Enable dynamic dispatching for Merge3

modules/gapi/src/backends/fluid/gfluidcore.cpp

@@ -2509,10 +2509,10 @@ GAPI_FLUID_KERNEL(GFluidSplit3, cv::gapi::core::GSplit3, false)
 
     static void run(const View &src, Buffer &dst1, Buffer &dst2, Buffer &dst3)
     {
-        const auto *in = src.InLine<uchar>(0);
-        auto *out1     = dst1.OutLine<uchar>();
-        auto *out2     = dst2.OutLine<uchar>();
-        auto *out3     = dst3.OutLine<uchar>();
+        const auto *in   = src.InLine<uchar>(0);
+              auto *out1 = dst1.OutLine<uchar>();
+              auto *out2 = dst2.OutLine<uchar>();
+              auto *out3 = dst3.OutLine<uchar>();
 
         GAPI_Assert(3 == src.meta().chan);
         int width = src.length();
@@ -2537,11 +2537,11 @@ GAPI_FLUID_KERNEL(GFluidSplit4, cv::gapi::core::GSplit4, false)
 
     static void run(const View &src, Buffer &dst1, Buffer &dst2, Buffer &dst3, Buffer &dst4)
     {
-        const auto *in = src.InLine<uchar>(0);
-        auto *out1     = dst1.OutLine<uchar>();
-        auto *out2     = dst2.OutLine<uchar>();
-        auto *out3     = dst3.OutLine<uchar>();
-        auto *out4     = dst4.OutLine<uchar>();
+        const auto *in   = src.InLine<uchar>(0);
+              auto *out1 = dst1.OutLine<uchar>();
+              auto *out2 = dst2.OutLine<uchar>();
+              auto *out3 = dst3.OutLine<uchar>();
+              auto *out4 = dst4.OutLine<uchar>();
 
         GAPI_Assert(4 == src.meta().chan);
         int width = src.length();
@@ -2574,18 +2574,10 @@ GAPI_FLUID_KERNEL(GFluidMerge3, cv::gapi::core::GMerge3, false)
 
         GAPI_Assert(3 == dst.meta().chan);
         int width = dst.length();
+        int w = 0;
 
-        int w = 0; // cycle counter
-
-    #if CV_SIMD128
-        for (; w <= width-16; w+=16)
-        {
-            v_uint8x16 a, b, c;
-            a = v_load(&in1[w]);
-            b = v_load(&in2[w]);
-            c = v_load(&in3[w]);
-            v_store_interleave(&out[3*w], a, b, c);
-        }
+    #if CV_SIMD
+        w = merge3_simd(in1, in2, in3, out, width);
     #endif
 
         for (; w < width; w++)

modules/gapi/src/backends/fluid/gfluidcore_func.dispatch.cpp

@@ -221,6 +221,13 @@ int split4_simd(const uchar in[], uchar out1[], uchar out2[],
                     CV_CPU_DISPATCH_MODES_ALL);
 }
 
+int merge3_simd(const uchar in1[], const uchar in2[], const uchar in3[],
+                uchar out[], const int width)
+{
+    CV_CPU_DISPATCH(merge3_simd, (in1, in2, in3, out, width),
+                    CV_CPU_DISPATCH_MODES_ALL);
+}
+
 } // namespace fluid
 } // namespace gapi
 } // namespace cv

modules/gapi/src/backends/fluid/gfluidcore_func.hpp

@@ -169,6 +169,9 @@ int split3_simd(const uchar in[], uchar out1[], uchar out2[],
 int split4_simd(const uchar in[], uchar out1[], uchar out2[],
                 uchar out3[], uchar out4[], const int width);
 
+int merge3_simd(const uchar in1[], const uchar in2[], const uchar in3[],
+               uchar out[], const int width);
+
 }  // namespace fluid
 }  // namespace gapi
 }  // namespace cv

modules/gapi/src/backends/fluid/gfluidcore_func.simd.hpp

@@ -190,6 +190,9 @@ int split3_simd(const uchar in[], uchar out1[], uchar out2[],
 int split4_simd(const uchar in[], uchar out1[], uchar out2[],
                 uchar out3[], uchar out4[], const int width);
 
+int merge3_simd(const uchar in1[], const uchar in2[], const uchar in3[],
+                uchar out[], const int width);
+
 #ifndef CV_CPU_OPTIMIZATION_DECLARATIONS_ONLY
 
 struct scale_tag {};
@@ -1643,6 +1646,40 @@ int split4_simd(const uchar in[], uchar out1[], uchar out2[],
     return x;
 }
 
+//-------------------------
+//
+// Fluid kernels: Merge3
+//
+//-------------------------
+
+int merge3_simd(const uchar in1[], const uchar in2[], const uchar in3[],
+                uchar out[], const int width)
+{
+    constexpr int nlanes = v_uint8::nlanes;
+    if (width < nlanes)
+        return 0;
+
+    int x = 0;
+    for (;;)
+    {
+        for (; x <= width - nlanes; x += nlanes)
+        {
+            v_uint8 a, b, c;
+            a = vx_load(&in1[x]);
+            b = vx_load(&in2[x]);
+            c = vx_load(&in3[x]);
+            v_store_interleave(&out[3 * x], a, b, c);
+        }
+        if (x < width)
+        {
+            x = width - nlanes;
+            continue;
+        }
+        break;
+    }
+    return x;
+}
+
 #endif  // CV_CPU_OPTIMIZATION_DECLARATIONS_ONLY
 
 CV_CPU_OPTIMIZATION_NAMESPACE_END