Arch Linux

Tid35

Member

Registered: 2024-10-13

Posts: 3

[Solved] Plasmashell crashing (X11 + Nvidia)

Hello everyone,

I use kde plasma with dracut, nvidia and X11.

I see many articles about this issue but always on wayland. I just updated my system, and around 218 packages got updated including linux, linux-headers, nvidia-dkms, plasma, etc. It went fine, but once I rebooted my computer and logged in using sddm into my X11 session, all I got was a black screen with no plasma panel. I could still move the cursor, open konsole with ctrl+alt+t and open applications like the browser, which all work perfectly.

But my desktop is blank. I cannot open the application launcher using the meta button, and I get the error message ‘plasmashell closed unexpectedly’ three times every time I log in. Wayland does not work either (never really has), but nvidia-smi works like a charm so the gpu is still recognized. I tried downgrading linux, linux-headers, nvidia-dkms, nvidia-utils and kwin to previous versions, tried reinstalling plasma group with the same end result. Adding nvidia_drm.fbdev=1 to my /etc/kernel/cmdline made no difference.

inxi -Fxxc0z

https://0x0.st/X6Ji.txt

journalctl -k -b -0

https://0x0.st/X6JZ.txt

This is the dump from the plasmashell crash error message:
https://0x0.st/X6Jq.txt

Xorg logs, if they help:
https://0x0.st/X6yo.txt

I see this line in my journalctl, might this have anything to do with the error?

nvidia: module verification failed: signature and/or required key missing - tainting kernel

Thank you.

Last edited by Tid35 (2024-10-13 15:03:30)

seth

Member

From: Don't DM me only for attention

Registered: 2012-09-03

Posts: 73,834

Re: [Solved] Plasmashell crashing (X11 + Nvidia)

I use kde plasma with dracut, nvidia and X11.

Thread 18 (Thread 0x795a268006c0 (LWP 1472)):
#0  0x0000795a4309fa19 in ??? () at /usr/lib/libc.so.6
#1  0x0000795a430a2479 in pthread_cond_wait () at /usr/lib/libc.so.6
#2  0x0000795a34ececae in ??? () at /usr/lib/libgallium-24.2.4-arch1.1.so
#3  0x0000795a34eab6bc in ??? () at /usr/lib/libgallium-24.2.4-arch1.1.so
#4  0x0000795a34ecebdd in ??? () at /usr/lib/libgallium-24.2.4-arch1.1.so
#5  0x0000795a430a339d in ??? () at /usr/lib/libc.so.6
#6  0x0000795a4312849c in ??? () at /usr/lib/libc.so.6

[     8.683] (--) PCI:*(0@0:2:0) 8086:9a60:1028:0a73 rev 1, Mem @ 0x622c000000/16777216, 0x4000000000/268435456, I/O @ 0x00005000/64, BIOS @ 0x????????/131072
[     8.683] (--) PCI: (1@0:0:0) 10de:24dd:1028:0a9c rev 161, Mem @ 0x83000000/16777216, 0x6000000000/8589934592, 0x6200000000/33554432, I/O @ 0x00004000/128, BIOS @ 0x????????/524288

You're running on the intel IGP, there're also no outputs connected to the nvidia GPU so this is likely what you want.

Debuginfod has been disabled.

"Not a good plan", though

Thread 1 (Thread 0x795a034006c0 (LWP 1652)):
[KCrash Handler]
#4  0x0000795a1d270ef7 in QQuick3DViewport::setupOffscreenRenderer(QSGNode*) () at /usr/lib/qt6/qml/QtQuick3D/../../../libQt6Quick3D.so.6
#5  0x0000795a1d27135a in QQuick3DViewport::updatePaintNode(QSGNode*, QQuickItem::UpdatePaintNodeData*) () at /usr/lib/qt6/qml/QtQuick3D/../../../libQt6Quick3D.so.6
#6  0x0000795a44e5ec54 in QQuickWindowPrivate::updateDirtyNode(QQuickItem*) () at /usr/lib/libQt6Quick.so.6
#7  0x0000795a44e5239d in QQuickWindowPrivate::updateDirtyNodes() () at /usr/lib/libQt6Quick.so.6
#8  0x0000795a44e52f36 in QQuickWindowPrivate::syncSceneGraph() () at /usr/lib/libQt6Quick.so.6
#9  0x0000795a45021bf3 in ??? () at /usr/lib/libQt6Quick.so.6
#10 0x0000795a438c53f7 in ??? () at /usr/lib/libQt6Core.so.6
#11 0x0000795a430a339d in ??? () at /usr/lib/libc.so.6
#12 0x0000795a4312849c in ??? () at /usr/lib/libc.so.6

but something has eaten the top of the stack:

Program terminated with signal SIGSEGV, Segmentation fault.
#0  0x0000795a430a53f4 in ?? () from /usr/lib/libc.so.6
[Current thread is 1 (Thread 0x795a034006c0 (LWP 1652))]
Cannot QML trace cores :(
/usr/share/drkonqi/gdb/python/gdb_preamble/preamble.py:516: DeprecationWarning: datetime.datetime.utcfromtimestamp() is deprecated and scheduled for removal in a future version. Use timezone-aware objects to represent datetimes in UTC: datetime.datetime.fromtimestamp(timestamp, datetime.UTC).
  boot_time = datetime.utcfromtimestamp(psutil.boot_time()).strftime('%Y-%m-%dT%H:%M:%S')
/usr/share/drkonqi/gdb/python/gdb_preamble/preamble.py:533: DeprecationWarning: datetime.datetime.utcnow() is deprecated and scheduled for removal in a future version. Use timezone-aware objects to represent datetimes in UTC: datetime.datetime.now(datetime.UTC).
  'timestamp': datetime.utcnow().isoformat(),
[Current thread is 1 (Thread 0x795a034006c0 (LWP 1652))]

QML might have printed some debug info, so look at the actual journal, not just the kernel ringbuffer.

---

How to upload text · How to boot w/o GUI · Disable Windows Fast-Start! · Fix your xinitrc

Tid35

Member

Registered: 2024-10-13

Posts: 3

Re: [Solved] Plasmashell crashing (X11 + Nvidia)

QML might have printed some debug info, so look at the actual journal, not just the kernel ringbuffer.

This is all the journal has to say regarding plasmashell:

https://0x0.st/X6yg.txt

The DE has always run on the IGP, the nvidia card is still recognized and can be used in nvidia-smi, python, etc. Something during the update messed things up w.r.t plasma / x11 and I'm not sure what.

Tid35

Member

Registered: 2024-10-13

Posts: 3

Re: [Solved] Plasmashell crashing (X11 + Nvidia)

Figured it out, a few widgets and the desktop theme I had installed are incompatible with plasma 6.2. Removing them fixed everything. Thank you.