projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 18f0950) | patch
[x86, InstCombine] transform x86 AVX2 masked stores to LLVM intrinsics
authorSanjay Patel <spatel@rotateright.com>
Fri, 26 Feb 2016 21:51:44 +0000 (21:51 +0000)
committerSanjay Patel <spatel@rotateright.com>
Fri, 26 Feb 2016 21:51:44 +0000 (21:51 +0000)
commitfc7e7ebf36d63bb65ba285b5f206ec106edc652e
tree2e51855e734f6c8d76b86c58039fecedb7ac2296tree | snapshot
parent18f09507831557a5d4bc8a6fd4ace1ef26a37f65commit | diff
[x86, InstCombine] transform x86 AVX2 masked stores to LLVM intrinsics

Replicate everything for integers...because x86.

Continuation of:
http://reviews.llvm.org/rL262064

llvm-svn: 262077
llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp diff | blob | history
llvm/test/Transforms/InstCombine/x86-masked-memops.ll diff | blob | history
Domain: System / Toolchain;