[all-commits] [llvm/llvm-project] 564142: [PowerPC] Use 'sync; ld; cmp; bc; isync' for atomi...

[Kai Luo via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 56414220dfeb274a15beb55ab3da757978a0255f
      https://github.com/llvm/llvm-project/commit/56414220dfeb274a15beb55ab3da757978a0255f
  Author: Kai Luo <lkail at cn.ibm.com>
  Date:   2023-12-20 (Wed, 20 Dec 2023)

  Changed paths:
    M llvm/lib/Target/PowerPC/PPCISelLowering.cpp
    M llvm/lib/Target/PowerPC/PPCInstrInfo.cpp
    M llvm/lib/Target/PowerPC/PPCInstrInfo.td
    M llvm/test/CodeGen/PowerPC/atomics-indexed.ll
    M llvm/test/CodeGen/PowerPC/atomics.ll

  Log Message:
  -----------
  [PowerPC] Use 'sync; ld; cmp; bc; isync' for atomic load seq-cst on 32-bit platform (#75905)

`cmp; bc; isync` is more performant than `lwsync` theoretically.

64-bit platform already features it, now implement it for 32-bit
platform.