diff --git a/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp b/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp
index fd762f35124b5..69f3367e475b2 100644
--- a/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp
+++ b/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp
@@ -1143,6 +1143,31 @@ InstructionCost RISCVTTIImpl::getInterleavedMemoryOpCost(
     return InstructionCost::getInvalid();
 
   auto *FVTy = cast<FixedVectorType>(VecTy);
+  // When gaps are only at the tail, for interleaved load, we can emit a wide
+  // masked load and shufflevectors. For interleaved store, we can emit
+  // shufflevectors and a wide masked store. The interleaved memory access pass
+  // will lower them into vlsseg/vssseg intrinsics.
+  if (UseMaskForGaps) {
+    unsigned NumOfFields = Indices.size();
+    bool IsTailGapOnly = NumOfFields > 1 && (NumOfFields == Indices.back() + 1);
+    if (IsTailGapOnly &&
+        NumOfFields <= TLI->getMaxSupportedInterleaveFactor()) {
+      std::pair<InstructionCost, MVT> LT = getTypeLegalizationCost(FVTy);
+      if (LT.second.isVector() &&
+          FVTy->getElementCount().isKnownMultipleOf(Factor)) {
+        auto *SubVecTy = VectorType::get(
+            FVTy->getElementType(),
+            FVTy->getElementCount().divideCoefficientBy(Factor));
+        if (TLI->isLegalInterleavedAccessType(SubVecTy, NumOfFields, Alignment,
+                                              AddressSpace, DL)) {
+          // The cost is proportional to the total number of element accesses.
+          unsigned NumAccesses = getEstimatedVLFor(FVTy);
+          return NumAccesses * TTI::TCC_Basic;
+        }
+      }
+    }
+  }
+
   InstructionCost MemCost =
       getMemoryOpCost(Opcode, VecTy, Alignment, AddressSpace, CostKind);
   unsigned VF = FVTy->getNumElements() / Factor;
diff --git a/llvm/test/Transforms/LoopVectorize/RISCV/tail-folding-interleave.ll b/llvm/test/Transforms/LoopVectorize/RISCV/tail-folding-interleave.ll
index aeab26ebc9f44..2a3ba57bbe1ea 100644
--- a/llvm/test/Transforms/LoopVectorize/RISCV/tail-folding-interleave.ll
+++ b/llvm/test/Transforms/LoopVectorize/RISCV/tail-folding-interleave.ll
@@ -502,30 +502,23 @@ define void @store_factor_4_with_tail_gap(i32 %n, ptr noalias %a) {
 ;
 ; NO-VP-LABEL: @store_factor_4_with_tail_gap(
 ; NO-VP-NEXT:  entry:
-; NO-VP-NEXT:    [[TMP0:%.*]] = call i32 @llvm.vscale.i32()
-; NO-VP-NEXT:    [[TMP1:%.*]] = shl nuw i32 [[TMP0]], 2
-; NO-VP-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i32 [[N:%.*]], [[TMP1]]
+; NO-VP-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i32 [[N:%.*]], 8
 ; NO-VP-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
 ; NO-VP:       vector.ph:
-; NO-VP-NEXT:    [[TMP2:%.*]] = call i32 @llvm.vscale.i32()
-; NO-VP-NEXT:    [[TMP9:%.*]] = shl nuw i32 [[TMP2]], 2
-; NO-VP-NEXT:    [[N_MOD_VF:%.*]] = urem i32 [[N]], [[TMP9]]
+; NO-VP-NEXT:    [[N_MOD_VF:%.*]] = urem i32 [[N]], 8
 ; NO-VP-NEXT:    [[N_VEC:%.*]] = sub i32 [[N]], [[N_MOD_VF]]
-; NO-VP-NEXT:    [[TMP7:%.*]] = call <vscale x 4 x i32> @llvm.stepvector.nxv4i32()
-; NO-VP-NEXT:    [[BROADCAST_SPLATINSERT2:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[TMP9]], i64 0
-; NO-VP-NEXT:    [[BROADCAST_SPLAT3:%.*]] = shufflevector <vscale x 4 x i32> [[BROADCAST_SPLATINSERT2]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
 ; NO-VP-NEXT:    br label [[VECTOR_BODY:%.*]]
 ; NO-VP:       vector.body:
 ; NO-VP-NEXT:    [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
-; NO-VP-NEXT:    [[VEC_IND4:%.*]] = phi <vscale x 4 x i32> [ [[TMP7]], [[VECTOR_PH]] ], [ [[VEC_IND_NEXT5:%.*]], [[VECTOR_BODY]] ]
-; NO-VP-NEXT:    [[TMP11:%.*]] = getelementptr inbounds [4 x i32], ptr [[A:%.*]], <vscale x 4 x i32> [[VEC_IND4]], i32 0
-; NO-VP-NEXT:    call void @llvm.masked.scatter.nxv4i32.nxv4p0(<vscale x 4 x i32> [[VEC_IND4]], <vscale x 4 x ptr> align 4 [[TMP11]], <vscale x 4 x i1> splat (i1 true))
-; NO-VP-NEXT:    [[TMP12:%.*]] = getelementptr inbounds [4 x i32], ptr [[A]], <vscale x 4 x i32> [[VEC_IND4]], i32 1
-; NO-VP-NEXT:    call void @llvm.masked.scatter.nxv4i32.nxv4p0(<vscale x 4 x i32> [[VEC_IND4]], <vscale x 4 x ptr> align 4 [[TMP12]], <vscale x 4 x i1> splat (i1 true))
-; NO-VP-NEXT:    [[TMP13:%.*]] = getelementptr inbounds [4 x i32], ptr [[A]], <vscale x 4 x i32> [[VEC_IND4]], i32 2
-; NO-VP-NEXT:    call void @llvm.masked.scatter.nxv4i32.nxv4p0(<vscale x 4 x i32> [[VEC_IND4]], <vscale x 4 x ptr> align 4 [[TMP13]], <vscale x 4 x i1> splat (i1 true))
-; NO-VP-NEXT:    [[INDEX_NEXT]] = add nuw i32 [[INDEX]], [[TMP9]]
-; NO-VP-NEXT:    [[VEC_IND_NEXT5]] = add nuw nsw <vscale x 4 x i32> [[VEC_IND4]], [[BROADCAST_SPLAT3]]
+; NO-VP-NEXT:    [[VEC_IND:%.*]] = phi <8 x i32> [ <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
+; NO-VP-NEXT:    [[TMP0:%.*]] = getelementptr inbounds [4 x i32], ptr [[A:%.*]], i32 [[INDEX]], i32 0
+; NO-VP-NEXT:    [[TMP1:%.*]] = shufflevector <8 x i32> [[VEC_IND]], <8 x i32> [[VEC_IND]], <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
+; NO-VP-NEXT:    [[TMP2:%.*]] = shufflevector <8 x i32> [[VEC_IND]], <8 x i32> poison, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
+; NO-VP-NEXT:    [[TMP3:%.*]] = shufflevector <16 x i32> [[TMP1]], <16 x i32> [[TMP2]], <32 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 16, i32 17, i32 18, i32 19, i32 20, i32 21, i32 22, i32 23, i32 24, i32 25, i32 26, i32 27, i32 28, i32 29, i32 30, i32 31>
+; NO-VP-NEXT:    [[INTERLEAVED_VEC:%.*]] = shufflevector <32 x i32> [[TMP3]], <32 x i32> poison, <32 x i32> <i32 0, i32 8, i32 16, i32 24, i32 1, i32 9, i32 17, i32 25, i32 2, i32 10, i32 18, i32 26, i32 3, i32 11, i32 19, i32 27, i32 4, i32 12, i32 20, i32 28, i32 5, i32 13, i32 21, i32 29, i32 6, i32 14, i32 22, i32 30, i32 7, i32 15, i32 23, i32 31>
+; NO-VP-NEXT:    call void @llvm.masked.store.v32i32.p0(<32 x i32> [[INTERLEAVED_VEC]], ptr align 4 [[TMP0]], <32 x i1> <i1 true, i1 true, i1 true, i1 false, i1 true, i1 true, i1 true, i1 false, i1 true, i1 true, i1 true, i1 false, i1 true, i1 true, i1 true, i1 false, i1 true, i1 true, i1 true, i1 false, i1 true, i1 true, i1 true, i1 false, i1 true, i1 true, i1 true, i1 false, i1 true, i1 true, i1 true, i1 false>)
+; NO-VP-NEXT:    [[INDEX_NEXT]] = add nuw i32 [[INDEX]], 8
+; NO-VP-NEXT:    [[VEC_IND_NEXT]] = add nuw nsw <8 x i32> [[VEC_IND]], splat (i32 8)
 ; NO-VP-NEXT:    [[TMP14:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
 ; NO-VP-NEXT:    br i1 [[TMP14]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP10:![0-9]+]]
 ; NO-VP:       middle.block: