[llvm] [VectorCombine] Try to scalarize vector loads feeding bitcast instructions. (PR #164682)

[Florian Hahn via llvm-commits]

================
@@ -0,0 +1,32 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 6
+; RUN: opt -passes=vector-combine,dce,vector-combine -mtriple=arm64-apple-darwinos -S %s | FileCheck %s
----------------
fhahn wrote:

do we need to run `vector-combine` twice to get the transform? If so, it would probably better to add this as PhaseOrdering test, making sure we get the expected fold with -O3

https://github.com/llvm/llvm-project/pull/164682