projects / oota-llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: e3f14b0) | patch
[X86][SSE] Added support for lowering to ADDSUBPS/ADDSUBPD with commuted inputs
authorSimon Pilgrim <llvm-dev@redking.me.uk>
Sun, 29 Nov 2015 16:41:04 +0000 (16:41 +0000)
committerSimon Pilgrim <llvm-dev@redking.me.uk>
Sun, 29 Nov 2015 16:41:04 +0000 (16:41 +0000)
commit86ccb922275baa98620d96673071144a0f7d71ec
tree20a4f0a85d9ef6de5d2d898861b96c2d8be8c8b6tree | snapshot
parente3f14b0e8c863c216d98928044c01ee8a2ee1d73commit | diff
[X86][SSE] Added support for lowering to ADDSUBPS/ADDSUBPD with commuted inputs

We could already recognise shuffle(FSUB, FADD) -> ADDSUB, this allow us to recognise shuffle(FADD, FSUB) -> ADDSUB by commuting the shuffle mask prior to matching.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@254259 91177308-0d34-0410-b5e6-96231b3b80d8
lib/Target/X86/X86ISelLowering.cpp diff | blob | history
test/CodeGen/X86/sse3-avx-addsub.ll diff | blob | history
C/C++ LLVM-based compilers that forbids OOTA behaviors