[llvm] [SLPVectorizer] Widen constant strided loads. (PR #162324)

[via llvm-commits]

github-actions[bot] wrote:

<!--LLVM CODE FORMAT COMMENT: {clang-format}-->


:warning: C/C++ code formatter, clang-format found issues in your code. :warning:

<details>
<summary>
You can test this locally with the following command:
</summary>

``````````bash
git-clang-format --diff origin/main HEAD --extensions cpp -- llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp
``````````

:warning:
The reproduction instructions above might return results for more than one PR
in a stack if you are using a stacked PR workflow. You can limit the results by
changing `origin/main` to the base branch/commit you want to compare against.
:warning:

</details>

<details>
<summary>
View the diff from clang-format here.
</summary>

``````````diff
diff --git a/llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp b/llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp
index 385cc54f1..28676dc46 100644
--- a/llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp
+++ b/llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp
@@ -2247,10 +2247,9 @@ public:
   /// Return true if an array of scalar loads can be replaced with a strided
   ///  load (with constant stride).
   ///
-  ///  It is possible that the load gets "widened". Suppose that originally each load loads `k` bytes and `PointerOps` can be arranged as follows (`%s` is constant):
-  ///  %b + 0 * %s + 0
-  ///  %b + 0 * %s + 1
-  ///  %b + 0 * %s + 2
+  ///  It is possible that the load gets "widened". Suppose that originally each
+  ///  load loads `k` bytes and `PointerOps` can be arranged as follows (`%s` is
+  ///  constant): %b + 0 * %s + 0 %b + 0 * %s + 1 %b + 0 * %s + 2
   ///  ...
   ///  %b + 0 * %s + (w - 1)
   ///
@@ -2272,17 +2271,18 @@ public:
   /// \param PointerOps list of pointer arguments of loads.
   /// \param ElemTy original scalar type of loads.
   /// \param Alignment alignment of the first load.
-  /// \param SortedIndices is the order of PointerOps as returned by `sortPtrAccesses`
-  /// \param Diff Pointer difference between the lowest and the highes pointer in `PointerOps` as returned by `getPointersDiff`.
+  /// \param SortedIndices is the order of PointerOps as returned by
+  /// `sortPtrAccesses`
+  /// \param Diff Pointer difference between the lowest and the highes pointer
+  /// in `PointerOps` as returned by `getPointersDiff`.
   /// \param Ptr0 first pointer in `PointersOps`.
   /// \param PtrN last pointer in `PointersOps`.
   /// \param SPtrInfo If the function return `true`, it also sets all the fields
   /// of `SPtrInfo` necessary to generate the strided load later.
-  bool analyzeConstantStrideCandidate(const ArrayRef<Value *> PointerOps,
-                                      Type *ElemTy, Align Alignment,
-                                      const SmallVectorImpl<unsigned> &SortedIndices,
-                                      const int64_t Diff, Value *Ptr0, Value *PtrN,
-                                      StridedPtrInfo &SPtrInfo) const;
+  bool analyzeConstantStrideCandidate(
+      const ArrayRef<Value *> PointerOps, Type *ElemTy, Align Alignment,
+      const SmallVectorImpl<unsigned> &SortedIndices, const int64_t Diff,
+      Value *Ptr0, Value *PtrN, StridedPtrInfo &SPtrInfo) const;
 
   /// Return true if an array of scalar loads can be replaced with a strided
   /// load (with run-time stride).
@@ -6910,8 +6910,8 @@ bool BoUpSLP::isStridedLoad(ArrayRef<Value *> PointerOps, Type *ScalarTy,
 
 bool BoUpSLP::analyzeConstantStrideCandidate(
     const ArrayRef<Value *> PointerOps, Type *ElemTy, Align CommonAlignment,
-    const SmallVectorImpl<unsigned> &SortedIndices, const int64_t Diff, Value *Ptr0,
-    Value *PtrN, StridedPtrInfo &SPtrInfo) const {
+    const SmallVectorImpl<unsigned> &SortedIndices, const int64_t Diff,
+    Value *Ptr0, Value *PtrN, StridedPtrInfo &SPtrInfo) const {
   const unsigned Sz = PointerOps.size();
   SmallVector<int64_t> SortedOffsetsFromBase(Sz);
   // Go through `PointerOps` in sorted order and record offsets from `Ptr0`.

``````````

</details>


https://github.com/llvm/llvm-project/pull/162324