projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 8e65f8d) | patch
AVX-512: Truncating store for i1 vectors
authorElena Demikhovsky <elena.demikhovsky@intel.com>
Mon, 4 Apr 2016 07:17:47 +0000 (07:17 +0000)
committerElena Demikhovsky <elena.demikhovsky@intel.com>
Mon, 4 Apr 2016 07:17:47 +0000 (07:17 +0000)
commite99c5613915db76eb47a04544aca430d86a69368
tree938508613e516f36a9b487262405e0fcc6a1b0f1tree | snapshot
parent8e65f8ddfd88fa6b88152ee914b36a67c051f610commit | diff
AVX-512: Truncating store for i1 vectors
Implemented truncstore for KNL and skylake-avx512.
Covered vectors from v2i1 to v64i1. We save the value in bits (not in bytes) - v32i1 is saved in 4 bytes.

Differential Revision: http://reviews.llvm.org/D18740

llvm-svn: 265283
llvm/include/llvm/Target/TargetLowering.h diff | blob | history
llvm/lib/Target/X86/X86ISelLowering.cpp diff | blob | history
llvm/test/CodeGen/X86/avx512-mask-op.ll diff | blob | history
llvm/test/CodeGen/X86/vector-compare-results.ll diff | blob | history
Domain: System / Toolchain;