Bug 526433

Summary: [KMS] X crashes on ATI Radeon X800 (R420)
Product: [Fedora] Fedora Reporter: Joachim Frieben <jfrieben>
Component: kernelAssignee: Kernel Maintainer List <kernel-maint>
Status: CLOSED RAWHIDE QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: medium Docs Contact:
Priority: low    
Version: rawhideCC: dougsland, gansalmon, itamar, kernel-maint, pebolle
Target Milestone: ---   
Target Release: ---   
Hardware: x86_64   
OS: Linux   
Whiteboard:
Fixed In Version: xorg-x11-server-Xorg-1.6.99.903-2.fc12.x86_64 Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2009-10-04 19:05:35 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Attachments:
Description Flags
Xorg.0.log for ATI Radeon X800 (R420) with KMS
none
Output of dmesg for ATI Radeon X800 (R420) with KMS and drm.debug=0x0f none

Description Joachim Frieben 2009-09-30 08:57:28 UTC
Description of problem:
For a current "rawhide" system sporting an "ATI Radeon X800 SE (R430) (PCIE)", X is broken when KMS is enabled. The system boots up as expected but gets stuck in a cycle of failed attempts to bring up X/GDM. Log file /var/log/messages exhibits the following recurrent section:

  ..
  kernel: integrated sync not supported
  kernel: [drm:drm_mode_getfb] *ERROR* invalid framebuffer id
  kernel: [drm] DAC-10: set mode 1400x1050@75 2d
  kernel: integrated sync not supported
  kernel: integrated sync not supported
  kernel: integrated sync not supported
  abrtd: Directory 'ccpp-1254287712-1984' creation detected
  abrtd: Lock file '/var/cache/abrt/ccpp-1254287712-1984.lock' is locked
           by process 2024
  abrt: saved core dump of pid 1984 to
           /var/cache/abrt/ccpp-1254287712-1984/coredump
  gdm-simple-greeter[2022]: Gtk-WARNING: cannot open display: :3.0
  gdm-simple-slave[1983]: WARNING: Child process -2005 was already dead.
  gdm-simple-slave[1983]: WARNING: Unable to kill D-Bus daemon
  gdm-binary[1790]: WARNING: GdmDisplay: display lasted 2.660207 seconds
  abrtd: Getting local universal unique identification...
  kernel: integrated sync not supported
  abrtd: Crash is in database already
  abrtd: Already saved crash, deleting...
  kernel: integrated sync not supported
  ..

Version-Release number of selected component (if applicable):
kernel-2.6.31.1-56.fc12.x86_64

How reproducible:
Always.

Steps to Reproduce:
1. Boot kernel 2.6.31.1-56.fc12 with active KMS.
  
Actual results:
When X is supposed to be launched, the system enters a cycle of failed attempts to bring up GDM.

Expected results:
X starts up successfully, and the GDM login panel appears on the screen.

Additional info:
Booting the system with option "nomodeset" allows to start X without issues.

Comment 1 Joachim Frieben 2009-09-30 09:27:16 UTC
Created attachment 363158 [details]
Xorg.0.log for ATI Radeon X800 (R420) with KMS

After booting into run level 3, executing 'startx' leads to a mouse pointer appearing in front of a black background. However, after a moment, the monitor (21" CRT HP A4576A) powers off. The system does not respond to ctrl-alt-del. Nevertheless, it is still possible to shut down the system in a controlled manner by pressing the soft power button.

Comment 2 Paul Bolle 2009-09-30 23:03:41 UTC
0) Similar issue with a "Radeon RV250 [Mobility FireGL 9000]" on an i686 system. "nomodeset" seems to be a workaround here too.

1) Not sure if its kernel related per se. I now see identical issues on both kernel-2.6.31.1-48.fc12.i686 and kernel-2.6.31.1-56.fc12.i686. At least I first noticed this with kernel-2.6.31.1-56.fc12.i686, but as kernel-2.6.31.1-48.fc12.i686 didn't trigger this issue before, I suppose one of the xorg packages that came in the same batch of updates as kernel-2.6.31.1-56.fc12.i686 (e.g. xorg-x11-drv-ati-6.13.0-0.6.20090929git7968e1fb8.fc12, xorg-x11-server-Xorg-1.6.99.903-1.fc12, and xorg-x11-server-common-1.6.99.903-1.fc12) could also be the culprit.

Comment 3 Joachim Frieben 2009-10-03 08:57:48 UTC
Created attachment 363545 [details]
Output of dmesg for ATI Radeon X800 (R420) with KMS and drm.debug=0x0f

Comment 4 Paul Bolle 2009-10-04 08:14:12 UTC
(In reply to comment #2)
> 1) [...] I suppose one of
> the xorg packages that came in the same batch of updates as
> kernel-2.6.31.1-56.fc12.i686 (e.g.
> xorg-x11-drv-ati-6.13.0-0.6.20090929git7968e1fb8.fc12,
> xorg-x11-server-Xorg-1.6.99.903-1.fc12, and
> xorg-x11-server-common-1.6.99.903-1.fc12) could also be the culprit.  

Issue seems to be resolved here (with a "Radeon RV250 [Mobility FireGL 9000]" on an i686 system) with xorg-x11-server-Xorg-1.6.99.903-2.fc12.i686 and friends (i.e., the current Rawhide packages). Can reporter confirm?