[Bug 97532] Regression: GLB 2.7 & Glmark-2 GLES versions segfault due to linker precision error (259fc505) on dead variable

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=97532

--- Comment #31 from Tomasz Figa <tfiga at chromium.org> ---
(In reply to Eero Tamminen from comment #30)
> (In reply to Tomasz Figa from comment #29)
> > This is also affecting Forge of Empires on Android. With Kenneth's patch and
> > the drirc option enabled for <application name="Default"> the game starts
> > working again.
> 
> IMHO it would be good the check whether the problem is just with a dead
> variable, or is there a real functional issue with the shader.
> 
[...]
> 
> Or dump & inspect the shader code directly.

I have a full trace captured with GAPID, which includes shaders code. Can't
share here due to potential legal issues, but I can share my observations:

- vertex shader starts with "precision highp float",
- fragment shader starts with "precision mediump float",
- both shaders define "uniform mat4 M1" after that,
- both shaders define "uniform mat4 M2" after that,
- "M1" is not used in either of the shaders,
- "M2" is used only in vertex shader,
- the link fails with "error: declarations for uniform `M1` have mismatching
precision qualifiers"

-- 
You are receiving this mail because:
You are the QA Contact for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/intel-3d-bugs/attachments/20170831/9a0038ca/attachment.html>