[PATCH] D26333: [AVX-512] Add unmasked version of shift by immediate and shift by single element in XMM.

[Craig Topper via llvm-commits]

craig.topper added a comment.

Elena, this the first patch of a four step plan.

1. Add new unmasked intrinsics
2. Add support for new unmasked intrinsics to InstCombineCalls
3. Switch clang to new unmasked intrinsics
4. Remove masked intrinsics and auto upgrade them to the unmasked instrinsic plus select.

So in the end there wont' be duplicate intrinsics. But there will be quite a bit of autoupgrade cases.


https://reviews.llvm.org/D26333