55ce06af6f1196c238581d08f29f99f7475ff270 - platform_external_llvm

commit	55ce06af6f1196c238581d08f29f99f7475ff270	[log] [tgz]
author	Simon Pilgrim <llvm-dev@redking.me.uk>	Sat Oct 01 15:33:01 2016 +0000
committer	Simon Pilgrim <llvm-dev@redking.me.uk>	Sat Oct 01 15:33:01 2016 +0000
tree	abcd007ff8a5376b1c7cad68ef36eb7a885b97be
parent	1c8d24e339205522c3e73d183d500c14aeb34098 [diff]

[X86][SSE] Always combine target shuffles to MOVSD/MOVSS

Now we can commute to BLENDPD/BLENDPS on SSE41+ targets if necessary, so simplify the combine matching where we can.

This required me to add a couple of scalar math movsd/moss fold patterns that hadn't been needed in the past.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@283038 91177308-0d34-0410-b5e6-96231b3b80d8

lib/Target/X86/X86ISelLowering.cpp[diff]
lib/Target/X86/X86InstrAVX512.td[diff]
lib/Target/X86/X86InstrSSE.td[diff]
test/CodeGen/X86/sse-scalar-fp-arith.ll[diff]
test/CodeGen/X86/sse41.ll[diff]
test/CodeGen/X86/vector-blend.ll[diff]
test/CodeGen/X86/vector-shuffle-128-v2.ll[diff]
test/CodeGen/X86/vector-shuffle-combining-ssse3.ll[diff]
test/CodeGen/X86/vector-shuffle-combining-xop.ll[diff]

9 files changed

tree: abcd007ff8a5376b1c7cad68ef36eb7a885b97be