[all-commits] [llvm/llvm-project] 921e89: [SVE] Only combine (fneg (fma)) => FNMLA with nsz

[Peter Waller via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 921e89c59a71ca3487e175164afba15e76ae2e09
      https://github.com/llvm/llvm-project/commit/921e89c59a71ca3487e175164afba15e76ae2e09
  Author: Peter Waller <peter.waller at arm.com>
  Date:   2021-12-13 (Mon, 13 Dec 2021)

  Changed paths:
    M llvm/lib/CodeGen/SelectionDAG/SelectionDAGBuilder.cpp
    M llvm/lib/Target/AArch64/AArch64ISelLowering.cpp
    M llvm/lib/Target/AArch64/AArch64SVEInstrInfo.td
    M llvm/test/CodeGen/AArch64/sve-fp-combine.ll

  Log Message:
  -----------
  [SVE] Only combine (fneg (fma)) => FNMLA with nsz

-(Za + Zm * Zn) != (-Za + Zm * (-Zn))
when the FMA produces a zero output (e.g. all zero inputs can produce -0
output)

Add a PatFrag to check presence of nsz on the fneg, add tests which
ensure the combine does not fire in the absense of nsz.

See https://reviews.llvm.org/D90901 for a similar discussion on X86.

Differential Revision: https://reviews.llvm.org/D109525