[PATCH] D113609: [X86][Costmodel] `getReplicationShuffleCost()`: promote 16 bit-wide elements to 32 bit when no AVX512BW

[Roman Lebedev via Phabricator via llvm-commits]

lebedev.ri added a comment.

In D113609#3127794 <https://reviews.llvm.org/D113609#3127794>, @RKSimon wrote:

> Can you give more examples of the bad cast costs?  trunc <32 x i32> to <32 x i8> -> https://wide.godbolt.org/z/Y3rznPEnY seems to be OK

Do you want me to produce standalone snippets (just sext/trunc) from these examples, and update `llvm/test/Analysis/CostModel/X86/extend.ll` test coverage?
Do we consider these to be blockers? Note that in reality `X86TTIImpl::getReplicationShuffleCost()` will mostly only be asked about i1 elt type.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D113609/new/

https://reviews.llvm.org/D113609