[PATCH] D133850: [AArch64] Improve codegen for "trunc <4 x i64> to <4 x i8>" for all cases

[Florian Hahn via Phabricator via llvm-commits]

fhahn added a comment.

In this context, D133495 <https://reviews.llvm.org/D133495> may also be interesting. It's using `tbl` to lower i32->i8 truncation and could also be extended to handle i64->i8. This would allow to do the conversion with one instructions plus a load that materializes the mask, which is why D133495 <https://reviews.llvm.org/D133495> limits this to cases in loops, where the load can be hoisted out.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D133850/new/

https://reviews.llvm.org/D133850