08
Posts: 3

[SOLVED] System Freeze - task kworker blocked for more than x seconds

I recently installed Arch with btrfs on my SSD and I have been experiencing system freezes at random times. The display just does not update and any audio playing stops after a few seconds. None of the keyboard keys seem to work so unable to drop in to a different TTY. I guess the problem may be related to either my btrfs configuration or my SSD since I have another Arch install on my HDD with ext4 which I have been using for quite some time and it has never froze. I am attaching the journalctl output from the time I believe the freeze started and then I had no option but to do a hard shutdown. Please let me know if I need to provide any additional details to pinpoint the issue.

Kernel

5.10.4-arch2-1

Journalctl output

Jan 08 12:09:08 arch systemd[1]: Starting DBus interface for snapper...
Jan 08 12:09:08 arch dbus-daemon[369]: [system] Successfully activated service 'org.opensuse.Snapper'
Jan 08 12:09:08 arch audit[1]: SERVICE_START pid=1 uid=0 auid=4294967295 ses=4294967295 msg='unit=snapperd comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=success'
Jan 08 12:09:08 arch systemd[1]: Started DBus interface for snapper.
Jan 08 12:09:08 arch kernel: audit: type=1130 audit(1610087948.105:158): pid=1 uid=0 auid=4294967295 ses=4294967295 msg='unit=snapperd comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=success'
Jan 08 12:09:08 arch kernel: BTRFS warning (device sdb2): qgroup rescan init failed, qgroup is not enabled
Jan 08 12:10:01 arch wpa_supplicant[424]: wlp2s0: Reject scan trigger since one is already pending
Jan 08 12:10:08 arch systemd[1]: snapperd.service: Succeeded.
Jan 08 12:10:08 arch audit[1]: SERVICE_STOP pid=1 uid=0 auid=4294967295 ses=4294967295 msg='unit=snapperd comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=success'
Jan 08 12:10:08 arch kernel: audit: type=1131 audit(1610088008.156:159): pid=1 uid=0 auid=4294967295 ses=4294967295 msg='unit=snapperd comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=success'
Jan 08 12:27:42 arch kernel: INFO: task kworker/1:1H:117 blocked for more than 122 seconds.
Jan 08 12:27:42 arch kernel:       Tainted: P           OE     5.10.4-arch2-1 #1
Jan 08 12:27:42 arch kernel: "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
Jan 08 12:27:42 arch kernel: task:kworker/1:1H    state:D stack:    0 pid:  117 ppid:     2 flags:0x00004000
Jan 08 12:27:42 arch kernel: Workqueue: events_highpri intel_atomic_cleanup_work [i915]
Jan 08 12:27:42 arch kernel: Call Trace:
Jan 08 12:27:42 arch kernel:  __schedule+0x295/0x810
Jan 08 12:27:42 arch kernel:  schedule+0x5b/0xc0
Jan 08 12:27:42 arch kernel:  schedule_preempt_disabled+0x11/0x20
Jan 08 12:27:42 arch kernel:  __ww_mutex_lock.constprop.0+0x4bd/0x810
Jan 08 12:27:42 arch kernel:  ? dequeue_entity+0xc6/0x460
Jan 08 12:27:42 arch kernel:  intel_unpin_fb_vma+0x25/0xa0 [i915]
Jan 08 12:27:42 arch kernel:  drm_atomic_helper_cleanup_planes+0x52/0x70 [drm_kms_helper]
Jan 08 12:27:42 arch kernel:  intel_atomic_cleanup_work+0x67/0x110 [i915]
Jan 08 12:27:42 arch kernel:  process_one_work+0x1d6/0x3a0
Jan 08 12:27:42 arch kernel:  worker_thread+0x4d/0x3d0
Jan 08 12:27:42 arch kernel:  ? rescuer_thread+0x410/0x410
Jan 08 12:27:42 arch kernel:  kthread+0x133/0x150
Jan 08 12:27:42 arch kernel:  ? __kthread_bind_mask+0x60/0x60
Jan 08 12:27:42 arch kernel:  ret_from_fork+0x22/0x30
Jan 08 12:29:45 arch kernel: INFO: task kworker/1:1H:117 blocked for more than 245 seconds.
Jan 08 12:29:45 arch kernel:       Tainted: P           OE     5.10.4-arch2-1 #1
Jan 08 12:29:45 arch kernel: "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
Jan 08 12:29:45 arch kernel: task:kworker/1:1H    state:D stack:    0 pid:  117 ppid:     2 flags:0x00004000
Jan 08 12:29:45 arch kernel: Workqueue: events_highpri intel_atomic_cleanup_work [i915]
Jan 08 12:29:45 arch kernel: Call Trace:
Jan 08 12:29:45 arch kernel:  __schedule+0x295/0x810
Jan 08 12:29:45 arch kernel:  schedule+0x5b/0xc0
Jan 08 12:29:45 arch kernel:  schedule_preempt_disabled+0x11/0x20
Jan 08 12:29:45 arch kernel:  __ww_mutex_lock.constprop.0+0x4bd/0x810
Jan 08 12:29:45 arch kernel:  ? dequeue_entity+0xc6/0x460
Jan 08 12:29:45 arch kernel:  intel_unpin_fb_vma+0x25/0xa0 [i915]
Jan 08 12:29:45 arch kernel:  drm_atomic_helper_cleanup_planes+0x52/0x70 [drm_kms_helper]
Jan 08 12:29:45 arch kernel:  intel_atomic_cleanup_work+0x67/0x110 [i915]
Jan 08 12:29:45 arch kernel:  process_one_work+0x1d6/0x3a0
Jan 08 12:29:45 arch kernel:  worker_thread+0x4d/0x3d0
Jan 08 12:29:45 arch kernel:  ? rescuer_thread+0x410/0x410
Jan 08 12:29:45 arch kernel:  kthread+0x133/0x150
Jan 08 12:29:45 arch kernel:  ? __kthread_bind_mask+0x60/0x60
Jan 08 12:29:45 arch kernel:  ret_from_fork+0x22/0x30

Smartctl -a output for SSD

=== START OF INFORMATION SECTION ===
Model Family:     Crucial/Micron Client SSDs
Device Model:     CT240BX500SSD1
Serial Number:    1938E19AE458
LU WWN Device Id: 0 000000 000000000
Firmware Version: M6CR013
User Capacity:    240,057,409,536 bytes [240 GB]
Sector Size:      512 bytes logical/physical
Rotation Rate:    Solid State Device
Form Factor:      2.5 inches
TRIM Command:     Available, deterministic, zeroed
Device is:        In smartctl database [for details use: -P show]
ATA Version is:   ACS-2 T13/2015-D revision 3
SATA Version is:  SATA 3.2, 6.0 Gb/s (current: 6.0 Gb/s)
Local Time is:    Fri Jan  8 16:19:43 2021 IST
SMART support is: Available - device has SMART capability.
SMART support is: Enabled

=== START OF READ SMART DATA SECTION ===
SMART overall-health self-assessment test result: PASSED

Btrfs layout and mount options

/               /dev/sdb2[/@root]       btrfs  rw,noatime,compress=zstd:3,ssd,space_cache,commit=120,subvolid=256,subvol=/@root
├─/home         /dev/sdb2[/@home]       btrfs  rw,noatime,compress=zstd:3,ssd,space_cache,commit=120,subvolid=257,subvol=/@home
│ └─/home/games /dev/sdb2[/@games]      btrfs  rw,noatime,compress=zstd:3,ssd,space_cache,commit=120,subvolid=260,subvol=/@games
├─/tmp          /dev/sdb2[/@tmp]        btrfs  rw,noatime,compress=zstd:3,ssd,space_cache,commit=120,subvolid=258,subvol=/@tmp
└─/.snapshots   /dev/sdb2[/@.snapshots] btrfs  rw,noatime,compress=zstd:3,ssd,space_cache,commit=120,subvolid=259,subvol=/@.snapshots

I would also like to add that I have a Nvidia GPU (GeForce 940MX) and have nvidia package installed (version 455.45.01)

Last edited by TheArc07 (2021-01-11 05:03:05)

Offline

#2 2021-01-08 11:15:21

V1del
Forum Moderator
Registered: 2012-10-16
Posts: 21,425

Re: [SOLVED] System Freeze - task kworker blocked for more than x seconds

Remove xf86-video-intel and any configurations using it and check whether you can reproduce.

Offline

#3 2021-01-08 12:02:54

TheArc07
Member
Registered: 2021-01-08
Posts: 3

Re: [SOLVED] System Freeze - task kworker blocked for more than x seconds

Thanks for the reply! Uninstalled xf86-video-intel with -Rsn options, do I have to remove anything else?

Offline

#4 2021-01-08 12:45:14

V1del
Forum Moderator
Registered: 2012-10-16
Posts: 21,425

Re: [SOLVED] System Freeze - task kworker blocked for more than x seconds

no, xorg should automatically fall back to the  modesetting driver, which tends to be more stable on modern chipsets.

Offline

#5 2021-01-08 14:52:54

lervag
Member
Registered: 2013-11-04
Posts: 25

Re: [SOLVED] System Freeze - task kworker blocked for more than x seconds

I have the same problem, I think. I get errors like this in my "sudo journalctl -b-1" output:

Jan 08 15:14:32 lotti kernel: INFO: task kworker/1:1H:226 blocked for more than 122 seconds.
Jan 08 15:14:32 lotti kernel:       Tainted: G S   U     OE     5.10.4-arch2-1 #1
Jan 08 15:14:32 lotti kernel: "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
Jan 08 15:14:32 lotti kernel: task:kworker/1:1H    state:D stack:    0 pid:  226 ppid:     2 flags:0x00004000
Jan 08 15:14:32 lotti kernel: Workqueue: events_highpri intel_atomic_cleanup_work [i915]
Jan 08 15:14:32 lotti kernel: Call Trace:
Jan 08 15:14:32 lotti kernel:  __schedule+0x295/0x810
Jan 08 15:14:32 lotti kernel:  schedule+0x5b/0xc0
Jan 08 15:14:32 lotti kernel:  schedule_preempt_disabled+0x11/0x20
Jan 08 15:14:32 lotti kernel:  __ww_mutex_lock.constprop.0+0x4bd/0x810
Jan 08 15:14:32 lotti kernel:  intel_unpin_fb_vma+0x25/0xa0 [i915]
Jan 08 15:14:32 lotti kernel:  drm_atomic_helper_cleanup_planes+0x52/0x70 [drm_kms_helper]
Jan 08 15:14:32 lotti kernel:  intel_atomic_cleanup_work+0x67/0x110 [i915]
Jan 08 15:14:32 lotti kernel:  process_one_work+0x1d6/0x3a0
Jan 08 15:14:32 lotti kernel:  worker_thread+0x4d/0x3d0
Jan 08 15:14:32 lotti kernel:  ? rescuer_thread+0x410/0x410
Jan 08 15:14:32 lotti kernel:  kthread+0x133/0x150
Jan 08 15:14:32 lotti kernel:  ? __kthread_bind_mask+0x60/0x60
Jan 08 15:14:32 lotti kernel:  ret_from_fork+0x22/0x30

I'll try the suggestion by V1del, but I'm curious what happened that might have triggered this. I don't think I've changed anything on my side, so it might be introduced by the kernel? It is the third time it happens, I think, and the first time was within the last couple of weeks.

Offline

#6 2021-01-08 15:28:03

seth
Member
Registered: 2012-09-03
Posts: 49,981

Re: [SOLVED] System Freeze - task kworker blocked for more than x seconds

https://bbs.archlinux.org/viewtopic.php?id=262283
https://gitlab.freedesktop.org/drm/intel/-/issues/2905 (linked there)
https://bbs.archlinux.org/viewtopic.php?id=254942

xf86-video-intel is some sort of zombie mode (I lost track on whether it's currently being phased ot or revived) but the general projection is to move fuctionality into the kernel module and use the generic modesetting interface.

Offline

#7 2021-01-08 15:57:55

V1del
Forum Moderator
Registered: 2012-10-16
Posts: 21,425

Re: [SOLVED] System Freeze - task kworker blocked for more than x seconds

Well if you just go by commit activity there always seems to be "something" that's happening, but if you look at the actual commits they basically never contain any substantial changes or if they do they are reverted a few days after because something broke. It's definitely a weird limbo it's in. The way I see it it's main merit is for GPUs that are so old that they can't do modesetting. Anything newer and you are just setting yourself up for weirdness.

Offline

#8 2021-01-08 16:51:40

lervag
Member
Registered: 2013-11-04
Posts: 25

Re: [SOLVED] System Freeze - task kworker blocked for more than x seconds

Ok, so, if I understand correctly you are saying I should not need to install xf86-video-intel on modern machines (such as my Thinkpad X1 Carbon 6th gen). That's useful to know, thanks.

I have the same issue mentioned here that my computer froze during boot; I'm trying to add the boot options now to see if it helps.

Edit: No, still no luck. Things freeze on startup. Not sure why; I get the
following output in "journalctl -b-1", not sure if it makes anyone wiser...

Jan 08 17:52:10 lotti sddm[599]: Initializing...
Jan 08 17:52:10 lotti sddm[599]: Starting...
Jan 08 17:52:10 lotti sddm[599]: Logind interface found
Jan 08 17:52:10 lotti sddm[599]: Adding new display on vt 1 ...
Jan 08 17:52:10 lotti sddm[599]: Loading theme configuration from ""
Jan 08 17:52:10 lotti sddm[599]: Display server starting...
Jan 08 17:52:10 lotti sddm[599]: Adding cookie to "/var/run/sddm/{ab68acfe-cee6-4998-975d-e5e3106f25b2}"
Jan 08 17:52:10 lotti sddm[599]: Running: /usr/bin/X -nolisten tcp -background none -seat seat0 vt1 -auth /var/run/sddm/{ab68acfe-cee6-4998-975d-e5e3106f25b2} -noreset -displayfd 17
... clip ...
Jan 08 17:52:10 lotti systemd[1]: Reached target Multi-User System.
Jan 08 17:52:10 lotti systemd[1]: Reached target Graphical Interface.
...
Jan 08 17:52:10 lotti sddm[599]: Failed to read display number from pipe
Jan 08 17:52:10 lotti sddm[599]: Display server stopping...
Jan 08 17:52:10 lotti sddm[599]: Attempt 1 starting the Display server on vt 1 failed

Last edited by lervag (2021-01-08 17:03:33)

Offline

#9 2021-01-08 17:17:27

V1del
Forum Moderator
Registered: 2012-10-16
Posts: 21,425

Re: [SOLVED] System Freeze - task kworker blocked for more than x seconds

Freezes "on-startup" are/can be a different issue and a different ball park, those can usually be fixed by enabling early KMS: https://wiki.archlinux.org/index.php/Ke … _KMS_start

Last edited by V1del (2021-01-08 17:17:52)

Offline

#10 2021-01-08 17:54:02

lervag
Member
Registered: 2013-11-04
Posts: 25

Re: [SOLVED] System Freeze - task kworker blocked for more than x seconds

Thanks! That also did not work, but I noticed that `Xorg.0.log` said that it did not find the "intel" driver. Then I noticed I had earlier created `/etc/X11/xorg..confd/20-intel.conf`. Blah. When I removed that, things now work. Sorry for the noise!

And thanks for the quick and helpful comments/answers!

Now I'm crossing my fingers that this also fixes my original issue smile

Offline

#11 2021-01-10 13:14:45

vsraier
Member
Registered: 2021-01-10
Posts: 2
Website

Re: [SOLVED] System Freeze - task kworker blocked for more than x seconds

Hi, I've got the same issue as OP with the exact same call trace in system log. I isolated to issue to kernel, because it never happens with LTS kernel, but it always hangs with both "linux" and "linux-zen" kernels. Switching to LTS might therefore be a short term workaround.

As of now, I have uninstalled the "xf86-video-intel" package from my system and I hope it will fix the issue. Was that fix successful for anybody?

Offline

#12 2021-01-10 20:00:23

lervag
Member
Registered: 2013-11-04
Posts: 25

Re: [SOLVED] System Freeze - task kworker blocked for more than x seconds

@vsraier It is difficult to say, because there is no way to specifically reproduce the issue. It happened perhaps between 3 to 5 times on my computer the last 2-3 weeks, I think. I can only say that I have not had it occuring after removing `xf86-video-intel`, which is only partial evidence of success.

Offline

#13 2021-01-10 21:14:12

vsraier
Member
Registered: 2021-01-10
Posts: 2
Website

Re: [SOLVED] System Freeze - task kworker blocked for more than x seconds

@lervag Ok, thanks for the info.

On my laptop, the crashes were a bit more common. Let's say it happened about 4 times during a 3h video call. I could not get the system to hang reliably, but it happened often enough to be very annoying. Other symptom of the same issue was probably that some windows kept freezing. Especially the Alacritty terminal stopped rendering it's window (I am running i3wm). Firefox, for example, was always fine, so not every application using the GPU was affected.

Since I've uninstalled the `xf86-video-intel` package, I haven't seen any issues whatsoever. No freezing windows, no system hangs. The issue is very likely fixed. So thank you @V1del for spotting the issue with such limited information. smile

Offline

#14 2021-01-11 05:02:39

TheArc07
Member
Registered: 2021-01-08
Posts: 3

Re: [SOLVED] System Freeze - task kworker blocked for more than x seconds

Yeah, I have also not experienced any freezes since I uninstalled the 'xf86-video-intel' package. I checked on my other arch install and the 'xf86-video-intel' package was not installed so it's probably this package that was causing the issue.

Thank you @V1del and @seth! smile

Offline

#15 2021-01-12 15:30:55

beachcoder
Member
Registered: 2009-10-08
Posts: 130

Re: [SOLVED] System Freeze - task kworker blocked for more than x seconds

Hi there! I had the same issue, and I uninstalled xf86-video-intel as well in hope that the problem doesn't raise again.

Anyway,

journalctl -b-1

didn't show anything from intel as it is the case in the previous posts, therefore I'm not sure if this is the solution which also works for me.

Last edited by beachcoder (2021-01-12 15:31:42)

Offline

#16 2021-09-12 17:31:42

lorn10
Member
Registered: 2021-03-10
Posts: 3

Re: [SOLVED] System Freeze - task kworker blocked for more than x seconds

Sorry to warm up this solved topic but here follows my solution regarding the same problem. Maybe this will be some kind of an alternative help for other users. ;-)

I had this error "INFO: task kworker blocked for more than 120 seconds." at an older iMac computer. It turns out that there is a problem with the GLX interface (in conjunction with older Radeon GPU models) which produces a high "overhead" or "system load" which effectively blocked the whole KDE environment (X11) for around 2 Minutes. This always happened right after the login when the "notification message" about the WiFi status was displayed.

The solution was in my case to switch the whole compositing from OpenGL / GLX to OpenGL ES / EGL. This can be done through a manual setting in the corresponding config file.

So I added the following line:

export KWIN_COMPOSE=O2ES

To the following file:

sudo nano /etc/profile.d/kwin.sh

And everything is working now absolutely perfectly! I can confirm a really big performance improvement in almost all parts of the UI behavior. Even Firefox which uses software based Webrender runs smoother, awesome.

Offline

Board footer

Powered by FluxBB