Bug 21781 - [965GM S4] vblank error when suspending and graphics corruption on resume
Summary: [965GM S4] vblank error when suspending and graphics corruption on resume
Status: RESOLVED FIXED
Alias: None
Product: xorg
Classification: Unclassified
Component: Driver/intel (show other bugs)
Version: git
Hardware: Other All
: medium normal
Assignee: Jesse Barnes
QA Contact: Xorg Project Team
URL:
Whiteboard:
Keywords: NEEDINFO
Depends on:
Blocks:
 
Reported: 2009-05-17 09:14 UTC by James Newton
Modified: 2009-06-23 17:38 UTC (History)
0 users

See Also:
i915 platform:
i915 features:


Attachments
xorg.conf (1.48 KB, text/plain)
2009-05-17 09:14 UTC, James Newton
no flags Details
dmesg (49.65 KB, text/plain)
2009-05-17 09:14 UTC, James Newton
no flags Details
xorg.log (32.15 KB, text/plain)
2009-05-17 09:15 UTC, James Newton
no flags Details
reg dump before (11.00 KB, text/plain)
2009-05-17 09:17 UTC, James Newton
no flags Details
reg dump after (11.00 KB, text/plain)
2009-05-17 09:17 UTC, James Newton
no flags Details

Description James Newton 2009-05-17 09:14:23 UTC
Created attachment 25934 [details]
xorg.conf

After putting the laptop into sleep to disk and resuming got some corruption of the display. In updated areas of windows. Mainly the top window bar, popup menus, and changed text. After 10 minutes it disappeared on its own and is now working correctly. Looking in the logs I see these two messages. The first one was during the resume of the kernel and the second was when the corruption stopped.

[drm:i915_get_vblank_counter] *ERROR* trying to get vblank count for disabled pipe 0

[drm:drm_wait_vblank] *ERROR* failed to acquire vblank counter, -22

While trying to reproduce this bug and get intel_reg_dump output the first message appeared again but the corruption did not. nor did the second message happen.

Environment details:

chipset: 965GM
arch: i686
versions: I think I got the git commit ids right. But if not these are all current against the freedesktop repositories as of 9am MDT 04-17-2009.
  drm: f57d7f4b0b14972f92a83f155ae8033478aa7729
  xf86-video-intel: 87332a7cc16af82aa47e07fbf90da3635b071dbf
  mesa: 08ec7e0d329a72433b427e8167b2c3442d1f53b4
  drm-intel-next: 7c7327d9664280cad833da1f14bad13a3ea8f0bf
kernel: 2.6.30-rc5-15812-g7c7327d
distribution: Gentoo (up to date)
machine: Lenovo X61s
display connector: LVDS laptop display
kernel config: KMS is on
Comment 1 James Newton 2009-05-17 09:14:45 UTC
Created attachment 25935 [details]
dmesg
Comment 2 James Newton 2009-05-17 09:15:43 UTC
Created attachment 25936 [details]
xorg.log
Comment 3 James Newton 2009-05-17 09:17:23 UTC
Created attachment 25937 [details]
reg dump before
Comment 4 James Newton 2009-05-17 09:17:43 UTC
Created attachment 25938 [details]
reg dump after
Comment 5 Jesse Barnes 2009-06-23 17:31:57 UTC
Hm, regdumps don't have anything interesting.  Do you still see this with more recent kernel bits?  Best if you could try git://git.kernel.org/pub/scm/linux/kernel/git/anholt/drm-intel.git (drm-intel-next branch).
Comment 6 James Newton 2009-06-23 17:36:46 UTC
Oh I'm sorry, I forgot about this bug report. I'm running 2.6.30-rc8-16278-g03d6069 now and its been working like a champ on suspend/resume. So problem fixed. :)
Comment 7 Jesse Barnes 2009-06-23 17:38:56 UTC
Yay, thanks for the update.


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.