[all-commits] [llvm/llvm-project] 204a41: AutoUpgrade: Use syncscope("agent") atomic.inc/dec...

[Matt Arsenault via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 204a417d5155c4944b8e6e3033f5b6072e723553
      https://github.com/llvm/llvm-project/commit/204a417d5155c4944b8e6e3033f5b6072e723553
  Author: Matt Arsenault <Matthew.Arsenault at amd.com>
  Date:   2023-08-10 (Thu, 10 Aug 2023)

  Changed paths:
    M llvm/lib/IR/AutoUpgrade.cpp
    M llvm/test/Bitcode/amdgcn-atomic.ll

  Log Message:
  -----------
  AutoUpgrade: Use syncscope("agent") atomic.inc/dec intrinsic upgrade

The old syncscope parameter never really worked correctly, but
effectively gave "workgroup" scope. Use something faster than system
but more correct than before.

https://reviews.llvm.org/D157389


  Commit: 7575ee7167a9c6071f34cb81c3f2c2861d644e5e
      https://github.com/llvm/llvm-project/commit/7575ee7167a9c6071f34cb81c3f2c2861d644e5e
  Author: Matt Arsenault <Matthew.Arsenault at amd.com>
  Date:   2023-08-10 (Thu, 10 Aug 2023)

  Changed paths:
    M llvm/test/CodeGen/AMDGPU/flat_atomics_i32_system.ll
    M llvm/test/CodeGen/AMDGPU/flat_atomics_i64_system.ll
    M llvm/test/CodeGen/AMDGPU/global_atomics_i32_system.ll
    M llvm/test/CodeGen/AMDGPU/global_atomics_i64_system.ll
    M llvm/test/Transforms/AtomicExpand/AMDGPU/expand-atomic-i16-system.ll
    M llvm/test/Transforms/AtomicExpand/AMDGPU/expand-atomic-i16.ll
    M llvm/test/Transforms/AtomicExpand/AMDGPU/expand-atomic-rmw-fadd.ll
    M llvm/test/Transforms/AtomicExpand/AMDGPU/expand-atomic-rmw-fsub.ll

  Log Message:
  -----------
  AMDGPU: Add more test coverage for FP-typed atomicrmw xchg


Compare: https://github.com/llvm/llvm-project/compare/9e94f7dcb55a...7575ee7167a9