[all-commits] [llvm/llvm-project] fba082: [X86] Fold FMSUBADD into VFCMULC for fp16 complex ...

[Rohan Shenoy via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: fba0822290721dd8144708d6c28902a87e3bb830
      https://github.com/llvm/llvm-project/commit/fba0822290721dd8144708d6c28902a87e3bb830
  Author: Rohan Shenoy <RohanRadhesh.Shenoy at amd.com>
  Date:   2026-09-23 (Wed, 23 Sep 2026)

  Changed paths:
    M llvm/lib/Target/X86/X86ISelLowering.cpp
    A llvm/test/CodeGen/X86/avx512fp16-combine-fmsubadd.ll

  Log Message:
  -----------
  [X86] Fold FMSUBADD into VFCMULC for fp16 complex conjugate multiply (#224533)

We reuse isCFMulFromFMSUBADD from #222896 to fold vfmsubadd into vfcmulc.
As a result using this pattern-match in combineFaddCFmul becomes
redundant as it (c += a * ~b) still becomes vfcmaddcph through the
existing fadd + vfcmulc combine.



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