[llvm] ab9078f - [SLP]Fix PR52756: SLPVectorizer crashes with assertion VecTy == FinalVecTy.

[Alexey Bataev via llvm-commits]

Author: Alexey Bataev
Date: 2021-12-20T07:21:20-08:00
New Revision: ab9078f3d369656c09b5e34bcda6a5d62699a3fb

URL: https://github.com/llvm/llvm-project/commit/ab9078f3d369656c09b5e34bcda6a5d62699a3fb
DIFF: https://github.com/llvm/llvm-project/commit/ab9078f3d369656c09b5e34bcda6a5d62699a3fb.diff

LOG: [SLP]Fix PR52756: SLPVectorizer crashes with assertion VecTy == FinalVecTy.

Need to check for the number of the unique non-constant values since the
unique values may include several constants.

Differential Revision: https://reviews.llvm.org/D115939

Added: 
    llvm/test/Transforms/SLPVectorizer/X86/reused-undefs.ll

Modified: 
    llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp

Removed: 
    


################################################################################
diff  --git a/llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp b/llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp
index cbcdcc3cf325b..37ae13666f7a2 100644
--- a/llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp
+++ b/llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp
@@ -3568,8 +3568,11 @@ void BoUpSLP::buildTree_rec(ArrayRef<Value *> VL, unsigned Depth,
     } else {
       LLVM_DEBUG(dbgs() << "SLP: Shuffle for reused scalars.\n");
       if (NumUniqueScalarValues <= 1 ||
-          (NumUniqueScalarValues == 2 &&
-           any_of(UniqueValues, UndefValue::classof)) ||
+          (UniquePositions.size() == 1 && all_of(UniqueValues,
+                                                 [](Value *V) {
+                                                   return isa<UndefValue>(V) ||
+                                                          !isConstant(V);
+                                                 })) ||
           !llvm::isPowerOf2_32(NumUniqueScalarValues)) {
         LLVM_DEBUG(dbgs() << "SLP: Scalar used twice in bundle.\n");
         newTreeEntry(VL, None /*not vectorized*/, S, UserTreeIdx);
@@ -9812,10 +9815,15 @@ bool SLPVectorizerPass::vectorizeChainsInBlock(BasicBlock *BB, BoUpSLP &R) {
       return true;
     if (Opcodes1.size() > Opcodes2.size())
       return false;
+    Optional<bool> ConstOrder;
     for (int I = 0, E = Opcodes1.size(); I < E; ++I) {
       // Undefs are compatible with any other value.
-      if (isa<UndefValue>(Opcodes1[I]) || isa<UndefValue>(Opcodes2[I]))
+      if (isa<UndefValue>(Opcodes1[I]) || isa<UndefValue>(Opcodes2[I])) {
+        if (!ConstOrder)
+          ConstOrder =
+              !isa<UndefValue>(Opcodes1[I]) && isa<UndefValue>(Opcodes2[I]);
         continue;
+      }
       if (auto *I1 = dyn_cast<Instruction>(Opcodes1[I]))
         if (auto *I2 = dyn_cast<Instruction>(Opcodes2[I])) {
           DomTreeNodeBase<BasicBlock> *NodeI1 = DT->getNode(I1->getParent());
@@ -9834,14 +9842,17 @@ bool SLPVectorizerPass::vectorizeChainsInBlock(BasicBlock *BB, BoUpSLP &R) {
             continue;
           return I1->getOpcode() < I2->getOpcode();
         }
-      if (isa<Constant>(Opcodes1[I]) && isa<Constant>(Opcodes2[I]))
+      if (isa<Constant>(Opcodes1[I]) && isa<Constant>(Opcodes2[I])) {
+        if (!ConstOrder)
+          ConstOrder = Opcodes1[I]->getValueID() < Opcodes2[I]->getValueID();
         continue;
+      }
       if (Opcodes1[I]->getValueID() < Opcodes2[I]->getValueID())
         return true;
       if (Opcodes1[I]->getValueID() > Opcodes2[I]->getValueID())
         return false;
     }
-    return false;
+    return ConstOrder && *ConstOrder;
   };
   auto AreCompatiblePHIs = [&PHIToOpcodes](Value *V1, Value *V2) {
     if (V1 == V2)

diff  --git a/llvm/test/Transforms/SLPVectorizer/X86/reused-undefs.ll b/llvm/test/Transforms/SLPVectorizer/X86/reused-undefs.ll
new file mode 100644
index 0000000000000..c587adfa9d11f
--- /dev/null
+++ b/llvm/test/Transforms/SLPVectorizer/X86/reused-undefs.ll
@@ -0,0 +1,40 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
+; RUN: opt -S -slp-vectorizer -mtriple=x86_64-unknown-linux-gnu -slp-threshold=-1000 < %s | FileCheck %s
+
+define i32 @main(i32 %0) {
+; CHECK-LABEL: @main(
+; CHECK-NEXT:  for.cond.preheader:
+; CHECK-NEXT:    br i1 false, label [[FOR_END:%.*]], label [[FOR_INC_PREHEADER:%.*]]
+; CHECK:       for.inc.preheader:
+; CHECK-NEXT:    [[TMP1:%.*]] = insertelement <8 x i32> <i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 poison, i32 poison>, i32 [[TMP0:%.*]], i32 6
+; CHECK-NEXT:    br i1 false, label [[FOR_END]], label [[L1_PREHEADER:%.*]]
+; CHECK:       for.end:
+; CHECK-NEXT:    [[DOTPR:%.*]] = phi i32 [ 0, [[FOR_INC_PREHEADER]] ], [ 0, [[FOR_COND_PREHEADER:%.*]] ]
+; CHECK-NEXT:    [[TMP2:%.*]] = insertelement <8 x i32> poison, i32 [[DOTPR]], i32 0
+; CHECK-NEXT:    [[SHUFFLE:%.*]] = shufflevector <8 x i32> [[TMP2]], <8 x i32> poison, <8 x i32> <i32 undef, i32 undef, i32 undef, i32 0, i32 0, i32 0, i32 0, i32 0>
+; CHECK-NEXT:    br label [[L1_PREHEADER]]
+; CHECK:       L1.preheader:
+; CHECK-NEXT:    [[TMP3:%.*]] = phi <8 x i32> [ [[SHUFFLE]], [[FOR_END]] ], [ [[TMP1]], [[FOR_INC_PREHEADER]] ]
+; CHECK-NEXT:    ret i32 0
+;
+for.cond.preheader:
+  br i1 false, label %for.end, label %for.inc.preheader
+
+for.inc.preheader:
+  br i1 false, label %for.end, label %L1.preheader
+
+for.end:
+  %.pr = phi i32 [ 0, %for.inc.preheader ], [ 0, %for.cond.preheader ]
+  br label %L1.preheader
+
+L1.preheader:
+  %1 = phi i32 [ %.pr, %for.end ], [ %0, %for.inc.preheader ]
+  %2 = phi i32 [ %.pr, %for.end ], [ 0, %for.inc.preheader ]
+  %3 = phi i32 [ %.pr, %for.end ], [ 0, %for.inc.preheader ]
+  %4 = phi i32 [ %.pr, %for.end ], [ undef, %for.inc.preheader ]
+  %j.2.ph1 = phi i32 [ %.pr, %for.end ], [ 0, %for.inc.preheader ]
+  %k.0.ph = phi i32 [ undef, %for.end ], [ 0, %for.inc.preheader ]
+  %o.1.ph = phi i32 [ undef, %for.end ], [ 0, %for.inc.preheader ]
+  %n.1.ph = phi i32 [ undef, %for.end ], [ 0, %for.inc.preheader ]
+  ret i32 0
+}