OSDN Git Service

ARM Cost model: Use the size of vector registers and widest vectorizable instruction...
authorNadav Rotem <nrotem@apple.com>
Wed, 9 Jan 2013 22:29:00 +0000 (22:29 +0000)
committerNadav Rotem <nrotem@apple.com>
Wed, 9 Jan 2013 22:29:00 +0000 (22:29 +0000)
commit14925e6b885f8bd8cf448627386d412831f4bf1b
treef87fac291a2f8740eac16cadbd051b395dd1a200
parentd2f8216ba496b139815aee29052093c1b0d7a996
ARM Cost model: Use the size of vector registers and widest vectorizable instruction to determine the max vectorization factor.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@172010 91177308-0d34-0410-b5e6-96231b3b80d8
include/llvm/Analysis/TargetTransformInfo.h
lib/Analysis/TargetTransformInfo.cpp
lib/CodeGen/BasicTargetTransformInfo.cpp
lib/Target/ARM/ARMTargetTransformInfo.cpp
lib/Target/X86/X86TargetTransformInfo.cpp
lib/Transforms/Vectorize/LoopVectorize.cpp
test/Transforms/LoopVectorize/ARM/gcc-examples.ll [new file with mode: 0644]
test/Transforms/LoopVectorize/X86/avx1.ll
test/Transforms/LoopVectorize/X86/conversion-cost.ll