Bug#1025312: libgl1-mesa-dri: multiple packages FTBFS and have autopkgtest regressions running test programs under Xvfb
Package: libgl1-mesa-dri
Version: 22.3.0-1
Severity: serious
Justification: FTBFS and autopkgtest regression in affected packages
To reproduce:
- Install a minimal amd64 VM with Debian testing and mesa_22.2.4-1
- apt --no-install-recommends build-dep gtk+3.0
- Build source package gtk+3.0_3.24.35-1 (it succeeds)
- Upgrade all packages except the ones from src:mesa to sid
- Build source package gtk+3.0_3.24.35-1 (it succeeds)
- Upgrade binary packages from src:mesa
- Build source package gtk+3.0_3.24.35-1
or alternatively:
- Install a chroot/container with Debian testing
- Run the autopkgtests for gtk+3.0 (it succeeds)
- Upgrade binary packages from src:mesa to 22.3.0-1
- Run the autopkgtests for gtk+3.0
or more minimally:
- Install a VM/chroot/container with Debian testing
- Install the dependencies of gtk+3.0's autopkgtests
(adwaita-icon-theme-full at-spi2-core dbus-daemon gnome-desktop-testing
gtk-3-examples librsvg2-common xauth xvfb)
- xvfb-run -e /dev/stderr -a /usr/libexec/installed-tests/gtk+/builder
Expected result: successful build from source, or successful autopkgtest
Actual result: FTBFS or failing autopkgtest; the test programs are unable
to connect to the Xvfb server.
The autopkgtests for clutter-1.0, cogl, gtk4, juce, kodi, muffin, mutter,
njplot, openscad and pyopengl show similar symptoms, affecting at least
amd64 and arm64.
The kodi autopkgtest shows this backtrace which might be helpful:
autopkgtest [17:19:52]: test gui: [-----------------------
(EE)
(EE) Backtrace:
(EE) 0: Xvfb (OsLookupColor+0x139) [0x562b8573d239]
(EE) 1: /lib/x86_64-linux-gnu/libc.so.6 (__sigaction+0x40) [0x7fee8f45af90]
(EE) unw_get_proc_name failed: no unwind info found [-10]
(EE) 2: /lib64/ld-linux-x86-64.so.2 (?+0x0) [0x7fee8fdbe029]
(EE) 3: /lib/x86_64-linux-gnu/libc.so.6 (_dl_catch_exception+0x7a) [0x7fee8f56de9a]
(EE) 4: /lib/x86_64-linux-gnu/libc.so.6 (_dl_catch_error+0x2f) [0x7fee8f56df4f]
(EE) 5: /lib/x86_64-linux-gnu/libc.so.6 (dlerror+0x297) [0x7fee8f4a3dc7]
(EE) 6: /lib/x86_64-linux-gnu/libc.so.6 (dlclose+0x36) [0x7fee8f4a3b26]
(EE) 7: /usr/lib/x86_64-linux-gnu/dri/swrast_dri.so (nouveau_drm_screen_create+0x1d50f4) [0x7fee8d70ff44]
(EE) 8: /usr/lib/x86_64-linux-gnu/dri/swrast_dri.so (nouveau_drm_screen_create+0x1d4340) [0x7fee8d70f190]
(EE) unw_get_proc_name failed: no unwind info found [-10]
(EE) 9: /usr/lib/x86_64-linux-gnu/dri/swrast_dri.so (?+0x0) [0x7fee8ccd2179]
(EE) 10: /usr/lib/x86_64-linux-gnu/dri/swrast_dri.so (__driDriverGetExtensions_d3d12+0x60ae16) [0x7fee8d2dd156]
(EE) 11: /usr/lib/x86_64-linux-gnu/dri/swrast_dri.so (__driDriverGetExtensions_d3d12+0x60ad84) [0x7fee8d2dd0c4]
(EE) 12: /usr/lib/x86_64-linux-gnu/dri/swrast_dri.so (__driDriverGetExtensions_d3d12+0x6f4) [0x7fee8ccd2a34]
(EE) 13: /usr/lib/x86_64-linux-gnu/dri/swrast_dri.so (__driDriverGetExtensions_d3d12+0xa175) [0x7fee8ccdc4b5]
(EE) 14: Xvfb (ht_dump_contents+0x81a7) [0x562b856397e7]
(EE) 15: Xvfb (ht_dump_contents+0x6faf) [0x562b856385ef]
(EE) 16: Xvfb (_CallCallbacks+0x34) [0x562b856e8b64]
(EE) 17: Xvfb (ht_dump_contents+0x2863f) [0x562b85659c7f]
(EE) 18: Xvfb (InitExtensions+0x89) [0x562b855fa919]
(EE) 19: Xvfb (InitFonts+0x1e8) [0x562b856e7628]
(EE) 20: /lib/x86_64-linux-gnu/libc.so.6 (__libc_init_first+0x8a) [0x7fee8f44618a]
(EE) 21: /lib/x86_64-linux-gnu/libc.so.6 (__libc_start_main+0x85) [0x7fee8f446245]
(EE) 22: Xvfb (_start+0x21) [0x562b855f8921]
(EE)
(EE) Segmentation fault at address 0x337
(EE)
Fatal server error:
(EE) Caught signal 11 (Segmentation fault). Server aborting
(EE)
xvfb-run: error: Xvfb failed to start
I get a similar backtrace when I try to run
xvfb-run -e /dev/stderr -a /usr/libexec/installed-tests/gtk+/builder
in a qemu VM.
smcv
Reply to: