Bug 64413 - [SNA]Rendercheck triangles fail with SNA enable
Summary: [SNA]Rendercheck triangles fail with SNA enable
Status: RESOLVED NOTOURBUG
Alias: None
Product: xorg
Classification: Unclassified
Component: Driver/intel (show other bugs)
Version: unspecified
Hardware: All Linux (All)
: medium normal
Assignee: Chris Wilson
QA Contact: Intel GFX Bugs mailing list
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2013-05-10 06:19 UTC by lu hua
Modified: 2015-05-19 10:05 UTC (History)
0 users

See Also:
i915 platform:
i915 features:

Attachments
output (170.09 KB, text/plain)
2013-05-10 06:19 UTC, lu hua 		no flags Details
View All

Description lu hua 2013-05-10 06:19:35 UTC 
Created attachment 79073 [details]
output

System Environment:
--------------------------
Libdrm:		(master)2.4.44-6-g011999927f76a7e9ba8f047fae4b4e084da6c2c3
Mesa:		(master)00035670de83f96499415bb8ef65374312505f5c
Xserver:	(master)xorg-server-1.14.99.1-76-g6993f8b459e25b1a7a0a03e209688a28ce6c2c56
Xf86_video_intel:(master)2.21.6-26-g262ee1ef1e98cb84fa0af6e679c8cd61dc93f008
Cairo:		(master)1704292e493b3c635e115df59d07330d19b39514
Libva:		(staging)4853e4a3a6cc5218e032a19d729096862db82bc6
Libva_intel_driver:(staging)5bd630d4ea732818728b0a464f45739c86e08be3
Kernel:	(drm-intel-nightly) c169e7eefc95e2c2df94766cad67efb0fe6284ea

Bug detailed description:
-----------------------------
It fails with SNA driver, and works well with EXA driver.

Reproduce steps:
----------------------------
1. xinit
2. ./rendercheck -o src,over,overreverse,xor -t triangles
Comment 1 Chris Wilson 2013-05-10 06:52:17 UTC 
rendercheck changed.
Comment 2 Chris Wilson 2013-05-10 06:59:17 UTC 
Perhaps to be more precise, the semantics of pixman_triangles does not match the existing semantics of CompositeTriangles (and friends) and rendercheck was modified so that pixman passed and everything else failed.
Comment 3 Gordon Jin 2015-05-14 02:37:34 UTC 
Sounds like there's still bug. 
Chris, or do you mean this is rendercheck bug by NOTOURBUG?
Comment 4 Chris Wilson 2015-05-15 13:29:57 UTC 
Yes, they changed the semantics of the render operation and then changed rendercheck such that it passed again.
Comment 5 Gordon Jin 2015-05-18 04:32:09 UTC 
(In reply to Chris Wilson from comment #4)
> Yes, they changed the semantics of the render operation and then changed
> rendercheck such that it passed again.

If it's rendercheck bug, how do we track the bug?
Comment 6 Chris Wilson 2015-05-19 10:05:28 UTC 
Also important to recognise is that triangles does not pass with either new or old Xvfb (or -modesetting). For rasterisation accuracy, I compared against cairo and prefer explicit comparison against Xvfb.

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.