[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:424-425
               }
-            } else if (NewLLT.getNumElements() == 2) {
-              // We need to pad a <1 x S> type to <2 x S>. Since we don't have
-              // <1 x S> vector types in GISel we use a build_vector instead
-              // of a vector merge/concat.
-              auto Undef = MIRBuilder.buildUndef({OldLLT});
+            } else if (NewLLT.getNumElements() >= 2 &&
+                       NewLLT.getNumElements() <= 8) {
+              // We need to pad a <1 x S> type to <2/4/8 x S>. Since we don't
----------------
This whole function is bad and I'd prefer to stop making the situation worse. Can you try switching to the helpers in MachineIRBuilder? e.g. buildPadVectorWithUndefElements


Repository:
  rG LLVM Github Monorepo

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

https://reviews.llvm.org/D153300