[PATCH] D34046: [AMDGPU] Add intrinsics for alignbit and alignbyte instructions

[Stanislav Mekhanoshin via Phabricator via llvm-commits]

rampitec added a comment.

In https://reviews.llvm.org/D34046#776849, @b-sumner wrote:

> This looks fine to me.  I was concerned about how successful we would be pattern matching them.


We do not match it after numerous of optimizations, it is better to use intrinsic from the library.


Repository:
  rL LLVM

https://reviews.llvm.org/D34046