[PATCH] D119728: [AArch64][SVE] Handle more cases in findMoreOptimalIndexType.

[Caroline via Phabricator via llvm-commits]

CarolineConcatto added inline comments.


================
Comment at: llvm/test/CodeGen/AArch64/sve-gather-scatter-addr-opts.ll:302
+  %add1 = add <vscale x 4 x i64> %splat.offset, %step
+  %add2 = add <vscale x 4 x i64> %add1, %splat.offset2
+  %gep = getelementptr [8 x half], [8 x half]* %base, <vscale x 4 x i64> %add2
----------------
If I switch this the fold does not happen.  I know this was not supported before, but should we try to support now?


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D119728/new/

https://reviews.llvm.org/D119728