[PATCH] D76514: [ARM] Avoid pointless vrev of element-wise vmov

[Dave Green via Phabricator via llvm-commits]

dmgreen added a comment.

We added a VECTOR_REG_CAST, which it like a bitcast but doesn't change the bits. Similar to the AArch64 NVCAST.

Not that you have to do this here, but it also might allow more patterns to be converted, removing move vrev's. Essentially if the immediate after it has been bitcast/vrev'd is still a legal immediate, we can just generate the new immediate and VECTOR_REG_CAST it.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D76514/new/

https://reviews.llvm.org/D76514