[SLP]Fix PR102279: check the tracked values for extractelements, not the original values

If the reduced value was replaced by the extractelement instruction
during vectorization and we attempt to check if this is so, need to
check the tracked value, not the original (deleted) instruction.
Otherwise, the compiler may crash

Fixes #102279

Author: alexey-bataev

llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp

@@ -17336,27 +17336,23 @@ class HorizontalReduction {
       // Try to handle shuffled extractelements.
       if (S.getOpcode() == Instruction::ExtractElement && !S.isAltShuffle() &&
           I + 1 < E) {
-        InstructionsState NextS = getSameOpcode(ReducedVals[I + 1], TLI);
-        if (NextS.getOpcode() == Instruction::ExtractElement &&
-            !NextS.isAltShuffle()) {
-          SmallVector<Value *> CommonCandidates(Candidates);
-          for (Value *RV : ReducedVals[I + 1]) {
-            Value *RdxVal = TrackedVals.find(RV)->second;
-            // Check if the reduction value was not overriden by the
-            // extractelement instruction because of the vectorization and
-            // exclude it, if it is not compatible with other values.
-            if (auto *Inst = dyn_cast<Instruction>(RdxVal))
-              if (!NextS.getOpcode() || !NextS.isOpcodeOrAlt(Inst))
-                continue;
-            CommonCandidates.push_back(RdxVal);
-            TrackedToOrig.try_emplace(RdxVal, RV);
-          }
-          SmallVector<int> Mask;
-          if (isFixedVectorShuffle(CommonCandidates, Mask)) {
-            ++I;
-            Candidates.swap(CommonCandidates);
-            ShuffledExtracts = true;
-          }
+        SmallVector<Value *> CommonCandidates(Candidates);
+        for (Value *RV : ReducedVals[I + 1]) {
+          Value *RdxVal = TrackedVals.find(RV)->second;
+          // Check if the reduction value was not overriden by the
+          // extractelement instruction because of the vectorization and
+          // exclude it, if it is not compatible with other values.
+          auto *Inst = dyn_cast<ExtractElementInst>(RdxVal);
+          if (!Inst)
+            continue;
+          CommonCandidates.push_back(RdxVal);
+          TrackedToOrig.try_emplace(RdxVal, RV);
+        }
+        SmallVector<int> Mask;
+        if (isFixedVectorShuffle(CommonCandidates, Mask)) {
+          ++I;
+          Candidates.swap(CommonCandidates);
+          ShuffledExtracts = true;
         }
       }
 

llvm/test/Transforms/SLPVectorizer/X86/reduced-value-replace-extractelement.ll

@@ -0,0 +1,34 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 5
+; RUN: opt -S --passes=slp-vectorizer -slp-threshold=-99999 < %s -mtriple=x86_64-unknown-linux-gnu | FileCheck %s
+
+define void @test() {
+; CHECK-LABEL: define void @test() {
+; CHECK-NEXT:  [[BB:.*]]:
+; CHECK-NEXT:    br label %[[BB1:.*]]
+; CHECK:       [[BB1]]:
+; CHECK-NEXT:    [[TMP0:%.*]] = phi <2 x i32> [ zeroinitializer, %[[BB]] ], [ [[TMP4:%.*]], %[[BB1]] ]
+; CHECK-NEXT:    [[TMP1:%.*]] = extractelement <2 x i32> [[TMP0]], i32 1
+; CHECK-NEXT:    [[TMP2:%.*]] = call i8 @llvm.vector.reduce.mul.v4i8(<4 x i8> zeroinitializer)
+; CHECK-NEXT:    [[TMP3:%.*]] = zext i8 [[TMP2]] to i32
+; CHECK-NEXT:    [[OP_RDX:%.*]] = mul i32 [[TMP3]], [[TMP1]]
+; CHECK-NEXT:    [[OP_RDX1:%.*]] = mul i32 [[OP_RDX]], 0
+; CHECK-NEXT:    [[TMP4]] = insertelement <2 x i32> <i32 0, i32 poison>, i32 [[OP_RDX1]], i32 1
+; CHECK-NEXT:    br label %[[BB1]]
+;
+bb:
+  br label %bb1
+
+bb1:
+  %phi = phi i32 [ 0, %bb ], [ %mul9, %bb1 ]
+  %phi2 = phi i32 [ 0, %bb ], [ 0, %bb1 ]
+  %trunc = trunc i64 0 to i32
+  %mul = mul i32 0, %trunc
+  %mul3 = mul i32 %trunc, %phi
+  %mul4 = mul i32 %mul3, %mul
+  %mul5 = mul i32 %mul4, %mul
+  %trunc6 = trunc i64 0 to i32
+  %mul7 = mul i32 0, %trunc6
+  %mul8 = mul i32 %mul5, %mul7
+  %mul9 = mul i32 %mul8, %mul7
+  br label %bb1
+}

llvm/test/Transforms/SLPVectorizer/X86/reduction-gather-non-scheduled-extracts.ll

@@ -10,7 +10,8 @@ define void @tes() {
 ; CHECK-NEXT:    [[TMP3:%.*]] = shufflevector <2 x i1> zeroinitializer, <2 x i1> [[TMP0]], <4 x i32> <i32 0, i32 0, i32 0, i32 2>
 ; CHECK-NEXT:    [[TMP4:%.*]] = call i1 @llvm.vector.reduce.and.v4i1(<4 x i1> [[TMP3]])
 ; CHECK-NEXT:    [[OP_RDX:%.*]] = select i1 false, i1 [[TMP4]], i1 false
-; CHECK-NEXT:    br i1 [[OP_RDX]], label [[TMP6:%.*]], label [[TMP5:%.*]]
+; CHECK-NEXT:    [[OP_RDX1:%.*]] = select i1 false, i1 [[OP_RDX]], i1 false
+; CHECK-NEXT:    br i1 [[OP_RDX1]], label [[TMP6:%.*]], label [[TMP5:%.*]]
 ; CHECK:       4:
 ; CHECK-NEXT:    ret void
 ; CHECK:       5: