[LV] Fold Add/Mul with neutral elements.

[fhahn]

Summary

Building on D152659, fold add/mul with their neutral elements.

There is a large number of tests that need updating, which I will do
if/once there's agreement on the direction. The missing fold was pointed
out in D147783

Depends on D152659.

[fhahn]

Interestingly, this is also causing some binary changes, so is presumably catching same missed simplifications or enabling additional/earlier simplifications: https://llvm-compile-time-tracker.com/compare.php?from=e3753cc0f9c8404f21dd7486febfb4766f4c9659&to=d34adaeac19f4555a08b38e65b393999c421237c&stat=size-text

[Ayal]

llvm/lib/Transforms/Vectorize/LoopVectorize.cpp

2517

Note the attempt here to apply add-zero and multiply-by-one foldings.

llvm/test/Transforms/LoopVectorize/RISCV/strided-accesses.ll

21

Redundant code in preheader, optimize the recipe that (will/not) generate it?

228

More mul-by-one opportunities to fold?

715

More mul-by-zero opportunities to fold?

llvm/lib/Transforms/Vectorize/LoopVectorize.cpp

 class GeneratedRTChecks;
 
 using SCEV2ValueTy = DenseMap<const SCEV *, Value *>;
 
 /// Folder to simplify common patterns generated by LV codegen.
 class LVFolder final : public ConstantFolder {
   void anchor() override {}
 
+  Value *SimplifyBinOp(Instruction::BinaryOps Opc, Value *LHS,
+                       Value *RHS) const {
+    using namespace llvm::PatternMatch;
+    switch (Opc) {
+    default:
+      return nullptr;
+    case Instruction::Add:
+      if (match(RHS, m_SpecificInt(0)))
+        return LHS;
+      if (match(LHS, m_SpecificInt(0)))
+        return RHS;
+      return nullptr;
+    case Instruction::Mul:
+      if (match(RHS, m_SpecificInt(1)))
+        return LHS;
+      if (match(LHS, m_SpecificInt(1)))
+        return RHS;
+      return nullptr;
+    }
+  }
+
 public:
   LVFolder() {}
 
+  Value *FoldNoWrapBinOp(Instruction::BinaryOps Opc, Value *LHS, Value *RHS,
+                         bool HasNUW, bool HasNSW) const override {
+    if (auto *V = SimplifyBinOp(Opc, LHS, RHS))
+      return V;
+    return ConstantFolder::FoldNoWrapBinOp(Opc, LHS, RHS, HasNUW, HasNSW);
+  }
+
+  Value *FoldBinOp(Instruction::BinaryOps Opc, Value *LHS,
+                   Value *RHS) const override {
+    if (auto *V = SimplifyBinOp(Opc, LHS, RHS))
+      return V;
+    return ConstantFolder::FoldBinOp(Opc, LHS, RHS);
+  }
+
   Value *FoldGEP(Type *Ty, Value *Ptr, ArrayRef<Value *> IdxList,
                  bool IsInBounds = false) const override {
     auto *CI = dyn_cast<ConstantInt>(IdxList[0]);
     if (IdxList.size() == 1 && CI && CI->isNullValue())
       return Ptr;
     return ConstantFolder::FoldGEP(Ty, Ptr, IdxList, IsInBounds);
   }
 };
auto CreateAdd = [&B](Value *X, Value *Y) {
     if (auto *CX = dyn_cast<ConstantInt>(X))
       if (CX->isZero())
         return Y;
     if (auto *CY = dyn_cast<ConstantInt>(Y))
       if (CY->isZero())
         return X;
     return B.CreateAdd(X, Y);
   };
 
   // We allow X to be a vector type, in which case Y will potentially be
   // splatted into a vector with the same element count.
   auto CreateMul = [&B](Value *X, Value *Y) {
     assert(X->getType()->getScalarType() == Y->getType() &&
            "Types don't match!");
     if (auto *CX = dyn_cast<ConstantInt>(X))
       if (CX->isOne())
         return Y;

llvm/test/Transforms/LoopVectorize/RISCV/strided-accesses.ll

 ; CHECK:       vector.ph:
 ; CHECK-NEXT:    [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
 ; CHECK-NEXT:    [[TMP3:%.*]] = mul i64 [[TMP2]], 4
 ; CHECK-NEXT:    [[N_MOD_VF:%.*]] = urem i64 1024, [[TMP3]]
 ; CHECK-NEXT:    [[TMP4:%.*]] = icmp eq i64 [[N_MOD_VF]], 0
 ; CHECK-NEXT:    [[TMP5:%.*]] = select i1 [[TMP4]], i64 [[TMP3]], i64 [[N_MOD_VF]]
 ; CHECK-NEXT:    [[N_VEC:%.*]] = sub i64 1024, [[TMP5]]
 ; CHECK-NEXT:    [[TMP6:%.*]] = call <vscale x 4 x i64> @llvm.experimental.stepvector.nxv4i64()
-; CHECK-NEXT:    [[TMP7:%.*]] = add <vscale x 4 x i64> [[TMP6]], zeroinitializer
-; CHECK-NEXT:    [[TMP8:%.*]] = mul <vscale x 4 x i64> [[TMP7]], shufflevector (<vscale x 4 x i64> insertelement (<vscale x 4 x i64> poison, i64 1, i64 0), <vscale x 4 x i64> poison, <vscale x 4 x i32> zeroinitializer)
-; CHECK-NEXT:    [[INDUCTION:%.*]] = add <vscale x 4 x i64> zeroinitializer, [[TMP8]]
-; CHECK-NEXT:    [[TMP9:%.*]] = call i64 @llvm.vscale.i64()
-; CHECK-NEXT:    [[TMP10:%.*]] = mul i64 [[TMP9]], 4
-; CHECK-NEXT:    [[TMP11:%.*]] = mul i64 1, [[TMP10]]
-; CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 4 x i64> poison, i64 [[TMP11]], i64 0
+; CHECK-NEXT:    [[TMP7:%.*]] = call i64 @llvm.vscale.i64()
+; CHECK-NEXT:    [[TMP8:%.*]] = mul i64 [[TMP7]], 4
+; CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 4 x i64> poison, i64 [[TMP8]], i64 0
 ; CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 4 x i64> [[DOTSPLATINSERT]], <vscale x 4 x i64> poison, <vscale x 4 x i32> zeroinitializer
 ; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
 ; CHECK:       vector.body:
 ; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
-; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <vscale x 4 x i64> [ [[INDUCTION]], [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
-; CHECK-NEXT:    [[TMP12:%.*]] = mul nuw nsw <vscale x 4 x i64> [[VEC_IND]], shufflevector (<vscale x 4 x i64> insertelement (<vscale x 4 x i64> poison, i64 8, i64 0), <vscale x 4 x i64> poison, <vscale x 4 x i32> zeroinitializer)
-; CHECK-NEXT:    [[TMP13:%.*]] = getelementptr i32, ptr [[P:%.*]], <vscale x 4 x i64> [[TMP12]]
-; CHECK-NEXT:    [[WIDE_MASKED_GATHER:%.*]] = call <vscale x 4 x i32> @llvm.masked.gather.nxv4i32.nxv4p0(<vscale x 4 x ptr> [[TMP13]], i32 4, <vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer), <vscale x 4 x i32> poison)
-; CHECK-NEXT:    [[TMP14:%.*]] = add <vscale x 4 x i32> [[WIDE_MASKED_GATHER]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 1, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
-; CHECK-NEXT:    call void @llvm.masked.scatter.nxv4i32.nxv4p0(<vscale x 4 x i32> [[TMP14]], <vscale x 4 x ptr> [[TMP13]], i32 4, <vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer))
-; CHECK-NEXT:    [[TMP15:%.*]] = call i64 @llvm.vscale.i64()
-; CHECK-NEXT:    [[TMP16:%.*]] = mul i64 [[TMP15]], 4
-; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP16]]
+; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <vscale x 4 x i64> [ [[TMP6]], [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[TMP9:%.*]] = mul nuw nsw <vscale x 4 x i64> [[VEC_IND]], shufflevector (<vscale x 4 x i64> insertelement (<vscale x 4 x i64> poison, i64 8, i64 0), <vscale x 4 x i64> poison, <vscale x 4 x i32> zeroinitializer)
+; CHECK-NEXT:    [[TMP10:%.*]] = getelementptr i32, ptr [[P:%.*]], <vscale x 4 x i64> [[TMP9]]
+; CHECK-NEXT:    [[WIDE_MASKED_GATHER:%.*]] = call <vscale x 4 x i32> @llvm.masked.gather.nxv4i32.nxv4p0(<vscale x 4 x ptr> [[TMP10]], i32 4, <vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer), <vscale x 4 x i32> poison)
+; CHECK-NEXT:    [[TMP11:%.*]] = add <vscale x 4 x i32> [[WIDE_MASKED_GATHER]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 1, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
+; CHECK-NEXT:    call void @llvm.masked.scatter.nxv4i32.nxv4p0(<vscale x 4 x i32> [[TMP11]], <vscale x 4 x ptr> [[TMP10]], i32 4, <vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer))
+; CHECK-NEXT:    [[TMP12:%.*]] = call i64 @llvm.vscale.i64()
+; CHECK-NEXT:    [[TMP13:%.*]] = mul i64 [[TMP12]], 4
+; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP13]]
 ; CHECK-NEXT:    [[VEC_IND_NEXT]] = add <vscale x 4 x i64> [[VEC_IND]], [[DOTSPLAT]]
-; CHECK-NEXT:    [[TMP17:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
-; CHECK-NEXT:    br i1 [[TMP17]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
+; CHECK-NEXT:    [[TMP14:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; CHECK-NEXT:    br i1 [[TMP14]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
 ; CHECK:       middle.block:
 ; CHECK-NEXT:    br label [[SCALAR_PH]]
 ; CHECK:       scalar.ph:
 ; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
 ; CHECK-NEXT:    br label [[LOOP:%.*]]
 ; CHECK:       loop:
 ; CHECK-NEXT:    [[I:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[NEXTI:%.*]], [[LOOP]] ]
 ; CHECK-NEXT:    [[OFFSET:%.*]] = mul nuw nsw i64 [[I]], 8

 ; CHECK-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
 ; CHECK:       vector.ph:
 ; CHECK-NEXT:    [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
 ; CHECK-NEXT:    [[TMP3:%.*]] = mul i64 [[TMP2]], 4
 ; CHECK-NEXT:    [[N_MOD_VF:%.*]] = urem i64 1024, [[TMP3]]
 ; CHECK-NEXT:    [[N_VEC:%.*]] = sub i64 1024, [[N_MOD_VF]]
 ; CHECK-NEXT:    [[IND_END:%.*]] = mul i64 [[N_VEC]], 64
 ; CHECK-NEXT:    [[TMP4:%.*]] = call <vscale x 4 x i64> @llvm.experimental.stepvector.nxv4i64()
-; CHECK-NEXT:    [[TMP5:%.*]] = add <vscale x 4 x i64> [[TMP4]], zeroinitializer
-; CHECK-NEXT:    [[TMP6:%.*]] = mul <vscale x 4 x i64> [[TMP5]], shufflevector (<vscale x 4 x i64> insertelement (<vscale x 4 x i64> poison, i64 64, i64 0), <vscale x 4 x i64> poison, <vscale x 4 x i32> zeroinitializer)
-; CHECK-NEXT:    [[INDUCTION:%.*]] = add <vscale x 4 x i64> zeroinitializer, [[TMP6]]
-; CHECK-NEXT:    [[TMP7:%.*]] = call i64 @llvm.vscale.i64()
-; CHECK-NEXT:    [[TMP8:%.*]] = mul i64 [[TMP7]], 4
-; CHECK-NEXT:    [[TMP9:%.*]] = mul i64 64, [[TMP8]]
-; CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 4 x i64> poison, i64 [[TMP9]], i64 0
+; CHECK-NEXT:    [[TMP5:%.*]] = mul <vscale x 4 x i64> [[TMP4]], shufflevector (<vscale x 4 x i64> insertelement (<vscale x 4 x i64> poison, i64 64, i64 0), <vscale x 4 x i64> poison, <vscale x 4 x i32> zeroinitializer)
+; CHECK-NEXT:    [[TMP6:%.*]] = call i64 @llvm.vscale.i64()
+; CHECK-NEXT:    [[TMP7:%.*]] = mul i64 [[TMP6]], 4
+; CHECK-NEXT:    [[TMP8:%.*]] = mul i64 64, [[TMP7]]
+; CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 4 x i64> poison, i64 [[TMP8]], i64 0
 ; CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 4 x i64> [[DOTSPLATINSERT]], <vscale x 4 x i64> poison, <vscale x 4 x i32> zeroinitializer
 ; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
 ; CHECK:       vector.body:
 ; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
-; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <vscale x 4 x i64> [ [[INDUCTION]], [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
-; CHECK-NEXT:    [[TMP10:%.*]] = getelementptr i32, ptr [[P:%.*]], <vscale x 4 x i64> [[VEC_IND]]
-; CHECK-NEXT:    [[WIDE_MASKED_GATHER:%.*]] = call <vscale x 4 x i32> @llvm.masked.gather.nxv4i32.nxv4p0(<vscale x 4 x ptr> [[TMP10]], i32 4, <vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer), <vscale x 4 x i32> poison)
-; CHECK-NEXT:    [[TMP11:%.*]] = add <vscale x 4 x i32> [[WIDE_MASKED_GATHER]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 1, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
-; CHECK-NEXT:    call void @llvm.masked.scatter.nxv4i32.nxv4p0(<vscale x 4 x i32> [[TMP11]], <vscale x 4 x ptr> [[TMP10]], i32 4, <vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer))
-; CHECK-NEXT:    [[TMP12:%.*]] = call i64 @llvm.vscale.i64()
-; CHECK-NEXT:    [[TMP13:%.*]] = mul i64 [[TMP12]], 4
-; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP13]]
+; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <vscale x 4 x i64> [ [[TMP5]], [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[TMP9:%.*]] = getelementptr i32, ptr [[P:%.*]], <vscale x 4 x i64> [[VEC_IND]]
+; CHECK-NEXT:    [[WIDE_MASKED_GATHER:%.*]] = call <vscale x 4 x i32> @llvm.masked.gather.nxv4i32.nxv4p0(<vscale x 4 x ptr> [[TMP9]], i32 4, <vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer), <vscale x 4 x i32> poison)
+; CHECK-NEXT:    [[TMP10:%.*]] = add <vscale x 4 x i32> [[WIDE_MASKED_GATHER]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 1, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
+; CHECK-NEXT:    call void @llvm.masked.scatter.nxv4i32.nxv4p0(<vscale x 4 x i32> [[TMP10]], <vscale x 4 x ptr> [[TMP9]], i32 4, <vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer))
+; CHECK-NEXT:    [[TMP11:%.*]] = call i64 @llvm.vscale.i64()
+; CHECK-NEXT:    [[TMP12:%.*]] = mul i64 [[TMP11]], 4
+; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP12]]
 ; CHECK-NEXT:    [[VEC_IND_NEXT]] = add <vscale x 4 x i64> [[VEC_IND]], [[DOTSPLAT]]
-; CHECK-NEXT:    [[TMP14:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
-; CHECK-NEXT:    br i1 [[TMP14]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
+; CHECK-NEXT:    [[TMP13:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; CHECK-NEXT:    br i1 [[TMP13]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
 ; CHECK:       middle.block:
 ; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 1024, [[N_VEC]]
 ; CHECK-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
 ; CHECK:       scalar.ph:
 ; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
 ; CHECK-NEXT:    [[BC_RESUME_VAL1:%.*]] = phi i64 [ [[IND_END]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY]] ]
 ; CHECK-NEXT:    br label [[LOOP:%.*]]
 ; CHECK:       loop:

 ; CHECK:       vector.ph:
 ; CHECK-NEXT:    [[IND_END:%.*]] = getelementptr i8, ptr [[P:%.*]], i64 8128
 ; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
 ; CHECK:       vector.body:
 ; CHECK-NEXT:    [[POINTER_PHI:%.*]] = phi ptr [ [[P]], [[VECTOR_PH]] ], [ [[PTR_IND:%.*]], [[VECTOR_BODY]] ]
 ; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
 ; CHECK-NEXT:    [[TMP0:%.*]] = getelementptr i8, ptr [[POINTER_PHI]], <8 x i64> <i64 0, i64 8, i64 16, i64 24, i64 32, i64 40, i64 48, i64 56>
 ; CHECK-NEXT:    [[TMP1:%.*]] = extractelement <8 x ptr> [[TMP0]], i32 0
-; CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i32, ptr [[TMP1]], i32 0
-; CHECK-NEXT:    [[WIDE_VEC:%.*]] = load <16 x i32>, ptr [[TMP2]], align 4
+; CHECK-NEXT:    [[WIDE_VEC:%.*]] = load <16 x i32>, ptr [[TMP1]], align 4
 ; CHECK-NEXT:    [[STRIDED_VEC:%.*]] = shufflevector <16 x i32> [[WIDE_VEC]], <16 x i32> poison, <8 x i32> <i32 0, i32 2, i32 4, i32 6, i32 8, i32 10, i32 12, i32 14>
-; CHECK-NEXT:    [[TMP3:%.*]] = add <8 x i32> [[STRIDED_VEC]], <i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1>
-; CHECK-NEXT:    call void @llvm.masked.scatter.v8i32.v8p0(<8 x i32> [[TMP3]], <8 x ptr> [[TMP0]], i32 4, <8 x i1> <i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true>)
+; CHECK-NEXT:    [[TMP2:%.*]] = add <8 x i32> [[STRIDED_VEC]], <i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1>
+; CHECK-NEXT:    call void @llvm.masked.scatter.v8i32.v8p0(<8 x i32> [[TMP2]], <8 x ptr> [[TMP0]], i32 4, <8 x i1> <i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true>)
 ; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 8
 ; CHECK-NEXT:    [[PTR_IND]] = getelementptr i8, ptr [[POINTER_PHI]], i64 64
-; CHECK-NEXT:    [[TMP4:%.*]] = icmp eq i64 [[INDEX_NEXT]], 1016
-; CHECK-NEXT:    br i1 [[TMP4]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP6:![0-9]+]]
+; CHECK-NEXT:    [[TMP3:%.*]] = icmp eq i64 [[INDEX_NEXT]], 1016
+; CHECK-NEXT:    br i1 [[TMP3]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP6:![0-9]+]]
 ; CHECK:       middle.block:
 ; CHECK-NEXT:    br label [[SCALAR_PH]]
 ; CHECK:       scalar.ph:
 ; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 1016, [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
 ; CHECK-NEXT:    [[BC_RESUME_VAL1:%.*]] = phi ptr [ [[IND_END]], [[MIDDLE_BLOCK]] ], [ [[P]], [[ENTRY]] ]
 ; CHECK-NEXT:    br label [[LOOP:%.*]]
 ; CHECK:       loop:
 ; CHECK-NEXT:    [[I:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[NEXTI:%.*]], [[LOOP]] ]

 ; NOSTRIDED:       vector.ph:
 ; NOSTRIDED-NEXT:    [[TMP3:%.*]] = call i64 @llvm.vscale.i64()
 ; NOSTRIDED-NEXT:    [[TMP4:%.*]] = mul i64 [[TMP3]], 4
 ; NOSTRIDED-NEXT:    [[N_MOD_VF:%.*]] = urem i64 1024, [[TMP4]]
 ; NOSTRIDED-NEXT:    [[N_VEC:%.*]] = sub i64 1024, [[N_MOD_VF]]
 ; NOSTRIDED-NEXT:    br label [[VECTOR_BODY:%.*]]
 ; NOSTRIDED:       vector.body:
 ; NOSTRIDED-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
-; NOSTRIDED-NEXT:    [[TMP5:%.*]] = add i64 [[INDEX]], 0
-; NOSTRIDED-NEXT:    [[TMP6:%.*]] = mul nuw nsw i64 [[TMP5]], 1
-; NOSTRIDED-NEXT:    [[TMP7:%.*]] = getelementptr i32, ptr [[P:%.*]], i64 [[TMP6]]
-; NOSTRIDED-NEXT:    [[TMP8:%.*]] = getelementptr i32, ptr [[TMP7]], i32 0
-; NOSTRIDED-NEXT:    [[WIDE_LOAD:%.*]] = load <vscale x 4 x i32>, ptr [[TMP8]], align 4
-; NOSTRIDED-NEXT:    [[TMP9:%.*]] = add <vscale x 4 x i32> [[WIDE_LOAD]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 1, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
-; NOSTRIDED-NEXT:    store <vscale x 4 x i32> [[TMP9]], ptr [[TMP8]], align 4
-; NOSTRIDED-NEXT:    [[TMP10:%.*]] = call i64 @llvm.vscale.i64()
-; NOSTRIDED-NEXT:    [[TMP11:%.*]] = mul i64 [[TMP10]], 4
-; NOSTRIDED-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP11]]
-; NOSTRIDED-NEXT:    [[TMP12:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
-; NOSTRIDED-NEXT:    br i1 [[TMP12]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP8:![0-9]+]]
+; NOSTRIDED-NEXT:    [[TMP5:%.*]] = mul nuw nsw i64 [[INDEX]], 1
+; NOSTRIDED-NEXT:    [[TMP6:%.*]] = getelementptr i32, ptr [[P:%.*]], i64 [[TMP5]]
+; NOSTRIDED-NEXT:    [[WIDE_LOAD:%.*]] = load <vscale x 4 x i32>, ptr [[TMP6]], align 4
+; NOSTRIDED-NEXT:    [[TMP7:%.*]] = add <vscale x 4 x i32> [[WIDE_LOAD]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 1, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
+; NOSTRIDED-NEXT:    store <vscale x 4 x i32> [[TMP7]], ptr [[TMP6]], align 4
+; NOSTRIDED-NEXT:    [[TMP8:%.*]] = call i64 @llvm.vscale.i64()
+; NOSTRIDED-NEXT:    [[TMP9:%.*]] = mul i64 [[TMP8]], 4
+; NOSTRIDED-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP9]]
+; NOSTRIDED-NEXT:    [[TMP10:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; NOSTRIDED-NEXT:    br i1 [[TMP10]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP8:![0-9]+]]
 ; NOSTRIDED:       middle.block:
 ; NOSTRIDED-NEXT:    [[CMP_N:%.*]] = icmp eq i64 1024, [[N_VEC]]
 ; NOSTRIDED-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
 ; NOSTRIDED:       scalar.ph:
 ; NOSTRIDED-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ], [ 0, [[VECTOR_SCEVCHECK]] ]
 ; NOSTRIDED-NEXT:    br label [[LOOP:%.*]]
 ; NOSTRIDED:       loop:
 ; NOSTRIDED-NEXT:    [[I:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[NEXTI:%.*]], [[LOOP]] ]

 ; NOSTRIDED-NEXT:    [[TMP3:%.*]] = call i64 @llvm.vscale.i64()
 ; NOSTRIDED-NEXT:    [[TMP4:%.*]] = mul i64 [[TMP3]], 4
 ; NOSTRIDED-NEXT:    [[N_MOD_VF:%.*]] = urem i64 1024, [[TMP4]]
 ; NOSTRIDED-NEXT:    [[N_VEC:%.*]] = sub i64 1024, [[N_MOD_VF]]
 ; NOSTRIDED-NEXT:    [[IND_END:%.*]] = mul i64 [[N_VEC]], [[STRIDE]]
 ; NOSTRIDED-NEXT:    br label [[VECTOR_BODY:%.*]]
 ; NOSTRIDED:       vector.body:
 ; NOSTRIDED-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
-; NOSTRIDED-NEXT:    [[TMP5:%.*]] = add i64 [[INDEX]], 0
-; NOSTRIDED-NEXT:    [[TMP6:%.*]] = getelementptr i32, ptr [[P:%.*]], i64 [[TMP5]]
-; NOSTRIDED-NEXT:    [[TMP7:%.*]] = getelementptr i32, ptr [[TMP6]], i32 0
-; NOSTRIDED-NEXT:    [[WIDE_LOAD:%.*]] = load <vscale x 4 x i32>, ptr [[TMP7]], align 4
-; NOSTRIDED-NEXT:    [[TMP8:%.*]] = add <vscale x 4 x i32> [[WIDE_LOAD]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 1, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
-; NOSTRIDED-NEXT:    store <vscale x 4 x i32> [[TMP8]], ptr [[TMP7]], align 4
-; NOSTRIDED-NEXT:    [[TMP9:%.*]] = call i64 @llvm.vscale.i64()
-; NOSTRIDED-NEXT:    [[TMP10:%.*]] = mul i64 [[TMP9]], 4
-; NOSTRIDED-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP10]]
-; NOSTRIDED-NEXT:    [[TMP11:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
-; NOSTRIDED-NEXT:    br i1 [[TMP11]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP10:![0-9]+]]
+; NOSTRIDED-NEXT:    [[TMP5:%.*]] = getelementptr i32, ptr [[P:%.*]], i64 [[INDEX]]
+; NOSTRIDED-NEXT:    [[WIDE_LOAD:%.*]] = load <vscale x 4 x i32>, ptr [[TMP5]], align 4
+; NOSTRIDED-NEXT:    [[TMP6:%.*]] = add <vscale x 4 x i32> [[WIDE_LOAD]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 1, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
+; NOSTRIDED-NEXT:    store <vscale x 4 x i32> [[TMP6]], ptr [[TMP5]], align 4
+; NOSTRIDED-NEXT:    [[TMP7:%.*]] = call i64 @llvm.vscale.i64()
+; NOSTRIDED-NEXT:    [[TMP8:%.*]] = mul i64 [[TMP7]], 4
+; NOSTRIDED-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP8]]
+; NOSTRIDED-NEXT:    [[TMP9:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; NOSTRIDED-NEXT:    br i1 [[TMP9]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP10:![0-9]+]]
 ; NOSTRIDED:       middle.block:
 ; NOSTRIDED-NEXT:    [[CMP_N:%.*]] = icmp eq i64 1024, [[N_VEC]]
 ; NOSTRIDED-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
 ; NOSTRIDED:       scalar.ph:
 ; NOSTRIDED-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ], [ 0, [[VECTOR_SCEVCHECK]] ]
 ; NOSTRIDED-NEXT:    [[BC_RESUME_VAL1:%.*]] = phi i64 [ [[IND_END]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY]] ], [ 0, [[VECTOR_SCEVCHECK]] ]
 ; NOSTRIDED-NEXT:    br label [[LOOP:%.*]]
 ; NOSTRIDED:       loop:

 ; NOSTRIDED:       vector.ph:
 ; NOSTRIDED-NEXT:    [[TMP7:%.*]] = call i64 @llvm.vscale.i64()
 ; NOSTRIDED-NEXT:    [[TMP8:%.*]] = mul i64 [[TMP7]], 4
 ; NOSTRIDED-NEXT:    [[N_MOD_VF:%.*]] = urem i64 1024, [[TMP8]]
 ; NOSTRIDED-NEXT:    [[N_VEC:%.*]] = sub i64 1024, [[N_MOD_VF]]
 ; NOSTRIDED-NEXT:    br label [[VECTOR_BODY:%.*]]
 ; NOSTRIDED:       vector.body:
 ; NOSTRIDED-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
-; NOSTRIDED-NEXT:    [[TMP9:%.*]] = add i64 [[INDEX]], 0
-; NOSTRIDED-NEXT:    [[TMP10:%.*]] = mul nuw nsw i64 [[TMP9]], 1
-; NOSTRIDED-NEXT:    [[TMP11:%.*]] = getelementptr i32, ptr [[P]], i64 [[TMP10]]
-; NOSTRIDED-NEXT:    [[TMP12:%.*]] = getelementptr i32, ptr [[TMP11]], i32 0
-; NOSTRIDED-NEXT:    [[WIDE_LOAD:%.*]] = load <vscale x 4 x i32>, ptr [[TMP12]], align 4
-; NOSTRIDED-NEXT:    [[TMP13:%.*]] = add <vscale x 4 x i32> [[WIDE_LOAD]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 1, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
-; NOSTRIDED-NEXT:    [[TMP14:%.*]] = getelementptr i32, ptr [[P2]], i64 [[TMP10]]
-; NOSTRIDED-NEXT:    [[TMP15:%.*]] = getelementptr i32, ptr [[TMP14]], i32 0
-; NOSTRIDED-NEXT:    store <vscale x 4 x i32> [[TMP13]], ptr [[TMP15]], align 4
-; NOSTRIDED-NEXT:    [[TMP16:%.*]] = call i64 @llvm.vscale.i64()
-; NOSTRIDED-NEXT:    [[TMP17:%.*]] = mul i64 [[TMP16]], 4
-; NOSTRIDED-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP17]]
-; NOSTRIDED-NEXT:    [[TMP18:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
-; NOSTRIDED-NEXT:    br i1 [[TMP18]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP12:![0-9]+]]
+; NOSTRIDED-NEXT:    [[TMP9:%.*]] = mul nuw nsw i64 [[INDEX]], 1
+; NOSTRIDED-NEXT:    [[TMP10:%.*]] = getelementptr i32, ptr [[P]], i64 [[TMP9]]
+; NOSTRIDED-NEXT:    [[WIDE_LOAD:%.*]] = load <vscale x 4 x i32>, ptr [[TMP10]], align 4
+; NOSTRIDED-NEXT:    [[TMP11:%.*]] = add <vscale x 4 x i32> [[WIDE_LOAD]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 1, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
+; NOSTRIDED-NEXT:    [[TMP12:%.*]] = getelementptr i32, ptr [[P2]], i64 [[TMP9]]
+; NOSTRIDED-NEXT:    store <vscale x 4 x i32> [[TMP11]], ptr [[TMP12]], align 4
+; NOSTRIDED-NEXT:    [[TMP13:%.*]] = call i64 @llvm.vscale.i64()
+; NOSTRIDED-NEXT:    [[TMP14:%.*]] = mul i64 [[TMP13]], 4
+; NOSTRIDED-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP14]]
+; NOSTRIDED-NEXT:    [[TMP15:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; NOSTRIDED-NEXT:    br i1 [[TMP15]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP12:![0-9]+]]
 ; NOSTRIDED:       middle.block:
 ; NOSTRIDED-NEXT:    [[CMP_N:%.*]] = icmp eq i64 1024, [[N_VEC]]
 ; NOSTRIDED-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
 ; NOSTRIDED:       scalar.ph:
 ; NOSTRIDED-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ], [ 0, [[VECTOR_SCEVCHECK]] ], [ 0, [[VECTOR_MEMCHECK]] ]
 ; NOSTRIDED-NEXT:    br label [[LOOP:%.*]]
 ; NOSTRIDED:       loop:
 ; NOSTRIDED-NEXT:    [[I:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[NEXTI:%.*]], [[LOOP]] ]

 ; STRIDED-NEXT:    [[FOUND_CONFLICT:%.*]] = and i1 [[BOUND0]], [[BOUND1]]
 ; STRIDED-NEXT:    br i1 [[FOUND_CONFLICT]], label [[SCALAR_PH]], label [[VECTOR_PH:%.*]]
 ; STRIDED:       vector.ph:
 ; STRIDED-NEXT:    [[TMP8:%.*]] = call i64 @llvm.vscale.i64()
 ; STRIDED-NEXT:    [[TMP9:%.*]] = mul i64 [[TMP8]], 4
 ; STRIDED-NEXT:    [[N_MOD_VF:%.*]] = urem i64 1024, [[TMP9]]
 ; STRIDED-NEXT:    [[N_VEC:%.*]] = sub i64 1024, [[N_MOD_VF]]
 ; STRIDED-NEXT:    [[TMP10:%.*]] = call <vscale x 4 x i64> @llvm.experimental.stepvector.nxv4i64()
-; STRIDED-NEXT:    [[TMP11:%.*]] = add <vscale x 4 x i64> [[TMP10]], zeroinitializer
-; STRIDED-NEXT:    [[TMP12:%.*]] = mul <vscale x 4 x i64> [[TMP11]], shufflevector (<vscale x 4 x i64> insertelement (<vscale x 4 x i64> poison, i64 1, i64 0), <vscale x 4 x i64> poison, <vscale x 4 x i32> zeroinitializer)
-; STRIDED-NEXT:    [[INDUCTION:%.*]] = add <vscale x 4 x i64> zeroinitializer, [[TMP12]]
-; STRIDED-NEXT:    [[TMP13:%.*]] = call i64 @llvm.vscale.i64()
-; STRIDED-NEXT:    [[TMP14:%.*]] = mul i64 [[TMP13]], 4
-; STRIDED-NEXT:    [[TMP15:%.*]] = mul i64 1, [[TMP14]]
-; STRIDED-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 4 x i64> poison, i64 [[TMP15]], i64 0
+; STRIDED-NEXT:    [[TMP11:%.*]] = call i64 @llvm.vscale.i64()
+; STRIDED-NEXT:    [[TMP12:%.*]] = mul i64 [[TMP11]], 4
+; STRIDED-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 4 x i64> poison, i64 [[TMP12]], i64 0
 ; STRIDED-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 4 x i64> [[DOTSPLATINSERT]], <vscale x 4 x i64> poison, <vscale x 4 x i32> zeroinitializer
 ; STRIDED-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <vscale x 4 x i64> poison, i64 [[STRIDE]], i64 0
 ; STRIDED-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <vscale x 4 x i64> [[BROADCAST_SPLATINSERT]], <vscale x 4 x i64> poison, <vscale x 4 x i32> zeroinitializer
 ; STRIDED-NEXT:    br label [[VECTOR_BODY:%.*]]
 ; STRIDED:       vector.body:
 ; STRIDED-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
-; STRIDED-NEXT:    [[VEC_IND:%.*]] = phi <vscale x 4 x i64> [ [[INDUCTION]], [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
-; STRIDED-NEXT:    [[TMP16:%.*]] = mul nuw nsw <vscale x 4 x i64> [[VEC_IND]], [[BROADCAST_SPLAT]]
-; STRIDED-NEXT:    [[TMP17:%.*]] = getelementptr i32, ptr [[P]], <vscale x 4 x i64> [[TMP16]]
-; STRIDED-NEXT:    [[WIDE_MASKED_GATHER:%.*]] = call <vscale x 4 x i32> @llvm.masked.gather.nxv4i32.nxv4p0(<vscale x 4 x ptr> [[TMP17]], i32 4, <vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer), <vscale x 4 x i32> poison), !alias.scope !8
-; STRIDED-NEXT:    [[TMP18:%.*]] = add <vscale x 4 x i32> [[WIDE_MASKED_GATHER]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 1, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
-; STRIDED-NEXT:    [[TMP19:%.*]] = getelementptr i32, ptr [[P2]], <vscale x 4 x i64> [[TMP16]]
-; STRIDED-NEXT:    call void @llvm.masked.scatter.nxv4i32.nxv4p0(<vscale x 4 x i32> [[TMP18]], <vscale x 4 x ptr> [[TMP19]], i32 4, <vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer)), !alias.scope !11, !noalias !8
-; STRIDED-NEXT:    [[TMP20:%.*]] = call i64 @llvm.vscale.i64()
-; STRIDED-NEXT:    [[TMP21:%.*]] = mul i64 [[TMP20]], 4
-; STRIDED-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP21]]
+; STRIDED-NEXT:    [[VEC_IND:%.*]] = phi <vscale x 4 x i64> [ [[TMP10]], [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
+; STRIDED-NEXT:    [[TMP13:%.*]] = mul nuw nsw <vscale x 4 x i64> [[VEC_IND]], [[BROADCAST_SPLAT]]
+; STRIDED-NEXT:    [[TMP14:%.*]] = getelementptr i32, ptr [[P]], <vscale x 4 x i64> [[TMP13]]
+; STRIDED-NEXT:    [[WIDE_MASKED_GATHER:%.*]] = call <vscale x 4 x i32> @llvm.masked.gather.nxv4i32.nxv4p0(<vscale x 4 x ptr> [[TMP14]], i32 4, <vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer), <vscale x 4 x i32> poison), !alias.scope !8
+; STRIDED-NEXT:    [[TMP15:%.*]] = add <vscale x 4 x i32> [[WIDE_MASKED_GATHER]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 1, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
+; STRIDED-NEXT:    [[TMP16:%.*]] = getelementptr i32, ptr [[P2]], <vscale x 4 x i64> [[TMP13]]
+; STRIDED-NEXT:    call void @llvm.masked.scatter.nxv4i32.nxv4p0(<vscale x 4 x i32> [[TMP15]], <vscale x 4 x ptr> [[TMP16]], i32 4, <vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer)), !alias.scope !11, !noalias !8
+; STRIDED-NEXT:    [[TMP17:%.*]] = call i64 @llvm.vscale.i64()
+; STRIDED-NEXT:    [[TMP18:%.*]] = mul i64 [[TMP17]], 4
+; STRIDED-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP18]]
 ; STRIDED-NEXT:    [[VEC_IND_NEXT]] = add <vscale x 4 x i64> [[VEC_IND]], [[DOTSPLAT]]
-; STRIDED-NEXT:    [[TMP22:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
-; STRIDED-NEXT:    br i1 [[TMP22]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP13:![0-9]+]]
+; STRIDED-NEXT:    [[TMP19:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; STRIDED-NEXT:    br i1 [[TMP19]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP13:![0-9]+]]
 ; STRIDED:       middle.block:
 ; STRIDED-NEXT:    [[CMP_N:%.*]] = icmp eq i64 1024, [[N_VEC]]
 ; STRIDED-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
 ; STRIDED:       scalar.ph:
 ; STRIDED-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ], [ 0, [[VECTOR_MEMCHECK]] ]
 ; STRIDED-NEXT:    br label [[LOOP:%.*]]
 ; STRIDED:       loop:
 ; STRIDED-NEXT:    [[I:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[NEXTI:%.*]], [[LOOP]] ]

 ; NOSTRIDED-NEXT:    [[TMP3:%.*]] = call i64 @llvm.vscale.i64()
 ; NOSTRIDED-NEXT:    [[TMP4:%.*]] = mul i64 [[TMP3]], 4
 ; NOSTRIDED-NEXT:    [[N_MOD_VF:%.*]] = urem i64 1024, [[TMP4]]
 ; NOSTRIDED-NEXT:    [[N_VEC:%.*]] = sub i64 1024, [[N_MOD_VF]]
 ; NOSTRIDED-NEXT:    [[IND_END:%.*]] = mul i64 [[N_VEC]], [[STRIDE]]
 ; NOSTRIDED-NEXT:    br label [[VECTOR_BODY:%.*]]
 ; NOSTRIDED:       vector.body:
 ; NOSTRIDED-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
-; NOSTRIDED-NEXT:    [[TMP5:%.*]] = add i64 [[INDEX]], 0
-; NOSTRIDED-NEXT:    [[TMP6:%.*]] = getelementptr i32, ptr [[P:%.*]], i64 [[TMP5]]
-; NOSTRIDED-NEXT:    [[TMP7:%.*]] = getelementptr i32, ptr [[TMP6]], i32 0
-; NOSTRIDED-NEXT:    [[WIDE_LOAD:%.*]] = load <vscale x 4 x i32>, ptr [[TMP7]], align 4
-; NOSTRIDED-NEXT:    [[TMP8:%.*]] = add <vscale x 4 x i32> [[WIDE_LOAD]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 1, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
-; NOSTRIDED-NEXT:    store <vscale x 4 x i32> [[TMP8]], ptr [[TMP7]], align 4
-; NOSTRIDED-NEXT:    [[TMP9:%.*]] = call i64 @llvm.vscale.i64()
-; NOSTRIDED-NEXT:    [[TMP10:%.*]] = mul i64 [[TMP9]], 4
-; NOSTRIDED-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP10]]
-; NOSTRIDED-NEXT:    [[TMP11:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
-; NOSTRIDED-NEXT:    br i1 [[TMP11]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP14:![0-9]+]]
+; NOSTRIDED-NEXT:    [[TMP5:%.*]] = getelementptr i32, ptr [[P:%.*]], i64 [[INDEX]]
+; NOSTRIDED-NEXT:    [[WIDE_LOAD:%.*]] = load <vscale x 4 x i32>, ptr [[TMP5]], align 4
+; NOSTRIDED-NEXT:    [[TMP6:%.*]] = add <vscale x 4 x i32> [[WIDE_LOAD]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 1, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
+; NOSTRIDED-NEXT:    store <vscale x 4 x i32> [[TMP6]], ptr [[TMP5]], align 4
+; NOSTRIDED-NEXT:    [[TMP7:%.*]] = call i64 @llvm.vscale.i64()
+; NOSTRIDED-NEXT:    [[TMP8:%.*]] = mul i64 [[TMP7]], 4
+; NOSTRIDED-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP8]]
+; NOSTRIDED-NEXT:    [[TMP9:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; NOSTRIDED-NEXT:    br i1 [[TMP9]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP14:![0-9]+]]
 ; NOSTRIDED:       middle.block:
 ; NOSTRIDED-NEXT:    [[CMP_N:%.*]] = icmp eq i64 1024, [[N_VEC]]
 ; NOSTRIDED-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
 ; NOSTRIDED:       scalar.ph:
 ; NOSTRIDED-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ], [ 0, [[VECTOR_SCEVCHECK]] ]
 ; NOSTRIDED-NEXT:    [[BC_RESUME_VAL1:%.*]] = phi i64 [ [[IND_END]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY]] ], [ 0, [[VECTOR_SCEVCHECK]] ]
 ; NOSTRIDED-NEXT:    br label [[LOOP:%.*]]
 ; NOSTRIDED:       loop:

 ; STRIDED-NEXT:    [[IND_END7:%.*]] = getelementptr i8, ptr [[P2]], i64 [[TMP11]]
 ; STRIDED-NEXT:    br label [[VECTOR_BODY:%.*]]
 ; STRIDED:       vector.body:
 ; STRIDED-NEXT:    [[POINTER_PHI:%.*]] = phi ptr [ [[P]], [[VECTOR_PH]] ], [ [[PTR_IND:%.*]], [[VECTOR_BODY]] ]
 ; STRIDED-NEXT:    [[POINTER_PHI11:%.*]] = phi ptr [ [[P2]], [[VECTOR_PH]] ], [ [[PTR_IND12:%.*]], [[VECTOR_BODY]] ]
 ; STRIDED-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
 ; STRIDED-NEXT:    [[TMP12:%.*]] = call i64 @llvm.vscale.i64()
 ; STRIDED-NEXT:    [[TMP13:%.*]] = mul i64 [[TMP12]], 4
-; STRIDED-NEXT:    [[TMP14:%.*]] = mul i64 [[TMP13]], 1
-; STRIDED-NEXT:    [[TMP15:%.*]] = mul i64 [[STRIDE]], [[TMP14]]
-; STRIDED-NEXT:    [[TMP16:%.*]] = mul i64 [[TMP13]], 0
-; STRIDED-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 4 x i64> poison, i64 [[TMP16]], i64 0
+; STRIDED-NEXT:    [[TMP14:%.*]] = mul i64 [[STRIDE]], [[TMP13]]
+; STRIDED-NEXT:    [[TMP15:%.*]] = mul i64 [[TMP13]], 0
+; STRIDED-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 4 x i64> poison, i64 [[TMP15]], i64 0
 ; STRIDED-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 4 x i64> [[DOTSPLATINSERT]], <vscale x 4 x i64> poison, <vscale x 4 x i32> zeroinitializer
-; STRIDED-NEXT:    [[TMP17:%.*]] = call <vscale x 4 x i64> @llvm.experimental.stepvector.nxv4i64()
-; STRIDED-NEXT:    [[TMP18:%.*]] = add <vscale x 4 x i64> [[DOTSPLAT]], [[TMP17]]
+; STRIDED-NEXT:    [[TMP16:%.*]] = call <vscale x 4 x i64> @llvm.experimental.stepvector.nxv4i64()
+; STRIDED-NEXT:    [[TMP17:%.*]] = add <vscale x 4 x i64> [[DOTSPLAT]], [[TMP16]]
 ; STRIDED-NEXT:    [[DOTSPLATINSERT9:%.*]] = insertelement <vscale x 4 x i64> poison, i64 [[STRIDE]], i64 0
 ; STRIDED-NEXT:    [[DOTSPLAT10:%.*]] = shufflevector <vscale x 4 x i64> [[DOTSPLATINSERT9]], <vscale x 4 x i64> poison, <vscale x 4 x i32> zeroinitializer
-; STRIDED-NEXT:    [[VECTOR_GEP:%.*]] = mul <vscale x 4 x i64> [[TMP18]], [[DOTSPLAT10]]
-; STRIDED-NEXT:    [[TMP19:%.*]] = getelementptr i8, ptr [[POINTER_PHI]], <vscale x 4 x i64> [[VECTOR_GEP]]
-; STRIDED-NEXT:    [[TMP20:%.*]] = call i64 @llvm.vscale.i64()
-; STRIDED-NEXT:    [[TMP21:%.*]] = mul i64 [[TMP20]], 4
-; STRIDED-NEXT:    [[TMP22:%.*]] = mul i64 [[TMP21]], 1
-; STRIDED-NEXT:    [[TMP23:%.*]] = mul i64 [[STRIDE]], [[TMP22]]
-; STRIDED-NEXT:    [[TMP24:%.*]] = mul i64 [[TMP21]], 0
-; STRIDED-NEXT:    [[DOTSPLATINSERT13:%.*]] = insertelement <vscale x 4 x i64> poison, i64 [[TMP24]], i64 0
+; STRIDED-NEXT:    [[VECTOR_GEP:%.*]] = mul <vscale x 4 x i64> [[TMP17]], [[DOTSPLAT10]]
+; STRIDED-NEXT:    [[TMP18:%.*]] = getelementptr i8, ptr [[POINTER_PHI]], <vscale x 4 x i64> [[VECTOR_GEP]]
+; STRIDED-NEXT:    [[TMP19:%.*]] = call i64 @llvm.vscale.i64()
+; STRIDED-NEXT:    [[TMP20:%.*]] = mul i64 [[TMP19]], 4
+; STRIDED-NEXT:    [[TMP21:%.*]] = mul i64 [[STRIDE]], [[TMP20]]
+; STRIDED-NEXT:    [[TMP22:%.*]] = mul i64 [[TMP20]], 0
+; STRIDED-NEXT:    [[DOTSPLATINSERT13:%.*]] = insertelement <vscale x 4 x i64> poison, i64 [[TMP22]], i64 0
 ; STRIDED-NEXT:    [[DOTSPLAT14:%.*]] = shufflevector <vscale x 4 x i64> [[DOTSPLATINSERT13]], <vscale x 4 x i64> poison, <vscale x 4 x i32> zeroinitializer
-; STRIDED-NEXT:    [[TMP25:%.*]] = call <vscale x 4 x i64> @llvm.experimental.stepvector.nxv4i64()
-; STRIDED-NEXT:    [[TMP26:%.*]] = add <vscale x 4 x i64> [[DOTSPLAT14]], [[TMP25]]
-; STRIDED-NEXT:    [[VECTOR_GEP17:%.*]] = mul <vscale x 4 x i64> [[TMP26]], [[DOTSPLAT10]]
-; STRIDED-NEXT:    [[TMP27:%.*]] = getelementptr i8, ptr [[POINTER_PHI11]], <vscale x 4 x i64> [[VECTOR_GEP17]]
-; STRIDED-NEXT:    [[WIDE_MASKED_GATHER:%.*]] = call <vscale x 4 x i32> @llvm.masked.gather.nxv4i32.nxv4p0(<vscale x 4 x ptr> [[TMP19]], i32 4, <vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer), <vscale x 4 x i32> poison), !alias.scope !15
-; STRIDED-NEXT:    [[TMP28:%.*]] = add <vscale x 4 x i32> [[WIDE_MASKED_GATHER]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 1, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
-; STRIDED-NEXT:    call void @llvm.masked.scatter.nxv4i32.nxv4p0(<vscale x 4 x i32> [[TMP28]], <vscale x 4 x ptr> [[TMP27]], i32 4, <vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer)), !alias.scope !18, !noalias !15
-; STRIDED-NEXT:    [[TMP29:%.*]] = call i64 @llvm.vscale.i64()
-; STRIDED-NEXT:    [[TMP30:%.*]] = mul i64 [[TMP29]], 4
-; STRIDED-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP30]]
-; STRIDED-NEXT:    [[PTR_IND]] = getelementptr i8, ptr [[POINTER_PHI]], i64 [[TMP15]]
-; STRIDED-NEXT:    [[PTR_IND12]] = getelementptr i8, ptr [[POINTER_PHI11]], i64 [[TMP23]]
-; STRIDED-NEXT:    [[TMP31:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
-; STRIDED-NEXT:    br i1 [[TMP31]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP20:![0-9]+]]
+; STRIDED-NEXT:    [[TMP23:%.*]] = call <vscale x 4 x i64> @llvm.experimental.stepvector.nxv4i64()
+; STRIDED-NEXT:    [[TMP24:%.*]] = add <vscale x 4 x i64> [[DOTSPLAT14]], [[TMP23]]
+; STRIDED-NEXT:    [[VECTOR_GEP17:%.*]] = mul <vscale x 4 x i64> [[TMP24]], [[DOTSPLAT10]]
+; STRIDED-NEXT:    [[TMP25:%.*]] = getelementptr i8, ptr [[POINTER_PHI11]], <vscale x 4 x i64> [[VECTOR_GEP17]]
+; STRIDED-NEXT:    [[WIDE_MASKED_GATHER:%.*]] = call <vscale x 4 x i32> @llvm.masked.gather.nxv4i32.nxv4p0(<vscale x 4 x ptr> [[TMP18]], i32 4, <vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer), <vscale x 4 x i32> poison), !alias.scope !15
+; STRIDED-NEXT:    [[TMP26:%.*]] = add <vscale x 4 x i32> [[WIDE_MASKED_GATHER]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 1, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
+; STRIDED-NEXT:    call void @llvm.masked.scatter.nxv4i32.nxv4p0(<vscale x 4 x i32> [[TMP26]], <vscale x 4 x ptr> [[TMP25]], i32 4, <vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer)), !alias.scope !18, !noalias !15
+; STRIDED-NEXT:    [[TMP27:%.*]] = call i64 @llvm.vscale.i64()
+; STRIDED-NEXT:    [[TMP28:%.*]] = mul i64 [[TMP27]], 4
+; STRIDED-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP28]]
+; STRIDED-NEXT:    [[PTR_IND]] = getelementptr i8, ptr [[POINTER_PHI]], i64 [[TMP14]]
+; STRIDED-NEXT:    [[PTR_IND12]] = getelementptr i8, ptr [[POINTER_PHI11]], i64 [[TMP21]]
+; STRIDED-NEXT:    [[TMP29:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; STRIDED-NEXT:    br i1 [[TMP29]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP20:![0-9]+]]
 ; STRIDED:       middle.block:
 ; STRIDED-NEXT:    [[CMP_N:%.*]] = icmp eq i64 1024, [[N_VEC]]
 ; STRIDED-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
 ; STRIDED:       scalar.ph:
 ; STRIDED-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ], [ 0, [[VECTOR_MEMCHECK]] ]
 ; STRIDED-NEXT:    [[BC_RESUME_VAL6:%.*]] = phi ptr [ [[IND_END]], [[MIDDLE_BLOCK]] ], [ [[P]], [[ENTRY]] ], [ [[P]], [[VECTOR_MEMCHECK]] ]
 ; STRIDED-NEXT:    [[BC_RESUME_VAL8:%.*]] = phi ptr [ [[IND_END7]], [[MIDDLE_BLOCK]] ], [ [[P2]], [[ENTRY]] ], [ [[P2]], [[VECTOR_MEMCHECK]] ]
 ; STRIDED-NEXT:    br label [[LOOP:%.*]]