[all-commits] [llvm/llvm-project] 6526c5: [AMDGPU] Avoid errors with 24-bit div/rem (#209711)

[LU-JOHN via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 6526c5ba127123f2c81b21b3f27bfa69dd623f9e
      https://github.com/llvm/llvm-project/commit/6526c5ba127123f2c81b21b3f27bfa69dd623f9e
  Author: LU-JOHN <John.Lu at amd.com>
  Date:   2026-07-15 (Wed, 15 Jul 2026)

  Changed paths:
    M llvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp
    M llvm/lib/Target/AMDGPU/AMDGPUISelLowering.h
    M llvm/test/CodeGen/AMDGPU/div-rem-fast-path.ll
    M llvm/test/CodeGen/AMDGPU/sdiv.ll
    M llvm/test/CodeGen/AMDGPU/sdivrem24.ll
    M llvm/test/CodeGen/AMDGPU/sdivrem64.r600.ll
    M llvm/test/CodeGen/AMDGPU/udiv.ll
    M llvm/test/CodeGen/AMDGPU/udivrem24.ll
    M llvm/test/CodeGen/AMDGPU/udivrem64.r600.ll

  Log Message:
  -----------
  [AMDGPU] Avoid errors with 24-bit div/rem (#209711)

Avoid errors when using floating-point reciprocal to calculate Y/X when
Y = (0x7FFFFF/X)*X-1. Limit expansion to 23-bit signed and 22-bit
unsigned to avoid this issue.

This change is analogous to the change done in
https://github.com/llvm/llvm-project/pull/202753, but in
AMDGPUISelLowering.cpp.

Signed-off-by: John Lu <John.Lu at amd.com>



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