[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

Eero Tamminen <eero.t.tamminen at intel.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |REOPENED
             Blocks|                            |98335
         Resolution|NOTOURBUG                   |---

--- Comment #19 from Eero Tamminen <eero.t.tamminen at intel.com> ---
With Mesa bug 97804 and glmark2 upstream code fixed, GLB 2.7 is the only
currently known test-suite/program this affects on *Linux desktop* [1].

Kenneth marked this early on as "NOTOURBUG" which is valid, but not necessarily
best option.  GLB 2.7 is still used (despite its problems, it's probably most
widely used GLES 2.x benchmark) and Mesa now enforcing GLES 3.x error for
harmless & earlier accepted inconsistency in GLES 2.0 programs, is a
functionality regression (although Intel backend doesn't yet even support
FP16).

Kenneth, I'm opening this to get (your or somebody else) opinion on adding user
workaround for this, e.g. the attached trivial patch.


[1] Linux desktop has only handful of GLES programs (and most of them
compositors i.e. not really using complex shaders).  There could be other
regressing cases on ChromeOS (WebGL www-pages) or Android.


Referenced Bugs:

https://bugs.freedesktop.org/show_bug.cgi?id=98335
[Bug 98335] [TRACKER] Intel release tracker for Mesa 13.0
-- 
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/20161102/0915e235/attachment.html>