Package: gnome-shell
Version: 47.3-1
Severity: normal
Dear Debian folks,
Due to a regression [1], I downgraded from unstable to testing, and put
*gnome-shell* on hold.
run0 apt-mark hold gnome-shell
Now after working again, the user reported again the GNOME fail screen:
Mär 18 11:27:41 laptop gnome-shell[973]: Running GNOME Shell (using
mutter 47.4) as a Wayland display server
Mär 18 11:27:42 laptop gnome-shell[973]: Thread 'KMS thread' will
be using high priority scheduling
Mär 18 11:27:42 laptop gnome-shell[973]: Device '/dev/dri/card0'
prefers shadow buffer
Mär 18 11:27:42 laptop gnome-shell[973]: Added device
'/dev/dri/card0' (i915) using atomic mode setting.
Mär 18 11:27:42 laptop gnome-shell[973]: Failed to initialize
accelerated iGPU/dGPU framebuffer sharing: KMS CRTC doesn't support GBM
format
Mär 18 11:27:42 laptop gnome-shell[973]: Created gbm renderer for
'/dev/dri/card0'
Mär 18 11:27:42 laptop gnome-shell[973]: GPU /dev/dri/card0
selected primary from builtin panel presence
Mär 18 11:27:42 laptop gnome-shell[973]: Obtained a high priority
EGL context
Mär 18 11:27:43 laptop gnome-shell[973]: g_variant_get_string:
assertion 'value != NULL' failed
Mär 18 11:27:43 laptop gnome-shell[973]: == Stack trace for context
0x56071af654b0 ==
Mär 18 11:27:43 laptop gnome-shell[973]: The mapping function given
to g_settings_get_mapped() for key 'overlay-key' in schema
'org.gnome.mutter' returned FALSE when given a NULL value.
Mär 18 11:27:43 laptop systemd-coredump[1013]: [🡕] Process 973
(gnome-shell) of user 1001 dumped core.
laptop Stack trace of
thread 973:
laptop #0
0x00007ff395afd8ac n/a (n/a + 0x0)
laptop ELF object binary
architecture: AMD x86-64
Mär 18 11:27:43 laptop systemd[812]:
org.gnome.Shell@wayland.service: Main process exited, code=dumped,
status=5/TRAP
Mär 18 11:27:43 laptop systemd[812]:
org.gnome.Shell@wayland.service: Failed with result 'core-dump'.
Mär 18 11:27:43 laptop systemd[812]: Failed to start
org.gnome.Shell@wayland.service - GNOME Shell on Wayland.
Mär 18 11:27:43 laptop systemd[812]:
org.gnome.Shell@wayland.service: Consumed 817ms CPU time, 135.7M memory
peak.
(The same with X11.)
Looking at the dpkg logs, *mutter-common* seems to have been upgraded to
48~rc-4, that entered testing on March 14th.
Upgrading to the stack in the suite *unstable*, fixed the issue in the
mean time, but it looks like more dependencies need to be denoted to
prevent such a problem in the future.
Kind regards,
Paul
[1]: https://gitlab.gnome.org/GNOME/mutter/-/issues/3954
