OSDN Git Service

[AMDGPU] Preserve operand order in SIFoldOperands
authorStanislav Mekhanoshin <Stanislav.Mekhanoshin@amd.com>
Sat, 3 Jun 2017 00:41:52 +0000 (00:41 +0000)
committerStanislav Mekhanoshin <Stanislav.Mekhanoshin@amd.com>
Sat, 3 Jun 2017 00:41:52 +0000 (00:41 +0000)
commitced381c0388fe1635f6dfe5e65f80b980aa7207e
tree6cdcc2520643fce5e3dd5744a0341852d8f25780
parent239d957e9eb0958df956294cfcf7697fd34f0e73
[AMDGPU] Preserve operand order in SIFoldOperands

SIFoldOperands can commute operands even if no folding was done.
This change is to preserve IR is no folding was done.

Differential Revision: https://reviews.llvm.org/D33802

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@304625 91177308-0d34-0410-b5e6-96231b3b80d8
lib/Target/AMDGPU/SIFoldOperands.cpp
test/CodeGen/AMDGPU/commute-compares.ll
test/CodeGen/AMDGPU/insert_vector_elt.v2i16.ll
test/CodeGen/AMDGPU/sub.i16.ll