llvm-project
8062a40b - AMDGPU/GlobalISel: Fold (trunc (lshr x, 16)) to unmerge high half

Commit
78 days ago
AMDGPU/GlobalISel: Fold (trunc (lshr x, 16)) to unmerge high half Huge performance impact in some shaders with high register pressure. What this really does is selects _hi16 register classes, and true16 machinery takes over later. Trunc + shift ends up in _lo16! This requires allocating an extra register when the _lo16 half of the original real 32-bit VGPR is still live.
Committer
Parents
Loading