[VPlan] Update type of cloned instruction in scalarizeInstruction.

The operands of the replicate recipe may have been narrowed, resulting
in a narrower result type. Update the type of the cloned instruction to
the correct type.

Fixes llvm#151392.

Author: fhahn

llvm/lib/Transforms/Vectorize/VPlanRecipes.cpp

@@ -2852,12 +2852,12 @@ static void scalarizeInstruction(const Instruction *Instr,
   Instruction *Cloned = Instr->clone();
   if (!IsVoidRetTy) {
     Cloned->setName(Instr->getName() + ".cloned");
-#if !defined(NDEBUG)
-    // Verify that VPlan type inference results agree with the type of the
-    // generated values.
-    assert(State.TypeAnalysis.inferScalarType(RepRecipe) == Cloned->getType() &&
-           "inferred type and type from generated instructions do not match");
-#endif
+    Type *ResultTy = State.TypeAnalysis.inferScalarType(RepRecipe);
+    // The operands of the replicate recipe may have been narrowed, resulting in
+    // a narrower result type. Update the type of the cloned instruction to the
+    // correct type.
+    if (ResultTy != Cloned->getType())
+      Cloned->mutateType(ResultTy);
   }
 
   RepRecipe->applyFlags(*Cloned);

llvm/test/Transforms/LoopVectorize/single-scalar-cast-minbw.ll

@@ -81,8 +81,12 @@ define void @single_scalar_cast_stored(ptr %src, ptr %dst, i32 %n) {
 ; CHECK-NEXT:    [[N_MOD_VF:%.*]] = urem i32 [[N]], 4
 ; CHECK-NEXT:    [[N_VEC:%.*]] = sub i32 [[N]], [[N_MOD_VF]]
 ; CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr [[SRC]], align 2, !alias.scope [[META4:![0-9]+]]
-; CHECK-NEXT:    [[TMP3:%.*]] = icmp eq i16 [[TMP0]], 0
-; CHECK-NEXT:    [[TMP4:%.*]] = and i16 [[TMP0]], 15
+; CHECK-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <4 x i16> poison, i16 [[TMP0]], i64 0
+; CHECK-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <4 x i16> [[BROADCAST_SPLATINSERT]], <4 x i16> poison, <4 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP1:%.*]] = icmp eq <4 x i16> [[BROADCAST_SPLAT]], zeroinitializer
+; CHECK-NEXT:    [[TMP2:%.*]] = and <4 x i16> [[BROADCAST_SPLAT]], splat (i16 15)
+; CHECK-NEXT:    [[TMP3:%.*]] = extractelement <4 x i1> [[TMP1]], i32 0
+; CHECK-NEXT:    [[TMP4:%.*]] = extractelement <4 x i16> [[TMP2]], i32 0
 ; CHECK-NEXT:    [[TMP5:%.*]] = select i1 [[TMP3]], i16 0, i16 [[TMP4]]
 ; CHECK-NEXT:    br label %[[VECTOR_BODY:.*]]
 ; CHECK:       [[VECTOR_BODY]]: