AVX512: Fix truncate v32i8 to v32i1 lowering implementation.
authorIgor Breger <igor.breger@intel.com>
Thu, 28 Jan 2016 13:19:25 +0000 (13:19 +0000)
committerIgor Breger <igor.breger@intel.com>
Thu, 28 Jan 2016 13:19:25 +0000 (13:19 +0000)
commitfca0a34398fda3966ad930d9ec18af12a54c2f23
tree675856915cb084ae6a89923921fcd32703c40838
parent92d4c328d11d533725e68c085d49cab8641a048d
AVX512: Fix truncate v32i8 to v32i1 lowering implementation.
Enable truncate 128/256bit packed byte/word with AVX512BW but without AVX512VL, use 512bit instructions.

Differential Revision: http://reviews.llvm.org/D16531

llvm-svn: 259044
llvm/lib/Target/X86/X86ISelLowering.cpp
llvm/lib/Target/X86/X86InstrAVX512.td
llvm/test/CodeGen/X86/avx512-bugfix-26264.ll [new file with mode: 0644]
llvm/test/CodeGen/X86/avx512-calling-conv.ll
llvm/test/CodeGen/X86/avx512-ext.ll
llvm/test/CodeGen/X86/avx512-mask-op.ll
llvm/test/CodeGen/X86/avx512-skx-insert-subvec.ll
llvm/test/CodeGen/X86/masked_gather_scatter.ll
llvm/test/CodeGen/X86/vector-shuffle-v1.ll