projects / oota-llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 3aa9b49) | patch
Vectorize intrinsic math function calls in SLPVectorizer.
authorKarthik Bhat <kv.bhat@samsung.com>
Sat, 3 May 2014 09:59:54 +0000 (09:59 +0000)
committerKarthik Bhat <kv.bhat@samsung.com>
Sat, 3 May 2014 09:59:54 +0000 (09:59 +0000)
commit486ad6262e4d513d426c19448b55103b3fff0aad
treef4f4fcccfaf2061ccaac6ffa550fcca74af5fd71tree | snapshot
parent3aa9b4911c18d06369285ff35ae6c99d6f168828commit | diff
Vectorize intrinsic math function calls in SLPVectorizer.
This patch adds support to recognize and vectorize intrinsic math functions in SLPVectorizer.
Review: http://reviews.llvm.org/D3560 and http://reviews.llvm.org/D3559

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@207901 91177308-0d34-0410-b5e6-96231b3b80d8
include/llvm/Transforms/Utils/VectorUtils.h diff | blob | history
lib/Transforms/Vectorize/LoopVectorize.cpp diff | blob | history
lib/Transforms/Vectorize/SLPVectorizer.cpp diff | blob | history
test/Transforms/SLPVectorizer/X86/call.ll [new file with mode: 0644] blob
C/C++ LLVM-based compilers that forbids OOTA behaviors