[all-commits] [llvm/llvm-project] 63fd5f: [AMDGPU] Optimize i64 uitofp for unsigned byte values

[Harrison Hao via All-commits]

  Branch: refs/heads/users/harrisonGPU/optimize-uitofp-ubyte
  Home:   https://github.com/llvm/llvm-project
  Commit: 63fd5fe3387f3c1ee8c5bebb60b7f87b6b308c9a
      https://github.com/llvm/llvm-project/commit/63fd5fe3387f3c1ee8c5bebb60b7f87b6b308c9a
  Author: Harrison Hao <tsworld1314 at gmail.com>
  Date:   2026-09-11 (Fri, 11 Sep 2026)

  Changed paths:
    M llvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp
    M llvm/lib/Target/AMDGPU/AMDGPUISelLowering.h
    M llvm/test/CodeGen/AMDGPU/cvt_f32_ubyte.ll

  Log Message:
  -----------
  [AMDGPU] Optimize i64 uitofp for unsigned byte values

Use v_cvt_f32_ubyte0 when the i64 source is known to fit in an unsigned
byte.
For example:

  uitofp (and i64 %x, 255) to float

This avoids the generic i64 to f32 expansion.



To unsubscribe from these emails, change your notification settings at https://github.com/llvm/llvm-project/settings/notifications