Arnold Schwaighofer a3fb330d05 LoopVectorizer: Recognize min/max reductions ... A min/max operation is represented by a select(cmp(lt/le/gt/ge, X, Y), X, Y) sequence in LLVM. If we see such a sequence we can treat it just as any other commutative binary instruction and reduce it. This appears to help bzip2 by about 1.5% on an imac12,2. radar://12960601 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179773 91177308-0d34-0410-b5e6-96231b3b80d8		2013-04-18 17:22:34 +00:00
..
BBVectorize.cpp	BBVectorize: Fixup debugging statements	2013-03-10 20:57:42 +00:00
CMakeLists.txt	Add support for bottom-up SLP vectorization infrastructure.	2013-04-09 19:44:35 +00:00
LLVMBuild.txt
LoopVectorize.cpp	LoopVectorizer: Recognize min/max reductions	2013-04-18 17:22:34 +00:00
Makefile
SLPVectorizer.cpp	SLPVectorizer: Make it a function pass and add code for hoisting the vector-gather sequence out of loops.	2013-04-15 22:00:26 +00:00
Vectorize.cpp	Rename the C function to create a SLPVectorizerPass to something sane and expose it in the header file.	2013-04-11 11:36:36 +00:00
VecUtils.cpp	SLPVectorizer: Make it a function pass and add code for hoisting the vector-gather sequence out of loops.	2013-04-15 22:00:26 +00:00
VecUtils.h	SLPVectorizer: Make it a function pass and add code for hoisting the vector-gather sequence out of loops.	2013-04-15 22:00:26 +00:00