[Bug 92743] Centroid shouldn't have to match between the FS and the VS

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=92743

            Bug ID: 92743
           Summary: Centroid shouldn't have to match between the FS and
                    the VS
           Product: Mesa
           Version: 10.5
          Hardware: x86-64 (AMD64)
                OS: Linux (All)
            Status: NEW
          Severity: normal
          Priority: medium
         Component: glsl-compiler
          Assignee: idr at freedesktop.org
          Reporter: corentin at wallez.net
        QA Contact: intel-3d-bugs at lists.freedesktop.org

While running the GLES3 dEQP tests on ANGLE there was a failure that seemed
related to a misinterpretation of an underspecified part of the GLSL spec.

This was running on an Ubuntu variant with the followin:
OpenGL renderer string: Mesa DRI Intel(R) HD Graphics 5500 (Broadwell GT2) 
OpenGL core profile version string: 3.3 (Core Profile) Mesa 10.5.9
OpenGL core profile shading language version string: 3.30

The problem is that glLinkProgram enforces varyings to have the same centroid
qualifier in the VS and FS. The GLSL 330 spec isn't clear on the this but the
GLSL ES 3.1 spec explicitly says centroid doesn't have to match. Also starting
from the GLSL 420 the list of interpolation qualifier is explicitly given and
centroid is not one of them. To finish there was a Khronos bug about that issue
which resolution was that centroid didn't have to match.

The relevant dEQP test is
dEQP-GLES3.functional.shaders.linkage.varying.rules.differing_interpolation_2,
you can find it by searching for "differing_interpolation" in
https://android.googlesource.com/platform/external/deqp/+/master/data/gles3/shaders/linkage.test

-- 
You are receiving this mail because:
You are the QA Contact for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/intel-3d-bugs/attachments/20151030/9ad1601f/attachment.html>