[llvm] [AArch64] Unrolling of loops with vector instructions. (PR #147420)

[Florian Hahn via llvm-commits]

================
@@ -0,0 +1,131 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 5
+; RUN: opt -p loop-unroll -mtriple=arm64-apple-macosx -mcpu=apple-m1 -S %s | FileCheck --check-prefix=APPLE %s
----------------
fhahn wrote:

The patch as-is impacts any AArch64 CPU, we also need tests without a `-mcpu`, including ones that don't require runtime unrolling.

https://github.com/llvm/llvm-project/pull/147420