[NVPTX] Improve handling of FP fusion
authorJustin Holewinski <jholewinski@nvidia.com>
Thu, 17 Jul 2014 18:10:09 +0000 (18:10 +0000)
committerJustin Holewinski <jholewinski@nvidia.com>
Thu, 17 Jul 2014 18:10:09 +0000 (18:10 +0000)
commit11ae250ec9d1adbd15fec36ccaf6e01a0fb1e518
tree4eeed8daf9971fb6defbc9fbddb6f33186eb47f1
parent15865afcf777b2323d56e02a8d62e295fd21b7ed
[NVPTX] Improve handling of FP fusion

We now consider the FPOpFusion flag when determining whether
to fuse ops.  We also explicitly emit add.rn when fusion is
disabled to prevent ptxas from fusing the operations on its
own.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@213287 91177308-0d34-0410-b5e6-96231b3b80d8
lib/Target/NVPTX/NVPTXISelDAGToDAG.cpp
lib/Target/NVPTX/NVPTXISelDAGToDAG.h
lib/Target/NVPTX/NVPTXISelLowering.cpp
lib/Target/NVPTX/NVPTXISelLowering.h
lib/Target/NVPTX/NVPTXInstrInfo.td
test/CodeGen/NVPTX/arithmetic-fp-sm20.ll
test/CodeGen/NVPTX/fma.ll
test/CodeGen/NVPTX/fp-contract.ll [new file with mode: 0644]
test/CodeGen/NVPTX/fp-literals.ll
test/CodeGen/NVPTX/implicit-def.ll