Skip to content

glsl-1.10: Avoid NIR optimization in gl_FrontFacing test

Ian Romanick requested to merge idr/piglit:review/mesa-issue-5876 into main

Use some non-obvious arithmetic to try to avoid some NIR optimizations that hide a bug in the Intel compiler backend for Gfx12+ (Tigerlake and newer) GPUs. Basically, when NIR sees

   x = (gl_FrontFacing ? -1.0 : 1.0) * a;

it converts it to

    x = gl_FrontFacing ? -a : a;

This avoids the bad code generation for the (gl_FrontFacing ? -1.0 : 1.0) case.

See also mesa#5876 (closed)

Edited by Ian Romanick

Merge request reports