[AVX-512] Replace masked 16-bit element variable shift intrinsics with new unmasked...
authorCraig Topper <craig.topper@gmail.com>
Fri, 18 Nov 2016 05:04:44 +0000 (05:04 +0000)
committerCraig Topper <craig.topper@gmail.com>
Fri, 18 Nov 2016 05:04:44 +0000 (05:04 +0000)
commit02b5a1b50f6d5b1730680fecf78a35e029e51526
treefa630f9efa7d9a0827bf0a6177de6950bc1c7bc9
parenteff1ad8d8e044a5859afcd604f06f353359aa739
[AVX-512] Replace masked 16-bit element variable shift intrinsics with new unmasked versions and selects.

The same thing was done to 32-bit and 64-bit element sizes previously.

This will allow us to support these shuffls in InstCombineCalls along with the other variable shift intrinsics.

llvm-svn: 287312
llvm/include/llvm/IR/IntrinsicsX86.td
llvm/lib/IR/AutoUpgrade.cpp
llvm/lib/Target/X86/X86IntrinsicsInfo.h
llvm/test/CodeGen/X86/avx512bw-intrinsics.ll
llvm/test/CodeGen/X86/avx512bwvl-intrinsics.ll