[all-commits] [llvm/llvm-project] 376f9b: [AMDGPU] Fold 24 bit multiply with zero low bits

[Harrison Hao via All-commits]

  Branch: refs/heads/users/harrisonGPU/i24-dag-opt
  Home:   https://github.com/llvm/llvm-project
  Commit: 376f9b62f944deadcbe9186fbc0ee63f1a153970
      https://github.com/llvm/llvm-project/commit/376f9b62f944deadcbe9186fbc0ee63f1a153970
  Author: Harrison Hao <tsworld1314 at gmail.com>
  Date:   2026-09-19 (Sat, 19 Sep 2026)

  Changed paths:
    M llvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp
    M llvm/test/CodeGen/AMDGPU/llvm.amdgcn.mul.i24.ll

  Log Message:
  -----------
  [AMDGPU] Fold 24 bit multiply with zero low bits

Fold `MUL_I24` and `MUL_U24` to zero when either operand has known zero
low 24 bits.

For example:
```
  llvm.amdgcn.mul.i24(x, 0x01000000) -> 0
```



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