[all-commits] [llvm/llvm-project] 5ae9d2: [AMDGPU][AtomicExpand] Use full flat emulation if ...

[Shilei Tian via All-commits]

  Branch: refs/heads/users/shiltian/use-full-flat-emulation-for-double-if-feasible
  Home:   https://github.com/llvm/llvm-project
  Commit: 5ae9d26e786d2f0e6c058d239b059b5248903b20
      https://github.com/llvm/llvm-project/commit/5ae9d26e786d2f0e6c058d239b059b5248903b20
  Author: Shilei Tian <i at tianshilei.me>
  Date:   2025-06-04 (Wed, 04 Jun 2025)

  Changed paths:
    M llvm/lib/Target/AMDGPU/SIISelLowering.cpp
    M llvm/test/CodeGen/AMDGPU/atomicrmw-expand.ll
    M llvm/test/CodeGen/AMDGPU/flat-atomicrmw-fadd.ll
    M llvm/test/CodeGen/AMDGPU/gep-const-address-space.ll
    M llvm/test/CodeGen/AMDGPU/infer-addrspace-flat-atomic.ll
    M llvm/test/Transforms/AtomicExpand/AMDGPU/expand-atomic-rmw-fadd.ll
    M llvm/test/Transforms/AtomicExpand/AMDGPU/expand-atomicrmw-flat-noalias-addrspace.ll

  Log Message:
  -----------
  [AMDGPU][AtomicExpand] Use full flat emulation if feasible

If a target supports f64 atomic add instruction, we can also use full flat
emulation.



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