[all-commits] [llvm/llvm-project] 3a62ed: [AArch64] MachineCombiner msub matching

[Allen via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 3a62edcf527bfc927f148c99b9d37d3d199db611
      https://github.com/llvm/llvm-project/commit/3a62edcf527bfc927f148c99b9d37d3d199db611
  Author: zhongyunde 00443407 <zhongyunde at huawei.com>
  Date:   2024-03-08 (Fri, 08 Mar 2024)

  Changed paths:
    M llvm/lib/Target/AArch64/AArch64InstrInfo.cpp
    A llvm/test/CodeGen/AArch64/scalar-mla-mls.ll

  Log Message:
  -----------
  [AArch64] MachineCombiner msub matching

Pattern should be sorted in priority order since the pattern evalutor
stops checking as soon as it finds a faster sequence.
so for a * b - c * d, we prefer to match the 2nd operands of sub,
which can be use msub to fold them.

Refer to https://www.slideshare.net/chimerawang/instruction-combine-in-llvm

Fix https://github.com/llvm/llvm-project/issues/84152


  Commit: a110a1c0ed9850be168cd0e29f05179e80941b04
      https://github.com/llvm/llvm-project/commit/a110a1c0ed9850be168cd0e29f05179e80941b04
  Author: zhongyunde 00443407 <zhongyunde at huawei.com>
  Date:   2024-03-08 (Fri, 08 Mar 2024)

  Changed paths:
    M llvm/lib/Target/AArch64/AArch64InstrInfo.cpp
    M llvm/test/CodeGen/AArch64/scalar-mla-mls.ll

  Log Message:
  -----------
  [AArch64] MachineCombiner msub matching for i64


Compare: https://github.com/llvm/llvm-project/compare/d36d805373c2...a110a1c0ed98

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