[LV][REVEC][AArch64] Proof of concept for re-vectorisation

[gbossu]

⚠️ DRAFT ⚠️

This shows the changes required to enable basic re-vectorisation support in LoopVectorizer. Most of the diff comes from the added tests, the changes to LoopVectorizer files are rather minimal. This proof-of-concept has obvious limitations and only represents the first building block.

My hope is that this helps discussions and complements the RFC at https://discourse.llvm.org/t/rfc-re-vectorisation-to-wider-vectors-in-loopvectorizer/91071.

Support for re-vectorisation is hidden behind a -vectorize-vector-loops flag and LV will bail out if it encounters constructs that are not yet supported. For example:

• shufflevectors
• gather/scatter and interleaved accesses
• target intrinsics
• reductions
• if-conversion or tail folding

Note: This is part of a stacked PR to create a proof-of-concept for re-vectorisation in LoopVectorizer.

• [IR][REVEC] Define llvm.vector.broadcast intrinsic #208212
• [LV][REVEC][AArch64] Proof of concept for re-vectorisation #208213 (this)

[github-actions]

⚠️ LLVM ABI annotation checker, ids-check found issues in your code. ⚠️

You can test this locally with the following command:

Build idt from compnerd/ids, then for each changed header:
    idt -p build/ --main-file <matching-source.cpp> \
        --apply-fixits --inplace <header>

View the diff from ids-check here.

diff --git a/llvm/include/llvm/Transforms/Vectorize/LoopVectorizationLegality.h b/llvm/include/llvm/Transforms/Vectorize/LoopVectorizationLegality.h
index 25ec10e71..cb1cf4784 100644
--- a/llvm/include/llvm/Transforms/Vectorize/LoopVectorizationLegality.h
+++ b/llvm/include/llvm/Transforms/Vectorize/LoopVectorizationLegality.h
@@ -48,7 +48,7 @@ class TargetLibraryInfo;
 class TargetTransformInfo;
 class Type;
 
-extern cl::opt<bool> VectorizeVectorLoops;
+extern LLVM_ABI cl::opt<bool> VectorizeVectorLoops;
 
 /// Utility class for getting and setting loop vectorizer hints in the form
 /// of loop metadata.