diff --git a/modules/core/include/opencv2/core/hal/intrin.hpp b/modules/core/include/opencv2/core/hal/intrin.hpp
index 0041030dad..d79c995a27 100644
--- a/modules/core/include/opencv2/core/hal/intrin.hpp
+++ b/modules/core/include/opencv2/core/hal/intrin.hpp
@@ -695,236 +695,14 @@ namespace CV__SIMD_NAMESPACE {
     /** @brief SIMD processing state cleanup call */
     inline void vx_cleanup() { VXPREFIX(_cleanup)(); }
 
-#if CV_SIMD
+#if !CV_SIMD_SCALABLE
     // Compatibility layer
-    #define CV_SIMD_SCALABLE 0
-    #define CV_SIMD_SCALABLE_64F 0
 
-    template <class T>
-    struct VTraits;
-#if CV_SIMD512 && (!defined(CV__SIMD_FORCE_WIDTH) || CV__SIMD_FORCE_WIDTH == 512)
-    template <>
-    struct VTraits<v_uint8>
-    {
-        static inline int vlanes() { return v_uint8::nlanes; }
-        enum { nlanes = 64, max_nlanes = nlanes };
-        using lane_type = uchar;
+    template<typename T> struct VTraits {
+        static inline int vlanes() { return T::nlanes; }
+        enum { nlanes = T::nlanes, max_nlanes = T::nlanes };
+        using lane_type = typename T::lane_type;
     };
-    template <>
-    struct VTraits<v_int8>
-    {
-        static inline int vlanes() { return v_int8::nlanes; }
-        enum { nlanes = 64, max_nlanes = nlanes };
-        using lane_type = schar;
-    };
-    template <>
-    struct VTraits<v_uint16>
-    {
-        static inline int vlanes() { return v_uint16::nlanes; }
-        enum { nlanes = 32, max_nlanes = nlanes };
-        using lane_type = ushort;
-    };
-    template <>
-    struct VTraits<v_int16>
-    {
-        static inline int vlanes() { return v_int16::nlanes;  }
-        enum { nlanes = 32, max_nlanes = nlanes };
-        using lane_type = short;
-    };
-    template <>
-    struct VTraits<v_uint32>
-    {
-        static inline int vlanes() { return v_uint32::nlanes; }
-        enum { nlanes = 16, max_nlanes = nlanes };
-        using lane_type = uint;
-    };
-    template <>
-    struct VTraits<v_int32>
-    {
-        static inline int vlanes() { return v_int32::nlanes; }
-        enum { nlanes = 16, max_nlanes = nlanes };
-        using lane_type = int;
-    };
-
-    template <>
-    struct VTraits<v_float32>
-    {
-        static inline int vlanes() { return v_float32::nlanes; }
-        enum { nlanes = 16, max_nlanes = nlanes };
-        using lane_type = float;
-    };
-    template <>
-    struct VTraits<v_uint64>
-    {
-        static inline int vlanes() { return v_uint64::nlanes; }
-        enum { nlanes = 8, max_nlanes = nlanes };
-        using lane_type = uint64;
-    };
-    template <>
-    struct VTraits<v_int64>
-    {
-        static inline int vlanes() { return v_int64::nlanes; }
-        enum { nlanes = 8, max_nlanes = nlanes };
-        using lane_type = int64;
-    };
-    #if CV_SIMD_64F
-    template <>
-    struct VTraits<v_float64>
-    {
-        static inline int vlanes() { return v_float64::nlanes; }
-        enum { nlanes = 8, max_nlanes = nlanes };
-        using lane_type = double;
-    };
-    #endif
-#elif CV_SIMD256 && (!defined(CV__SIMD_FORCE_WIDTH) || CV__SIMD_FORCE_WIDTH == 256)
-    template <>
-    struct VTraits<v_uint8>
-    {
-        static inline int vlanes() { return v_uint8::nlanes; }
-        enum { nlanes = 32, max_nlanes = nlanes };
-        using lane_type = uchar;
-    };
-    template <>
-    struct VTraits<v_int8>
-    {
-        static inline int vlanes() { return v_int8::nlanes; }
-        enum { nlanes = 32, max_nlanes = nlanes };
-        using lane_type = schar;
-    };
-    template <>
-    struct VTraits<v_uint16>
-    {
-        static inline int vlanes() { return v_uint16::nlanes; }
-        enum { nlanes = 16, max_nlanes = nlanes };
-        using lane_type = ushort;
-    };
-    template <>
-    struct VTraits<v_int16>
-    {
-        static inline int vlanes() { return v_int16::nlanes;  }
-        enum { nlanes = 16, max_nlanes = nlanes };
-        using lane_type = short;
-    };
-    template <>
-    struct VTraits<v_uint32>
-    {
-        static inline int vlanes() { return v_uint32::nlanes; }
-        enum { nlanes = 8, max_nlanes = nlanes };
-        using lane_type = uint;
-    };
-    template <>
-    struct VTraits<v_int32>
-    {
-        static inline int vlanes() { return v_int32::nlanes; }
-        enum { nlanes = 8, max_nlanes = nlanes };
-        using lane_type = int;
-    };
-
-    template <>
-    struct VTraits<v_float32>
-    {
-        static inline int vlanes() { return v_float32::nlanes; }
-        enum { nlanes = 8, max_nlanes = nlanes };
-        using lane_type = float;
-    };
-    template <>
-    struct VTraits<v_uint64>
-    {
-        static inline int vlanes() { return v_uint64::nlanes; }
-        enum { nlanes = 4, max_nlanes = nlanes };
-        using lane_type = uint64;
-    };
-    template <>
-    struct VTraits<v_int64>
-    {
-        static inline int vlanes() { return v_int64::nlanes; }
-        enum { nlanes = 4, max_nlanes = nlanes };
-        using lane_type = int64;
-    };
-    #if CV_SIMD_64F
-    template <>
-    struct VTraits<v_float64>
-    {
-        static inline int vlanes() { return v_float64::nlanes; }
-        enum { nlanes = 4, max_nlanes = nlanes };
-        using lane_type = double;
-    };
-    #endif
-#elif CV_SIMD128 && (!defined(CV__SIMD_FORCE_WIDTH) || CV__SIMD_FORCE_WIDTH == 128)
-    template <>
-    struct VTraits<v_uint8>
-    {
-        static inline int vlanes() { return v_uint8::nlanes; }
-        enum { nlanes = 16, max_nlanes = nlanes };
-        using lane_type = uchar;
-    };
-    template <>
-    struct VTraits<v_int8>
-    {
-        static inline int vlanes() { return v_int8::nlanes; }
-        enum { nlanes = 16, max_nlanes = nlanes };
-        using lane_type = schar;
-    };
-    template <>
-    struct VTraits<v_uint16>
-    {
-        static inline int vlanes() { return v_uint16::nlanes; }
-        enum { nlanes = 8, max_nlanes = nlanes };
-        using lane_type = ushort;
-    };
-    template <>
-    struct VTraits<v_int16>
-    {
-        static inline int vlanes() { return v_int16::nlanes;  }
-        enum { nlanes = 8, max_nlanes = nlanes };
-        using lane_type = short;
-    };
-    template <>
-    struct VTraits<v_uint32>
-    {
-        static inline int vlanes() { return v_uint32::nlanes; }
-        enum { nlanes = 4, max_nlanes = nlanes };
-        using lane_type = uint;
-    };
-    template <>
-    struct VTraits<v_int32>
-    {
-        static inline int vlanes() { return v_int32::nlanes; }
-        enum { nlanes = 4, max_nlanes = nlanes };
-        using lane_type = int;
-    };
-
-    template <>
-    struct VTraits<v_float32>
-    {
-        static inline int vlanes() { return v_float32::nlanes; }
-        enum { nlanes = 4, max_nlanes = nlanes };
-        using lane_type = float;
-    };
-    template <>
-    struct VTraits<v_uint64>
-    {
-        static inline int vlanes() { return v_uint64::nlanes; }
-        enum { nlanes = 2, max_nlanes = nlanes };
-        using lane_type = uint64;
-    };
-    template <>
-    struct VTraits<v_int64>
-    {
-        static inline int vlanes() { return v_int64::nlanes; }
-        enum { nlanes = 2, max_nlanes = nlanes };
-        using lane_type = int64;
-    };
-    #if CV_SIMD_64F
-    template <>
-    struct VTraits<v_float64>
-    {
-        static inline int vlanes() { return v_float64::nlanes; }
-        enum { nlanes = 2, max_nlanes = nlanes };
-        using lane_type = double;
-    };
-    #endif
-#endif
 
     #define OPENCV_HAL_WRAP_BIN_OP_ADDSUB(_Tpvec) \
     inline _Tpvec v_add(const _Tpvec& a, const _Tpvec& b) \
diff --git a/modules/core/include/opencv2/core/hal/intrin_cpp.hpp b/modules/core/include/opencv2/core/hal/intrin_cpp.hpp
index 46222140e6..633a58bca4 100644
--- a/modules/core/include/opencv2/core/hal/intrin_cpp.hpp
+++ b/modules/core/include/opencv2/core/hal/intrin_cpp.hpp
@@ -48,6 +48,7 @@
 #include <limits>
 #include <cstring>
 #include <algorithm>
+#include "opencv2/core/utility.hpp"
 #include "opencv2/core/saturate.hpp"
 
 //! @cond IGNORED
diff --git a/modules/core/include/opencv2/core/saturate.hpp b/modules/core/include/opencv2/core/saturate.hpp
index 8127e3d9ef..e0cc965ab6 100644
--- a/modules/core/include/opencv2/core/saturate.hpp
+++ b/modules/core/include/opencv2/core/saturate.hpp
@@ -46,6 +46,7 @@
 #define OPENCV_CORE_SATURATE_HPP
 
 #include "opencv2/core/cvdef.h"
+#include <climits>
 #include "opencv2/core/fast_math.hpp"
 
 namespace cv
diff --git a/modules/core/test/test_intrin_utils.hpp b/modules/core/test/test_intrin_utils.hpp
index 5d4442f111..7477a004ef 100644
--- a/modules/core/test/test_intrin_utils.hpp
+++ b/modules/core/test/test_intrin_utils.hpp
@@ -25,61 +25,7 @@ void test_hal_intrin_float16();
 template <typename R> struct Data;
 template <int N> struct initializer;
 
-#if CV_SIMD
-template <> struct initializer<64>
-{
-    template <typename R> static R init(const Data<R> & d)
-    {
-        return R(d[0], d[1], d[2], d[3], d[4], d[5], d[6], d[7], d[8], d[9], d[10], d[11], d[12], d[13], d[14], d[15],
-        d[16], d[17], d[18], d[19], d[20], d[21], d[22], d[23], d[24], d[25], d[26], d[27], d[28], d[29], d[30], d[31],
-        d[32], d[33], d[34], d[35], d[36], d[37], d[38], d[39], d[40], d[41], d[42], d[43], d[44], d[45], d[46], d[47],
-        d[48], d[49], d[50], d[51], d[52], d[53], d[54], d[55], d[56], d[57], d[58], d[59], d[60], d[61], d[62], d[63]);
-    }
-};
-
-template <> struct initializer<32>
-{
-    template <typename R> static R init(const Data<R> & d)
-    {
-        return R(d[0], d[1], d[2], d[3], d[4], d[5], d[6], d[7], d[8], d[9], d[10], d[11], d[12], d[13], d[14], d[15],
-        d[16], d[17], d[18], d[19], d[20], d[21], d[22], d[23], d[24], d[25], d[26], d[27], d[28], d[29], d[30], d[31]);
-    }
-};
-
-template <> struct initializer<16>
-{
-    template <typename R> static R init(const Data<R> & d)
-    {
-        return R(d[0], d[1], d[2], d[3], d[4], d[5], d[6], d[7], d[8], d[9], d[10], d[11], d[12], d[13], d[14], d[15]);
-    }
-};
-
-template <> struct initializer<8>
-{
-    template <typename R> static R init(const Data<R> & d)
-    {
-        return R(d[0], d[1], d[2], d[3], d[4], d[5], d[6], d[7]);
-    }
-};
-
-template <> struct initializer<4>
-{
-    template <typename R> static R init(const Data<R> & d)
-    {
-        return R(d[0], d[1], d[2], d[3]);
-    }
-};
-
-template <> struct initializer<2>
-{
-    template <typename R> static R init(const Data<R> & d)
-    {
-        return R(d[0], d[1]);
-    }
-};
-
-#else
-
+#if CV_SIMD_SCALABLE
 template <> struct initializer<128>
 {
     template <typename R> static R init(const Data<R> & d)
@@ -146,6 +92,59 @@ template <> struct initializer<2>
         return v_load({d[0], d[1]});
     }
 };
+
+#else
+template <> struct initializer<64>
+{
+    template <typename R> static R init(const Data<R> & d)
+    {
+        return R(d[0], d[1], d[2], d[3], d[4], d[5], d[6], d[7], d[8], d[9], d[10], d[11], d[12], d[13], d[14], d[15],
+        d[16], d[17], d[18], d[19], d[20], d[21], d[22], d[23], d[24], d[25], d[26], d[27], d[28], d[29], d[30], d[31],
+        d[32], d[33], d[34], d[35], d[36], d[37], d[38], d[39], d[40], d[41], d[42], d[43], d[44], d[45], d[46], d[47],
+        d[48], d[49], d[50], d[51], d[52], d[53], d[54], d[55], d[56], d[57], d[58], d[59], d[60], d[61], d[62], d[63]);
+    }
+};
+
+template <> struct initializer<32>
+{
+    template <typename R> static R init(const Data<R> & d)
+    {
+        return R(d[0], d[1], d[2], d[3], d[4], d[5], d[6], d[7], d[8], d[9], d[10], d[11], d[12], d[13], d[14], d[15],
+        d[16], d[17], d[18], d[19], d[20], d[21], d[22], d[23], d[24], d[25], d[26], d[27], d[28], d[29], d[30], d[31]);
+    }
+};
+
+template <> struct initializer<16>
+{
+    template <typename R> static R init(const Data<R> & d)
+    {
+        return R(d[0], d[1], d[2], d[3], d[4], d[5], d[6], d[7], d[8], d[9], d[10], d[11], d[12], d[13], d[14], d[15]);
+    }
+};
+
+template <> struct initializer<8>
+{
+    template <typename R> static R init(const Data<R> & d)
+    {
+        return R(d[0], d[1], d[2], d[3], d[4], d[5], d[6], d[7]);
+    }
+};
+
+template <> struct initializer<4>
+{
+    template <typename R> static R init(const Data<R> & d)
+    {
+        return R(d[0], d[1], d[2], d[3]);
+    }
+};
+
+template <> struct initializer<2>
+{
+    template <typename R> static R init(const Data<R> & d)
+    {
+        return R(d[0], d[1]);
+    }
+};
 #endif
 //==================================================================================================
 
@@ -1726,8 +1725,122 @@ template<typename R> struct TheTest
 #endif
 };
 
+#if CV_SIMD_SCALABLE //Temporary
+#define DUMP_ENTRY(type) printf("SIMD: %s\n", CV__TRACE_FUNCTION);
+
+
+//=============  8-bit integer =====================================================================
+
+void test_hal_intrin_uint8()
+{
+    DUMP_ENTRY(v_uint8);
+    // typedef v_uint8 R;
+    TheTest<v_uint8>()
+        .test_loadstore()
+        .test_min_max()
+        ;
+}
+
+void test_hal_intrin_int8()
+{
+    DUMP_ENTRY(v_int8);
+    // typedef v_int8 R;
+    TheTest<v_int8>()
+        .test_loadstore()
+        .test_min_max()
+        ;
+}
+
+//============= 16-bit integer =====================================================================
+
+void test_hal_intrin_uint16()
+{
+    DUMP_ENTRY(v_uint16);
+    // typedef v_uint16 R;
+    TheTest<v_uint16>()
+        .test_loadstore()
+        .test_min_max()
+        ;
+}
+
+void test_hal_intrin_int16()
+{
+    DUMP_ENTRY(v_int16);
+    // typedef v_int16 R;
+    TheTest<v_int16>()
+        .test_loadstore()
+        .test_min_max()
+        ;
+}
+
+//============= 32-bit integer =====================================================================
+
+void test_hal_intrin_uint32()
+{
+    DUMP_ENTRY(v_uint32);
+    // typedef v_uint32 R;
+    TheTest<v_uint32>()
+        .test_loadstore()
+        .test_min_max()
+        ;
+}
+
+void test_hal_intrin_int32()
+{
+    DUMP_ENTRY(v_int32);
+    // typedef v_int32 R;
+    TheTest<v_int32>()
+        .test_loadstore()
+        .test_min_max()
+        ;
+}
+
+//============= 64-bit integer =====================================================================
+
+void test_hal_intrin_uint64()
+{
+    DUMP_ENTRY(v_uint64);
+    // typedef v_uint64 R;
+    TheTest<v_uint64>()
+        .test_loadstore()
+        ;
+}
+
+void test_hal_intrin_int64()
+{
+    DUMP_ENTRY(v_int64);
+    // typedef v_int64 R;
+    TheTest<v_int64>()
+        .test_loadstore()
+        ;
+}
+
+//============= Floating point =====================================================================
+void test_hal_intrin_float32()
+{
+    DUMP_ENTRY(v_float32);
+    // typedef v_float32 R;
+    TheTest<v_float32>()
+        .test_loadstore()
+        .test_min_max()
+        ;
+}
+
+void test_hal_intrin_float64()
+{
+    DUMP_ENTRY(v_float64);
+#if CV_SIMD_64F
+    // typedef v_float64 R;
+    TheTest<v_float64>()
+        .test_loadstore()
+        .test_min_max()
+        ;
+
+#endif
+}
+
+#else
 
-#if CV_SIMD
 #define DUMP_ENTRY(type) printf("SIMD%d: %s\n", 8*(int)sizeof(v_uint8), CV__TRACE_FUNCTION);
 //=============  8-bit integer =====================================================================
 
@@ -2075,119 +2188,6 @@ void test_hal_intrin_float16()
     std::cout << "SKIP: CV_FP16 is not available" << std::endl;
 #endif
 }
-#elif CV_SIMD_SCALABLE //Temporary
-#define DUMP_ENTRY(type) printf("SIMD: %s\n", CV__TRACE_FUNCTION);
-
-
-//=============  8-bit integer =====================================================================
-
-void test_hal_intrin_uint8()
-{
-    DUMP_ENTRY(v_uint8);
-    // typedef v_uint8 R;
-    TheTest<v_uint8>()
-        .test_loadstore()
-        .test_min_max()
-        ;
-}
-
-void test_hal_intrin_int8()
-{
-    DUMP_ENTRY(v_int8);
-    // typedef v_int8 R;
-    TheTest<v_int8>()
-        .test_loadstore()
-        .test_min_max()
-        ;
-}
-
-//============= 16-bit integer =====================================================================
-
-void test_hal_intrin_uint16()
-{
-    DUMP_ENTRY(v_uint16);
-    // typedef v_uint16 R;
-    TheTest<v_uint16>()
-        .test_loadstore()
-        .test_min_max()
-        ;
-}
-
-void test_hal_intrin_int16()
-{
-    DUMP_ENTRY(v_int16);
-    // typedef v_int16 R;
-    TheTest<v_int16>()
-        .test_loadstore()
-        .test_min_max()
-        ;
-}
-
-//============= 32-bit integer =====================================================================
-
-void test_hal_intrin_uint32()
-{
-    DUMP_ENTRY(v_uint32);
-    // typedef v_uint32 R;
-    TheTest<v_uint32>()
-        .test_loadstore()
-        .test_min_max()
-        ;
-}
-
-void test_hal_intrin_int32()
-{
-    DUMP_ENTRY(v_int32);
-    // typedef v_int32 R;
-    TheTest<v_int32>()
-        .test_loadstore()
-        .test_min_max()
-        ;
-}
-
-//============= 64-bit integer =====================================================================
-
-void test_hal_intrin_uint64()
-{
-    DUMP_ENTRY(v_uint64);
-    // typedef v_uint64 R;
-    TheTest<v_uint64>()
-        .test_loadstore()
-        ;
-}
-
-void test_hal_intrin_int64()
-{
-    DUMP_ENTRY(v_int64);
-    // typedef v_int64 R;
-    TheTest<v_int64>()
-        .test_loadstore()
-        ;
-}
-
-//============= Floating point =====================================================================
-void test_hal_intrin_float32()
-{
-    DUMP_ENTRY(v_float32);
-    // typedef v_float32 R;
-    TheTest<v_float32>()
-        .test_loadstore()
-        .test_min_max()
-        ;
-}
-
-void test_hal_intrin_float64()
-{
-    DUMP_ENTRY(v_float64);
-#if CV_SIMD_64F
-    // typedef v_float64 R;
-    TheTest<v_float64>()
-        .test_loadstore()
-        .test_min_max()
-        ;
-
-#endif
-}
 
 #endif
 
diff --git a/modules/dnn/src/layers/fast_convolution/depthwise_convolution.cpp b/modules/dnn/src/layers/fast_convolution/depthwise_convolution.cpp
index c98c3d6549..4eb47c46b2 100644
--- a/modules/dnn/src/layers/fast_convolution/depthwise_convolution.cpp
+++ b/modules/dnn/src/layers/fast_convolution/depthwise_convolution.cpp
@@ -19,7 +19,7 @@ static void depthWiseBlock(const float *inptr, float *outptr, const float *weigh
                            int dilation_y, int stride_x, int stride_y, int inner_xleft, int inner_xright, int inner_ytop,
                            int inner_ybottom, bool ifMinMaxAct, bool useSIMD, bool is3x3)
 {
-#ifdef CV_SIMD128
+#if CV_SIMD128
     v_float32x4 vminval = v_setall_f32(minval), vmaxval = v_setall_f32(maxval);
 
     v_float32x4 w0 = v_setall_f32(
@@ -44,7 +44,7 @@ static void depthWiseBlock(const float *inptr, float *outptr, const float *weigh
     int dy0 = 1;
     for (int y0 = 0; y0 < H0; y0 += dy0, outptr += W0 * dy0)
     {
-#ifdef CV_SIMD128
+#if CV_SIMD128
         dy0 = inner_ytop <= y0 && y0 + 3 < inner_ybottom && is3x3 && stride_y == 1 && dilation_y == 1
               ? 3 : 1;
 #endif
@@ -103,7 +103,7 @@ static void depthWiseBlock(const float *inptr, float *outptr, const float *weigh
             if (x0 == W0)
                 break;
             x1 = inner_xright;
-#ifdef CV_SIMD128
+#if CV_SIMD128
             if (useSIMD)
             {
                 if (is3x3)