[all-commits] [llvm/llvm-project] b2497e: [PowerPC] Add generic fnmsub intrinsic

[Qiu Chaofan via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: b2497e54356d454d0e16d8f44086bf6db6aff0e3
      https://github.com/llvm/llvm-project/commit/b2497e54356d454d0e16d8f44086bf6db6aff0e3
  Author: Qiu Chaofan <qiucofan at cn.ibm.com>
  Date:   2022-03-07 (Mon, 07 Mar 2022)

  Changed paths:
    M clang/lib/CodeGen/CGBuiltin.cpp
    M clang/test/CodeGen/PowerPC/builtins-ppc-fma.c
    M clang/test/CodeGen/PowerPC/builtins-ppc-fpconstrained.c
    M clang/test/CodeGen/PowerPC/builtins-ppc-vsx.c
    M clang/test/CodeGen/PowerPC/builtins-ppc-xlcompat-math.c
    M llvm/include/llvm/IR/IntrinsicsPowerPC.td
    M llvm/lib/Target/PowerPC/PPCISelLowering.cpp
    M llvm/lib/Target/PowerPC/PPCInstrInfo.td
    M llvm/lib/Target/PowerPC/PPCInstrVSX.td
    M llvm/test/CodeGen/PowerPC/builtins-ppc-xlcompat-math.ll

  Log Message:
  -----------
  [PowerPC] Add generic fnmsub intrinsic

Currently in Clang, we have two types of builtins for fnmsub operation:
one for float/double vector, they'll be transformed into IR operations;
one for float/double scalar, they'll generate corresponding intrinsics.

But for the vector version of builtin, the 3 op chain may be recognized
as expensive by some passes (like early cse). We need some way to keep
the fnmsub form until code generation.

This patch introduces ppc.fnmsub.* intrinsic to unify four fnmsub
intrinsics.

Reviewed By: shchenz

Differential Revision: https://reviews.llvm.org/D116015