Desktop / system frequently unresponsive

Celsiuss · 2026-03-13 13:14:18

I've had this problem today and yesterday. The system sporadically becomes unresponsive.

Some symptoms:
- Boot takes longer, or hangs indefinitely
- Plasma goes into a black screen after login (with the mouse pointer showing and moving)
- Graphical applications never launch, or takes up to a couple of minutes to launch. CLI only applications seem to launch normally.
- Shutdown sequence hangs forever

Using plasma (plasma-desktop 6.6.2), linux 6.19.6, plasma-login-manager 6.6.2. Wayland session
Machine is a Lenovo T16 Gen 2.
AMD Ryzen 7 PRO 7840U w/ Radeon 780M Graphics

glxinfo, this also took a good minute or two to run

$ glxinfo -B
name of display: :0
display: :0  screen: 0
direct rendering: Yes
Extended renderer info (GLX_MESA_query_renderer):
    Vendor: AMD (0x1002)
    Device: AMD Radeon 780M Graphics (radeonsi, phoenix, ACO, DRM 3.64, 6.19.6-arch1-1) (0x15bf)
    Version: 26.0.1
    Accelerated: yes
    Video memory: 1024MB
    Unified memory: no
    Preferred profile: core (0x1)
    Max core profile version: 4.6
    Max compat profile version: 4.6
    Max GLES1 profile version: 1.1
    Max GLES[23] profile version: 3.2
Memory info (GL_ATI_meminfo):
    VBO free memory - total: 233 MB, largest block: 233 MB
    VBO free aux. memory - total: 14173 MB, largest block: 14173 MB
    Texture free memory - total: 233 MB, largest block: 233 MB
    Texture free aux. memory - total: 14173 MB, largest block: 14173 MB
    Renderbuffer free memory - total: 233 MB, largest block: 233 MB
    Renderbuffer free aux. memory - total: 14173 MB, largest block: 14173 MB
Memory info (GL_NVX_gpu_memory_info):
    Dedicated video memory: 1024 MB
    Total available memory: 16384 MB
    Currently available dedicated video memory: 233 MB
OpenGL vendor string: AMD
OpenGL renderer string: AMD Radeon 780M Graphics (radeonsi, phoenix, ACO, DRM 3.64, 6.19.6-arch1-1)
OpenGL core profile version string: 4.6 (Core Profile) Mesa 26.0.1-arch1.1
OpenGL core profile shading language version string: 4.60
OpenGL core profile context flags: (none)
OpenGL core profile profile mask: core profile

OpenGL version string: 4.6 (Compatibility Profile) Mesa 26.0.1-arch1.1
OpenGL shading language version string: 4.60
OpenGL context flags: (none)
OpenGL profile mask: compatibility profile

OpenGL ES profile version string: OpenGL ES 3.2 Mesa 26.0.1-arch1.1
OpenGL ES profile shading language version string: OpenGL ES GLSL ES 3.20

System logs from when the system was completely unresponsive: https://0x0.st/P_y-.txt

Lines I notice frequently now are ones such as:

INFO: task kworker/14:0:63 blocked for more than 122 seconds.

With this below the trace:

INFO: task kworker/14:0:63 is blocked on a mutex likely owned by task kworker/R-kbloc:126.

These seem to be very frequent during complete system hang/unresponsiveness.

I do not see anything interesting while waiting for an application to launch. Here are logs from when launching system settings, which took nearly 2 minutes:

mars 13 14:03:50 arch-T16 systemd[1272]: Started System Settings - System Settings.
mars 13 14:05:10 arch-T16 systemd[1272]: app-org.kde.dolphin@219da0eb9f634a4c91b74cf7a36367f7.service: Consumed 1.345s CPU time over 2min 47.345s wall clock time, 120.5M memory peak.
mars 13 14:05:11 arch-T16 systemsettings[9736]: Failed to register with host portal QDBusError("org.freedesktop.portal.Error.Failed", "Could not register app ID: Connection already associated with an application ID")

What I have tried:
- Updating the system
- Reinstalling all packages (pacman -Qq | xargs pacman -S)
- memtest (passed)
- Investigating logs?

I don't know how to troubleshoot this. It started when I installed i3 for funsies, and when I launched it via plasma-login-manager, the UI just blackscreened. My assumption was that I uninstalled xorg some time ago.
I uninstalled i3, but since then my system has been very unreliable with these issues. But I assume the i3 hiccup and this issue is just a coincidence.

Last edited by Celsiuss (2026-03-13 14:38:21)

seth · 2026-03-13 14:26:26

Mar 13 12:56:36 arch-T16 kernel: nxp-nci_i2c i2c-NXP1001:00: NFC: Read failed with error -121
Mar 13 12:56:48 arch-T16 org_kde_powerdevil[1960]: Failed to find connector name for /dev/i2c-13, set_connector_for_businfo_using_edid at line 1411 in file i2c_bus_core.c.
Mar 13 12:56:48 arch-T16 org_kde_powerdevil[1960]: Failed to find connector name for /dev/i2c-13, set_connector_for_businfo_using_edid at line 1411 in file i2c_bus_core.c.

Did "I installed i3 for funsies" result in a kernel update or reboot?
Do you get this w/ the LTS kernel?

x-ref, https://bbs.archlinux.org/viewtopic.php?id=312592

Celsiuss · 2026-03-13 14:42:05

seth wrote:

Mar 13 12:56:36 arch-T16 kernel: nxp-nci_i2c i2c-NXP1001:00: NFC: Read failed with error -121
Mar 13 12:56:48 arch-T16 org_kde_powerdevil[1960]: Failed to find connector name for /dev/i2c-13, set_connector_for_businfo_using_edid at line 1411 in file i2c_bus_core.c.
Mar 13 12:56:48 arch-T16 org_kde_powerdevil[1960]: Failed to find connector name for /dev/i2c-13, set_connector_for_businfo_using_edid at line 1411 in file i2c_bus_core.c.

Did "I installed i3 for funsies" result in a kernel update or reboot?
Do you get this w/ the LTS kernel?

x-ref, https://bbs.archlinux.org/viewtopic.php?id=312592

i3 install only reboot.

Tried with LTS kernel, seems like symptoms are the same. Logs from one boot where it got stuck on a black-screen before reaching the login manager:
https://0x0.st/P_w5.txt

Last edited by Celsiuss (2026-03-13 14:43:04)

seth · 2026-03-13 14:59:49

There's a lot of network noise, since the worker thread is waiting for IO, what if you disable dockerd/containerd (any the virtual devices and bridges by inference)?

Celsiuss · 2026-03-13 16:51:50

seth wrote:

There's a lot of network noise, since the worker thread is waiting for IO, what if you disable dockerd/containerd (any the virtual devices and bridges by inference)?

Yeah you're right, should have done that from the start.
Here is another one stuck before I get to the login manager:
https://0x0.st/P_68.txt

seth · 2026-03-13 20:59:43

Here is another one stuck before I get to the login manager:

Mar 13 17:40:30 arch-T16 plasmalogin-helper[919]: pam_unix(plasmalogin-greeter:session): session opened for user plasmalogin(uid=955) by (uid=0)
Mar 13 17:44:00 arch-T16 systemd[1]: user@955.service: Main process exited, code=killed, status=9/KILL
Mar 13 17:44:26 arch-T16 kernel: INFO: task khugepaged:125 is blocked on a mutex likely owned by task (environment-ge:936.
Mar 13 17:44:26 arch-T16 kernel: INFO: task (environment-ge:936 blocked for more than 122 seconds.
Mar 13 17:44:26 arch-T16 kernel:       Not tainted 6.18.17-1-lts #1
Mar 13 17:44:26 arch-T16 kernel: "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
Mar 13 17:44:26 arch-T16 kernel: task:(environment-ge state:D stack:0     pid:936   tgid:936   ppid:1      task_flags:0x400040 flags:0x00080003
Mar 13 17:46:01 arch-T16 systemd[1]: user@955.service: Processes still around after final SIGKILL. Entering failed mode.
Mar 13 17:46:01 arch-T16 systemd[1]: user@955.service: Failed with result 'timeout'.
Mar 13 17:46:01 arch-T16 systemd[1]: Failed to start User Manager for UID 955.

Looks like it is the new plasmalogin thing, do you have any issues w/ sddm?

Celsiuss · 2026-03-14 13:49:26

seth wrote:

Looks like it is the new plasmalogin thing, do you have any issues w/ sddm?

Thanks, I switched back to sddm, and initially it seemed better, but after some reboots, issues are still present.
I made a few tries with both linux and linux-lts. It seems like it does not matter which one I use.

Some logs of when I experienced issues.

Latest linux, desktop launched but applications are slow/fails to launch, plus stuck in shutdown sequence:
https://0x0.st/PLHv.txt

linux-lts, black screen before I can see the login manager:
https://0x0.st/PLH0.txt

EDIT: corrected link

Last edited by Celsiuss (2026-03-14 13:54:40)

seth · 2026-03-14 15:23:13

Last journal has SDDM crash

Mar 14 14:37:50 arch-T16 sddm[873]: Auth: sddm-helper (--socket /tmp/sddm-auth-7e10ff2c-2d24-48ba-9909-2ca79eb06024 --id 2 --start /usr/bin/sddm-greeter-qt6 --socket /tmp/sddm--OJKxeu --theme /usr/share/sddm/themes/breeze --user sddm --display-server kwin_wayland --drm --no-lockscreen --no-global-shortcuts --locale1 --greeter) crashed (exit code 1)

but restart

Mar 14 14:37:52 arch-T16 sddm-greeter-qt6[954]: Adding view for "eDP-1" QRect(0,0 1536x960)

and then you login

Mar 14 14:37:54 arch-T16 sddm-greeter-qt6[954]: Message received from daemon: LoginSucceeded
Mar 14 14:37:54 arch-T16 sddm-helper[1068]: pam_kwallet5(sddm:setcred): pam_kwallet5: pam_sm_setcred
Mar 14 14:37:54 arch-T16 sddm-helper[1068]: pam_unix(sddm:session): session opened for user daniel(uid=1000) by daniel(uid=0)
Mar 14 14:37:54 arch-T16 systemd[1]: Created slice User Slice of UID 1000.
Mar 14 14:37:54 arch-T16 systemd[1]: Starting User Runtime Directory /run/user/1000...
Mar 14 14:37:54 arch-T16 systemd-logind[808]: New session '2' of user 'daniel' with class 'user' and type 'wayland'.

Since you don't seem to have amdgpu in the initramfs, does it help to add it there (explicitly or using the kms hook)?
Could be a simple race condition (sddm certainly starts while amdgpu is still initializing)

Celsiuss · 2026-03-15 16:10:59

seth wrote:

Last journal has SDDM crash

Mar 14 14:37:50 arch-T16 sddm[873]: Auth: sddm-helper (--socket /tmp/sddm-auth-7e10ff2c-2d24-48ba-9909-2ca79eb06024 --id 2 --start /usr/bin/sddm-greeter-qt6 --socket /tmp/sddm--OJKxeu --theme /usr/share/sddm/themes/breeze --user sddm --display-server kwin_wayland --drm --no-lockscreen --no-global-shortcuts --locale1 --greeter) crashed (exit code 1)

but restart

Mar 14 14:37:52 arch-T16 sddm-greeter-qt6[954]: Adding view for "eDP-1" QRect(0,0 1536x960)

and then you login

Mar 14 14:37:54 arch-T16 sddm-greeter-qt6[954]: Message received from daemon: LoginSucceeded
Mar 14 14:37:54 arch-T16 sddm-helper[1068]: pam_kwallet5(sddm:setcred): pam_kwallet5: pam_sm_setcred
Mar 14 14:37:54 arch-T16 sddm-helper[1068]: pam_unix(sddm:session): session opened for user daniel(uid=1000) by daniel(uid=0)
Mar 14 14:37:54 arch-T16 systemd[1]: Created slice User Slice of UID 1000.
Mar 14 14:37:54 arch-T16 systemd[1]: Starting User Runtime Directory /run/user/1000...
Mar 14 14:37:54 arch-T16 systemd-logind[808]: New session '2' of user 'daniel' with class 'user' and type 'wayland'.

Since you don't seem to have amdgpu in the initramfs, does it help to add it there (explicitly or using the kms hook)?
Could be a simple race condition (sddm certainly starts while amdgpu is still initializing)

Sorry I can't remember precisely the events that happened there.

However I added the kms hook to mkinitcpio and rebuilt initramfs, but the issues are still present.

First I got a black screen instead of seeing sddm, mouse pointer showing. Switched to tty3 and suddenly sddm displayed. Tried logging in but it resulted in another black screen. Switched tty again and logged in (took 1 minute or 2 for the login). Reboot hanged forever.
https://0x0.st/PLOR.txt

Same black screen as the previous one, killed the power after a while.
https://0x0.st/PLO5.txt

seth · 2026-03-15 22:20:46

Mar 15 16:35:12 arch-T16 sddm-helper-start-wayland[960]: Starting Wayland process "kwin_wayland --drm --no-lockscreen --no-global-shortcuts --locale1" "sddm"
Mar 15 16:35:12 arch-T16 sddm-helper-start-wayland[960]: started succesfully "kwin_wayland --drm --no-lockscreen --no-global-shortcuts --locale1"
Mar 15 16:35:12 arch-T16 sddm-helper-start-wayland[960]: Directory "/run/user/970" has changed, checking for Wayland socket
Mar 15 16:35:12 arch-T16 sddm-helper-start-wayland[960]: Found Wayland socket "/run/user/970/wayland-0"
Mar 15 16:35:12 arch-T16 sddm-helper-start-wayland[960]: "Accepting client connections on sockets: QList(\"wayland-0\")\n"
Mar 15 16:35:12 arch-T16 sddm-greeter-qt6[963]: High-DPI autoscaling Enabled
Mar 15 16:35:13 arch-T16 sddm-greeter-qt6[963]: Reading from "/usr/local/share/wayland-sessions/plasma.desktop"
Mar 15 16:35:13 arch-T16 sddm-greeter-qt6[963]: Reading from "/usr/share/wayland-sessions/plasma.desktop"
Mar 15 16:35:13 arch-T16 sddm-greeter-qt6[963]: Loading theme configuration from "/usr/share/sddm/themes/breeze/theme.conf"
Mar 15 16:35:13 arch-T16 sddm-greeter-qt6[963]: Connected to the daemon.
Mar 15 16:35:13 arch-T16 sddm[906]: Message received from greeter: Connect
Mar 15 16:41:11 arch-T16 sddm-greeter-qt6[963]: Loading file:///usr/share/sddm/themes/breeze/Main.qml...
Mar 15 16:41:11 arch-T16 sddm-greeter-qt6[963]: QObject::installEventFilter(): Cannot filter events for objects in a different thread.
Mar 15 16:41:12 arch-T16 sddm-greeter-qt6[963]: Adding view for "eDP-1" QRect(0,0 1536x960)

Does SDDM on X11 work?

Celsiuss · Yesterday 12:55:14

seth wrote:

Does SDDM on X11 work?

Issues still present with sddm on x11. I'm still using plasma on wayland session, as xorg is being phased out of plasma afaik.

Here is one session when I logged in and the DE seemingly worked fine, until I rebooted and the reboot sequence locked up.
https://0x0.st/PLUO.txt

And also another one where I got presented with a black screen instead of sddm.
https://0x0.st/PLU6.txt

Here it looks like sddm says I authenticated, while I did not as I was never even presented with the login screen. Not sure what is going on here really, I do not use the autologin feature.

mars 16 13:23:36 arch-T16 sddm-helper[1061]: pam_kwallet5(sddm:auth): pam_kwallet5: pam_sm_authenticate
mars 16 13:23:36 arch-T16 sddm[910]: Authentication for user  "daniel"  successful
mars 16 13:23:36 arch-T16 sddm-helper[1061]: pam_kwallet5(sddm:setcred): pam_kwallet5: pam_sm_setcred
mars 16 13:23:36 arch-T16 sddm-helper[1061]: pam_unix(sddm:session): session opened for user daniel(uid=1000) by daniel(uid=0)
mars 16 13:23:36 arch-T16 sddm-greeter-qt6[988]: Message received from daemon: LoginSucceeded
mars 16 13:23:36 arch-T16 systemd-logind[849]: New session '2' of user 'daniel' with class 'user' and type 'wayland'.
mars 16 13:23:36 arch-T16 systemd[1]: Created slice User Slice of UID 1000.
mars 16 13:23:36 arch-T16 systemd[1]: Starting User Runtime Directory /run/user/1000...

EDIT:
This one might be interesting, sddm was never even called here before the system froze.
https://0x0.st/PL0W.txt

Last edited by Celsiuss (Yesterday 13:45:36)

seth · Yesterday 14:39:49

Mar 16 14:28:17 arch-T16 systemd-udevd[667]: intel_rapl_msr.0: Worker [702] processing SEQNUM=5232 is taking a long time.
Mar 16 14:28:47 arch-T16 systemd[1]: systemd-logind.service: start operation timed out. Terminating.
Mar 16 14:30:16 arch-T16 systemd[1]: systemd-timesyncd.service: Watchdog timeout (limit 3min)!
Mar 16 14:30:16 arch-T16 systemd[1]: systemd-timesyncd.service: Killing process 643 (systemd-timesyn) with signal SIGABRT.
Mar 16 14:30:16 arch-T16 systemd[1]: systemd-resolved.service: Watchdog timeout (limit 3min)!
Mar 16 14:30:16 arch-T16 systemd[1]: systemd-resolved.service: Killing process 642 (systemd-resolve) with signal SIGABRT.
Mar 16 14:30:17 arch-T16 systemd[1]: systemd-logind.service: State 'stop-sigterm' timed out. Killing.
Mar 16 14:30:17 arch-T16 systemd[1]: systemd-logind.service: Killing process 848 (systemd-logind) with signal SIGKILL.
Mar 16 14:30:27 arch-T16 systemd-udevd[667]: intel_rapl_msr.0: Worker [702] processing SEQNUM=5232 killed.
Mar 16 14:31:12 arch-T16 kernel: INFO: task khugepaged:125 blocked for more than 122 seconds.

https://www.kernelconfig.io/CONFIG_INTEL_RAPL
"module_blacklist=intel_rapl_msr", https://wiki.archlinux.org/title/Kernel … and_line_2

Celsiuss · Yesterday 15:09:52

seth wrote:

https://www.kernelconfig.io/CONFIG_INTEL_RAPL
"module_blacklist=intel_rapl_msr", https://wiki.archlinux.org/title/Kernel … and_line_2

Blacklisted intel_rapl_msr, still experiencing issues.

sddm started according to the logs, but the system froze before any GUI showed.
https://0x0.st/PLD-.txt

seth · Yesterday 15:36:42

Mar 16 15:53:34 arch-T16 systemd-resolved[645]: Using degraded feature set UDP instead of UDP+EDNS0 for DNS server 10.11.11.1.
Mar 16 15:53:34 arch-T16 systemd-timesyncd[646]: Contacted time server 152.65.32.101:123 (3.arch.pool.ntp.org).
Mar 16 15:53:34 arch-T16 systemd-timesyncd[646]: Initial clock synchronization to Mon 2026-03-16 15:53:35.087894 CET.
Mar 16 15:53:35 arch-T16 systemd[1]: systemd-hostnamed.service: Deactivated successfully.
Mar 16 15:56:59 arch-T16 kernel: INFO: task kworker/6:0:39 blocked for more than 122 seconds.
Mar 16 15:56:59 arch-T16 kernel:       Not tainted 6.18.17-1-lts #1
Mar 16 15:56:59 arch-T16 kernel: "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
Mar 16 15:56:59 arch-T16 kernel: task:kworker/6:0     state:D stack:0     pid:39    tgid:39    ppid:2      task_flags:0x4208060 flags:0x00080000
Mar 16 15:56:59 arch-T16 kernel: Workqueue: pm pm_runtime_work
Mar 16 15:56:59 arch-T16 kernel: Call Trace:
Mar 16 15:56:59 arch-T16 kernel:  <TASK>

https://www.kernelconfig.io/CONFIG_DRM_ACCEL_AMDXDNA
"module_blacklist=intel_rapl_msr,amdxdna"

Celsiuss · Yesterday 16:25:42

seth wrote:

https://www.kernelconfig.io/CONFIG_DRM_ACCEL_AMDXDNA
"module_blacklist=intel_rapl_msr,amdxdna"

Added blacklist, still issues.

System freeze after sddm login:
https://0x0.st/PLk9.txt

System freeze before I saw sddm (not too much logs here, maybe I did not wait long enough before power cyling):
https://0x0.st/PLkV.txt

seth · Yesterday 16:36:05

First one hangs in

Mar 16 17:08:40 arch-T16 systemd[1150]: plasma-ksplash.service: Consumed 1.598s CPU time over 30.779s wall clock time, 51.9M memory peak.
Mar 16 17:08:59 arch-T16 /usr/lib/xdg-desktop-portal[1231]: Failed to create settings proxy: Error calling StartServiceByName for org.freedesktop.impl.portal.desktop.kde: Timeout was reached
Mar 16 17:09:24 arch-T16 /usr/lib/xdg-desktop-portal[1231]: GError set over the top of a previous GError or uninitialized memory.
                                                            This indicates a bug in someone's code. You must ensure an error is NULL before it's set.
                                                            The overwriting error message was: Error calling StartServiceByName for org.freedesktop.impl.portal.desktop.gtk: Timeout was reached
Mar 16 17:09:24 arch-T16 /usr/lib/xdg-desktop-portal[1231]: Failed to create settings proxy: Error calling StartServiceByName for org.freedesktop.impl.portal.desktop.kde: Timeout was reached
Mar 16 17:09:24 arch-T16 /usr/lib/xdg-desktop-portal[1231]: No skeleton to export
Mar 16 17:09:39 arch-T16 systemd[1150]: xdg-desktop-portal.service: start operation timed out. Terminating.
Mar 16 17:09:39 arch-T16 systemd[1150]: xdg-desktop-portal.service: Failed with result 'timeout'.
Mar 16 17:09:39 arch-T16 systemd[1150]: Failed to start Portal service.

pacman -Qs portal

Second one is entirely unsuspicious

Mar 16 17:14:40 arch-T16 NetworkManager[858]: <info>  [1773677680.5801] manager: NetworkManager state is now CONNECTED_GLOBAL
Mar 16 17:14:42 arch-T16 systemd[1]: NetworkManager-dispatcher.service: Deactivated successfully.
Mar 16 17:14:42 arch-T16 NetworkManager[858]: <info>  [1773677682.8903] manager: startup complete
Mar 16 17:15:02 arch-T16 systemd-resolved[655]: Using degraded feature set UDP instead of UDP+EDNS0 for DNS server 10.11.11.1.
Mar 16 17:15:02 arch-T16 systemd[1]: systemd-hostnamed.service: Deactivated successfully.
Mar 16 17:15:04 arch-T16 systemd-resolved[655]: Clock change detected. Flushing caches.
Mar 16 17:15:04 arch-T16 systemd-timesyncd[656]: Contacted time server 193.150.22.36:123 (3.arch.pool.ntp.org).
Mar 16 17:15:04 arch-T16 systemd-timesyncd[656]: Initial clock synchronization to Mon 2026-03-16 17:15:04.957605 CET.
Mar 16 17:16:25 arch-T16 kernel: psmouse serio1: trackpoint: Elan TrackPoint firmware: 0x00, buttons: 3/3
Mar 16 17:16:25 arch-T16 kernel: input: TPPS/2 Elan TrackPoint as /devices/platform/i8042/serio1/input/input25
Mar 16 17:16:26 arch-T16 sddm[918]: Setting default cursor
Mar 16 17:16:26 arch-T16 sddm[918]: Could not setup default cursor
Mar 16 17:16:26 arch-T16 sddm[918]: Running display setup script  "/usr/share/sddm/scripts/Xsetup"

Is also start to wonder whether the ntp synchronization could introduce the time gaps - try to disable NM to boot w/o network access.

Celsiuss · Yesterday 19:10:13

seth wrote:

First one hangs in

Mar 16 17:08:40 arch-T16 systemd[1150]: plasma-ksplash.service: Consumed 1.598s CPU time over 30.779s wall clock time, 51.9M memory peak.
Mar 16 17:08:59 arch-T16 /usr/lib/xdg-desktop-portal[1231]: Failed to create settings proxy: Error calling StartServiceByName for org.freedesktop.impl.portal.desktop.kde: Timeout was reached
Mar 16 17:09:24 arch-T16 /usr/lib/xdg-desktop-portal[1231]: GError set over the top of a previous GError or uninitialized memory.
                                                            This indicates a bug in someone's code. You must ensure an error is NULL before it's set.
                                                            The overwriting error message was: Error calling StartServiceByName for org.freedesktop.impl.portal.desktop.gtk: Timeout was reached
Mar 16 17:09:24 arch-T16 /usr/lib/xdg-desktop-portal[1231]: Failed to create settings proxy: Error calling StartServiceByName for org.freedesktop.impl.portal.desktop.kde: Timeout was reached
Mar 16 17:09:24 arch-T16 /usr/lib/xdg-desktop-portal[1231]: No skeleton to export
Mar 16 17:09:39 arch-T16 systemd[1150]: xdg-desktop-portal.service: start operation timed out. Terminating.
Mar 16 17:09:39 arch-T16 systemd[1150]: xdg-desktop-portal.service: Failed with result 'timeout'.
Mar 16 17:09:39 arch-T16 systemd[1150]: Failed to start Portal service.

pacman -Qs portal

Second one is entirely unsuspicious

Mar 16 17:14:40 arch-T16 NetworkManager[858]: <info>  [1773677680.5801] manager: NetworkManager state is now CONNECTED_GLOBAL
Mar 16 17:14:42 arch-T16 systemd[1]: NetworkManager-dispatcher.service: Deactivated successfully.
Mar 16 17:14:42 arch-T16 NetworkManager[858]: <info>  [1773677682.8903] manager: startup complete
Mar 16 17:15:02 arch-T16 systemd-resolved[655]: Using degraded feature set UDP instead of UDP+EDNS0 for DNS server 10.11.11.1.
Mar 16 17:15:02 arch-T16 systemd[1]: systemd-hostnamed.service: Deactivated successfully.
Mar 16 17:15:04 arch-T16 systemd-resolved[655]: Clock change detected. Flushing caches.
Mar 16 17:15:04 arch-T16 systemd-timesyncd[656]: Contacted time server 193.150.22.36:123 (3.arch.pool.ntp.org).
Mar 16 17:15:04 arch-T16 systemd-timesyncd[656]: Initial clock synchronization to Mon 2026-03-16 17:15:04.957605 CET.
Mar 16 17:16:25 arch-T16 kernel: psmouse serio1: trackpoint: Elan TrackPoint firmware: 0x00, buttons: 3/3
Mar 16 17:16:25 arch-T16 kernel: input: TPPS/2 Elan TrackPoint as /devices/platform/i8042/serio1/input/input25
Mar 16 17:16:26 arch-T16 sddm[918]: Setting default cursor
Mar 16 17:16:26 arch-T16 sddm[918]: Could not setup default cursor
Mar 16 17:16:26 arch-T16 sddm[918]: Running display setup script  "/usr/share/sddm/scripts/Xsetup"

Is also start to wonder whether the ntp synchronization could introduce the time gaps - try to disable NM to boot w/o network access.

I've got xdg-desktop-portal-kde installed through plasma-meta.

$ pacman -Q | grep portal
libportal 0.9.1-2
libportal-qt6 0.9.1-2
xdg-desktop-portal 1.20.3-2
xdg-desktop-portal-gtk 1.15.3-1
xdg-desktop-portal-kde 6.6.2-1

I disabled networkmanager now.

Freeze before sddm shows:
https://0x0.st/PL5s.txt

Another one froze after sddm login:
https://0x0.st/PL5T.txt

seth · Yesterday 20:39:39

Mar 16 19:38:11 arch-T16 boltd[833]: probing: started [1000]
Mar 16 19:38:11 arch-T16 kernel: ath11k_pci 0000:02:00.0: chip_id 0x12 chip_family 0xb board_id 0xff soc_id 0x400c1211
Mar 16 19:38:11 arch-T16 kernel: ath11k_pci 0000:02:00.0: fw_version 0x11088c35 fw_build_timestamp 2024-04-17 08:34 fw_build_id WLAN.HSP.1.1-03125-QCAHSPSWPL_V1_V2_SILICONZ_LITE-3.6510.41
Mar 16 19:38:12 arch-T16 kernel: ath11k_pci 0000:02:00.0 wlp2s0: renamed from wlan0
Mar 16 19:38:12 arch-T16 kernel: ath11k_pci 0000:02:00.0: Failed to set the requested Country regulatory setting
Mar 16 19:38:12 arch-T16 kernel: ath11k_pci 0000:02:00.0: failed to process regulatory info -22
Mar 16 19:38:12 arch-T16 kernel: ath11k_pci 0000:02:00.0: Failed to set the requested Country regulatory setting
Mar 16 19:38:12 arch-T16 kernel: ath11k_pci 0000:02:00.0: failed to process regulatory info -22
Mar 16 19:38:14 arch-T16 boltd[833]: probing: timeout, done: [2641541] (2000000)
Mar 16 19:38:17 arch-T16 systemd[1]: systemd-rfkill.service: Deactivated successfully.
Mar 16 19:38:41 arch-T16 systemd[1]: systemd-hostnamed.service: Deactivated successfully.
Mar 16 19:42:00 arch-T16 kernel: INFO: task khugepaged:125 blocked for more than 122 seconds.
Mar 16 19:42:00 arch-T16 kernel:       Not tainted 6.18.17-1-lts #1
Mar 16 19:42:00 arch-T16 kernel: "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
Mar 16 19:42:00 arch-T16 kernel: task:khugepaged      state:D stack:0     pid:125   tgid:125   ppid:2      task_flags:0x200040 flags:0x00080000
Mar 16 19:42:00 arch-T16 kernel: Call Trace:
Mar 16 19:42:00 arch-T16 kernel:  <TASK>
Mar 16 19:42:00 arch-T16 kernel:  __schedule+0x418/0x1320
Mar 16 19:42:00 arch-T16 kernel:  ? srso_alias_return_thunk+0x5/0xfbef5
Mar 16 19:42:00 arch-T16 kernel:  ? srso_alias_return_thunk+0x5/0xfbef5
Mar 16 19:42:00 arch-T16 kernel:  ? psi_group_change+0x10c/0x2c0
Mar 16 19:42:00 arch-T16 kernel:  ? srso_alias_return_thunk+0x5/0xfbef5
Mar 16 19:42:00 arch-T16 kernel:  ? srso_alias_return_thunk+0x5/0xfbef5
Mar 16 19:42:00 arch-T16 kernel:  ? __pick_eevdf+0x133/0x260
Mar 16 19:42:00 arch-T16 kernel:  schedule+0x27/0xd0
Mar 16 19:42:00 arch-T16 kernel:  schedule_timeout+0xbd/0x100

Well, let's cut more fat then - disable boltd and blacklist ath11k_pci and though I'm not sure it not just a victim "transparent_hugepage=never" and "zswap.enabled=0"

Celsiuss · Yesterday 22:20:43

seth wrote:

Well, let's cut more fat then - disable boltd and blacklist ath11k_pci and though I'm not sure it not just a victim "transparent_hugepage=never" and "zswap.enabled=0"

Cut some fat.

I got two boots here where it crashed before sddm was shown.
The first one don't show much, but the second one shows some errors at the end.

https://0x0.st/PL7d.txt
https://0x0.st/PL7I.txt

This one I could log in and see the DE, but applications would not launch.
https://0x0.st/PL74.txt

Last edited by Celsiuss (Yesterday 22:24:00)

seth · Yesterday 23:37:12

Mar 16 23:06:01 arch-T16 sddm[920]: Display server starting...
Mar 16 23:06:01 arch-T16 sddm[920]: Writing cookie to "/run/sddm/xauth_TlXmyL"
Mar 16 23:06:01 arch-T16 sddm[920]: Running: /usr/bin/X -nolisten tcp -background none -seat seat0 vt2 -auth /run/sddm/xauth_TlXmyL -noreset -displayfd 16
Mar 16 23:06:02 arch-T16 polkitd[926]: Started polkitd version 127
Mar 16 23:06:02 arch-T16 systemd[1]: Started Authorization Manager.
Mar 16 23:06:02 arch-T16 systemd[1]: Started Power Profiles daemon.
Mar 16 23:06:02 arch-T16 systemd[1]: Reached target Graphical Interface.
Mar 16 23:06:02 arch-T16 systemd[1]: Startup finished in 9.342s (firmware) + 2.117s (loader) + 12.759s (kernel) + 18.686s (userspace) = 42.905s.
Mar 16 23:06:16 arch-T16 systemd[1]: systemd-hostnamed.service: Deactivated successfully.
Mar 16 23:09:36 arch-T16 kernel: INFO: task kworker/R-mm_pe:14 blocked for more than 122 seconds.
Mar 16 23:09:36 arch-T16 kernel:       Not tainted 6.18.17-1-lts #1
Mar 16 23:09:36 arch-T16 kernel: "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
Mar 16 23:09:36 arch-T16 kernel: task:kworker/R-mm_pe state:D stack:0     pid:14    tgid:14    ppid:2      task_flags:0x4208060 flags:0x00080000
Mar 16 23:09:36 arch-T16 kernel: Call Trace:
Mar 16 23:09:36 arch-T16 kernel:  <TASK>
Mar 16 23:09:36 arch-T16 kernel:  __schedule+0x418/0x1320
Mar 16 23:09:36 arch-T16 kernel:  ? srso_alias_return_thunk+0x5/0xfbef5
Mar 16 23:09:36 arch-T16 kernel:  ? psi_task_switch+0x16f/0x2a0
Mar 16 23:09:36 arch-T16 kernel:  schedule+0x27/0xd0
Mar 16 23:09:36 arch-T16 kernel:  schedule_preempt_disabled+0x15/0x30

This is getting ridiculous - I fear it might be the CPU.
Test the behavior w/ some live distro (https://grml.org/) and notably look out for hung tasks in the journal/dmesg.
https://wiki.archlinux.org/title/Ryzen# … k_freezing and in doubt just limit "processor.max_cstate=1"

Celsiuss · Today 13:49:49

seth wrote:

This is getting ridiculous - I fear it might be the CPU.
Test the behavior w/ some live distro (https://grml.org/) and notably look out for hung tasks in the journal/dmesg.
https://wiki.archlinux.org/title/Ryzen# … k_freezing and in doubt just limit "processor.max_cstate=1"

Indeed.
I tried out grml, spent a while with it over many reboots. Worked perfectly fine. Also tried Linux mint for a more fleshed DE to see if that mattered. Also no issues.
processor.max_cstate=1, did nothing as well.

This morning I was doing some work while my system was semi-stable. Applications barely launched, so I made sure to not close them.
I noticed a process named irq/104-nxp-nci_i2c was constantly maxing out one of the CPU cores. Apparently this is related to some NFC driver. Not sure if my laptop even has NFC, maybe? Never used it.

According to the wiki, some NFC chips are affected by bugs in some kernel modules: https://wiki.archlinux.org/title/NFC#NX … t_detected

Checking my logs

mars 17 10:22:38 arch-T16 kernel: nxp-nci_i2c i2c-NXP1001:00: NFC: Read failed with error -121

it looks like my laptop has one of these affected chips, NXP1001.

The page has a link to a fix, but I ended up removing the module instead

modprobe -r nxp_nci_i2c

Which immediately removed the process, and the system turned stable!

I then blacklisted nxp_nci_i2c, and the system is stable over many reboots and prolonged use with no symptoms showing.
Maybe I'll try the nfc fix a bit later too.

I just want to thank you seth for being incredibly helpful! It's great to be able to get this level of support here :)

Arch Linux

#1 2026-03-13 13:14:18

Desktop / system frequently unresponsive

#2 2026-03-13 14:26:26

Re: Desktop / system frequently unresponsive

#3 2026-03-13 14:42:05

Re: Desktop / system frequently unresponsive

#4 2026-03-13 14:59:49

Re: Desktop / system frequently unresponsive

#5 2026-03-13 16:51:50

Re: Desktop / system frequently unresponsive

#6 2026-03-13 20:59:43

Re: Desktop / system frequently unresponsive

#7 2026-03-14 13:49:26

Re: Desktop / system frequently unresponsive

#8 2026-03-14 15:23:13

Re: Desktop / system frequently unresponsive

#9 2026-03-15 16:10:59

Re: Desktop / system frequently unresponsive

#10 2026-03-15 22:20:46

Re: Desktop / system frequently unresponsive

#11 Yesterday 12:55:14

Re: Desktop / system frequently unresponsive

#12 Yesterday 14:39:49

Re: Desktop / system frequently unresponsive

#13 Yesterday 15:09:52

Re: Desktop / system frequently unresponsive

#14 Yesterday 15:36:42

Re: Desktop / system frequently unresponsive

#15 Yesterday 16:25:42

Re: Desktop / system frequently unresponsive

#16 Yesterday 16:36:05

Re: Desktop / system frequently unresponsive

#17 Yesterday 19:10:13

Re: Desktop / system frequently unresponsive

#18 Yesterday 20:39:39

Re: Desktop / system frequently unresponsive

#19 Yesterday 22:20:43

Re: Desktop / system frequently unresponsive

#20 Yesterday 23:37:12

Re: Desktop / system frequently unresponsive

#21 Today 13:49:49

Re: Desktop / system frequently unresponsive

Board footer