[all-commits] [llvm/llvm-project] 884acb: AMDGPU: Enhancement on FDIV lowering in AMDGPUCode...

[Changpeng Fang via All-commits]

  Branch: refs/heads/master
  Home:   https://github.com/llvm/llvm-project
  Commit: 884acbb9e167d5668e43581630239d688edec8ad
      https://github.com/llvm/llvm-project/commit/884acbb9e167d5668e43581630239d688edec8ad
  Author: Changpeng Fang <changpeng.fang at gmail.com>
  Date:   2020-02-07 (Fri, 07 Feb 2020)

  Changed paths:
    M llvm/lib/Target/AMDGPU/AMDGPUCodeGenPrepare.cpp
    M llvm/lib/Target/AMDGPU/SIISelLowering.cpp
    M llvm/test/CodeGen/AMDGPU/amdgpu-codegenprepare-fdiv.ll
    M llvm/test/CodeGen/AMDGPU/fdiv.f16.ll

  Log Message:
  -----------
  AMDGPU: Enhancement on FDIV lowering in AMDGPUCodeGenPrepare

Summary:
  The accuracy limit to use rcp is adjusted to 1.0 ulp from 2.5 ulp.
Also, afn instead of arcp is used to allow inaccurate rcp to be used.

Reviewers:
  arsenm

Differential Revision: https://reviews.llvm.org/D73588