Bug 22056 - [UMS DRI2] vblank waits timeout
Summary: [UMS DRI2] vblank waits timeout
Status: CLOSED DUPLICATE of bug 22166
Alias: None
Product: Mesa
Classification: Unclassified
Component: Drivers/DRI/i915 (show other bugs)
Version: unspecified
Hardware: Other Linux (All)
: medium major
Assignee: Jesse Barnes
QA Contact:
URL:
Whiteboard:
Keywords:
: 23707 (view as bug list)
Depends on:
Blocks:
 
Reported: 2009-06-03 01:16 UTC by liuhaien
Modified: 2010-07-13 23:34 UTC (History)
1 user (show)

See Also:
i915 platform:
i915 features:

Attachments
xorg.0.log (407.74 KB, text/plain)
2009-06-03 01:16 UTC, liuhaien 		Details
xorg conf file (4.19 KB, text/plain)
2009-06-03 01:17 UTC, liuhaien 		Details
View All

Description liuhaien 2009-06-03 01:16:27 UTC 
Created attachment 26381 [details]
xorg.0.log

System Environment:
--------------------------
Host:		x-aspire1
Arch:		i386
Platform:	945GME	
Libdrm:		(master)dbffeadcfca7cb4e3905ec0b8573d10e0418295e
Mesa:		(mesa_7_5_branch)1aef032d438aaa40ec28bf279ad5c089370773f0
Xserver:	(server-1.6-branch)5cd5a01259ba349f1868ca4af04207cf120d69e4
Xf86_video_intel:		(master)704771f1c595df481d244a6f074e9f5620d4a720
Kernel:       (for-linus)07f4f3e8a24138ca2f3650723d670df25687cd05

Bug detailed description:
--------------------------
startx with UMS,all renderings under compiz become very slow.and I think it is caused by vlank, and only happens with LVDS(pipe 0?). it works well with VGA.
same issue happens on gm45.dmesg shows error like below:
[drm:i915_get_vblank_counter] *ERROR* trying to get vblank count for disabled pipe 0

Reproduce Steps:
-----------------------
1.xinit&
2.gnome-session&
3. enable compiz(direct)
4.run any application
Comment 1 liuhaien 2009-06-03 01:17:39 UTC 
Created attachment 26382 [details]
xorg conf file
Comment 2 Gordon Jin 2009-06-03 13:00:06 UTC 
you indicated it's fine with KMS, right?
Comment 3 liuhaien 2009-06-03 17:36:40 UTC 
(In reply to comment #2)
> you indicated it's fine with KMS, right?
> 

yes, KMS is ok.
Comment 4 Gordon Jin 2009-06-09 22:18:10 UTC 
Jesse, this is the bug I just mentioned to you on the car.
Comment 5 Jesse Barnes 2009-06-10 08:00:06 UTC 
Can you bisect it?
Comment 6 liuhaien 2009-06-28 20:40:59 UTC 
I have returned to our Q4 release image, but this issue still exists with below commits. 
xf86_video_intel   2.6.0 release
mesa   intel-2008-q4 branch
       commit 200fa9165d7078a6f36c5c9d3e0c997c2438bde3 (before and close to 7.3)
kernel  drm-intel-2.6.28 branch
        commit e1a6fcee467556a7e955fe1f7ccc134dd2f974e7 (2.6.28 + 6 patches)
libdrm  2.4.4 release
xserver 1.6

I think this issue comes out because we use direct rendering for compiz since 2009 Q1. and indirect rendering was utilized before, so that we didn't meet the issue.
Comment 7 Jesse Barnes 2009-10-05 10:30:23 UTC 
Updated testing instructions for this code at http://dri.freedesktop.org/wiki/CompositeSwap.  glsync should work better with the new DRI2 features.
Comment 8 Jesse Barnes 2009-10-05 10:33:28 UTC 
*** Bug 23707 has been marked as a duplicate of this bug. ***
Comment 9 Jesse Barnes 2009-10-05 10:37:51 UTC 

*** This bug has been marked as a duplicate of bug 22166 ***

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.