Bug 91459

Summary: Hard crash when logging out of X
Product: xorg Reporter: Chris Rankin <rankincj>
Component: Driver/RadeonAssignee: xf86-video-ati maintainers <xorg-driver-ati>
Status: RESOLVED MOVED QA Contact: Xorg Project Team <xorg-team>
Severity: normal    
Priority: medium    
Version: git   
Hardware: x86 (IA32)   
OS: Linux (All)   
Whiteboard:
i915 platform: i915 features:

Description Chris Rankin 2015-07-25 13:00:26 UTC
I have a Radeon HD4650 running in a Thinkpad docking station. The T60p laptop's built-in M66GL chip has been disabled so that the HD4650 is the only graphics device on the PCI bus:

lspci:
0c:00.0 0300: 1002:9498
0c:00.1 0403: 1002:aa38

The HD4650 has two DVI adapters, and one is attached to a 1920x1200 monitor. The other is attached to a HDTV via an HDMI adapter, and the TV is not always turned on.

When Fedora 22 was first released, I noticed that the laptop crashed (hard!) whenever I logged out of X / GNOME shell, and had to be power-cycled to restart. I have now narrowed the circumstances of this crash down to when the TV is connected to the DVI port but not powered.

The laptop does not crash if either the TV is on or on stand-by, or if the HDMI adapter is physically disconnected from the HD4650's DVI port.

I have not been able to find any log files relating to the crash.
Comment 1 Chris Rankin 2015-07-25 13:16:06 UTC
And then I log out, and the laptop immediately locks up again! So I'll amend my original report to say that disconnecting the HDMI adapter *doesn't* prevent the problem after all, despite appearing to with my initial tests!

I've currently reconnected the HDMI adapter and put the TV on standby. xrandr now reports the following:

Screen 0: minimum 320 x 200, current 3840 x 1200, maximum 8192 x 8192
DVI-0 connected 1920x1080+1920+0 (normal left inverted right x axis y axis) 698mm x 392mm
   1920x1080     50.00*+  60.00  
DIN disconnected (normal left inverted right x axis y axis)
DVI-1 connected primary 1920x1200+0+0 (normal left inverted right x axis y axis) 546mm x 352mm
   1920x1200     59.95*+
   1600x1200     60.00  
   1680x1050     59.88  
   1600x1000     60.01  
   1280x1024     75.02  
   1280x960      60.00  
   1152x864      75.00  
   1024x768      75.08    60.00  
   832x624       74.55  
   800x600       75.00    60.32  
   640x480       75.00    60.00  
   720x400       70.08  

I don't know whether GNOME shell remembers any "state" between sessions relating to numbers and size of screens - something like that could have affected my previous tests with DVI-0 appearing and disappearing.
Comment 2 Chris Rankin 2015-08-06 00:19:48 UTC
On closer inspection, the "hard lockup" looks more like a broken attempt at "Suspend to RAM". The last time I reproduced this, I'm sure I briefly glimpsed the GNOME3 "screensaver" lock page before the screen went blank and turned itself off. However, I've no idea why GNOME3 should try to suspend itself to RAM whenever I log out with my HDTV off either!? Nor does the laptop even try to restore itself when I press the "power" button.
Comment 3 Chris Rankin 2015-08-09 21:54:36 UTC
Of course, this could all be a Wayland bug really!?! I have discovered that Fedora 22 is running gdm using Wayland, and the bug does happen when logging out... or is it when restarting gdm to login again?
Comment 4 Chris Rankin 2015-10-31 21:57:33 UTC
I am no longing seeing this bug, and am wondering if this patch has fixed it:

commit 7186a8713ba004de4991f21c1a9fc4abc62aeff4
Author: Stephen Chandler Paul <cpaul@redhat.com>
Date:   Fri Oct 23 09:59:36 2015 -0400

    Handle failures in setting a CRTC to a DRM mode properly
    
    This fixes a bug where running the card out of PPLL's when hotplugging
    another monitor would result in all of the displays going blank and
    failing to work properly until X was restarted or the user switched to
    another VT.

But I'll keep checking before declaring this bug to be fixed, seeing as I've been wrong before.
Comment 5 Martin Peres 2019-11-19 07:51:25 UTC
-- GitLab Migration Automatic Message --

This bug has been migrated to freedesktop.org's GitLab instance and has been closed from further activity.

You can subscribe and participate further through the new bug through this link to our GitLab instance: https://gitlab.freedesktop.org/xorg/driver/xf86-video-ati/issues/135.

Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.