[PATCH] D153300: [AArch64][GlobalISel] IR translate support for a return instruction of type <1 x i8> or <1 x i16> when using GlobalISel.

[Matt Arsenault via Phabricator via llvm-commits]

arsenm added inline comments.


================
Comment at: llvm/lib/Target/AArch64/GISel/AArch64CallLowering.cpp:406-422
           if (NewVT.isVector()) {
             if (OldLLT.isVector()) {
               if (NewLLT.getNumElements() > OldLLT.getNumElements()) {
                 // We don't handle VA types which are not exactly twice the
                 // size, but can easily be done in future.
                 if (NewLLT.getNumElements() != OldLLT.getNumElements() * 2) {
                   LLVM_DEBUG(dbgs() << "Outgoing vector ret has too many elts");
----------------
Actually I think we have the same ugly code in the generic call lowering. I think at one point I was trying to delete this, maybe you'll have better luck now


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D153300/new/

https://reviews.llvm.org/D153300