[all-commits] [llvm/llvm-project] ccfe7e: [AArch64] Combine vector FNEG+FMA into `FNML[A|S]`...

[Damian Heaton via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: ccfe7e09540f94e0419c607396e270f8d456e608
      https://github.com/llvm/llvm-project/commit/ccfe7e09540f94e0419c607396e270f8d456e608
  Author: Damian Heaton <Damian.Heaton at arm.com>
  Date:   2025-12-05 (Fri, 05 Dec 2025)

  Changed paths:
    M llvm/lib/Target/AArch64/AArch64ISelLowering.cpp
    M llvm/lib/Target/AArch64/AArch64ISelLowering.h
    M llvm/lib/Target/AArch64/AArch64SVEInstrInfo.td
    A llvm/test/CodeGen/AArch64/sve-fmsub.ll

  Log Message:
  -----------
  [AArch64] Combine vector FNEG+FMA into `FNML[A|S]` (#167900)

This allows for FNEG + FMA sequences to be combined into a single
operation, with `FNML[A|S]`, `FNMAD`, or `FNMSB` selected depending on
the operand order, similarly to how `performSVEMulAddSubCombine` enables
generating `ML[A|S]` instructions by converting the ADD/SUB intrinsics
to scalable vectors.



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