[PATCH][AVX512] Add masked vector blend intrinsics

[Cameron McInally]

Hey guys,

Here is a patch to add intrinsics for the AVX512 masked vector blend
instructions.

Please note that these intrinsics accept the actual mask types (v8i1
and v16i1) instead of the generic i16 operand type.

Tx,
Cameron
-------------- next part --------------
A non-text attachment was scrubbed...
Name: blend.patch
Type: application/octet-stream
Size: 7885 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20131101/e9354672/attachment.obj>