[all-commits] [llvm/llvm-project] b9c394: [PowerPC] Generate inlined quadword lock free atom...

[bzEq via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: b9c3941cd61de1e1b9e4f3311ddfa92394475f4b
      https://github.com/llvm/llvm-project/commit/b9c3941cd61de1e1b9e4f3311ddfa92394475f4b
  Author: Kai Luo <lkail at cn.ibm.com>
  Date:   2021-07-15 (Thu, 15 Jul 2021)

  Changed paths:
    M llvm/include/llvm/IR/IntrinsicsPowerPC.td
    M llvm/lib/CodeGen/AtomicExpandPass.cpp
    M llvm/lib/Target/PowerPC/CMakeLists.txt
    M llvm/lib/Target/PowerPC/PPC.h
    M llvm/lib/Target/PowerPC/PPC.td
    A llvm/lib/Target/PowerPC/PPCExpandAtomicPseudoInsts.cpp
    M llvm/lib/Target/PowerPC/PPCISelLowering.cpp
    M llvm/lib/Target/PowerPC/PPCISelLowering.h
    M llvm/lib/Target/PowerPC/PPCInstr64Bit.td
    M llvm/lib/Target/PowerPC/PPCInstrInfo.td
    M llvm/lib/Target/PowerPC/PPCSubtarget.h
    M llvm/lib/Target/PowerPC/PPCTargetMachine.cpp
    M llvm/test/CodeGen/PowerPC/O3-pipeline.ll
    A llvm/test/CodeGen/PowerPC/atomics-i128.ll

  Log Message:
  -----------
  [PowerPC] Generate inlined quadword lock free atomic operations via AtomicExpand

This patch uses AtomicExpandPass to implement quadword lock free atomic operations. It adopts the method introduced in https://reviews.llvm.org/D47882, which expand atomic operations post RA to avoid spilling that might prevent LL/SC progress.

Reviewed By: jsji

Differential Revision: https://reviews.llvm.org/D103614