Bug 30847 - [945gm] Serious rendering issues when drawing to front buffer with GL_LIGHTING enabled
Summary: [945gm] Serious rendering issues when drawing to front buffer with GL_LIGHTIN...
Alias: None
Product: Mesa
Classification: Unclassified
Component: Drivers/DRI/i915 (show other bugs)
Version: git
Hardware: x86-64 (AMD64) Linux (All)
: medium normal
Assignee: Eric Anholt
QA Contact:
Depends on:
Reported: 2010-10-13 14:41 UTC by J-P Jokela
Modified: 2011-07-12 15:28 UTC (History)
0 users

See Also:
i915 platform:
i915 features:

Testcase to reproduce the problem. Can be compiled by executing as shell script. (2.18 KB, application/octet-stream)
2010-10-13 14:41 UTC, J-P Jokela

Description J-P Jokela 2010-10-13 14:41:21 UTC
Created attachment 39421 [details]
Testcase to reproduce the problem. Can be compiled by executing as shell script.

Test code fails on 7.9~git20100924-0ubuntu2 version running on Ubuntu 10.10. I haven't tested on other systems yet, and am not sure, in which component the bug actually is.

Expected result:
When running the test program on my nvidia system, it draws same triangle (one vertex at left, one right and third at the top, all centered) in various (constant for all vertices) colors.

Actual result:
When running the same program on my Asus 1215N using the Intel chipset, however, the first triangle is rendered ok, but next one (and all after that) have one red vertex at top left, another red at bottom center, and white top right.

This code works the same on both if I either:
1)Don't draw to front buffer
2)don't glEnable(GL_LIGHTING)

If glEnable(GL_DEPTH_TEST) line is uncommented, only the top line of upside down triangle will be drawn.

While drawing to front buffer probably isn't too common, it can result in whole program becoming useless, when rendering stops working.

The code itself doesn't make much sense, as it's based on bugles exetrace of a lot bigger project, with functions that don't change the outcome removed.
Comment 1 Marek Olšák 2011-03-02 09:26:25 UTC
Works for me with r300g and swrast. Reassigning to Intel. Please specify the driver you use, so that they can set the component correctly.
Comment 2 J-P Jokela 2011-03-09 00:50:34 UTC
Actually late last year, I got rid of the system I could reproduce the bug on (it was the only practically GMA-only system I had). Of course I can no longer get any information regarding the linux installation, but if it helps, it was Asus 1215N.
Comment 3 Eric Anholt 2011-07-12 15:28:11 UTC
While it was broken in Mesa 7.10.3, it works now in 7.11.

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.