OSDN Git Service

(root) / android-x86 / external-llvm-project.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 6520976) | patch
AMDGPU: Enhancement on FDIV lowering in AMDGPUCodeGenPrepare
authorChangpeng Fang <changpeng.fang@gmail.com>
Fri, 7 Feb 2020 19:46:23 +0000 (11:46 -0800)
committerChangpeng Fang <changpeng.fang@gmail.com>
Fri, 7 Feb 2020 19:46:23 +0000 (11:46 -0800)
commit884acbb9e167d5668e43581630239d688edec8ad
tree661150491e8998d41688c820acc2600066bbebb0tree | snapshot
parent6520976064c4c59397dd02e72d42e9cff3a2a8dccommit | diff
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
llvm/lib/Target/AMDGPU/AMDGPUCodeGenPrepare.cpp diff | blob | history
llvm/lib/Target/AMDGPU/SIISelLowering.cpp diff | blob | history
llvm/test/CodeGen/AMDGPU/amdgpu-codegenprepare-fdiv.ll diff | blob | history
llvm/test/CodeGen/AMDGPU/fdiv.f16.ll diff | blob | history
external/llvm-project
About OSDN
Find Software
Develop Software
Help