[SLP]Fix PR58956: fix insertpoint for reduced buildvector graphs.

If the graph is only the buildvector node without main operation, need
to inherit insrtpoint from the redution instruction. Otherwise the
compiler crashes trying to insert instruction at the entry block.

Author: alexey-bataev

llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp

@@ -970,7 +970,8 @@ class BoUpSLP {
   /// Vectorize the tree but with the list of externally used values \p
   /// ExternallyUsedValues. Values in this MapVector can be replaced but the
   /// generated extractvalue instructions.
-  Value *vectorizeTree(ExtraValueToDebugLocsMap &ExternallyUsedValues);
+  Value *vectorizeTree(ExtraValueToDebugLocsMap &ExternallyUsedValues,
+                       Instruction *ReductionRoot = nullptr);
 
   /// \returns the cost incurred by unwanted spills and fills, caused by
   /// holding live values over call sites.
@@ -9002,8 +9003,8 @@ struct ShuffledInsertData {
 };
 } // namespace
 
-Value *
-BoUpSLP::vectorizeTree(ExtraValueToDebugLocsMap &ExternallyUsedValues) {
+Value *BoUpSLP::vectorizeTree(ExtraValueToDebugLocsMap &ExternallyUsedValues,
+                              Instruction *ReductionRoot) {
   // All blocks must be scheduled before any instructions are inserted.
   for (auto &BSIter : BlocksSchedules) {
     scheduleBlock(BSIter.second.get());
@@ -9020,7 +9021,8 @@ BoUpSLP::vectorizeTree(ExtraValueToDebugLocsMap &ExternallyUsedValues) {
     EntryToLastInstruction.try_emplace(E.get(), LastInst);
   }
 
-  Builder.SetInsertPoint(&F->getEntryBlock().front());
+  Builder.SetInsertPoint(ReductionRoot ? ReductionRoot
+                                       : &F->getEntryBlock().front());
   auto *VectorRoot = vectorizeTree(VectorizableTree[0].get());
 
   // If the vectorized tree can be rewritten in a smaller type, we truncate the
@@ -11944,16 +11946,18 @@ class HorizontalReduction {
 
         Builder.setFastMathFlags(RdxFMF);
 
-        // Vectorize a tree.
-        Value *VectorizedRoot = V.vectorizeTree(LocalExternallyUsedValues);
-
         // Emit a reduction. If the root is a select (min/max idiom), the insert
         // point is the compare condition of that select.
         Instruction *RdxRootInst = cast<Instruction>(ReductionRoot);
+        Instruction *InsertPt = RdxRootInst;
         if (IsCmpSelMinMax)
-          Builder.SetInsertPoint(GetCmpForMinMaxReduction(RdxRootInst));
-        else
-          Builder.SetInsertPoint(RdxRootInst);
+          InsertPt = GetCmpForMinMaxReduction(RdxRootInst);
+
+        // Vectorize a tree.
+        Value *VectorizedRoot =
+            V.vectorizeTree(LocalExternallyUsedValues, InsertPt);
+
+        Builder.SetInsertPoint(InsertPt);
 
         // To prevent poison from leaking across what used to be sequential,
         // safe, scalar boolean logic operations, the reduction operand must be

llvm/test/Transforms/SLPVectorizer/AArch64/buildvector-reduce.ll

@@ -0,0 +1,32 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
+; RUN: opt -S -passes=slp-vectorizer < %s -mtriple=arm64-apple-macosx | FileCheck %s
+
+define i8 @test() {
+; CHECK-LABEL: @test(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
+; CHECK:       for.body:
+; CHECK-NEXT:    [[SUM:%.*]] = phi i32 [ [[TMP1:%.*]], [[FOR_BODY]] ], [ 0, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    [[CALL278:%.*]] = call i32 @fn(i32 [[SUM]])
+; CHECK-NEXT:    [[TMP0:%.*]] = insertelement <8 x i32> poison, i32 [[CALL278]], i32 0
+; CHECK-NEXT:    [[SHUFFLE:%.*]] = shufflevector <8 x i32> [[TMP0]], <8 x i32> poison, <8 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP1]] = call i32 @llvm.vector.reduce.add.v8i32(<8 x i32> [[SHUFFLE]])
+; CHECK-NEXT:    br label [[FOR_BODY]]
+;
+entry:
+  br label %for.body
+
+for.body:
+  %sum = phi i32 [ %add285.19, %for.body ], [ 0, %entry ]
+  %call278 = call i32 @fn(i32 %sum)
+  %add285.13 = add i32 %call278, %call278
+  %add285.14 = add i32 %add285.13, %call278
+  %add285.15 = add i32 %add285.14, %call278
+  %add285.16 = add i32 %add285.15, %call278
+  %add285.17 = add i32 %add285.16, %call278
+  %add285.18 = add i32 %add285.17, %call278
+  %add285.19 = add i32 %add285.18, %call278
+  br label %for.body
+}
+
+declare i32 @fn(i32)