projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 9748369) | patch
[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)
commit428cf0e49a26c256b0cb4992a00f748500d71bb4
treeba7557ff2a7f63daa24ae032998ddd52f85e2560tree | snapshot
parent97483694e7e7a9d07d39ac7fee66b10a7e5e2953commit | diff
[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.

llvm-svn: 213287
llvm/lib/Target/NVPTX/NVPTXISelDAGToDAG.cpp diff | blob | history
llvm/lib/Target/NVPTX/NVPTXISelDAGToDAG.h diff | blob | history
llvm/lib/Target/NVPTX/NVPTXISelLowering.cpp diff | blob | history
llvm/lib/Target/NVPTX/NVPTXISelLowering.h diff | blob | history
llvm/lib/Target/NVPTX/NVPTXInstrInfo.td diff | blob | history
llvm/test/CodeGen/NVPTX/arithmetic-fp-sm20.ll diff | blob | history
llvm/test/CodeGen/NVPTX/fma.ll diff | blob | history
llvm/test/CodeGen/NVPTX/fp-contract.ll [new file with mode: 0644] blob
llvm/test/CodeGen/NVPTX/fp-literals.ll diff | blob | history
llvm/test/CodeGen/NVPTX/implicit-def.ll diff | blob | history
Domain: System / Toolchain;