Bug 678115

Summary: [PALM] [KMS:HD6250:WRESTLER:FUSION] entering GNOME fails and hangs system - acer Aspire One 522 (AMD C-50)
Product: [Fedora] Fedora Reporter: Adam Williamson <awilliam>
Component: xorg-x11-drv-atiAssignee: Jérôme Glisse <jglisse>
Status: CLOSED CURRENTRELEASE QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: high Docs Contact:
Priority: unspecified    
Version: 15CC: airlied, xgl-maint
Target Milestone: ---Keywords: Triaged
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2011-06-01 06:36:36 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Bug Depends On:    
Bug Blocks: 678116    
Attachments:
Description Flags
Xorg.0.log from affected system (booted to runlevel 3, startx then run)
none
dmesg from affected system
none
/var/log/messages from affected system none

Description Adam Williamson 2011-02-16 19:04:28 UTC 
I have here a pretty new netbook, Acer Aspire One 522, which is based on a low-end implementation of the AMD Fusion stuff, the C-50 CPU with graphics identified in Xorg.0.log as HD 6250. If I boot F15 Alpha TC2 desktop live image, splash screen and GDM come up fine, but on logging in, I get a screen with a pointer and the background but nothing else, and after a bit the system hangs (can't move the pointer or switch to a VT). If I boot to runlevel 3 and run startx as liveuser, X will start up and be running in the background and as long as I stay in console VTs I can do stuff as long as I like, the system doesn't hang...but as soon as I actually switch to the VT where X went, I see the background and a pointer and nothing else, as in the direct-to-X boot case. I can move the cursor for a couple of seconds, but then it hangs just the same as in the direct-to-X boot case.

From /var/log/messages it appears that a full X session does actually run, it's just not visible. I can see messages from NetworkManager and PackageKit and all sorts of other things that clearly come from apps which have got started as part of the X session. .xsession-errors doesn't actually exist for liveuser, so there's nothing in there. /var/log/Xorg.0.log looks unhelpful also. I will include it, though. There's a lot of drm stuff in dmesg with drm.debug=15, so I'll include that. I'll also shortly test and see if a non-accelerated desktop like Xfce will run.
Comment 1 Adam Williamson 2011-02-16 19:05:24 UTC 
Created attachment 479195 [details]
Xorg.0.log from affected system (booted to runlevel 3, startx then run)
Comment 2 Adam Williamson 2011-02-16 19:05:55 UTC 
Created attachment 479196 [details]
dmesg from affected system
Comment 3 Adam Williamson 2011-02-16 19:06:21 UTC 
Created attachment 479197 [details]
/var/log/messages from affected system
Comment 4 Jérôme Glisse 2011-03-29 21:28:55 UTC 
Any more luck after updating to lastest stuff ?
Comment 5 Adam Williamson 2011-03-29 21:43:47 UTC 
hold on, i'll grab a nightly and check.
Comment 6 Adam Williamson 2011-03-29 22:31:40 UTC 
With a 2011-03-27 nightly, no joy. Boots direct to a background and movable cursor but no desktop at all, ps claims gnome-shell is running. .xsession-errors is full of telepathy spam but also has recurrent:

Clutter-CRITICAL **: Unable to make the stage window 0x200009 the current GLX drawable"

errors. 2011-03-27 nightly has kernel 2.6.38-1.fc15 , xorg-x11-drv-ati 6.14.0-5.20110204gitblahblah , and mesa 7.10-0.29; do you need me to test newer versions of any of those? if so i'll have to roll a custom live.
Comment 7 Adam Williamson 2011-03-29 22:49:32 UTC 
I'm building a custom live with all the latest kernel, xorg and mesa stuff to see if that's different.
Comment 8 Adam Williamson 2011-03-29 23:42:30 UTC 
with xorg-x11-server-1.10.0-6.fc15 , mesa-7.11-0.1.20110327.0.fc15 , kernel-2.6.38.2-8.fc15 , and xorg-x11-drv-ati-6.14.0-7.20110316gitcdfc007ec.fc15 , it works! Shell comes up and seems to run fine in quick testing. So we need to get all that stuff into dist-f15, I'll check if updates are filed.