projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 1ce0c37) | patch
[X86] AVX512: Enable it in the Loop Vectorizer
authorAdam Nemet <anemet@apple.com>
Wed, 9 Jul 2014 18:22:33 +0000 (18:22 +0000)
committerAdam Nemet <anemet@apple.com>
Wed, 9 Jul 2014 18:22:33 +0000 (18:22 +0000)
commit2820a5b9e974456ace273fbfeea2747dca94c212
treea48f09e226b02a1c99282fc4e218e5ae4222c736tree | snapshot
parent1ce0c37bf058d493ab320c6bd8d60693aebe92c9commit | diff
[X86] AVX512: Enable it in the Loop Vectorizer

This lets us experiment with 512-bit vectorization without passing
force-vector-width manually.

The code generated for a simple integer memset loop is properly vectorized.
Disassembly is still broken for it though :(.

llvm-svn: 212634
llvm/lib/Target/X86/X86TargetTransformInfo.cpp diff | blob | history
llvm/test/Transforms/LoopVectorize/X86/avx512.ll [new file with mode: 0644] blob
Domain: System / Toolchain;