[PATCH] D48487: [X86][AVX512] Lowering _mm512_[max|min]_p[s|d] to native IR

[David Greene via Phabricator via cfe-commits]

greened added a comment.

In https://reviews.llvm.org/D48487#1140703, @RKSimon wrote:

> I haven't used the AVX512 versions, but I do know plenty of cases that rely on the 'use the second op if there is a NAN or both zeros' special cases in the SSE version of these FMIN/FMAX intrinsics.


Yes, this behavior is important for AVX512 as well.  Any lowering must preserve the full semantics of the intrinsics.


Repository:
  rC Clang

https://reviews.llvm.org/D48487