Files
clang-p2996/llvm/lib/Target/AMDGPU/AMDGPUCodeGenPrepare.cpp
Pierre van Houtryve aacebaeab5 [AMDGPU] Do not promote uniform i16 operations to i32 in CGP (#140208)
For the majority of cases, this is a neutral or positive change.
There are even testcases that greatly benefit from it, but some regressions are possible.
There is #140040 for GlobalISel that'd need to be fixed but it's only a one instruction regression and I think it can be fixed later.

Solves #64591
2025-05-16 10:31:03 +02:00

83 KiB