Bug ID 1160807
Summary X fails to start after updating to version 202006 (segfault with radeon driver)
Classification openSUSE
Product openSUSE Tumbleweed
Version Current
Hardware i686
OS Other
Status NEW
Severity Major
Priority P5 - None
Component X.Org
Assignee xorg-maintainer-bugs@forge.provo.novell.com
Reporter schaefer.frank@gmx.net
QA Contact xorg-maintainer-bugs@forge.provo.novell.com
Found By ---
Blocker ---

After upgrading from an end-of October-version to 202006, X doesn't start
anymore.
Unfortunately I can't switch back to a working version.
Relevant lines in Xorg.o.log (see attachment) seem to be:

...
[    63.275] (EE) Failed to open authorization file
"/run/sddm/{cc28ba16-c77b-499a-a4d5-86c4ea584c2c}": No such file or directory
[    63.593] (EE) 
[    63.593] (EE) Backtrace:
[    63.593] (EE) 0: /usr/bin/X (xorg_backtrace+0x52) [0x66e0f2]
[    63.593] (EE) 1: /usr/bin/X (0x4c7000+0x1aafd2) [0x671fd2]
[    63.593] (EE) 2: linux-gate.so.1 (__kernel_rt_sigreturn+0x0) [0xb7f6faa4]
[    63.593] (EE) 3: /usr/lib/xorg/modules/drivers/radeon_drv.so
(0xb70e1000+0x4863f) [0xb712963f]
[    63.594] (EE) 4: /usr/bin/X (miSyncInitFence+0x6c) [0x642e9c]
[    63.594] (EE) 5: /usr/bin/X (0x4c7000+0x17c700) [0x643700]
[    63.594] (EE) 6: /usr/bin/X (0x4c7000+0xdf501) [0x5a6501]
[    63.594] (EE) 7: /usr/bin/X (0x4c7000+0x17813c) [0x63f13c]
[    63.594] (EE) 8: /usr/bin/X (0x4c7000+0x469f6) [0x50d9f6]
[    63.594] (EE) 9: /usr/bin/X (0x4c7000+0x4aeeb) [0x511eeb]
[    63.594] (EE) 10: /usr/bin/X (0x4c7000+0x33f5b) [0x4faf5b]
[    63.594] (EE) 11: /lib/libc.so.6 (__libc_start_main+0xf1) [0xb7738ec1]
[    63.594] (EE) 12: /usr/bin/X (_start+0x31) [0x4fafa1]
[    63.594] (EE) 
[    63.594] (EE) Segmentation fault at address 0x8
[    63.594] (EE) 
Fatal server error:
[    63.594] (EE) Caught signal 11 (Segmentation fault). Server aborting
...

Might be related to bug 1160637. Differences are:
- my graphics card (HD6570) uses the older "radeon" driver instead of "amdgpu"
- stack trace looks a bit different
- I'm using a 32 bit installation
- result is segfault (signal 11) instead of abort (signal 6)

After the segfault, X restart seems to be triggered automatically, resulting in
an endless loop (X restart - segfault - X restart - segfault - ...).
That might be a separate bug.


You are receiving this mail because: