[all-commits] [llvm/llvm-project] 3bffe6: [mlir][VectorOps] Lower vector.fma to llvm.fmuladd...

[Benjamin Kramer via All-commits]

  Branch: refs/heads/master
  Home:   https://github.com/llvm/llvm-project
  Commit: 3bffe6022cc96a38ad0f6ada5f5b7b41eca5796e
      https://github.com/llvm/llvm-project/commit/3bffe6022cc96a38ad0f6ada5f5b7b41eca5796e
  Author: Benjamin Kramer <benny.kra at googlemail.com>
  Date:   2020-07-13 (Mon, 13 Jul 2020)

  Changed paths:
    M mlir/lib/Conversion/VectorToLLVM/ConvertVectorToLLVM.cpp
    M mlir/test/Conversion/VectorToLLVM/vector-to-llvm.mlir

  Log Message:
  -----------
  [mlir][VectorOps] Lower vector.fma to llvm.fmuladd instead of llvm.fma

Summary:
These are semantically equivalent, but fmuladd allows decaying the op
into fmul+fadd if there is no fma instruction available. llvm.fma lowers
to scalar calls to libm fmaf, which is a lot slower.

Reviewers: nicolasvasilache, aartbik, ftynse

Subscribers: mehdi_amini, rriddle, jpienaar, shauheen, antiagainst, arpith-jacob, mgester, lucyrfox, liufengdb, stephenneuendorffer, Joonsoo, grosul1, Kayjukh, jurahul, msifontes

Tags: #mlir

Differential Revision: https://reviews.llvm.org/D83666