[llvm] [AArch64] Allow unrolling of scalar epilogue loops (PR #151164)

[Ramkumar Ramachandra via llvm-commits]

================
@@ -485,12 +485,206 @@ exit:                                 ; preds = %vector.body
 !0 = !{!0, !1}
 !1 = !{!"llvm.loop.isvectorized", i32 1}
 
+; On Cortex-A55 we should runtime unroll the scalar epilogue loop, but not the
+; vector loop.
+define void @scalar_epilogue(i64 %N, ptr %p, i8 %val) {
----------------
artagnon wrote:

```suggestion
define void @scalar_epilogue(ptr %p, i8 %splat.scalar, i8 %n) {
```

https://github.com/llvm/llvm-project/pull/151164