Bug 1160715

Summary: X crash with lightdm and amdgpu on latest TW (20200109-0)
Product: [openSUSE] openSUSE Tumbleweed Reporter: Chris H <raymanfx>
Component: X.OrgAssignee: E-mail List <xorg-maintainer-bugs>
Status: RESOLVED DUPLICATE QA Contact: E-mail List <xorg-maintainer-bugs>
Severity: Major    
Priority: P5 - None    
Version: Current   
Target Milestone: ---   
Hardware: x86-64   
OS: openSUSE Factory   
Whiteboard:
Found By: --- Services Priority:
Business Priority: Blocker: ---
Marketing QA Status: --- IT Deployment: ---
Attachments: Xorg.0.log
Log from a working X system in the older snapshot

Description Chris H 2020-01-11 16:21:56 UTC 
User-Agent:       Mozilla/5.0 (X11; Linux x86_64; rv:71.0) Gecko/20100101 Firefox/71.0
Build Identifier: 

Happens after a regular "zypper dup".

I had to roll back to 20200106-0, which works fine (lightdm comes up).

Reproducible: Always

Steps to Reproduce:
1. Update to 20200109-0
2. Set lightdm as default display manager, reboot to graphical target
3. ???
Actual Results:  
X crashes

Expected Results:  
X should be fine, lightdm should come up

I'm using the standard amdgpu driver with a RX Vega 56.
CPU is a Ryzen 1800x.

Desktop environment is Pantheon (stable branch).
Interestingly, attempting to start "gdm" results in the same crash.

The kernel I am using is self-compiled, but works after the rollback with the older system, so that can probably be ruled out.
Comment 1 Chris H 2020-01-11 16:23:35 UTC 
Created attachment 827373 [details]
Xorg.0.log

Doesn't look like the stacktrace is too helpful - although I wonder about this line:

"Setting screen physical size to 1016 x 571"

because I am running a 27" 4k LG monitor.
Comment 2 Chris H 2020-01-11 16:27:23 UTC 
Created attachment 827374 [details]
Log from a working X system in the older snapshot
Comment 3 Chris H 2020-01-11 16:30:21 UTC 
Uploaded a (good) X log to compare against.
Curious differences:

[GOOD]
[    12.804] (II) GLX: Initialized DRI2 GL provider for screen 0

[BAD]
[    12.628] (II) AIGLX: Screen 0 is not DRI2 capable

Moreover, there's no mention of "radeonsi" in the [BAD] log.
Comment 4 Stefan Dirsch 2020-01-12 16:11:15 UTC 
Looks like a duplicate.

*** This bug has been marked as a duplicate of bug 1160637 ***