COMPMID-1277 - Optimizing CLIm2ColKernel for NHWC.

This patch includes:

- Im2Col optimizations for NHWC using a new data layout
- Refactoring of CLIm2ColKernel adding validation method and auto-init
- Removed im2col_reduced from CLIm2ColKernel and created a new kernel CLFlattenLayerKernel

Change-Id: I1620640b6796baa268324b33ae92cdd8de53e27c
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/141241
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Giorgio Arena <giorgio.arena@arm.com>

1 files changed, 269 insertions, 260 deletions

diff --git a/src/core/CL/cl_kernels/im2col.cl b/src/core/CL/cl_kernels/im2col.cl
index d034b30b68..274ec20046 100644
--- a/src/core/CL/cl_kernels/im2col.cl
+++ b/src/core/CL/cl_kernels/im2col.cl
@@ -35,13 +35,12 @@
 #error "Element size not support"
 #endif // ELEMENT_SIZE
 
-#if defined(CONVOLVED_WIDTH) && defined(STRIDE_Y) && defined(KERNEL_DEPTH)
-/** This kernel performs a reshaping of the input tensor to a tensor used to perform convolution using GEMM when the kernel size is 1x1 and the stride_x = 1
+#if defined(CONVOLVED_WIDTH) && defined(STRIDE_Y) && defined(SRC_DEPTH)
+/** This opencl kernel performs im2col when the kernel size is 1x1, the stride_x = 1 and the data layout is NCHW
  *
- * @note This kernel computes 4 elements
  * @note The data type must be passed at compile time using -DDATA_TYPE: e.g. -DDATA_TYPE=float
  * @note The width of output tensor after matrix multiplication must be passed at compile time using -DCONVOLVED_WIDTH: e.g. -DCONVOLVED_WIDTH=34
- * @note The kernel depth must be passed at compile time using -DKERNEL_DEPTH: e.g. -DKERNEL_DEPTH=3
+ * @note The number of input channels must be passed at compile time using -DSRC_DEPTH: e.g. -DSRC_DEPTH=3
  * @note The stride along the Y direction must be passed at compile time using -DSTRIDE_Y: e.g. -DSTRIDE_Y=1
  * @note In case biases will be added to the convolution -DHAS_BIAS has to be passed to append the final matrix with 1 in each row.
  *
@@ -62,16 +61,16 @@
  * @param[in]  src_stride_w                      Stride of the source tensor in W dimension (in bytes).
  * @param[in]  dst_stride_w                      Stride of the destination tensor in W dimension (in bytes).
  */
-__kernel void im2col1x1_stridex1_dchw(
+__kernel void im2col1x1_stridex1_nchw(
     TENSOR3D_DECLARATION(src),
     IMAGE_DECLARATION(dst),
     uint src_stride_w,
     uint dst_stride_w)
 {
-    const uint xc    = get_global_id(0) * 4;            // x coordinate in the convolved tensor
-    const uint yc    = get_global_id(1);                // y coordinate in the convolved tensor
-    const uint ch    = get_global_id(2) % KERNEL_DEPTH; // input feature map
-    const uint batch = get_global_id(2) / KERNEL_DEPTH; // batch size
+    const uint xc    = get_global_id(0) * 4;         // x coordinate in the convolved tensor
+    const uint yc    = get_global_id(1);             // y coordinate in the convolved tensor
+    const uint ch    = get_global_id(2) % SRC_DEPTH; // input feature map
+    const uint batch = get_global_id(2) / SRC_DEPTH; // batch size
 
     // Clamp xc
     // The strategy clamps at "xc" as it will be a valid value for sure
@@ -107,7 +106,7 @@ __kernel void im2col1x1_stridex1_dchw(
     *(__global DATA_TYPE *)(output_ptr + yo.s3 * dst_stride_y) = data.s3;
 
 #ifdef HAS_BIAS
-    if(ch == (KERNEL_DEPTH - 1))
+    if(ch == (SRC_DEPTH - 1))
     {
         *((__global DATA_TYPE *)(output_ptr + yo.s0 * dst_stride_y) + 1) = 1.0f;
         *((__global DATA_TYPE *)(output_ptr + yo.s1 * dst_stride_y) + 1) = 1.0f;
@@ -116,18 +115,16 @@ __kernel void im2col1x1_stridex1_dchw(
     }
 #endif // HAS_BIAS
 }
-#endif // defined(CONVOLVED_WIDTH) && defined(STRIDE_Y) && defined(KERNEL_DEPTH)
+#endif // defined(CONVOLVED_WIDTH) && defined(STRIDE_Y) && defined(SRC_DEPTH)
 
-#define PTR_TO_VALUE(PTR, DATA_TYPE) *((__global DATA_TYPE *)(PTR))
-
-#if defined(CONVOLVED_WIDTH) && defined(SRC_WIDTH) && defined(SRC_HEIGHT) && defined(STRIDE_X) && defined(STRIDE_Y) && defined(KERNEL_DEPTH) && defined(PAD_LEFT) && defined(PAD_RIGHT) && defined(PAD_TOP) && defined(PAD_BOTTOM) && defined(PAD_VALUE)
-
-/** This kernel performs a reshaping of the input tensor to a tensor used to perform convolution using GEMM when the kernel size is 5x5
+#if defined(CONVOLVED_WIDTH) && defined(SRC_WIDTH) && defined(SRC_HEIGHT) && defined(STRIDE_X) && defined(STRIDE_Y) && defined(SRC_DEPTH) && defined(PAD_LEFT) && defined(PAD_RIGHT) && defined(PAD_TOP) && defined(PAD_BOTTOM) && defined(PAD_VALUE)
+#if defined(DILATION_X) && defined(DILATION_Y)
+/** This opencl kernel performs a generic im2col implementation when the data layout is NCHW
  *
  * @note The data type must be passed at compile time using -DDATA_TYPE: e.g. -DDATA_TYPE=float
  * @note The width and height of the input tensor must be passed at compile time using -DSRC_WIDTH and -DSRC_HEIGHT: e.g. -DSRC_WIDTH=128 and -DSRC_HEIGHT=128
  * @note The width of output tensor after matrix multiplication must be passed at compile time using -DCONVOLVED_WIDTH: e.g. -DCONVOLVED_WIDTH=34
- * @note The kernel depth must be passed at compile time using -DKERNEL_DEPTH: e.g. -DKERNEL_DEPTH=3
+ * @note The kernel width, height and depth must be passed at compile time using -DKERNEL_WIDTH, -DKERNEL_HEIGHT and -DSRC_DEPTH: e.g. -DKERNEL_WIDTH=3, -DKERNEL_HEIGHT=3 and -DSRC_DEPTH=64
  * @note The pad_left, pad_right, pad_top and pad_bottom must be passed at compile time using -DPAD_LEFT, -DPAD_RIGHT, -DPAD_TOP and -DPAD_BOTTOM: e.g. -DPAD_LEFT=1, -DPAD_RIGHT=2, -DPAD_TOP=3 and -DPAD_BOTTOM=2
  * @note The zero value to store in case we load values out-of-bounds must be passed at compile time using -DPAD_VALUE: e.g. -DPAD_VALUE=0.0
  * @note The stride along the X and Y directions must be passed at compile time using -DSTRIDE_X and -DSTRIDE_Y: e.g. -DSTRIDE_X=1 and -DSTRIDE_Y=1
@@ -151,183 +148,65 @@ __kernel void im2col1x1_stridex1_dchw(
  * @param[in]  src_stride_w                      Stride of the source tensor in W dimension (in bytes).
  * @param[in]  dst_stride_w                      Stride of the destination tensor in W dimension (in bytes).
  */
-__kernel void im2col_generic_nhwc(
+__kernel void im2col_generic_nchw(
     TENSOR3D_DECLARATION(src),
     IMAGE_DECLARATION(dst),
     uint src_stride_w,
     uint dst_stride_w)
 {
-    const int src_stride_y_int = (int)src_stride_y;
-    const int src_stride_z_int = (int)src_stride_z;
-    const int xc               = get_global_id(1);                    // x coordinate in the convolved tensor
-    const int yc               = get_global_id(2) % CONVOLVED_HEIGHT; // y coordinate in the convolved tensor
-    const int ch               = get_global_id(0);                    // input feature map
-    const int batch            = get_global_id(2) / CONVOLVED_HEIGHT; // batch size
+    const int xc    = get_global_id(0);             // x coordinate in the convolved tensor
+    const int yc    = get_global_id(1);             // y coordinate in the convolved tensor
+    const int ch    = get_global_id(2) % SRC_DEPTH; // input feature map
+    const int batch = get_global_id(2) / SRC_DEPTH; // batch size
 
     // Calculate input indices
     const int xi = xc * STRIDE_X - PAD_LEFT;
     const int yi = yc * STRIDE_Y - PAD_TOP;
 
     // Calculate output indices
-    const int xo = ch * KERNEL_HEIGHT * KERNEL_WIDTH;
+    const int xo = ch * KERNEL_WIDTH * KERNEL_HEIGHT;
     const int yo = xc + yc * CONVOLVED_WIDTH; // Index of the convolution
 
-    // Get input and output address
-    __global uchar *input_ptr  = src_ptr + src_offset_first_element_in_bytes + xi * src_stride_y_int + yi * src_stride_z_int + ch * src_stride_x + batch * src_stride_w;
-    __global uchar *output_ptr = dst_ptr + dst_offset_first_element_in_bytes + xo * dst_stride_x + yo * dst_stride_y + batch * dst_stride_w;
+    __global uchar *input_ptr      = src_ptr + src_offset_first_element_in_bytes + ch * src_stride_z + batch * src_stride_w;
+    __global DATA_TYPE *output_ptr = ((__global DATA_TYPE *)(dst_ptr + dst_offset_first_element_in_bytes + yo * dst_stride_y + batch * dst_stride_w)) + xo;
 
+    // Linearize convolution elements
     for(int yk = 0; yk < KERNEL_HEIGHT; ++yk)
     {
-        const int dilated_offset_y = yk * DILATION_Y;
-        const int y0               = yi + dilated_offset_y;
-        if(y0 >= 0 && y0 < SRC_HEIGHT)
+        int y = yi + yk * DILATION_Y;
+        for(int xk = 0; xk < KERNEL_WIDTH; ++xk, ++output_ptr)
         {
-            int xk;
-            for(xk = 0; xk < KERNEL_WIDTH; xk++)
+            int x = xi + xk * DILATION_X;
+#if PAD_LEFT == 0 && PAD_TOP == 0 && PAD_RIGHT == 0 && PAD_BOTTOM == 0
+            *output_ptr = *((__global DATA_TYPE *)(input_ptr + x * src_stride_x + y * src_stride_y));
+#else  // PAD_LEFT == 0 && PAD_TOP == 0 && PAD_RIGHT == 0 && PAD_BOTTOM == 0
+            if(x < 0 || x >= SRC_WIDTH || y < 0 || y >= SRC_HEIGHT)
             {
-                const int dilated_offset_x = xk * DILATION_X;
-                const int x0               = xi + dilated_offset_x;
-                if(x0 >= 0 && x0 < SRC_WIDTH)
-                {
-                    *((__global DATA_TYPE *)output_ptr) = PTR_TO_VALUE(input_ptr + dilated_offset_x * src_stride_y + dilated_offset_y * src_stride_z, DATA_TYPE);
-                }
-                else
-                {
-                    *((__global DATA_TYPE *)output_ptr) = PAD_VALUE;
-                }
-                output_ptr += 1 * sizeof(DATA_TYPE);
+                *output_ptr = PAD_VALUE;
             }
-        }
-        else
-        {
-            for(int xk = 0; xk < KERNEL_WIDTH; xk++)
+            else
             {
-                *((__global DATA_TYPE *)output_ptr) = (DATA_TYPE)PAD_VALUE;
-                output_ptr += 1 * dst_stride_x;
+                *output_ptr = *((__global DATA_TYPE *)(input_ptr + x * src_stride_x + y * src_stride_y));
             }
+#endif // PAD_LEFT == 0 && PAD_TOP == 0 && PAD_RIGHT == 0 && PAD_BOTTOM == 0
         }
     }
-#ifdef HAS_BIAS
-    if(ch == (KERNEL_DEPTH - 1))
-    {
-        *((__global DATA_TYPE *)output_ptr) = 1.0f;
-        output_ptr += 1 * dst_stride_x;
-    }
-#endif // HAS_BIAS
-}
-
-/** This kernel performs a reshaping of the input tensor (with layout NHWC) to a tensor used to perform convolution using GEMM when the kernel size is 3x3
- *
- * @note The data type must be passed at compile time using -DDATA_TYPE: e.g. -DDATA_TYPE=float
- * @note The width and height of the input tensor must be passed at compile time using -DSRC_WIDTH and -DSRC_HEIGHT: e.g. -DSRC_WIDTH=128 and -DSRC_HEIGHT=128
- * @note The width of output tensor after matrix multiplication must be passed at compile time using -DCONVOLVED_WIDTH: e.g. -DCONVOLVED_WIDTH=34
- * @note The kernel depth must be passed at compile time using -DKERNEL_DEPTH: e.g. -DKERNEL_DEPTH=3
- * @note The pad_left, pad_right, pad_top and pad_bottom must be passed at compile time using -DPAD_LEFT, -DPAD_RIGHT, -DPAD_TOP and -DPAD_BOTTOM: e.g. -DPAD_LEFT=1, -DPAD_RIGHT=2, -DPAD_TOP=3 and -DPAD_BOTTOM=2
- * @note The zero value to store in case we load values out-of-bounds must be passed at compile time using -DPAD_VALUE: e.g. -DPAD_VALUE=0.0
- * @note The stride along the X and Y directions must be passed at compile time using -DSTRIDE_X and -DSTRIDE_Y: e.g. -DSTRIDE_X=1 and -DSTRIDE_Y=1
- * @note In case biases will be added to the convolution -DHAS_BIAS has to be passed to append the final matrix with 1 in each row.
- *
- * @param[in]  src_ptr                           Pointer to the source tensor. Supported data types: QASYMM8/F16/F32
- * @param[in]  src_stride_x                      Stride of the source tensor in X dimension (in bytes)
- * @param[in]  src_step_x                        src_stride_x * number of elements along X processed per workitem(in bytes)
- * @param[in]  src_stride_y                      Stride of the source tensor in Y dimension (in bytes)
- * @param[in]  src_step_y                        src_stride_y * number of elements along Y processed per workitem(in bytes)
- * @param[in]  src_stride_z                      Stride of the source tensor in Z dimension (in bytes)
- * @param[in]  src_step_z                        src_stride_z * number of elements along Z processed per workitem(in bytes)
- * @param[in]  src_offset_first_element_in_bytes The offset of the first element in the source tensor
- * @param[out] dst_ptr                           Pointer to the destination tensor. Supported data types: same as @p src_ptr
- * @param[in]  dst_stride_x                      Stride of the destination tensor in X dimension (in bytes)
- * @param[in]  dst_step_x                        dst_stride_x * number of elements along X processed per workitem(in bytes)
- * @param[in]  dst_stride_y                      Stride of the destination tensor in Y dimension (in bytes)
- * @param[in]  dst_step_y                        dst_stride_y * number of elements along Y processed per workitem(in bytes)
- * @param[in]  dst_offset_first_element_in_bytes The offset of the first element in the destination tensor
- * @param[in]  src_stride_w                      Stride of the source tensor in W dimension (in bytes).
- * @param[in]  dst_stride_w                      Stride of the destination tensor in W dimension (in bytes).
- */
-__kernel void im2col3x3_nhwc(
-    TENSOR3D_DECLARATION(src),
-    IMAGE_DECLARATION(dst),
-    uint src_stride_w,
-    uint dst_stride_w)
-{
-    const int src_stride_y_int = (int)src_stride_y;
-    const int src_stride_z_int = (int)src_stride_z;
-    const int xc               = get_global_id(1);                    // x coordinate in the convolved tensor
-    const int yc               = get_global_id(2) % CONVOLVED_HEIGHT; // y coordinate in the convolved tensor
-    const int ch               = get_global_id(0);                    // input feature map
-    const int batch            = get_global_id(2) / CONVOLVED_HEIGHT; // batch size
-
-    // Calculate input indices
-    const int xi = xc * STRIDE_X - PAD_LEFT;
-    const int yi = yc * STRIDE_Y - PAD_TOP;
-
-    // Calculate output indices
-    const int xo = ch * 9;                    // 3x3
-    const int yo = xc + yc * CONVOLVED_WIDTH; // Index of the convolution
-
-    // Get input and output address
-    __global uchar *input_ptr  = src_ptr + src_offset_first_element_in_bytes + xi * src_stride_y_int + yi * src_stride_z_int + ch * src_stride_x + batch * src_stride_w;
-    __global uchar *output_ptr = dst_ptr + dst_offset_first_element_in_bytes + xo * dst_stride_x + yo * dst_stride_y + batch * dst_stride_w;
-
-    VEC_DATA_TYPE(DATA_TYPE, 3)
-    row0 = (VEC_DATA_TYPE(DATA_TYPE, 3))(PAD_VALUE);
-    VEC_DATA_TYPE(DATA_TYPE, 3)
-    row1 = (VEC_DATA_TYPE(DATA_TYPE, 3))(PAD_VALUE);
-    VEC_DATA_TYPE(DATA_TYPE, 3)
-    row2 = (VEC_DATA_TYPE(DATA_TYPE, 3))(PAD_VALUE);
-
-    const int3 y = (int3)yi + (int3)(0, 1, 2);
-    // Guard against reading outside the input buffer, there is no padding in Z so we check if ry is inside the buffer.
-    if(y.s0 >= 0 && y.s0 < SRC_HEIGHT)
-    {
-        row0 = (VEC_DATA_TYPE(DATA_TYPE, 3))(
-                   PTR_TO_VALUE(input_ptr + 0 * src_stride_y, DATA_TYPE),
-                   PTR_TO_VALUE(input_ptr + 1 * src_stride_y, DATA_TYPE),
-                   PTR_TO_VALUE(input_ptr + 2 * src_stride_y, DATA_TYPE));
-    }
-
-    if(y.s1 >= 0 && y.s1 < SRC_HEIGHT)
-    {
-        row1 = (VEC_DATA_TYPE(DATA_TYPE, 3))(
-                   PTR_TO_VALUE(input_ptr + 0 * src_stride_y + 1 * src_stride_z, DATA_TYPE),
-                   PTR_TO_VALUE(input_ptr + 1 * src_stride_y + 1 * src_stride_z, DATA_TYPE),
-                   PTR_TO_VALUE(input_ptr + 2 * src_stride_y + 1 * src_stride_z, DATA_TYPE));
-    }
-
-    if(y.s2 >= 0 && y.s2 < SRC_HEIGHT)
-    {
-        row2 = (VEC_DATA_TYPE(DATA_TYPE, 3))(
-                   PTR_TO_VALUE(input_ptr + 0 * src_stride_y + 2 * src_stride_z, DATA_TYPE),
-                   PTR_TO_VALUE(input_ptr + 1 * src_stride_y + 2 * src_stride_z, DATA_TYPE),
-                   PTR_TO_VALUE(input_ptr + 2 * src_stride_y + 2 * src_stride_z, DATA_TYPE));
-    }
-
-#if PAD_LEFT != 0 || PAD_TOP != 0 || PAD_RIGHT != 0 || PAD_BOTTOM != 0
-    // Put 0 if the value is out-of-bound
-    const int3 x = (int3)xi + (int3)(0, 1, 2);
-    VEC_DATA_TYPE(COND_DATA_TYPE, 3)
-    cond0 = CONVERT((x >= (int3)0 && x < (int3)SRC_WIDTH), VEC_DATA_TYPE(COND_DATA_TYPE, 3));
-    row0  = select((VEC_DATA_TYPE(DATA_TYPE, 3))PAD_VALUE, row0, cond0);
-    row1  = select((VEC_DATA_TYPE(DATA_TYPE, 3))PAD_VALUE, row1, cond0);
-    row2  = select((VEC_DATA_TYPE(DATA_TYPE, 3))PAD_VALUE, row2, cond0);
-#endif // PAD_LEFT != 0 || PAD_TOP != 0 || PAD_RIGHT != 0 || PAD_BOTTOM != 0
-    vstore8((VEC_DATA_TYPE(DATA_TYPE, 8))(row0.s012, row1.s012, row2.s01), 0, (__global DATA_TYPE *)output_ptr);
-    *((__global DATA_TYPE *)output_ptr + 8) = row2.s2;
 
 #ifdef HAS_BIAS
-    if(ch == (KERNEL_DEPTH - 1))
+    if(ch == (SRC_DEPTH - 1))
     {
-        *((__global DATA_TYPE *)output_ptr + 9) = 1.0f;
+        *output_ptr = 1.0f;
     }
 #endif // HAS_BIAS
 }
+#endif // defined(DILATION_X) && defined(DILATION_Y)
 
-/** This kernel performs a reshaping of the input tensor to a tensor used to perform convolution using GEMM when the kernel size is 3x3
+/** This opencl kernel performs im2col when the kernel size is 3x3 and the data layout is NCHW
  *
  * @note The data type must be passed at compile time using -DDATA_TYPE: e.g. -DDATA_TYPE=float
  * @note The width and height of the input tensor must be passed at compile time using -DSRC_WIDTH and -DSRC_HEIGHT: e.g. -DSRC_WIDTH=128 and -DSRC_HEIGHT=128
  * @note The width of output tensor after matrix multiplication must be passed at compile time using -DCONVOLVED_WIDTH: e.g. -DCONVOLVED_WIDTH=34
- * @note The kernel depth must be passed at compile time using -DKERNEL_DEPTH: e.g. -DKERNEL_DEPTH=3
+ * @note The number of input channels must be passed at compile time using -DSRC_DEPTH: e.g. -DSRC_DEPTH=3
  * @note The pad_left, pad_right, pad_top and pad_bottom must be passed at compile time using -DPAD_LEFT, -DPAD_RIGHT, -DPAD_TOP and -DPAD_BOTTOM: e.g. -DPAD_LEFT=1, -DPAD_RIGHT=2, -DPAD_TOP=3 and -DPAD_BOTTOM=2
  * @note The zero value to store in case we load values out-of-bounds must be passed at compile time using -DPAD_VALUE: e.g. -DPAD_VALUE=0.0
  * @note The stride along the X and Y directions must be passed at compile time using -DSTRIDE_X and -DSTRIDE_Y: e.g. -DSTRIDE_X=1 and -DSTRIDE_Y=1
@@ -350,16 +229,16 @@ __kernel void im2col3x3_nhwc(
  * @param[in]  src_stride_w                      Stride of the source tensor in W dimension (in bytes).
  * @param[in]  dst_stride_w                      Stride of the destination tensor in W dimension (in bytes).
  */
-__kernel void im2col3x3_dchw(
+__kernel void im2col3x3_nchw(
     TENSOR3D_DECLARATION(src),
     IMAGE_DECLARATION(dst),
     uint src_stride_w,
     uint dst_stride_w)
 {
-    const int xc    = get_global_id(0);                // x coordinate in the convolved tensor
-    const int yc    = get_global_id(1);                // y coordinate in the convolved tensor
-    const int ch    = get_global_id(2) % KERNEL_DEPTH; // input feature map
-    const int batch = get_global_id(2) / KERNEL_DEPTH; // batch size
+    const int xc    = get_global_id(0);             // x coordinate in the convolved tensor
+    const int yc    = get_global_id(1);             // y coordinate in the convolved tensor
+    const int ch    = get_global_id(2) % SRC_DEPTH; // input feature map
+    const int batch = get_global_id(2) / SRC_DEPTH; // batch size
 
     // Calculate input indices
     const int xi = xc * STRIDE_X - PAD_LEFT;
@@ -402,19 +281,19 @@ __kernel void im2col3x3_dchw(
     *((__global DATA_TYPE *)output_ptr + 8) = row2.s2;
 
 #ifdef HAS_BIAS
-    if(ch == (KERNEL_DEPTH - 1))
+    if(ch == (SRC_DEPTH - 1))
     {
         *((__global DATA_TYPE *)output_ptr + 9) = 1.0f;
     }
 #endif // HAS_BIAS
 }
 
-/** This kernel performs a reshaping of the input tensor to a tensor used to perform convolution using GEMM when the kernel size is 5x5
+/** This opencl kernel performs im2col when the kernel size is 5x5 and the data layout is NCHW
  *
  * @note The data type must be passed at compile time using -DDATA_TYPE: e.g. -DDATA_TYPE=float
  * @note The width and height of the input tensor must be passed at compile time using -DSRC_WIDTH and -DSRC_HEIGHT: e.g. -DSRC_WIDTH=128 and -DSRC_HEIGHT=128
  * @note The width of output tensor after matrix multiplication must be passed at compile time using -DCONVOLVED_WIDTH: e.g. -DCONVOLVED_WIDTH=34
- * @note The kernel depth must be passed at compile time using -DKERNEL_DEPTH: e.g. -DKERNEL_DEPTH=3
+ * @note The number of input channels must be passed at compile time using -DSRC_DEPTH: e.g. -DSRC_DEPTH=3
  * @note The pad_left, pad_right, pad_top and pad_bottom must be passed at compile time using -DPAD_LEFT, -DPAD_RIGHT, -DPAD_TOP and -DPAD_BOTTOM: e.g. -DPAD_LEFT=1, -DPAD_RIGHT=2, -DPAD_TOP=3 and -DPAD_BOTTOM=2
  * @note The zero value to store in case we load values out-of-bounds must be passed at compile time using -DPAD_VALUE: e.g. -DPAD_VALUE=0.0
  * @note The stride along the X and Y directions must be passed at compile time using -DSTRIDE_X and -DSTRIDE_Y: e.g. -DSTRIDE_X=1 and -DSTRIDE_Y=1
@@ -437,16 +316,16 @@ __kernel void im2col3x3_dchw(
  * @param[in]  src_stride_w                      Stride of the source tensor in W dimension (in bytes).
  * @param[in]  dst_stride_w                      Stride of the destination tensor in W dimension (in bytes).
  */
-__kernel void im2col5x5_dchw(
+__kernel void im2col5x5_nchw(
     TENSOR3D_DECLARATION(src),
     IMAGE_DECLARATION(dst),
     uint src_stride_w,
     uint dst_stride_w)
 {
-    const int xc    = get_global_id(0);                // x coordinate in the convolved tensor
-    const int yc    = get_global_id(1);                // y coordinate in the convolved tensor
-    const int ch    = get_global_id(2) % KERNEL_DEPTH; // input feature map
-    const int batch = get_global_id(2) / KERNEL_DEPTH; // batch size
+    const int xc    = get_global_id(0);             // x coordinate in the convolved tensor
+    const int yc    = get_global_id(1);             // y coordinate in the convolved tensor
+    const int ch    = get_global_id(2) % SRC_DEPTH; // input feature map
+    const int batch = get_global_id(2) / SRC_DEPTH; // batch size
 
     // Calculate input indices
     const int xi = xc * STRIDE_X - PAD_LEFT;
@@ -576,20 +455,20 @@ __kernel void im2col5x5_dchw(
     }
 
 #ifdef HAS_BIAS
-    if(ch == (KERNEL_DEPTH - 1))
+    if(ch == (SRC_DEPTH - 1))
     {
         *((__global DATA_TYPE *)output_ptr) = 1.0f;
     }
 #endif // HAS_BIAS
 }
-#endif // defined(CONVOLVED_WIDTH) && defined(SRC_WIDTH) && defined(SRC_HEIGHT) && defined(STRIDE_X) && defined(STRIDE_Y) && defined(KERNEL_DEPTH) && defined(PAD_LEFT) && defined(PAD_RIGHT) && defined(PAD_TOP) && defined(PAD_BOTTOM) && defined(PAD_VALUE)
+#endif // defined(CONVOLVED_WIDTH) && defined(SRC_WIDTH) && defined(SRC_HEIGHT) && defined(STRIDE_X) && defined(STRIDE_Y) && defined(SRC_DEPTH) && defined(PAD_LEFT) && defined(PAD_RIGHT) && defined(PAD_TOP) && defined(PAD_BOTTOM) && defined(PAD_VALUE)
 
-#if defined(CONVOLVED_WIDTH) && defined(STRIDE_X) && defined(STRIDE_Y) && defined(KERNEL_DEPTH)
-/** This kernel performs a reshaping of the input tensor to a tensor used to perform convolution using GEMM when the kernel size is 11x11
+#if defined(CONVOLVED_WIDTH) && defined(STRIDE_X) && defined(STRIDE_Y) && defined(SRC_DEPTH)
+/** This opencl kernel performs im2col when the kernel size is 11x11, we do not have paddings and the data layout is NCHW
  *
  * @note The data type must be passed at compile time using -DDATA_TYPE: e.g. -DDATA_TYPE=float
  * @note The width of output tensor after matrix multiplication must be passed at compile time using -DCONVOLVED_WIDTH: e.g. -DCONVOLVED_WIDTH=34
- * @note The kernel depth must be passed at compile time using -DKERNEL_DEPTH: e.g. -DKERNEL_DEPTH=3
+ * @note The number of input channels must be passed at compile time using -DSRC_DEPTH: e.g. -DSRC_DEPTH=3
  * @note The stride along the X and Y directions must be passed at compile time using -DSTRIDE_X and -DSTRIDE_Y: e.g. -DSTRIDE_X=1 and -DSTRIDE_Y=1
  * @note In case biases will be added to the convolution -DHAS_BIAS has to be passed to append the final matrix with 1 in each row.
  *
@@ -610,16 +489,16 @@ __kernel void im2col5x5_dchw(
  * @param[in]  src_stride_w                      Stride of the source tensor in W dimension (in bytes).
  * @param[in]  dst_stride_w                      Stride of the destination tensor in W dimension (in bytes).
  */
-__kernel void im2col11x11_padx0_pady0_dchw(
+__kernel void im2col11x11_padx0_pady0_nchw(
     TENSOR3D_DECLARATION(src),
     IMAGE_DECLARATION(dst),
     uint src_stride_w,
     uint dst_stride_w)
 {
-    const int xc    = get_global_id(0);                // x coordinate in the convolved tensor
-    const int yc    = get_global_id(1);                // y coordinate in the convolved tensor
-    const int ch    = get_global_id(2) % KERNEL_DEPTH; // input feature map
-    const int batch = get_global_id(2) / KERNEL_DEPTH; // batch size
+    const int xc    = get_global_id(0);             // x coordinate in the convolved tensor
+    const int yc    = get_global_id(1);             // y coordinate in the convolved tensor
+    const int ch    = get_global_id(2) % SRC_DEPTH; // input feature map
+    const int batch = get_global_id(2) / SRC_DEPTH; // batch size
 
     // Calculate input indices
     const int xi = xc * STRIDE_X;
@@ -776,21 +655,21 @@ __kernel void im2col11x11_padx0_pady0_dchw(
     }
 
 #ifdef HAS_BIAS
-    if(ch == (KERNEL_DEPTH - 1))
+    if(ch == (SRC_DEPTH - 1))
     {
         *((__global DATA_TYPE *)output_ptr) = 1.0f;
     }
 #endif // HAS_BIAS
 }
-#endif // defined(CONVOLVED_WIDTH) && defined(STRIDE_X) && defined(STRIDE_Y) && defined(KERNEL_DEPTH)
+#endif // defined(CONVOLVED_WIDTH) && defined(STRIDE_X) && defined(STRIDE_Y) && defined(SRC_DEPTH)
 
-#if defined(CONVOLVED_WIDTH) && defined(STRIDE_X) && defined(STRIDE_Y) && defined(KERNEL_WIDTH) && defined(KERNEL_HEIGHT) && defined(KERNEL_DEPTH) && defined(SRC_WIDTH) && defined(SRC_HEIGHT) && defined(VECTOR_SIZE) && defined(WIDTH_MOD_VECTOR_SIZE)
-/** This kernel reshapes the input tensor to a tensor used to perform convolution using GEMM when
- * the kernel width is greater than 1 (except when the kernel size is 3x3) and pad_x == pad_y == 0.
+#if defined(CONVOLVED_WIDTH) && defined(STRIDE_X) && defined(STRIDE_Y) && defined(KERNEL_WIDTH) && defined(KERNEL_HEIGHT) && defined(SRC_DEPTH) && defined(SRC_WIDTH) && defined(SRC_HEIGHT) && defined(VECTOR_SIZE) && defined(WIDTH_MOD_VECTOR_SIZE)
+/** This opencl kernel performs im2col when the kernel size is greater than 1x1, we do not have paddings and the data layout is NCHW
  *
  * @note The data type must be passed at compile time using -DDATA_TYPE e.g. -DDATA_TYPE=float.
  * @note The vector size must be passed at compile time using -DVECTOR_SIZE e.g. -DVECTOR_SIZE=4.
  * @note The width modulo vector size must be passed at compile time using -DWIDTH_MOD_VECTOR_SIZE e.g. -DWIDTH_MOD_VECTOR_SIZE=3.
+ * @note The stride along the X and Y directions must be passed at compile time using -DSTRIDE_X and -DSTRIDE_Y: e.g. -DSTRIDE_X=1 and -DSTRIDE_Y=1
  * @note In case biases will be added to the convolution -DHAS_BIAS has to be passed to append the final matrix with 1 in each row.
  *
  * @param[in]  src_ptr                           Pointer to the source tensor. Supported data types: F16/F32
@@ -810,16 +689,16 @@ __kernel void im2col11x11_padx0_pady0_dchw(
  * @param[in]  src_stride_w                      Stride of the source tensor in W dimension (in bytes).
  * @param[in]  dst_stride_w                      Stride of the destination tensor in W dimension (in bytes).
  */
-__kernel void im2col_generic_padx0_pady0_dchw(
+__kernel void im2col_generic_padx0_pady0_nchw(
     TENSOR3D_DECLARATION(src),
     IMAGE_DECLARATION(dst),
     uint src_stride_w,
     uint dst_stride_w)
 {
-    const int xc    = get_global_id(0);                // x coordinate in the convolved tensor
-    const int yc    = get_global_id(1);                // y coordinate in the convolved tensor
-    const int ch    = get_global_id(2) % KERNEL_DEPTH; // input feature map
-    const int batch = get_global_id(2) / KERNEL_DEPTH; // batch size
+    const int xc    = get_global_id(0);             // x coordinate in the convolved tensor
+    const int yc    = get_global_id(1);             // y coordinate in the convolved tensor
+    const int ch    = get_global_id(2) % SRC_DEPTH; // input feature map
+    const int batch = get_global_id(2) / SRC_DEPTH; // batch size
 
     // Calculate input indices
     const int xi = xc * STRIDE_X;
@@ -855,25 +734,25 @@ __kernel void im2col_generic_padx0_pady0_dchw(
     } /* End of loop over KERNEL_HEIGHT */
 
 #ifdef HAS_BIAS
-    if(ch == (KERNEL_DEPTH - 1))
+    if(ch == (SRC_DEPTH - 1))
     {
         *output_ptr = 1.0f;
     }
 #endif // HAS_BIAS
 }
-#endif //defined(CONVOLVED_WIDTH) && defined(STRIDE_X) && defined(STRIDE_Y) && defined(PAD_LEFT) && defined(PAD_TOP) && defined(PAD_RIGHT) && defined(PAD_BOTTOM) && defined(KERNEL_WIDTH) && defined(KERNEL_HEIGHT) && defined(KERNEL_DEPTH) && defined(SRC_WIDTH) && defined(SRC_HEIGHT) && defined(VECTOR_SIZE) && defined(WIDTH_MOD_VECTOR_SIZE)
+#endif //defined(CONVOLVED_WIDTH) && defined(STRIDE_X) && defined(STRIDE_Y) && defined(PAD_LEFT) && defined(PAD_TOP) && defined(PAD_RIGHT) && defined(PAD_BOTTOM) && defined(KERNEL_WIDTH) && defined(KERNEL_HEIGHT) && defined(SRC_DEPTH) && defined(SRC_WIDTH) && defined(SRC_HEIGHT) && defined(VECTOR_SIZE) && defined(WIDTH_MOD_VECTOR_SIZE)
+
+#if defined(CONVOLVED_WIDTH) && defined(SRC_WIDTH) && defined(SRC_HEIGHT) && defined(STRIDE_X) && defined(STRIDE_Y) && defined(KERNEL_WIDTH) && defined(KERNEL_HEIGHT) && defined(SRC_DEPTH) && defined(PAD_LEFT) && defined(PAD_RIGHT) && defined(PAD_TOP) && defined(PAD_BOTTOM) && defined(PAD_VALUE) && defined(VECTOR_SIZE) && defined(LAST_ACCESSED)
 
-#if defined(CONVOLVED_WIDTH) && defined(SRC_WIDTH) && defined(SRC_HEIGHT) && defined(STRIDE_X) && defined(STRIDE_Y) && defined(KERNEL_WIDTH) && defined(KERNEL_HEIGHT) && defined(KERNEL_DEPTH) && defined(PAD_LEFT) && defined(PAD_RIGHT) && defined(PAD_TOP) && defined(PAD_BOTTOM) && defined(PAD_VALUE)
-/** This kernel performs a reshaping of the input tensor to a tensor used to perform convolution using GEMM.
+#define VECTOR_N VEC_DATA_TYPE(DATA_TYPE, VECTOR_SIZE)
+
+/** This kernel performs im2col when the kernel size is 3x3 and the data layout is NHWC
  *
+ * @note This kernel computes VECTOR_SIZE elements
  * @note The data type must be passed at compile time using -DDATA_TYPE: e.g. -DDATA_TYPE=float
- * @note The width and height of the input tensor must be passed at compile time using -DSRC_WIDTH and -DSRC_HEIGHT: e.g. -DSRC_WIDTH=128 and -DSRC_HEIGHT=128
  * @note The width of output tensor after matrix multiplication must be passed at compile time using -DCONVOLVED_WIDTH: e.g. -DCONVOLVED_WIDTH=34
- * @note The kernel width, height and depth must be passed at compile time using -DKERNEL_WIDTH, -DKERNEL_HEIGHT and -DKERNEL_DEPTH: e.g. -DKERNEL_WIDTH=3, -DKERNEL_HEIGHT=3 and -DKERNEL_DEPTH=64
- * @note The pad_left, pad_right, pad_top and pad_bottom must be passed at compile time using -DPAD_LEFT, -DPAD_RIGHT, -DPAD_TOP and -DPAD_BOTTOM: e.g. -DPAD_LEFT=1, -DPAD_RIGHT=2, -DPAD_TOP=3 and -DPAD_BOTTOM=2
- * @note The zero value to store in case we load values out-of-bounds must be passed at compile time using -DPAD_VALUE: e.g. -DPAD_VALUE=0.0
- * @note The stride along the X and Y directions must be passed at compile time using -DSTRIDE_X and -DSTRIDE_Y: e.g. -DSTRIDE_X=1 and -DSTRIDE_Y=1
- * @note The dilation_x and dilation_y must be passed at compile time using -DDILATION_X and -DDILATION_Y: e.g. -DDILATION_X=1, -DDILATION_Y=1
+ * @note The kernel depth must be passed at compile time using -DSRC_DEPTH: e.g. -DSRC_DEPTH=3
+ * @note The stride along the Y direction must be passed at compile time using -DSTRIDE_Y: e.g. -DSTRIDE_Y=1
  * @note In case biases will be added to the convolution -DHAS_BIAS has to be passed to append the final matrix with 1 in each row.
  *
  * @param[in]  src_ptr                           Pointer to the source tensor. Supported data types: QASYMM8/F16/F32
@@ -893,64 +772,154 @@ __kernel void im2col_generic_padx0_pady0_dchw(
  * @param[in]  src_stride_w                      Stride of the source tensor in W dimension (in bytes).
  * @param[in]  dst_stride_w                      Stride of the destination tensor in W dimension (in bytes).
  */
-__kernel void im2col_generic_dchw(
+__kernel void im2col3x3_nhwc(
     TENSOR3D_DECLARATION(src),
     IMAGE_DECLARATION(dst),
     uint src_stride_w,
     uint dst_stride_w)
 {
-    const int xc    = get_global_id(0);                // x coordinate in the convolved tensor
-    const int yc    = get_global_id(1);                // y coordinate in the convolved tensor
-    const int ch    = get_global_id(2) % KERNEL_DEPTH; // input feature map
-    const int batch = get_global_id(2) / KERNEL_DEPTH; // batch size
+    const int ch    = min((int)(get_global_id(0) * VECTOR_SIZE), LAST_ACCESSED); // input feature map
+    const int yo    = get_global_id(1);
+    const int batch = get_global_id(2); // batch size
 
     // Calculate input indices
-    const int xi = xc * STRIDE_X - PAD_LEFT;
-    const int yi = yc * STRIDE_Y - PAD_TOP;
+    const int xi = (get_global_id(1) % CONVOLVED_WIDTH) * STRIDE_X;
+    const int yi = (get_global_id(1) / (int)CONVOLVED_WIDTH) * STRIDE_Y;
 
-    // Calculate output indices
-    const int xo = ch * KERNEL_WIDTH * KERNEL_HEIGHT;
-    const int yo = xc + yc * CONVOLVED_WIDTH; // Index of the convolution
-
-    __global uchar *input_ptr      = src_ptr + src_offset_first_element_in_bytes + ch * src_stride_z + batch * src_stride_w;
-    __global DATA_TYPE *output_ptr = ((__global DATA_TYPE *)(dst_ptr + dst_offset_first_element_in_bytes + yo * dst_stride_y + batch * dst_stride_w)) + xo;
-
-    // Linearize convolution elements
-    for(int yk = 0; yk < KERNEL_HEIGHT; ++yk)
-    {
-        int y = yi + yk * DILATION_Y;
-        for(int xk = 0; xk < KERNEL_WIDTH; ++xk, ++output_ptr)
-        {
-            int x = xi + xk * DILATION_X;
-#if PAD_LEFT == 0 && PAD_TOP == 0 && PAD_RIGHT == 0 && PAD_BOTTOM == 0
-            *output_ptr = *((__global DATA_TYPE *)(input_ptr + x * src_stride_x + y * src_stride_y));
-#else  // PAD_LEFT == 0 && PAD_TOP == 0 && PAD_RIGHT == 0 && PAD_BOTTOM == 0
-            if(x < 0 || x >= SRC_WIDTH || y < 0 || y >= SRC_HEIGHT)
-            {
-                *output_ptr = PAD_VALUE;
-            }
-            else
-            {
-                *output_ptr = *((__global DATA_TYPE *)(input_ptr + x * src_stride_x + y * src_stride_y));
-            }
-#endif // PAD_LEFT == 0 && PAD_TOP == 0 && PAD_RIGHT == 0 && PAD_BOTTOM == 0
-        }
-    }
+    // Get input and output address
+    __global uchar *input_ptr  = src_ptr + src_offset_first_element_in_bytes + ch * sizeof(DATA_TYPE) + batch * (int)src_stride_w;
+    __global uchar *output_ptr = dst_ptr + dst_offset_first_element_in_bytes + ch * sizeof(DATA_TYPE) + yo * (int)dst_stride_y + batch * (int)dst_stride_w;
+
+    int  yi_coord = 0;
+    int3 offset   = 0;
+
+    // Clamp xi
+    int3 xi_offset = ((int3)xi + (int3)(0, 1, 2) * DILATION_X - (int3)PAD_LEFT);
+#if PAD_TOP != 0 || PAD_BOTTOM != 0
+#define CLAMP(x, min_val, max_val) min(max(x, min_val), max_val)
+    xi_offset = CLAMP(xi_offset, (int3)0, (int3)(SRC_WIDTH - 1));
+#endif // PAD_TOP != 0 || PAD_BOTTOM != 0
+    xi_offset *= (int3)src_stride_y;
+
+    // Out-of-bound condition for X
+    int3 x_cond = (((int3)xi + (int3)(0, 1, 2) * DILATION_X - (int3)PAD_LEFT) < (int3)0) || (((int3)xi + (int3)(0, 1, 2) * DILATION_X - (int3)PAD_LEFT) >= (int3)SRC_WIDTH);
+
+    // yi == 0
+    // Clamp yi
+    // yi_coord is casted to unsigned int in order to use just a min() operation
+    // A "-1" 32 bit signed variable converted to unsigned gives 4294967295
+    yi_coord = yi - (int)PAD_TOP;
+
+    // Clamp only if PAD_TOP or PAD_BOTTOM is not equal to 0
+#if PAD_TOP != 0 || PAD_BOTTOM != 0
+    yi_coord = min((uint)yi_coord, (uint)(SRC_HEIGHT - 1));
+#endif // PAD_TOP != 0 || PAD_BOTTOM != 0
+
+    // Compute offset
+    offset = xi_offset + (yi_coord * (int)src_stride_z);
+
+    // Load input values
+    VECTOR_N values0 = VLOAD(VECTOR_SIZE)(0, (__global DATA_TYPE *)(input_ptr + offset.s0));
+    VECTOR_N values1 = VLOAD(VECTOR_SIZE)(0, (__global DATA_TYPE *)(input_ptr + offset.s1));
+    VECTOR_N values2 = VLOAD(VECTOR_SIZE)(0, (__global DATA_TYPE *)(input_ptr + offset.s2));
+
+#if PAD_TOP != 0 || PAD_LEFT != 0 || PAD_BOTTOM != 0 || PAD_RIGHT != 0
+    // Replace invalid values with PAD_VALUE
+    int y_cond = (int)((uint)(yi - (int)PAD_TOP) >= (uint)(SRC_HEIGHT));
+    values0    = select(values0, (VECTOR_N)PAD_VALUE, (VEC_DATA_TYPE(COND_DATA_TYPE, VECTOR_SIZE))y_cond || (VEC_DATA_TYPE(COND_DATA_TYPE, VECTOR_SIZE))(x_cond.s0));
+    values1    = select(values1, (VECTOR_N)PAD_VALUE, (VEC_DATA_TYPE(COND_DATA_TYPE, VECTOR_SIZE))y_cond || (VEC_DATA_TYPE(COND_DATA_TYPE, VECTOR_SIZE))(x_cond.s1));
+    values2    = select(values2, (VECTOR_N)PAD_VALUE, (VEC_DATA_TYPE(COND_DATA_TYPE, VECTOR_SIZE))y_cond || (VEC_DATA_TYPE(COND_DATA_TYPE, VECTOR_SIZE))(x_cond.s2));
+#endif // PAD_TOP != 0 || PAD_LEFT != 0 || PAD_BOTTOM != 0 || PAD_RIGHT != 0
+
+    // yi == 1
+    // Clamp yi_coord (it can be negative if PAD_TOP > 1)
+    yi_coord = yi - (int)PAD_TOP + 1 * DILATION_Y;
+
+    // Clamp only if PAD_TOP or PAD_BOTTOM is not equal to 0
+#if PAD_TOP != 0 || PAD_BOTTOM != 0
+    yi_coord = min((uint)yi_coord, (uint)(SRC_HEIGHT - 1));
+#endif // PAD_TOP != 0 || PAD_BOTTOM != 0
+
+    // Compute offset
+    offset = xi_offset + (yi_coord * (int)src_stride_z);
+
+    // Load input values
+    VECTOR_N values3 = VLOAD(VECTOR_SIZE)(0, (__global DATA_TYPE *)(input_ptr + offset.s0));
+    VECTOR_N values4 = VLOAD(VECTOR_SIZE)(0, (__global DATA_TYPE *)(input_ptr + offset.s1));
+    VECTOR_N values5 = VLOAD(VECTOR_SIZE)(0, (__global DATA_TYPE *)(input_ptr + offset.s2));
+
+#if PAD_TOP != 0 || PAD_LEFT != 0 || PAD_BOTTOM != 0 || PAD_RIGHT != 0
+    // Replace invalid values with zeros
+    y_cond  = (int)((uint)(yi - (int)PAD_TOP + 1) >= (uint)(SRC_HEIGHT));
+    values3 = select(values3, (VECTOR_N)PAD_VALUE, (VEC_DATA_TYPE(COND_DATA_TYPE, VECTOR_SIZE))y_cond || (VEC_DATA_TYPE(COND_DATA_TYPE, VECTOR_SIZE))(x_cond.s0));
+    values4 = select(values4, (VECTOR_N)PAD_VALUE, (VEC_DATA_TYPE(COND_DATA_TYPE, VECTOR_SIZE))y_cond || (VEC_DATA_TYPE(COND_DATA_TYPE, VECTOR_SIZE))(x_cond.s1));
+    values5 = select(values5, (VECTOR_N)PAD_VALUE, (VEC_DATA_TYPE(COND_DATA_TYPE, VECTOR_SIZE))y_cond || (VEC_DATA_TYPE(COND_DATA_TYPE, VECTOR_SIZE))(x_cond.s2));
+#endif // PAD_TOP != 0 || PAD_LEFT != 0 || PAD_BOTTOM != 0 || PAD_RIGHT != 0
+
+    // yi == 2
+    // Clamp yi_coord
+    yi_coord = yi - (int)PAD_TOP + 2 * DILATION_Y;
+
+    // Clamp only if PAD_TOP or PAD_BOTTOM is not equal to 0
+#if PAD_TOP != 0 || PAD_BOTTOM != 0
+    yi_coord = min((uint)yi_coord, (uint)(SRC_HEIGHT - 1));
+#endif // PAD_TOP != 0 || PAD_BOTTOM != 0
+
+    // Compute offset
+    offset = xi_offset + (yi_coord * (int)src_stride_z);
+
+    // Load input values
+    VECTOR_N values6 = VLOAD(VECTOR_SIZE)(0, (__global DATA_TYPE *)(input_ptr + offset.s0));
+    VECTOR_N values7 = VLOAD(VECTOR_SIZE)(0, (__global DATA_TYPE *)(input_ptr + offset.s1));
+    VECTOR_N values8 = VLOAD(VECTOR_SIZE)(0, (__global DATA_TYPE *)(input_ptr + offset.s2));
+
+#if PAD_TOP != 0 || PAD_LEFT != 0 || PAD_BOTTOM != 0 || PAD_RIGHT != 0
+    // Replace invalid values with PAD_VALUE
+    y_cond  = (int)((uint)(yi - (int)PAD_TOP + 2) >= (uint)(SRC_HEIGHT));
+    values6 = select(values6, (VECTOR_N)PAD_VALUE, (VEC_DATA_TYPE(COND_DATA_TYPE, VECTOR_SIZE))y_cond || (VEC_DATA_TYPE(COND_DATA_TYPE, VECTOR_SIZE))(x_cond.s0));
+    values7 = select(values7, (VECTOR_N)PAD_VALUE, (VEC_DATA_TYPE(COND_DATA_TYPE, VECTOR_SIZE))y_cond || (VEC_DATA_TYPE(COND_DATA_TYPE, VECTOR_SIZE))(x_cond.s1));
+    values8 = select(values8, (VECTOR_N)PAD_VALUE, (VEC_DATA_TYPE(COND_DATA_TYPE, VECTOR_SIZE))y_cond || (VEC_DATA_TYPE(COND_DATA_TYPE, VECTOR_SIZE))(x_cond.s2));
+#endif // PAD_TOP != 0 || PAD_LEFT != 0 || PAD_BOTTOM != 0 || PAD_RIGHT != 0
+
+    // Store
+    VSTORE(VECTOR_SIZE)
+    (values0, 0, (__global DATA_TYPE *)(output_ptr) + 0 * SRC_DEPTH);
+    VSTORE(VECTOR_SIZE)
+    (values1, 0, (__global DATA_TYPE *)(output_ptr) + 1 * SRC_DEPTH);
+    VSTORE(VECTOR_SIZE)
+    (values2, 0, (__global DATA_TYPE *)(output_ptr) + 2 * SRC_DEPTH);
+    VSTORE(VECTOR_SIZE)
+    (values3, 0, (__global DATA_TYPE *)(output_ptr) + 3 * SRC_DEPTH);
+    VSTORE(VECTOR_SIZE)
+    (values4, 0, (__global DATA_TYPE *)(output_ptr) + 4 * SRC_DEPTH);
+    VSTORE(VECTOR_SIZE)
+    (values5, 0, (__global DATA_TYPE *)(output_ptr) + 5 * SRC_DEPTH);
+    VSTORE(VECTOR_SIZE)
+    (values6, 0, (__global DATA_TYPE *)(output_ptr) + 6 * SRC_DEPTH);
+    VSTORE(VECTOR_SIZE)
+    (values7, 0, (__global DATA_TYPE *)(output_ptr) + 7 * SRC_DEPTH);
+    VSTORE(VECTOR_SIZE)
+    (values8, 0, (__global DATA_TYPE *)(output_ptr) + 8 * SRC_DEPTH);
 
 #ifdef HAS_BIAS
-    if(ch == (KERNEL_DEPTH - 1))
+    if((ch + VECTOR_SIZE) >= SRC_DEPTH)
     {
-        *output_ptr = 1.0f;
+        *((__global DATA_TYPE *)(output_ptr) - ch + SRC_DEPTH * 9) = 1.0f;
     }
 #endif // HAS_BIAS
 }
-#endif // defined(CONVOLVED_WIDTH) && defined(SRC_WIDTH) && defined(SRC_HEIGHT) && defined(STRIDE_X) && defined(STRIDE_Y) && defined(KERNEL_WIDTH) && defined(KERNEL_HEIGHT) && defined(KERNEL_DEPTH) && defined(PAD_LEFT) && defined(PAD_RIGHT) && defined(PAD_TOP) && defined(PAD_BOTTOM) && defined(PAD_VALUE)
 
-/**This kernel reshapes the input tensor to a tensor used to perform convolution using GEMM when
- * the kernel width and height are the same of width and height of the input tensor
+/** This opencl kernel performs a generic im2col implementation when the data layout is NHWC
  *
- * @note Datatype should be given as a preprocessor argument using -DDATA_TYPE=type. e.g. -DDATA_TYPE=float
- * @note In case biases will be added in late stage, -DHAS_BIAS has to be passed to append the final matrix with 1 in each row.
+ * @note The data type must be passed at compile time using -DDATA_TYPE: e.g. -DDATA_TYPE=float
+ * @note The width and height of the input tensor must be passed at compile time using -DSRC_WIDTH and -DSRC_HEIGHT: e.g. -DSRC_WIDTH=128 and -DSRC_HEIGHT=128
+ * @note The width of output tensor after matrix multiplication must be passed at compile time using -DCONVOLVED_WIDTH: e.g. -DCONVOLVED_WIDTH=34
+ * @note The kernel width, height and depth must be passed at compile time using -DKERNEL_WIDTH, -DKERNEL_HEIGHT and -DSRC_DEPTH: e.g. -DKERNEL_WIDTH=3, -DKERNEL_HEIGHT=3 and -DSRC_DEPTH=64
+ * @note The pad_left, pad_right, pad_top and pad_bottom must be passed at compile time using -DPAD_LEFT, -DPAD_RIGHT, -DPAD_TOP and -DPAD_BOTTOM: e.g. -DPAD_LEFT=1, -DPAD_RIGHT=2, -DPAD_TOP=3 and -DPAD_BOTTOM=2
+ * @note The zero value to store in case we load values out-of-bounds must be passed at compile time using -DPAD_VALUE: e.g. -DPAD_VALUE=0.0
+ * @note The stride along the X and Y directions must be passed at compile time using -DSTRIDE_X and -DSTRIDE_Y: e.g. -DSTRIDE_X=1 and -DSTRIDE_Y=1
+ * @note The dilation_x and dilation_y must be passed at compile time using -DDILATION_X and -DDILATION_Y: e.g. -DDILATION_X=1, -DDILATION_Y=1
+ * @note In case biases will be added to the convolution -DHAS_BIAS has to be passed to append the final matrix with 1 in each row.
  *
  * @param[in]  src_ptr                           Pointer to the source tensor. Supported data types: QASYMM8/F16/F32
  * @param[in]  src_stride_x                      Stride of the source tensor in X dimension (in bytes)
@@ -958,35 +927,75 @@ __kernel void im2col_generic_dchw(
  * @param[in]  src_stride_y                      Stride of the source tensor in Y dimension (in bytes)
  * @param[in]  src_step_y                        src_stride_y * number of elements along Y processed per workitem(in bytes)
  * @param[in]  src_stride_z                      Stride of the source tensor in Z dimension (in bytes)
- * @param[in]  src_step_z                        src_stride_z * number of elements along Y processed per workitem(in bytes)
+ * @param[in]  src_step_z                        src_stride_z * number of elements along Z processed per workitem(in bytes)
  * @param[in]  src_offset_first_element_in_bytes The offset of the first element in the source tensor
- * @param[out] dst_ptr                           Pointer to the destination tensor. Same as @p src_ptr
+ * @param[out] dst_ptr                           Pointer to the destination tensor. Supported data types: same as @p src_ptr
  * @param[in]  dst_stride_x                      Stride of the destination tensor in X dimension (in bytes)
  * @param[in]  dst_step_x                        dst_stride_x * number of elements along X processed per workitem(in bytes)
+ * @param[in]  dst_stride_y                      Stride of the destination tensor in Y dimension (in bytes)
+ * @param[in]  dst_step_y                        dst_stride_y * number of elements along Y processed per workitem(in bytes)
  * @param[in]  dst_offset_first_element_in_bytes The offset of the first element in the destination tensor
- * @param[in]  width                             The width of the input tensor
- * @param[in]  height                            The height of the input tensor
+ * @param[in]  src_stride_w                      Stride of the source tensor in W dimension (in bytes).
+ * @param[in]  dst_stride_w                      Stride of the destination tensor in W dimension (in bytes).
  */
-__kernel void im2col_reduced_dchw(
+__kernel void im2col_generic_nhwc(
     TENSOR3D_DECLARATION(src),
-    VECTOR_DECLARATION(dst),
-    uint width, uint height)
+    IMAGE_DECLARATION(dst),
+    uint src_stride_w,
+    uint dst_stride_w)
 {
-    Tensor3D src = CONVERT_TO_TENSOR3D_STRUCT(src);
+    const int ch    = min((int)(get_global_id(0) * VECTOR_SIZE), LAST_ACCESSED); // input feature map
+    const int yo    = get_global_id(1);
+    const int batch = get_global_id(2); // batch size
 
-    const uint image_size = width * height;
+    // Calculate input indices
+    const int xi = (get_global_id(1) % CONVOLVED_WIDTH) * STRIDE_X;
+    const int yi = (get_global_id(1) / (int)CONVOLVED_WIDTH) * STRIDE_Y;
+
+    // Get input and output address
+    __global uchar *input_ptr  = src_ptr + src_offset_first_element_in_bytes + ch * sizeof(DATA_TYPE) + batch * (int)src_stride_w;
+    __global uchar *output_ptr = dst_ptr + dst_offset_first_element_in_bytes + ch * sizeof(DATA_TYPE) + yo * (int)dst_stride_y + batch * (int)dst_stride_w;
+
+    int i = 0;
+    for(int yk = 0; yk < KERNEL_HEIGHT; ++yk)
+    {
+        // Clamp yi_coord
+        int yi_coord = yi + yk * DILATION_Y - (int)PAD_TOP;
+        yi_coord     = CLAMP(yi_coord, (int)0, (int)(SRC_HEIGHT - 1));
 
-    __global uchar *tmp_out_ptr = dst_ptr + dst_offset_first_element_in_bytes + (get_global_id(0) + get_global_id(1) * width + get_global_id(2) * image_size) * dst_stride_x;
+        // Out-of-bound condition for Y
+        int y_border_condition = ((yi + yk * DILATION_Y - (int)PAD_TOP) < (int)0) || ((yi + yk * DILATION_Y - (int)PAD_TOP) >= (int)SRC_HEIGHT);
 
-    *((__global DATA_TYPE *)tmp_out_ptr) = *((__global DATA_TYPE *)src.ptr);
+        for(int xk = 0; xk < KERNEL_WIDTH; ++xk)
+        {
+            // Clamp xi_coord
+            int xi_coord = (xi + xk * DILATION_X - (int)PAD_LEFT);
+            xi_coord     = CLAMP(xi_coord, (int)0, (int)(SRC_WIDTH - 1));
+
+            // Out-of-bound condition for X
+            int x_border_condition = ((xi + xk * DILATION_X - (int)PAD_LEFT) < (int)0) || ((xi + xk * DILATION_X - (int)PAD_LEFT) >= (int)SRC_WIDTH);
+
+            int offset = xi_coord * (int)src_stride_y + (yi_coord * (int)src_stride_z);
+
+            VECTOR_N values0 = VLOAD(VECTOR_SIZE)(0, (__global DATA_TYPE *)(input_ptr + offset));
+
+            // Replace with PAD_VALUE if the value is out-of-bound
+            values0 = select(values0, (VECTOR_N)PAD_VALUE, (VEC_DATA_TYPE(COND_DATA_TYPE, VECTOR_SIZE))x_border_condition || (VEC_DATA_TYPE(COND_DATA_TYPE, VECTOR_SIZE))(y_border_condition));
+
+            // Store
+            VSTORE(VECTOR_SIZE)
+            (values0, 0, (__global DATA_TYPE *)(output_ptr) + i * (int)SRC_DEPTH);
+
+            i++;
+        }
+    }
 
 #ifdef HAS_BIAS
-    // If it is the last thread in the 3 dimensional workgroup
-    if(get_global_id(0) == (get_global_size(0) - 1) && get_global_id(1) == (get_global_size(1) - 1) && get_global_id(2) == (get_global_size(2) - 1))
+    if((ch + VECTOR_SIZE) >= SRC_DEPTH)
     {
-        tmp_out_ptr += dst_stride_x;
-        *((__global DATA_TYPE *)tmp_out_ptr) = (DATA_TYPE)1.0f;
+        *((__global DATA_TYPE *)(output_ptr) - ch + SRC_DEPTH * KERNEL_WIDTH * KERNEL_HEIGHT) = 1.0f;
     }
 #endif // HAS_BIAS
 }
+#endif // defined(CONVOLVED_WIDTH) && defined(SRC_WIDTH) && defined(SRC_HEIGHT) && defined(STRIDE_X) && defined(STRIDE_Y) && defined(KERNEL_WIDTH) && defined(KERNEL_HEIGHT) && defined(SRC_DEPTH) && defined(PAD_LEFT) && defined(PAD_RIGHT) && defined(PAD_TOP) && defined(PAD_BOTTOM) && defined(PAD_VALUE) && defined(VECTOR_SIZE) && defined(LAST_ACCESSED)
 #endif // defined(DATA_TYPE) && defined(ELEMENT_SIZE)