[PATCH] D114779: [LV][X86] Sink `LoopVectorizationCostModel::useEmulatedMaskMemRefHack()` further into TTI, disable for X86/AVX2+

[Simon Pilgrim via Phabricator via llvm-commits]

RKSimon added a comment.

In D114779#3260815 <https://reviews.llvm.org/D114779#3260815>, @lebedev.ri wrote:

> As per @fhahn's "someone more familiar with X86 cost modeling should review the claim the cost model is fixed :)"

Sorry - I haven't been following this ticket much at all. Where did @fhahn says this and in what context? I can't see that here or D43208 <https://reviews.llvm.org/D43208>


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D114779/new/

https://reviews.llvm.org/D114779