[PATCH] D134413: [AARCH64][CostModel] Modified the cost of mask vector load/store

[LiqinWeng via Phabricator via llvm-commits]

Miss_Grape added a comment.

In D134413#3808618 <https://reviews.llvm.org/D134413#3808618>, @david-arm wrote:

> I think the change seems sensible because the cost of a NEON load is probably similar to the cost of a SVE masked load on CPUs with SVE. However, I am curious to learn about the problem you're hitting. Are you looking at a specific benchmark or loop where the cost is too high to vectorise with SVE? Or is the cost preventing additional interleaving in the vectoriser?

In the test case of sve-intrinsics-mask-ldst-ext.ll, I found that the instruction of llvm.masked.load.nxv/llvm.masked.store.nxv only corresponds to one assembly, so I think its correct cost should be 1


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D134413/new/

https://reviews.llvm.org/D134413