[PATCH] D111222: [X86][Costmodel] Fix `X86TTIImpl::getGSScalarCost()`

[Roman Lebedev via Phabricator via llvm-commits]

lebedev.ri added a comment.

In D111222#3062480 <https://reviews.llvm.org/D111222#3062480>, @RKSimon wrote:

> Do you think we should add gather/scatter test coverage for x86 triples with 32-bit pointers (i686 or gnux32 ?)

Does anyone actually care about 32-bit performance, especially vectorized?
We could extract gather/scatter coverage out of `llvm/test/Analysis/CostModel/X86/masked-intrinsic-cost.ll` into a new file and either duplicate it or just duplicate runlines i suppose.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D111222/new/

https://reviews.llvm.org/D111222