ab3e6394a0
The mat4x2 portion of the test no longer checks its results, as there are bugs in the Intel and Radeon GPU drivers that prevent mat4x2s from being constructed properly. The SkSL optimizer ends up eliminating the 4x2 matrix entirely because it is unused by the rest of the code, as well at the 4x4 matrix which is calculated from the 4x2. At this point, I'm OK with this. Change-Id: If1464f9e4938b0a37b2ec180c686972389d94e83 Bug: skia:12003 Reviewed-on: https://skia-review.googlesource.com/c/skia/+/408900 Auto-Submit: John Stiles <johnstiles@google.com> Commit-Queue: Ethan Nicholas <ethannicholas@google.com> Reviewed-by: Ethan Nicholas <ethannicholas@google.com> |
||
|---|---|---|
| .. | ||
| blend | ||
| dslfp | ||
| errors | ||
| folding | ||
| fp | ||
| glsl | ||
| inliner | ||
| intrinsics | ||
| metal | ||
| runtime | ||
| runtime_errors | ||
| shared | ||
| spirv | ||
| workarounds | ||
| README.txt | ||
This directory contains source files for testing skslc compilation. The compiled output files are in the /tests/sksl/ directory.