[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 #33 from Eero Tamminen <eero.t.tamminen at intel.com> ---
Ok, so it's the same case as with GLB (and git version of fixed Glmark), the
shaders are violating the spec, but it's not a functionality problem because
variables in question are just declared with conflicting precision, but not
used with conflicting precision.

It would be interesting to know whether the other drivers that allow this
behavior, would fail when variables with mismatched precision are actually used
across shader stages.

And does that behavior change depending on whether application is using GLES
2.x or GLES 3.x context (GLES 2.x spec didn't require this check, it came
afterwards, but Mesa promotes all GLES 2.x contexts to 3.x).

-- 
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/20170831/213e0a88/attachment.html>