System Settings SDDM Theme Crashes - Plasma 6.7.0

After the Plasma 6.7.0 update on Tumbleweed, System Settings crashes when I click on Colors & Themes> Login Screen (SDDM). I’m running an Intel ARC A750 and Wayland. I need to know if this is a Plasma 6.7.0 bug or a Tumbleweed bug.

Running systemsettings kcm_sddm in a terminal starts System Settings at the Login Screen (SDDM) configuration page and I am able to navigate to other settings and return to the SDDM settings without issues. It also prints warnings about themes in the terminal, but no errors.

When I run just systemsettings in a terminal, it crashes like it does from an app launcher, except I get some warnings about themes and an error about X.

> systemsettings

org.kde.libkbolt: Failed to connect to Bolt manager DBus interface:
org.kde.libkbolt: Failed to connect to Bolt manager DBus interface:
Using fontconfig file: “/home/john/.config/fontconfig/fonts.conf”
kf.coreaddons: “Could not load plugin from /usr/share/plasma/systemsettings/externalmodules/YaST-systemsettings.desktop: Failed to extract plugin meta data from ‘/usr/share/plasma/systemsettings/externalmodules/YaST-systemsettings.desktop’: ‘/usr/share/plasma/systemsettings/externalmodules/YaST-systemsettings.desktop’ is not a valid ELF object (invalid signature)”
qt.qpa.services: Failed to register with host portal QDBusError(“org.freedesktop.portal.Error.Failed”, “Could not register app ID: Connection already associated with an application ID”)
kf.bluezqt: InitManagerJob Error: Job was deleted before finished!
KPackageStructure of KPluginMetaData(pluginId:“LightbulbSplash3D”, fileName: “/home/john/.local/share/plasma/look-and-feel/LightbulbSplash3D/metadata.json”) does not match requested format “Plasma/LookAndFeel”
qt.multimedia.ffmpeg: Using Qt multimedia with FFmpeg version 8.1.2 nonfree and unredistributable

X Error of failed request:  BadAccess (attempt to access private resource denied)
Major opcode of failed request:  151 (GLX)
Minor opcode of failed request:  5 (X_GLXMakeCurrent)
Serial number of failed request:  28
Current serial number in output stream:  28

NOTE: Installing the old System Settings fixes this issue.

sudo zypper in --oldpackage https://download-repositories.opensuse.org/history/20260616/tumbleweed/repo/oss/x86_64/sddm-kcm6-6.6.5-1.1.x86_64.rpm

2 Likes

@jsmith64:

On this machine –

 > kinfo
Operating System: openSUSE Tumbleweed 20260619
KDE Plasma Version: 6.7.0
KDE Frameworks Version: 6.27.0
Qt Version: 6.11.1
Kernel Version: 7.0.12-1-default (64-bit)
Graphics Platform: Wayland
Processors: 12 × AMD Ryzen 5 8600G w/ Radeon 760M Graphics
Memory: 34 GB of RAM (32.8 GB usable)
Graphics Processor: AMD Radeon 760M Graphics
 >

– executing “systemsettings” from a Konsole Bash CLI seems to be OK:

 > systemsettings 
org.kde.libkbolt: Failed to connect to Bolt manager DBus interface: 
org.kde.libkbolt: Failed to connect to Bolt manager DBus interface: 
Using fontconfig file: "/home/Users/xxx/.config/fontconfig/fonts.conf"
qrc:/kcm/kcm_landingpage/main.qml:278:21: Unable to assign [undefined] to QQuickItem*
qt.qpa.services: Failed to register with host portal QDBusError("org.freedesktop.portal.Error.Failed", "Could not register app ID: Connection already associated with an application ID")
 > 
 > LANG=C.UTF-8 systemsettings 
org.kde.libkbolt: Failed to connect to Bolt manager DBus interface: 
org.kde.libkbolt: Failed to connect to Bolt manager DBus interface: 
Using fontconfig file: "/home/Users/xxx/.config/fontconfig/fonts.conf"
qrc:/kcm/kcm_landingpage/main.qml:278:21: Unable to assign [undefined] to QQuickItem*
qt.qpa.services: Failed to register with host portal QDBusError("org.freedesktop.portal.Error.Failed", "Could not register app ID: Connection already associated with an application ID")
 > 

But, did you check the following?

 > ls -aldZ /usr/share/plasma/systemsettings/externalmodules/YaST-systemsettings.desktop
-rw-r--r--. 1 root root system_u:object_r:usr_t:s0 231 16. Sep 2025  /usr/share/plasma/systemsettings/externalmodules/YaST-systemsettings.desktop
 >
  • In particular, please check the SELinux security context …

If in doubt: “restorecon -F -R -v /usr” …

  • Do you have any “private” (machine local) security contexts defined?

semanage fcontext --locallist --list

I do not have Selinux installed.
YaST works from system settings.

> ls -aldZ /usr/share/plasma/systemsettings/externalmodules/YaST-systemsettings.desktop

-rw-r--r-- 1 root root ? 231 Sep 16  2025 /usr/share/plasma/systemsettings/externalmodules/YaST-systemsettings.desktop

System Settings launches from a terminal just fine, but clicking on Login Screen (SDDM) causes it to crash. Running systemsettings kcm_sddm in a terminal starts System Settings and takes me directly to the Login Screen (SDDM) configuration page.

Okay, also having this problem, running AppArmor. The fix of installing the old package works, thanks @jsmith64.

Wondering what to do to get this resolved. Yesterday I did run gdb and saw this was for me actually a SIG_SEGV but I do not have a good backtrace as debuginfo is missing.

Also should this be an OpenSuse bugzilla issue or a KDE systemsettings bug?

1 Like

Hit the 30 minutes time-out.

What is also remarkable is how I have to open the Login settings:

No Login screen, so instead I search for it and then it pops up:

With the newer package then the crash happens, with the older package things are fine. How are things working for others?

1 Like

I don’t know. I posted about it here and the KDE forum and nobody ever said if their’s works on either forum. You were the first I saw who had the problem.

It should show up in colors and themes.
I’m on my phone right now, can’t research much. I’m in the middle of a Fedora install and it doesn’t use SDDM. It’s using Plasma Login Manager and it does work in System Settings… But it’d be a hassle to test it in Fedora.

Plasma 7.0.1 should show up any day now and maybe that’ll fix things. I always say I’m going to skip the x.x.0 and I installed it again.

1 Like

Did you test with a fresh user profile? Yours seem broken.

No crash in the SDDM setting window or in systemsettings itself here.

Where is this splash screen from? This seems like you downloaded this external splash screen from the Plasma store or somewhere else. Files from the Plasma store or other external sources can cause instabilities including crashes like you see. That is why there is a big fat warning on top of systemsettings when you download external themes/color schemes/ and so on.

As this splash screen is not available from the openSUSE repos but downloaded from an external source, the crash is not caused by anything openSUSE related but the incompatibility of the 3rd party splash screen with Plasma.

As this external splash screen was updated the last time two years ago…not that surprising.

No systemsettings/SDDM setting crashes here with standard splash screens from the official repos and latest TW snapshot.

1 Like

@jsmith64 you should also test with a fresh user profile to rule out the influence of external files like the named splash screen.

1 Like

I totally understand, but! I did try and different user. Running systemsettings kcm_sddm in a terminal works and it gets the same theme warnings, but no X error. I’ll boot to Tumbleweed and see what happens and report back.

Removing the light bulb splash didn’t work. I’m convinced it’s the X error. I wonder what GPU marel is using? It makes no sense that clicking on it causes a crash, but opening system settings directly to it doesn’t crash it.

I just did and exactly the same problem.

Did run under gdb once more and now I see:

(gdb) run
Starting program: /usr/bin/systemsettings 
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib64/libthread_db.so.1".
[New Thread 0x7ffff03b06c0 (LWP 17768)]
[New Thread 0x7fffef28b6c0 (LWP 17769)]
[New Thread 0x7fffeea8a6c0 (LWP 17770)]
org.kde.libkbolt: Failed to connect to Bolt manager DBus interface: 
[New Thread 0x7fffde65d6c0 (LWP 17771)]
[New Thread 0x7fffdd2bc6c0 (LWP 17774)]
[Thread 0x7fffdd2bc6c0 (LWP 17774) exited]
[Thread 0x7fffde65d6c0 (LWP 17771) exited]
[New Thread 0x7fffde65d6c0 (LWP 17775)]
[New Thread 0x7fffdd2bc6c0 (LWP 17777)]
[New Thread 0x7fffd5d796c0 (LWP 17778)]
[New Thread 0x7fffd55786c0 (LWP 17779)]
[New Thread 0x7fffd4cf26c0 (LWP 17780)]
[New Thread 0x7fffcbfff6c0 (LWP 17781)]
[New Thread 0x7fffcb7fe6c0 (LWP 17782)]
[New Thread 0x7fffbfbff6c0 (LWP 17783)]
[New Thread 0x7fffbf3fe6c0 (LWP 17784)]
[New Thread 0x7fffbebfd6c0 (LWP 17785)]
[New Thread 0x7fffbe3fc6c0 (LWP 17786)]
[New Thread 0x7fffbdbfb6c0 (LWP 17787)]
[New Thread 0x7fffbd3fa6c0 (LWP 17788)]
[Thread 0x7fffbd3fa6c0 (LWP 17788) exited]
[Thread 0x7fffbdbfb6c0 (LWP 17787) exited]
[Thread 0x7fffbe3fc6c0 (LWP 17786) exited]
[Thread 0x7fffbebfd6c0 (LWP 17785) exited]
[New Thread 0x7fffbebfd6c0 (LWP 17789)]
[New Thread 0x7fffbe3fc6c0 (LWP 17790)]
[New Thread 0x7fffbdbfb6c0 (LWP 17791)]
[New Thread 0x7fffbd3fa6c0 (LWP 17792)]
[New Thread 0x7fffbcbf96c0 (LWP 17793)]
[New Thread 0x7fffaebff6c0 (LWP 17794)]
[Thread 0x7fffaebff6c0 (LWP 17794) exited]
[New Thread 0x7fffaebff6c0 (LWP 17795)]
[Thread 0x7fffaebff6c0 (LWP 17795) exited]
qt.qpa.services: Failed to register with host portal QDBusError("org.freedesktop.portal.Error.Failed", "Could not register app ID: Connection already associated with an application ID")
[Thread 0x7fffde65d6c0 (LWP 17775) exited]
[New Thread 0x7fffde65d6c0 (LWP 17800)]
qt.multimedia.ffmpeg: Using Qt multimedia with FFmpeg version 8.1.1 GPL version 3 or later
[New Thread 0x7fffaebff6c0 (LWP 17801)]
[New Thread 0x7fff8dfff6c0 (LWP 17802)]
[New Thread 0x7fff8d7fe6c0 (LWP 17803)]
[New Thread 0x7fff8cffd6c0 (LWP 17804)]
[New Thread 0x7fff7bfff6c0 (LWP 17805)]
[New Thread 0x7fff7b7fe6c0 (LWP 17806)]
X Error of failed request:  BadAccess (attempt to access private resource denied)
  Major opcode of failed request:  151 (GLX)
  Minor opcode of failed request:  5 (X_GLXMakeCurrent)
  Serial number of failed request:  28
  Current serial number in output stream:  28
[Thread 0x7fff7b7fe6c0 (LWP 17806) exited]
[Thread 0x7fff7bfff6c0 (LWP 17805) exited]
[Thread 0x7fff8cffd6c0 (LWP 17804) exited]
[Thread 0x7fff8d7fe6c0 (LWP 17803) exited]
[Thread 0x7fff8dfff6c0 (LWP 17802) exited]
[Thread 0x7fffaebff6c0 (LWP 17801) exited]
[Thread 0x7fffbd3fa6c0 (LWP 17792) exited]
[Thread 0x7fffbdbfb6c0 (LWP 17791) exited]
[Thread 0x7fffbe3fc6c0 (LWP 17790) exited]
[Thread 0x7fffbebfd6c0 (LWP 17789) exited]
[Thread 0x7fffbf3fe6c0 (LWP 17784) exited]
[Thread 0x7fffbfbff6c0 (LWP 17783) exited]
[Thread 0x7ffff03b06c0 (LWP 17768) exited]
[Thread 0x7fffde65d6c0 (LWP 17800) exited]
[Thread 0x7fffbcbf96c0 (LWP 17793) exited]
[Thread 0x7fffcbfff6c0 (LWP 17781) exited]
[Thread 0x7fffd4cf26c0 (LWP 17780) exited]
[Thread 0x7fffd55786c0 (LWP 17779) exited]
[Thread 0x7fffd5d796c0 (LWP 17778) exited]
[Thread 0x7fffdd2bc6c0 (LWP 17777) exited]
[Thread 0x7fffeea8a6c0 (LWP 17770) exited]
[Thread 0x7fffef28b6c0 (LWP 17769) exited]
[Thread 0x7ffff0689c80 (LWP 17767) exited]
[Thread 0x7fffcb7fe6c0 (LWP 17782) exited]
[New process 17767]
[Inferior 1 (process 17767) exited with code 01]

This is under Wayland, under X11 the problem did look different.

As things do crash the same way with another user and do not crash with the older version of sddm-kcm6 it looks to me like some regression or some sddm config confusing the newer code.

On that last, this is how with the old sddm-kcm6 the session selection looks like:

Why two times IceWM and Plasma (X11)?

That’s exactly the error I have. What’s your hardware? I’m using an Intel ARC A750 and Wayland. As far as I know I’ve updated regularly and didn’t miss anything.

Note that with the new kcm_sddm I can right click on System Settings in the Application Launcher (Kickoff) and select Login Screen (SDDM) and it’ll go directly to the configuration page without any issues.

run: sudo systemctl disable dispaly-manager
then run: sudo systemctl enable sddm
reboot and then sddm will show up in the KDE system settings

1 Like

The elegant and clean way would be:

systemctl enable --now --force sddm.service

Thanks @BlackBird.BB, that solved one part of the problem.

$ sudo systemctl status display-manager
Unit display-manager.service could not be found.

Checked also sddm and that was not running and not enabled. I did enable it rebooted and came in some web consolse, checked sddm.service and it was not running and not enabled so did start it and the GUI booted, so something changed, before the boot was normal with sddm.service disabled.

sddm is not perfectly happy:

> sudo systemctl status sddm.service
● sddm.service - Simple Desktop Display Manager
     Loaded: loaded (/usr/lib/systemd/system/sddm.service; enabled; preset: disabled)
     Active: active (running) since Thu 2026-06-25 09:47:58 CEST; 1min 14s ago
 Invocation: c2c270fce8b547659a52277b806a6113
       Docs: man:sddm(1)
             man:sddm.conf(5)
   Main PID: 2227 (sddm)
      Tasks: 11 (limit: 153564)
        CPU: 8.159s
     CGroup: /system.slice/sddm.service
             ├─2227 /usr/bin/sddm
             └─2267 /usr/bin/Xorg.bin -nolisten tcp -background none -seat seat0 vt2 -auth /run/sddm/xauth_TUlwuI -noreset -displayfd 16

Jun 25 09:47:59 shadow sddm-helper[2288]: [PAM] returning.
Jun 25 09:47:59 shadow sddm[2227]: Authentication for user  "marel"  successful
Jun 25 09:48:00 shadow sddm-helper[2288]: pam_unix(sddm-autologin:session): session opened for user marel(uid=1000) by marel(uid=0)
Jun 25 09:48:00 shadow sddm-helper[2288]: [PAM] Preparing to converse...
Jun 25 09:48:00 shadow sddm-helper[2288]: [PAM] Conversation with 1 messages
Jun 25 09:48:00 shadow sddm[2227]: Authentication information: SDDM::Auth::INFO_UNKNOWN "Last login: Thu Jun 25 09:31:36 CEST 2026 from :0 on tty2"
Jun 25 09:48:00 shadow sddm-helper[2288]: Writing cookie to "/tmp/xauth_kjEsDC"
Jun 25 09:48:00 shadow sddm-helper[2288]: Starting X11 session: "" "/usr/etc/X11/xdm/Xsession \"/usr/bin/startplasma-x11\""
Jun 25 09:48:00 shadow sddm[2227]: Session started true
Jun 25 09:48:00 shadow sddm-helper[2288]: Failed to write utmpx:  No such file or directory

Notice the preset: disabled and utmpx problem.

System Settings does now show the Login Screen under the Global Theme menu (good) but pressing it still triggers a crash. Back to X11, this is how the crash looks under X11:

[New Thread 0x7fff5adfe6c0 (LWP 5283)]
[New Thread 0x7fff5b5ff6c0 (LWP 5284)]
Failed to compile shader: 
Source was:
#version 150

struct buf
{
    mat4 matrix;
    float opacity;
};

uniform buf _19;

<way more lines>

uniform buf _25;

out vec2 qt_TexCoord;
in vec2 qt_VertexTexCoord;
in vec4 qt_VertexPosition;
in float _qt_order;

void main()
{
    qt_TexCoord = qt_VertexTexCoord;
    gl_Position = _25.qt_Matrix * qt_VertexPosition;
    gl_Position.z = _qt_order * gl_Position.w;
}


Failed to build graphics pipeline state
qrc:/kcm/kcm_sddm/main.qml:197:13: QML QQuickImage*: Protocol "" is unknown

Thread 1 "systemsettings" received signal SIGSEGV, Segmentation fault.
0x00007fffc4e8ed00 in ?? () from /lib64/libgallium-26.1.3.so

I should probably try with QT_QUICK_BACKEND DEFAUILT="software", see here.

Internal graphics:

~> inxi -G
Graphics:
  Device-1: Intel CoffeeLake-S GT2 [UHD Graphics 630] driver: i915 v: kernel
  Display: x11 server: X.Org v: 21.1.21 with: Xwayland v: 24.1.11 driver: X:
    loaded: intel unloaded: modesetting,vesa dri: iris gpu: i915
    resolution: 2560x1440~60Hz
  API: EGL v: 1.5 drivers: iris,swrast platforms: gbm,x11,surfaceless,device
  API: OpenGL v: 4.6 vendor: mesa v: 26.1.3 renderer: llvmpipe (LLVM 22.1.7
    256 bits)
  API: Vulkan v: 1.4.350 drivers: intel,llvmpipe surfaces: N/A
  Info: Tools: api: clinfo, eglinfo, glxinfo, vulkaninfo
    de: kscreen-console,kscreen-doctor gpu: amdgpu_top, gputop, intel_gpu_top,
    lsgpu wl: wayland-info x11: xdpyinfo, xprop, xrandr

Would be good to know if more people with Intel graphics hardware have this problem. Just open System Settings, search for SDDM and open Login Screen (SDDM), does it crash?

1 Like

i915 driver. That doesn’t surprise me.
My Intel ARC A750 also uses the i915 driver. There is an Intel Xe driver that works with it when it wants to. Sometimes it’s pretty stable and right now it’s awful.

Most of Intel’s work on the Xe and Xe2 drivers is for the newer cards. I think it’s all related to the 7.x.x kernel and possibly the new Plasma 6.7.0. I’m not sure when my issue started so I could be wrong on both assumptions.

I’m also having some kwin crashes but there’s no pattern that I’ve pinned down. Those started with 6.7.0.

I installed Fedora and it’s using 6.7.0 but not SDDM. It uses Plasma Login Manager and I do have kwin crashes in Fedora.

So I’m assuming it’s the i915 driver combined with kernel 7 or Plasma 6.7.0, or some combination. Malcolm runs ARC cards so it’d be nice if he chimed in. It’ll probably be fixed and disappear as fast as it appeared.

I waste a lot of time studying things that will get fixed, but it’s not really a waste. I learn a lot about how things work. I just wish that I could remember all of it like I could when I was younger.

I am running a graphical Java application and that now has suddenly it’s Graphics broken at least under X11, not completely but the the menu elements are black unless you mouse over it:

The log gives:

libEGL warning: DRI3 error: Could not get DRI3 device
libEGL warning: Ensure your X server supports DRI3 to get accelerated rendering

Checking:

> xdpyinfo | grep -i dri
    DRI2

No DRI3. will try enabling it using these instructions adding /etc/X11/xorg.conf.d/10-dri3.conf.

I understand Wayland by default using DRI3 so that is likely why the problem manifest itself there differently.

But yes, there seems to be some i915 driver changer or Mesa-libEGL1 change that could be related.

1 Like

Indeed under Wayland the Java application was working fine

Creating that did solve the X11 problem for the Java application.

> cat  /etc/X11/xorg.conf.d/10-dri3.conf
Section "Module"
  Load "dri3"
EndSection

Section "Device"
  Identifier "Intel"
  Option "DRI" "3"
EndSection

So a recent (up to two weeks) update did somehow disable DRI3 for i915…

For the System Settings menu it made the Login Screen (SDDM) menu entry present again under Global Theme but executing it still triggered the crash, “Failed to build graphics pipeline state”.

Yes, @malcolmlewis’s input would be nice.

Hi, no issues here, but I’m on GNOME and Wayland… Haven’t run X11 inprobably a few years here…

at /etc/environment 
## CURRENT ##
## Reduce Mutter debugging output
MUTTER_DEBUG_FORCE_KMS_MODE=simple
## FFMPEG encode/decode options
ANV_VIDEO_DECODE="1"
ANV_DEBUG="video-decode,video-encode"
## END CURRENT ##

I haven’t run the Xe driver in some time, but should be better with Mesa integration, application integration, not so sure.

1 Like