COMPMID-1338 Split winograd.cl

Change-Id: I583227fc1a38b1a34de253e383d71cca66007f18
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/138273
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>

1 files changed, 1152 insertions, 0 deletions

diff --git a/src/core/CL/cl_kernels/winograd_output_transform.cl b/src/core/CL/cl_kernels/winograd_output_transform.cl
new file mode 100644
index 0000000000..d195c14ccd
--- /dev/null
+++ b/src/core/CL/cl_kernels/winograd_output_transform.cl
@@ -0,0 +1,1152 @@
+/*
+ * Copyright (c) 2018 ARM Limited.
+ *
+ * SPDX-License-Identifier: MIT
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to
+ * deal in the Software without restriction, including without limitation the
+ * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
+ * sell copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in all
+ * copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#include "helpers.h"
+
+#if defined(NUM_TILES_X) && defined(OUTPUT_TILE_W) && defined(OUTPUT_TILE_H)
+/** This OpenCL kernel performs Winograd output transform when the output tile is 2x2/2x1 or 1x2, the filter size 3x3/3x1 or 1x3 and the data layout is NCHW
+ *
+ * @note The number of tiles along the X direction must be passed at compile time using -DNUM_TILES_X: e.g. -DNUM_TILES_X=16
+ * @note The width of the output tile must be passed at compile time using -DOUTPUT_TILE_W: e.g. -DOUTPUT_TILE_W=2
+ * @note The height of the output tile must be passed at compile time using -DOUTPUT_TILE_H: e.g. -DOUTPUT_TILE_H=2
+ * @note If this kernel is used to perform Winograd output transform 3x1, -DWINOGRAD_OUTPUT_TRANSFORM_HORIZONTAL has to be passed at compile time
+ * @note If this kernel is used to perform Winograd output transform 1x3, -DWINOGRAD_OUTPUT_TRANSFORM_VERTICAL has to be passed at compile time
+ *
+ * @param[in]  src_ptr                           Pointer to the source tensor. Supported data types: 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]  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]  dst_offset_first_element_in_bytes The offset of the first element in the destination tensor
+ */
+__kernel void winograd_output_transform_2x2_3x3_nchw(
+    TENSOR3D_DECLARATION(src),
+    TENSOR3D_DECLARATION(dst)
+#if defined(HAS_BIAS)
+    ,
+    VECTOR_DECLARATION(bias)
+#endif // defined(HAS_BIAS)
+)
+{
+    // Each thread stores a 2x2/2x1 or 1x2 tile accordingly with the filter size
+    Tensor3D src = CONVERT_TO_TENSOR3D_STRUCT(src);
+
+    const __global uchar *src_addr = tensor3D_offset(&src, 0, 0, 0);
+
+    // Load the values across the 16 or 4 channels to compose the 4x4 or 4x1 tile
+    float d00 = *((__global float *)(src_addr + 0 * src_stride_z));
+    float d01 = *((__global float *)(src_addr + 1 * src_stride_z));
+    float d02 = *((__global float *)(src_addr + 2 * src_stride_z));
+    float d03 = *((__global float *)(src_addr + 3 * src_stride_z));
+
+#if defined(WINOGRAD_OUTPUT_TRANSFORM_HORIZONTAL) || defined(WINOGRAD_OUTPUT_TRANSFORM_VERTICAL)
+    // Compute the 2x1 or 1x2 output tile
+    // out00 = d00 + d01 + d02
+    // out01 = d01 - d02 - d03
+
+    float out00 = d00 + d01 + d02;
+    float out01 = d01 - d02 - d03;
+#else  // defined(WINOGRAD_OUTPUT_TRANSFORM_HORIZONTAL) || defined(WINOGRAD_OUTPUT_TRANSFORM_VERTICAL)
+    float d10 = *((__global float *)(src_addr + 4 * src_stride_z));
+    float d11 = *((__global float *)(src_addr + 5 * src_stride_z));
+    float d12 = *((__global float *)(src_addr + 6 * src_stride_z));
+    float d13 = *((__global float *)(src_addr + 7 * src_stride_z));
+
+    float d20 = *((__global float *)(src_addr + 8 * src_stride_z));
+    float d21 = *((__global float *)(src_addr + 9 * src_stride_z));
+    float d22 = *((__global float *)(src_addr + 10 * src_stride_z));
+    float d23 = *((__global float *)(src_addr + 11 * src_stride_z));
+
+    float d30 = *((__global float *)(src_addr + 12 * src_stride_z));
+    float d31 = *((__global float *)(src_addr + 13 * src_stride_z));
+    float d32 = *((__global float *)(src_addr + 14 * src_stride_z));
+    float d33 = *((__global float *)(src_addr + 15 * src_stride_z));
+
+    // Compute the 2x2 output tile
+    float k0 = d01 + d11 + d21;
+    float k1 = d02 + d12 + d22;
+    float k2 = d11 - d21 - d31;
+    float k3 = d12 - d22 - d32;
+
+    // out00 = d00 + d10 + d20 + d01 + d11 + d21 + d02 + d12 + d22
+    // out01 = d01 + d11 + d21 - (d02 + d12 + d22) - (d03 + d13 + d23)
+    // out10 = d10 - d20 - d30 + (d11 - d21 - d31) + (d12 - d22 - d32)
+    // out11 = d11 - d21 - d31 - (d12 - d22 - d32) - (d13 - d23 - d33)
+
+    float out00 = d10;
+    float out01 = -d13;
+    float out10 = d10;
+    float out11 = -d13;
+
+    out00 += d00 + d20 + k0 + k1;
+    out01 += k0 - k1 - (d03 + d23);
+    out10 += -d20 - d30 + k2 + k3;
+    out11 += k2 - k3 + d23 + d33;
+#endif // defined(WINOGRAD_OUTPUT_TRANSFORM_HORIZONTAL) || defined(WINOGRAD_OUTPUT_TRANSFORM_VERTICAL)
+
+    int y_in  = get_global_id(1);
+    int x_out = (y_in % NUM_TILES_X) * OUTPUT_TILE_W;
+    int y_out = (y_in / NUM_TILES_X) * OUTPUT_TILE_H;
+    int z_out = get_global_id(0);
+
+#if defined(HAS_BIAS)
+    // Add bias
+    Vector bias = CONVERT_TO_VECTOR_STRUCT_NO_STEP(bias);
+
+    float b = (float) * ((__global float *)(vector_offset(&bias, z_out)));
+
+    out00 += (float)b;
+    out01 += (float)b;
+#endif // defined(HAS_BIAS)
+
+    // Get output address
+    __global uchar *dst_addr = dst_ptr + dst_offset_first_element_in_bytes + x_out * sizeof(float) + y_out * dst_stride_y + z_out * dst_stride_z;
+
+    // Store the output tile
+#if defined(WINOGRAD_OUTPUT_TRANSFORM_VERTICAL)
+    *((__global float *)(dst_addr + 0 * dst_stride_y)) = out00;
+    *((__global float *)(dst_addr + 1 * dst_stride_y)) = out01;
+#else  // defined(WINOGRAD_OUTPUT_TRANSFORM_VERTICAL)
+    vstore2((float2)(out00, out01), 0, (__global float *)(dst_addr + 0 * dst_stride_y));
+#endif // defined(WINOGRAD_OUTPUT_TRANSFORM_VERTICAL)
+
+#if !defined(WINOGRAD_OUTPUT_TRANSFORM_HORIZONTAL) && !defined(WINOGRAD_OUTPUT_TRANSFORM_VERTICAL)
+#if defined(HAS_BIAS)
+    // Add bias
+    out10 += (float)b;
+    out11 += (float)b;
+#endif // defined(HAS_BIAS)
+
+    vstore2((float2)(out10, out11), 0, (__global float *)(dst_addr + 1 * dst_stride_y));
+#endif // !defined(WINOGRAD_OUTPUT_TRANSFORM_HORIZONTAL) && !defined(WINOGRAD_OUTPUT_TRANSFORM_VERTICAL)
+}
+
+/** This OpenCL kernel performs Winograd output transform when the output tile is 4x4, the filter size 3x3 and the data layout is NCHW
+ *
+ * @note The number of tiles along the X direction must be passed at compile time using -DNUM_TILES_X: e.g. -DNUM_TILES_X=16
+ * @note The width of the output tile must be passed at compile time using -DOUTPUT_TILE_W: e.g. -DOUTPUT_TILE_W=4
+ * @note The height of the output tile must be passed at compile time using -DOUTPUT_TILE_H: e.g. -DOUTPUT_TILE_H=4
+ * @note If this kernel is used to perform Winograd output transform 3x1, -DWINOGRAD_OUTPUT_TRANSFORM_HORIZONTAL has to be passed at compile time
+ * @note If this kernel is used to perform Winograd output transform 1x3, -DWINOGRAD_OUTPUT_TRANSFORM_VERTICAL has to be passed at compile time
+ *
+ * @param[in]  src_ptr                           Pointer to the source tensor. Supported data types: 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]  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]  dst_offset_first_element_in_bytes The offset of the first element in the destination tensor
+ */
+__kernel void winograd_output_transform_4x4_3x3_nchw(
+    TENSOR3D_DECLARATION(src),
+    TENSOR3D_DECLARATION(dst)
+#if defined(HAS_BIAS)
+    ,
+    VECTOR_DECLARATION(bias)
+#endif // defined(HAS_BIAS)
+)
+{
+    // Each thread stores a 4x4/4x1 or 1x4 tile
+    Tensor3D src = CONVERT_TO_TENSOR3D_STRUCT(src);
+
+    const __global uchar *src_addr = tensor3D_offset(&src, 0, 0, 0);
+
+    // Load the values across the channels to compose the 6x6 or 6x1 tile
+    float d00 = *((__global float *)(src_addr + 0 * src_stride_z));
+    float d01 = *((__global float *)(src_addr + 1 * src_stride_z));
+    float d02 = *((__global float *)(src_addr + 2 * src_stride_z));
+    float d03 = *((__global float *)(src_addr + 3 * src_stride_z));
+    float d04 = *((__global float *)(src_addr + 4 * src_stride_z));
+    float d05 = *((__global float *)(src_addr + 5 * src_stride_z));
+
+#if defined(WINOGRAD_OUTPUT_TRANSFORM_HORIZONTAL) || defined(WINOGRAD_OUTPUT_TRANSFORM_VERTICAL)
+    // Compute out00, out01, out02 and out03
+    float out00 = d00 + d01 + d02 + d03 + d04;
+    float out01 = d01 - d02 + 2.0f * d03 - 2.0f * d04;
+    float out02 = d01 + d02 + 4.0f * d03 + 4.0f * d04;
+    float out03 = d01 - d02 + 8.0f * d03 - 8.0f * d04 + d05;
+#else  // defined(WINOGRAD_OUTPUT_TRANSFORM_HORIZONTAL) || defined(WINOGRAD_OUTPUT_TRANSFORM_VERTICAL)
+    float d10 = *((__global float *)(src_addr + 6 * src_stride_z));
+    float d11 = *((__global float *)(src_addr + 7 * src_stride_z));
+    float d12 = *((__global float *)(src_addr + 8 * src_stride_z));
+    float d13 = *((__global float *)(src_addr + 9 * src_stride_z));
+    float d14 = *((__global float *)(src_addr + 10 * src_stride_z));
+    float d15 = *((__global float *)(src_addr + 11 * src_stride_z));
+
+    float d20 = *((__global float *)(src_addr + 12 * src_stride_z));
+    float d21 = *((__global float *)(src_addr + 13 * src_stride_z));
+    float d22 = *((__global float *)(src_addr + 14 * src_stride_z));
+    float d23 = *((__global float *)(src_addr + 15 * src_stride_z));
+    float d24 = *((__global float *)(src_addr + 16 * src_stride_z));
+    float d25 = *((__global float *)(src_addr + 17 * src_stride_z));
+
+    float d30 = *((__global float *)(src_addr + 18 * src_stride_z));
+    float d31 = *((__global float *)(src_addr + 19 * src_stride_z));
+    float d32 = *((__global float *)(src_addr + 20 * src_stride_z));
+    float d33 = *((__global float *)(src_addr + 21 * src_stride_z));
+    float d34 = *((__global float *)(src_addr + 22 * src_stride_z));
+    float d35 = *((__global float *)(src_addr + 23 * src_stride_z));
+
+    float d40 = *((__global float *)(src_addr + 24 * src_stride_z));
+    float d41 = *((__global float *)(src_addr + 25 * src_stride_z));
+    float d42 = *((__global float *)(src_addr + 26 * src_stride_z));
+    float d43 = *((__global float *)(src_addr + 27 * src_stride_z));
+    float d44 = *((__global float *)(src_addr + 28 * src_stride_z));
+    float d45 = *((__global float *)(src_addr + 29 * src_stride_z));
+
+    float d50 = *((__global float *)(src_addr + 30 * src_stride_z));
+    float d51 = *((__global float *)(src_addr + 31 * src_stride_z));
+    float d52 = *((__global float *)(src_addr + 32 * src_stride_z));
+    float d53 = *((__global float *)(src_addr + 33 * src_stride_z));
+    float d54 = *((__global float *)(src_addr + 34 * src_stride_z));
+    float d55 = *((__global float *)(src_addr + 35 * src_stride_z));
+
+    // Compute out00, out01, out02 and out03
+    float out00 = d01 + d21 + d41 + d11 + d31;
+    float out01 = d01 + d21 + d41 + d11 + d31;
+    float out02 = d01 + d21 + d41 + d11 + d31;
+    float out03 = d01 + d21 + d41 + d11 + d31;
+
+    float k0 = d03 + d04 + d13 + d14 + d23 + d24 + d33 + d34 + d43 + d44;
+    float k1 = 2.0f * d03 - 2.0f * d04 + 2.0f * d13 - 2.0f * d14 + 2.0f * d23 - 2.0f * d24 + 2.0f * d33 - 2.0f * d34 + 2.0f * d43 - 2.0f * d44;
+
+    out00 += k0 + d00 + d02 + d10 + d12 + d20 + d22 + d30 + d32 + d40 + d42;
+    out01 += k1 - d02 - d12 - d22 - d32 - d42;
+    out02 += 4.0f * k0 + d02 + d12 + d22 + d32 + d42;
+    out03 += 4.0f * k1 - d02 - d12 - d22 - d32 - d42 + d05 + d15 + d25 + d35 + d45;
+
+    // Compute out10, out11, out12 and out13
+    float out10 = d11 - d21 + 2.0f * d31 - 2.0f * d41;
+    float out11 = d11 - d21 + 2.0f * d31 - 2.0f * d41;
+    float out12 = d11 - d21 + 2.0f * d31 - 2.0f * d41;
+    float out13 = d11 - d21 + 2.0f * d31 - 2.0f * d41;
+
+    k0 = d13 + d14 - d23 - d24 + 2.0f * d33 + 2.0f * d34 - 2.0f * d43 - 2.0f * d44;
+    k1 = 2.0f * d13 - 2.0f * d14 - 2.0f * d23 + 2.0f * d24 + 4.0f * d33 - 4.0f * d34 - 4.0f * d43 + 4.0f * d44;
+
+    out10 += k0 + d10 + d12 - d20 - d22 + 2.0f * d30 + 2.0f * d32 - 2.0f * d40 - 2.0f * d42;
+    out11 += k1 - d12 + d22 - 2.0f * d32 + 2.0f * d42;
+    out12 += 4.0f * k0 + d12 - d22 + 2.0f * d32 - 2.0f * d42;
+    out13 += 4.0f * k1 - d12 + d15 + d22 - d25 - 2.0f * d32 + 2.0f * d35 + 2.0f * d42 - 2.0f * d45;
+
+    // Compute out20, out21, out22 and out23
+    float out20 = d11 + d21 + 4.0f * d31 + 4.0f * d41;
+    float out21 = d11 + d21 + 4.0f * d31 + 4.0f * d41;
+    float out22 = d11 + d21 + 4.0f * d31 + 4.0f * d41;
+    float out23 = d11 + d21 + 4.0f * d31 + 4.0f * d41;
+
+    k0 = d13 + d14 + d23 + d24 + 4.0f * d33 + 4.0f * d34 + 4.0f * d43 + 4.0f * d44;
+    k1 = 2.0f * d13 - 2.0f * d14 + 2.0f * d23 - 2.0f * d24 + 8.0f * d33 - 8.0f * d34 + 8.0f * d43 - 8.0f * d44;
+
+    out20 += k0 + d10 + d12 + d20 + d22 + 4.0f * d30 + 4.0f * d32 + 4.0f * d40 + 4.0f * d42;
+    out21 += k1 - d12 - d22 - 4.0f * d32 - 4.0f * d42;
+    out22 += 4.0f * k0 + d12 + d22 + 4.0f * d32 + 4.0f * d42;
+    out23 += 4.0f * k1 - d12 + d15 - d22 + d25 - 4.0f * d32 + 4.0f * d35 - 4.0f * d42 + 4.0f * d45;
+
+    // Compute out30, out31, out32 and out33
+    float out30 = d11 - d21 + 8.0f * d31 - 8.0f * d41 + d51;
+    float out31 = d11 - d21 + 8.0f * d31 - 8.0f * d41 + d51;
+    float out32 = d11 - d21 + 8.0f * d31 - 8.0f * d41 + d51;
+    float out33 = d11 - d21 + 8.0f * d31 - 8.0f * d41 + d51;
+
+    k0 = d13 + d14 - d23 - d24 + 8.0f * d33 + 8.0f * d34 - 8.0f * d43 - 8.0f * d44 + d53 + d54;
+    k1 = 2.0f * d13 - 2.0f * d14 - 2.0f * d23 + 2.0f * d24 + 16.0f * d33 - 16.0f * d34 - 16.0f * d43 + 16.0f * d44 + 2.0f * d53 - 2.0f * d54;
+
+    out30 += k0 + d10 + d12 - d20 - d22 + 8.0f * d30 + 8.0f * d32 - 8.0f * d40 - 8.0f * d42 + d50 + d52;
+    out31 += k1 - d12 + d22 - 8.0f * d32 + 8.0f * d42 - d52;
+    out32 += 4.0f * k0 + d12 - d22 + 8.0f * d32 - 8.0f * d42 + d52;
+    out33 += 4.0f * k1 - d12 + d15 + d22 - d25 - 8.0f * d32 + 8.0f * d35 + 8.0f * d42 - 8.0f * d45 - d52 + d55;
+#endif // defined(WINOGRAD_OUTPUT_TRANSFORM_HORIZONTAL) || defined(WINOGRAD_OUTPUT_TRANSFORM_VERTICAL)
+
+    int y_in  = get_global_id(1);
+    int x_out = (y_in % NUM_TILES_X) * OUTPUT_TILE_W;
+    int y_out = (y_in / NUM_TILES_X) * OUTPUT_TILE_H;
+    int z_out = get_global_id(0);
+
+#if defined(HAS_BIAS)
+    // Add bias
+    Vector bias = CONVERT_TO_VECTOR_STRUCT_NO_STEP(bias);
+
+    float b = (float) * ((__global float *)(vector_offset(&bias, z_out)));
+
+    out00 += (float)b;
+    out01 += (float)b;
+    out02 += (float)b;
+    out03 += (float)b;
+#endif // defined(HAS_BIAS)
+
+    // Get output address
+    __global uchar *dst_addr = dst_ptr + dst_offset_first_element_in_bytes + x_out * sizeof(float) + y_out * dst_stride_y + z_out * dst_stride_z;
+
+    // Store the output tile
+#if defined(WINOGRAD_OUTPUT_TRANSFORM_VERTICAL)
+    *((__global float *)(dst_addr + 0 * dst_stride_y)) = out00;
+    *((__global float *)(dst_addr + 1 * dst_stride_y)) = out01;
+    *((__global float *)(dst_addr + 2 * dst_stride_y)) = out02;
+    *((__global float *)(dst_addr + 3 * dst_stride_y)) = out03;
+#else  // defined(WINOGRAD_OUTPUT_TRANSFORM_VERTICAL)
+    vstore4((float4)(out00, out01, out02, out03), 0, (__global float *)(dst_addr + 0 * dst_stride_y));
+#endif // defined(WINOGRAD_OUTPUT_TRANSFORM_VERTICAL)
+
+#if !defined(WINOGRAD_OUTPUT_TRANSFORM_HORIZONTAL) && !defined(WINOGRAD_OUTPUT_TRANSFORM_VERTICAL)
+#if defined(HAS_BIAS)
+    // Add bias
+    out10 += (float)b;
+    out11 += (float)b;
+    out12 += (float)b;
+    out13 += (float)b;
+
+    out20 += (float)b;
+    out21 += (float)b;
+    out22 += (float)b;
+    out23 += (float)b;
+
+    out30 += (float)b;
+    out31 += (float)b;
+    out32 += (float)b;
+    out33 += (float)b;
+#endif // defined(HAS_BIAS)
+    vstore4((float4)(out10, out11, out12, out13), 0, (__global float *)(dst_addr + 1 * dst_stride_y));
+    vstore4((float4)(out20, out21, out22, out23), 0, (__global float *)(dst_addr + 2 * dst_stride_y));
+    vstore4((float4)(out30, out31, out32, out33), 0, (__global float *)(dst_addr + 3 * dst_stride_y));
+#endif // !defined(WINOGRAD_OUTPUT_TRANSFORM_HORIZONTAL) && !defined(WINOGRAD_OUTPUT_TRANSFORM_VERTICAL)
+}
+
+#if defined(WINOGRAD_OUTPUT_TRANSFORM_HORIZONTAL)
+/** This OpenCL kernel performs Winograd output transform when the output tile is 2x1, the filter size 3x1 and the data layout is NCHW
+ *
+ * @note The number of tiles along the X direction must be passed at compile time using -DNUM_TILES_X: e.g. -DNUM_TILES_X=16
+ * @note The width of the output tile must be passed at compile time using -DOUTPUT_TILE_W: e.g. -DOUTPUT_TILE_W=2
+ * @note The height of the output tile must be passed at compile time using -DOUTPUT_TILE_H: e.g. -DOUTPUT_TILE_H=1
+ * @note -DWINOGRAD_OUTPUT_TRANSFORM_HORIZONTAL has to be passed at compile time
+ *
+ * @param[in]  src_ptr                           Pointer to the source tensor. Supported data types: 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]  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]  dst_offset_first_element_in_bytes The offset of the first element in the destination tensor
+ */
+__kernel void winograd_output_transform_2x1_3x1_nchw(
+    TENSOR3D_DECLARATION(src),
+    TENSOR3D_DECLARATION(dst)
+#if defined(HAS_BIAS)
+    ,
+    VECTOR_DECLARATION(bias)
+#endif // defined(HAS_BIAS)
+)
+{
+    winograd_output_transform_2x2_3x3_nchw(src_ptr,
+                                           src_stride_x,
+                                           src_step_x,
+                                           src_stride_y,
+                                           src_step_y,
+                                           src_stride_z,
+                                           src_step_z,
+                                           src_offset_first_element_in_bytes,
+                                           dst_ptr,
+                                           dst_stride_x,
+                                           dst_step_x,
+                                           dst_stride_y,
+                                           dst_step_y,
+                                           dst_stride_z,
+                                           dst_step_z,
+                                           dst_offset_first_element_in_bytes
+#if defined(HAS_BIAS)
+                                           ,
+                                           bias_ptr,
+                                           bias_stride_x,
+                                           bias_step_x,
+                                           bias_offset_first_element_in_bytes
+#endif // defined(HAS_BIAS)
+                                          );
+}
+
+/** This OpenCL kernel performs Winograd output transform when the output tile is 4x1, the filter size 3x1 and the data layout is NCHW
+ *
+ * @note The number of tiles along the X direction must be passed at compile time using -DNUM_TILES_X: e.g. -DNUM_TILES_X=16
+ * @note The width of the output tile must be passed at compile time using -DOUTPUT_TILE_W: e.g. -DOUTPUT_TILE_W=4
+ * @note The height of the output tile must be passed at compile time using -DOUTPUT_TILE_H: e.g. -DOUTPUT_TILE_H=1
+ * @note -DWINOGRAD_OUTPUT_TRANSFORM_HORIZONTAL has to be passed at compile time
+ *
+ * @param[in]  src_ptr                           Pointer to the source tensor. Supported data types: 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]  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]  dst_offset_first_element_in_bytes The offset of the first element in the destination tensor
+ */
+__kernel void winograd_output_transform_4x1_3x1_nchw(
+    TENSOR3D_DECLARATION(src),
+    TENSOR3D_DECLARATION(dst)
+#if defined(HAS_BIAS)
+    ,
+    VECTOR_DECLARATION(bias)
+#endif // defined(HAS_BIAS)
+)
+{
+    winograd_output_transform_4x4_3x3_nchw(src_ptr,
+                                           src_stride_x,
+                                           src_step_x,
+                                           src_stride_y,
+                                           src_step_y,
+                                           src_stride_z,
+                                           src_step_z,
+                                           src_offset_first_element_in_bytes,
+                                           dst_ptr,
+                                           dst_stride_x,
+                                           dst_step_x,
+                                           dst_stride_y,
+                                           dst_step_y,
+                                           dst_stride_z,
+                                           dst_step_z,
+                                           dst_offset_first_element_in_bytes
+#if defined(HAS_BIAS)
+                                           ,
+                                           bias_ptr,
+                                           bias_stride_x,
+                                           bias_step_x,
+                                           bias_offset_first_element_in_bytes
+#endif // defined(HAS_BIAS)
+                                          );
+}
+#endif // defined(WINOGRAD_OUTPUT_TRANSFORM_HORIZONTAL)
+
+#if defined(WINOGRAD_OUTPUT_TRANSFORM_VERTICAL)
+/** This OpenCL kernel performs Winograd output transform when the output tile is 1x2, the filter size 1x3 and the data layout is NCHW
+ *
+ * @note The number of tiles along the X direction must be passed at compile time using -DNUM_TILES_X: e.g. -DNUM_TILES_X=16
+ * @note The width of the output tile must be passed at compile time using -DOUTPUT_TILE_W: e.g. -DOUTPUT_TILE_W=1
+ * @note The height of the output tile must be passed at compile time using -DOUTPUT_TILE_H: e.g. -DOUTPUT_TILE_H=2
+ * @note -DWINOGRAD_OUTPUT_TRANSFORM_VERTICAL has to be passed at compile time
+ *
+ * @param[in]  src_ptr                           Pointer to the source tensor. Supported data types: 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]  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]  dst_offset_first_element_in_bytes The offset of the first element in the destination tensor
+ */
+__kernel void winograd_output_transform_1x2_1x3_nchw(
+    TENSOR3D_DECLARATION(src),
+    TENSOR3D_DECLARATION(dst)
+#if defined(HAS_BIAS)
+    ,
+    VECTOR_DECLARATION(bias)
+#endif // defined(HAS_BIAS)
+)
+{
+    winograd_output_transform_2x2_3x3_nchw(src_ptr,
+                                           src_stride_x,
+                                           src_step_x,
+                                           src_stride_y,
+                                           src_step_y,
+                                           src_stride_z,
+                                           src_step_z,
+                                           src_offset_first_element_in_bytes,
+                                           dst_ptr,
+                                           dst_stride_x,
+                                           dst_step_x,
+                                           dst_stride_y,
+                                           dst_step_y,
+                                           dst_stride_z,
+                                           dst_step_z,
+                                           dst_offset_first_element_in_bytes
+#if defined(HAS_BIAS)
+                                           ,
+                                           bias_ptr,
+                                           bias_stride_x,
+                                           bias_step_x,
+                                           bias_offset_first_element_in_bytes
+#endif // defined(HAS_BIAS)
+                                          );
+}
+
+/** This OpenCL kernel performs Winograd output transform when the output tile is 1x4, the filter size 1x3 and the data layout is NCHW
+ *
+ * @note The number of tiles along the X direction must be passed at compile time using -DNUM_TILES_X: e.g. -DNUM_TILES_X=16
+ * @note The width of the output tile must be passed at compile time using -DOUTPUT_TILE_W: e.g. -DOUTPUT_TILE_W=1
+ * @note The height of the output tile must be passed at compile time using -DOUTPUT_TILE_H: e.g. -DOUTPUT_TILE_H=4
+ * @note -DWINOGRAD_OUTPUT_TRANSFORM_VERTICAL has to be passed at compile time
+ *
+ * @param[in]  src_ptr                           Pointer to the source tensor. Supported data types: 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]  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]  dst_offset_first_element_in_bytes The offset of the first element in the destination tensor
+ */
+__kernel void winograd_output_transform_1x4_1x3_nchw(
+    TENSOR3D_DECLARATION(src),
+    TENSOR3D_DECLARATION(dst)
+#if defined(HAS_BIAS)
+    ,
+    VECTOR_DECLARATION(bias)
+#endif // defined(HAS_BIAS)
+)
+{
+    winograd_output_transform_4x4_3x3_nchw(src_ptr,
+                                           src_stride_x,
+                                           src_step_x,
+                                           src_stride_y,
+                                           src_step_y,
+                                           src_stride_z,
+                                           src_step_z,
+                                           src_offset_first_element_in_bytes,
+                                           dst_ptr,
+                                           dst_stride_x,
+                                           dst_step_x,
+                                           dst_stride_y,
+                                           dst_step_y,
+                                           dst_stride_z,
+                                           dst_step_z,
+                                           dst_offset_first_element_in_bytes
+#if defined(HAS_BIAS)
+                                           ,
+                                           bias_ptr,
+                                           bias_stride_x,
+                                           bias_step_x,
+                                           bias_offset_first_element_in_bytes
+#endif // defined(HAS_BIAS)
+                                          );
+}
+#endif // defined(WINOGRAD_OUTPUT_TRANSFORM_VERTICAL)
+
+/** This OpenCL kernel performs Winograd output transform when the output tile is 4x4, the filter size 3x3 and the data layout is NHWC
+ *
+ * @note The number of tiles along the X direction must be passed at compile time using -DNUM_TILES_X: e.g. -DNUM_TILES_X=16
+ *
+ * @param[in]  src_ptr                           Pointer to the source tensor. Supported data types: 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]  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]  dst_offset_first_element_in_bytes The offset of the first element in the destination tensor
+ * @param[in]  dst_size                          Size of the destination tensor, minus the last padding
+ */
+__kernel void winograd_output_transform_4x4_3x3_nhwc(
+    TENSOR3D_DECLARATION(src),
+    TENSOR3D_DECLARATION(dst),
+#if defined(HAS_BIAS)
+    VECTOR_DECLARATION(bias),
+#endif // defined(HAS_BIAS)
+    int dst_size)
+{
+    // Each thread stores a 4x4 tile
+    Tensor3D src = CONVERT_TO_TENSOR3D_STRUCT(src);
+
+    const __global uchar *src_addr = tensor3D_offset(&src, 0, 0, 0);
+
+    // Load the values across the 36 channels to compose the 6x6 tile
+    float d00 = *((__global float *)(src_addr + 0 * src_stride_z));
+    float d01 = *((__global float *)(src_addr + 1 * src_stride_z));
+    float d02 = *((__global float *)(src_addr + 2 * src_stride_z));
+    float d03 = *((__global float *)(src_addr + 3 * src_stride_z));
+    float d04 = *((__global float *)(src_addr + 4 * src_stride_z));
+    float d05 = *((__global float *)(src_addr + 5 * src_stride_z));
+
+    float d10 = *((__global float *)(src_addr + 6 * src_stride_z));
+    float d11 = *((__global float *)(src_addr + 7 * src_stride_z));
+    float d12 = *((__global float *)(src_addr + 8 * src_stride_z));
+    float d13 = *((__global float *)(src_addr + 9 * src_stride_z));
+    float d14 = *((__global float *)(src_addr + 10 * src_stride_z));
+    float d15 = *((__global float *)(src_addr + 11 * src_stride_z));
+
+    float d20 = *((__global float *)(src_addr + 12 * src_stride_z));
+    float d21 = *((__global float *)(src_addr + 13 * src_stride_z));
+    float d22 = *((__global float *)(src_addr + 14 * src_stride_z));
+    float d23 = *((__global float *)(src_addr + 15 * src_stride_z));
+    float d24 = *((__global float *)(src_addr + 16 * src_stride_z));
+    float d25 = *((__global float *)(src_addr + 17 * src_stride_z));
+
+    float d30 = *((__global float *)(src_addr + 18 * src_stride_z));
+    float d31 = *((__global float *)(src_addr + 19 * src_stride_z));
+    float d32 = *((__global float *)(src_addr + 20 * src_stride_z));
+    float d33 = *((__global float *)(src_addr + 21 * src_stride_z));
+    float d34 = *((__global float *)(src_addr + 22 * src_stride_z));
+    float d35 = *((__global float *)(src_addr + 23 * src_stride_z));
+
+    float d40 = *((__global float *)(src_addr + 24 * src_stride_z));
+    float d41 = *((__global float *)(src_addr + 25 * src_stride_z));
+    float d42 = *((__global float *)(src_addr + 26 * src_stride_z));
+    float d43 = *((__global float *)(src_addr + 27 * src_stride_z));
+    float d44 = *((__global float *)(src_addr + 28 * src_stride_z));
+    float d45 = *((__global float *)(src_addr + 29 * src_stride_z));
+
+    float d50 = *((__global float *)(src_addr + 30 * src_stride_z));
+    float d51 = *((__global float *)(src_addr + 31 * src_stride_z));
+    float d52 = *((__global float *)(src_addr + 32 * src_stride_z));
+    float d53 = *((__global float *)(src_addr + 33 * src_stride_z));
+    float d54 = *((__global float *)(src_addr + 34 * src_stride_z));
+    float d55 = *((__global float *)(src_addr + 35 * src_stride_z));
+
+    // Compute out00, out01, out02 and out03
+    float out00 = d01 + d21 + d41 + d11 + d31;
+    float out01 = d01 + d21 + d41 + d11 + d31;
+    float out02 = d01 + d21 + d41 + d11 + d31;
+    float out03 = d01 + d21 + d41 + d11 + d31;
+
+    float k0 = d03 + d04 + d13 + d14 + d23 + d24 + d33 + d34 + d43 + d44;
+    float k1 = 2.0f * d03 - 2.0f * d04 + 2.0f * d13 - 2.0f * d14 + 2.0f * d23 - 2.0f * d24 + 2.0f * d33 - 2.0f * d34 + 2.0f * d43 - 2.0f * d44;
+
+    out00 += k0 + d00 + d02 + d10 + d12 + d20 + d22 + d30 + d32 + d40 + d42;
+    out01 += k1 - d02 - d12 - d22 - d32 - d42;
+    out02 += 4.0f * k0 + d02 + d12 + d22 + d32 + d42;
+    out03 += 4.0f * k1 - d02 - d12 - d22 - d32 - d42 + d05 + d15 + d25 + d35 + d45;
+
+    // Compute out10, out11, out12 and out13
+    float out10 = d11 - d21 + 2.0f * d31 - 2.0f * d41;
+    float out11 = d11 - d21 + 2.0f * d31 - 2.0f * d41;
+    float out12 = d11 - d21 + 2.0f * d31 - 2.0f * d41;
+    float out13 = d11 - d21 + 2.0f * d31 - 2.0f * d41;
+
+    k0 = d13 + d14 - d23 - d24 + 2.0f * d33 + 2.0f * d34 - 2.0f * d43 - 2.0f * d44;
+    k1 = 2.0f * d13 - 2.0f * d14 - 2.0f * d23 + 2.0f * d24 + 4.0f * d33 - 4.0f * d34 - 4.0f * d43 + 4.0f * d44;
+
+    out10 += k0 + d10 + d12 - d20 - d22 + 2.0f * d30 + 2.0f * d32 - 2.0f * d40 - 2.0f * d42;
+    out11 += k1 - d12 + d22 - 2.0f * d32 + 2.0f * d42;
+    out12 += 4.0f * k0 + d12 - d22 + 2.0f * d32 - 2.0f * d42;
+    out13 += 4.0f * k1 - d12 + d15 + d22 - d25 - 2.0f * d32 + 2.0f * d35 + 2.0f * d42 - 2.0f * d45;
+
+    // Compute out20, out21, out22 and out23
+    float out20 = d11 + d21 + 4.0f * d31 + 4.0f * d41;
+    float out21 = d11 + d21 + 4.0f * d31 + 4.0f * d41;
+    float out22 = d11 + d21 + 4.0f * d31 + 4.0f * d41;
+    float out23 = d11 + d21 + 4.0f * d31 + 4.0f * d41;
+
+    k0 = d13 + d14 + d23 + d24 + 4.0f * d33 + 4.0f * d34 + 4.0f * d43 + 4.0f * d44;
+    k1 = 2.0f * d13 - 2.0f * d14 + 2.0f * d23 - 2.0f * d24 + 8.0f * d33 - 8.0f * d34 + 8.0f * d43 - 8.0f * d44;
+
+    out20 += k0 + d10 + d12 + d20 + d22 + 4.0f * d30 + 4.0f * d32 + 4.0f * d40 + 4.0f * d42;
+    out21 += k1 - d12 - d22 - 4.0f * d32 - 4.0f * d42;
+    out22 += 4.0f * k0 + d12 + d22 + 4.0f * d32 + 4.0f * d42;
+    out23 += 4.0f * k1 - d12 + d15 - d22 + d25 - 4.0f * d32 + 4.0f * d35 - 4.0f * d42 + 4.0f * d45;
+
+    // Compute out30, out31, out32 and out33
+    float out30 = d11 - d21 + 8.0f * d31 - 8.0f * d41 + d51;
+    float out31 = d11 - d21 + 8.0f * d31 - 8.0f * d41 + d51;
+    float out32 = d11 - d21 + 8.0f * d31 - 8.0f * d41 + d51;
+    float out33 = d11 - d21 + 8.0f * d31 - 8.0f * d41 + d51;
+
+    k0 = d13 + d14 - d23 - d24 + 8.0f * d33 + 8.0f * d34 - 8.0f * d43 - 8.0f * d44 + d53 + d54;
+    k1 = 2.0f * d13 - 2.0f * d14 - 2.0f * d23 + 2.0f * d24 + 16.0f * d33 - 16.0f * d34 - 16.0f * d43 + 16.0f * d44 + 2.0f * d53 - 2.0f * d54;
+
+    out30 += k0 + d10 + d12 - d20 - d22 + 8.0f * d30 + 8.0f * d32 - 8.0f * d40 - 8.0f * d42 + d50 + d52;
+    out31 += k1 - d12 + d22 - 8.0f * d32 + 8.0f * d42 - d52;
+    out32 += 4.0f * k0 + d12 - d22 + 8.0f * d32 - 8.0f * d42 + d52;
+    out33 += 4.0f * k1 - d12 + d15 + d22 - d25 - 8.0f * d32 + 8.0f * d35 + 8.0f * d42 - 8.0f * d45 - d52 + d55;
+
+    int y_in  = get_global_id(1);
+    int x_out = get_global_id(0);
+    int y_out = (y_in % NUM_TILES_X) * 4;
+    int z_out = (y_in / NUM_TILES_X) * 4;
+
+#if defined(HAS_BIAS)
+    // Add bias
+    Vector bias = CONVERT_TO_VECTOR_STRUCT_NO_STEP(bias);
+
+    float b = (float) * ((__global float *)(vector_offset(&bias, z_out)));
+
+    out00 += (float)b;
+    out01 += (float)b;
+    out02 += (float)b;
+    out03 += (float)b;
+
+    out10 += (float)b;
+    out11 += (float)b;
+    out12 += (float)b;
+    out13 += (float)b;
+
+    out20 += (float)b;
+    out21 += (float)b;
+    out22 += (float)b;
+    out23 += (float)b;
+
+    out30 += (float)b;
+    out31 += (float)b;
+    out32 += (float)b;
+    out33 += (float)b;
+
+#endif // defined(HAS_BIAS)
+
+    // Get output address
+    int4 offset = (int4)(dst_offset_first_element_in_bytes + x_out * sizeof(float) + y_out * dst_stride_y + z_out * dst_stride_z);
+    offset      = min(offset + (int4)(0, 1, 2, 3) * (int4)dst_stride_z, dst_size); // If address is beyond the last plane, clamp it to dst_size (which points to the last padding).
+    int4 mult_y = min(dst_size - offset, 1);                                       // If out of bound, we don't want to increase dst_stride_y, so we set the multiplier to 0. It will be 1 otherwise.
+
+    // Store the 4x4 output tile
+    *((__global float *)(dst_ptr + mult_y.s0 * 0 * dst_stride_y + offset.s0)) = out00;
+    *((__global float *)(dst_ptr + mult_y.s0 * 1 * dst_stride_y + offset.s0)) = out01;
+    *((__global float *)(dst_ptr + mult_y.s0 * 2 * dst_stride_y + offset.s0)) = out02;
+    *((__global float *)(dst_ptr + mult_y.s0 * 3 * dst_stride_y + offset.s0)) = out03;
+    *((__global float *)(dst_ptr + mult_y.s1 * 0 * dst_stride_y + offset.s1)) = out10;
+    *((__global float *)(dst_ptr + mult_y.s1 * 1 * dst_stride_y + offset.s1)) = out11;
+    *((__global float *)(dst_ptr + mult_y.s1 * 2 * dst_stride_y + offset.s1)) = out12;
+    *((__global float *)(dst_ptr + mult_y.s1 * 3 * dst_stride_y + offset.s1)) = out13;
+    *((__global float *)(dst_ptr + mult_y.s2 * 0 * dst_stride_y + offset.s2)) = out20;
+    *((__global float *)(dst_ptr + mult_y.s2 * 1 * dst_stride_y + offset.s2)) = out21;
+    *((__global float *)(dst_ptr + mult_y.s2 * 2 * dst_stride_y + offset.s2)) = out22;
+    *((__global float *)(dst_ptr + mult_y.s2 * 3 * dst_stride_y + offset.s2)) = out23;
+    *((__global float *)(dst_ptr + mult_y.s3 * 0 * dst_stride_y + offset.s3)) = out30;
+    *((__global float *)(dst_ptr + mult_y.s3 * 1 * dst_stride_y + offset.s3)) = out31;
+    *((__global float *)(dst_ptr + mult_y.s3 * 2 * dst_stride_y + offset.s3)) = out32;
+    *((__global float *)(dst_ptr + mult_y.s3 * 3 * dst_stride_y + offset.s3)) = out33;
+}
+
+#define COMPUTE_TMP_COL(col, d0, d1, d2, d3, d4, d5, d6, d7, comm_fact)  \
+    ({                                                                   \
+        comm_fact.s0 = d1 + d2;                                          \
+        comm_fact.s1 = d3 + d4;                                          \
+        comm_fact.s2 = d5 + d6;                                          \
+        \
+        col.s0 = comm_fact.s0 + comm_fact.s1 + 8.f * comm_fact.s2 + d0;  \
+        col.s2 = comm_fact.s0 + 4.f * comm_fact.s1 + 2.f * comm_fact.s2; \
+        \
+        comm_fact.s0 = d1 - d2;                                          \
+        comm_fact.s1 = d3 - d4;                                          \
+        comm_fact.s2 = d5 - d6;                                          \
+        \
+        col.s1 = comm_fact.s0 + 2.f * comm_fact.s1 + 4.f * comm_fact.s2; \
+        col.s3 = comm_fact.s0 + 8.f * comm_fact.s1 + comm_fact.s2 + d7;  \
+    })
+
+/** This OpenCL kernel performs Winograd output transform when the output tile is 4x4, the filter size 5x5 and the data layout is NCHW
+ *
+ * @note The number of tiles along the X direction must be passed at compile time using -DNUM_TILES_X: e.g. -DNUM_TILES_X=16
+ *
+ * @param[in]  src_ptr                           Pointer to the source tensor. Supported data types: 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]  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]  dst_offset_first_element_in_bytes The offset of the first element in the destination tensor
+ */
+__kernel void winograd_output_transform_4x4_5x5_nchw(
+    TENSOR3D_DECLARATION(src),
+    TENSOR3D_DECLARATION(dst)
+#if defined(HAS_BIAS)
+    ,
+    VECTOR_DECLARATION(bias)
+#endif // defined(HAS_BIAS)
+)
+{
+    // Each thread stores a 4x4 tile
+    Tensor3D src = CONVERT_TO_TENSOR3D_STRUCT(src);
+
+    const __global uchar *src_addr = tensor3D_offset(&src, 0, 0, 0);
+
+    // Load the values across the 64 channels to compose the 8x8 input tile
+    float d00 = *((__global float *)(src_addr + 0 * src_stride_z));
+    float d01 = *((__global float *)(src_addr + 1 * src_stride_z));
+    float d02 = *((__global float *)(src_addr + 2 * src_stride_z));
+    float d03 = *((__global float *)(src_addr + 3 * src_stride_z));
+    float d04 = *((__global float *)(src_addr + 4 * src_stride_z));
+    float d05 = *((__global float *)(src_addr + 5 * src_stride_z));
+    float d06 = *((__global float *)(src_addr + 6 * src_stride_z));
+    float d07 = *((__global float *)(src_addr + 7 * src_stride_z));
+
+    float d10 = *((__global float *)(src_addr + 8 * src_stride_z));
+    float d11 = *((__global float *)(src_addr + 9 * src_stride_z));
+    float d12 = *((__global float *)(src_addr + 10 * src_stride_z));
+    float d13 = *((__global float *)(src_addr + 11 * src_stride_z));
+    float d14 = *((__global float *)(src_addr + 12 * src_stride_z));
+    float d15 = *((__global float *)(src_addr + 13 * src_stride_z));
+    float d16 = *((__global float *)(src_addr + 14 * src_stride_z));
+    float d17 = *((__global float *)(src_addr + 15 * src_stride_z));
+
+    float d20 = *((__global float *)(src_addr + 16 * src_stride_z));
+    float d21 = *((__global float *)(src_addr + 17 * src_stride_z));
+    float d22 = *((__global float *)(src_addr + 18 * src_stride_z));
+    float d23 = *((__global float *)(src_addr + 19 * src_stride_z));
+    float d24 = *((__global float *)(src_addr + 20 * src_stride_z));
+    float d25 = *((__global float *)(src_addr + 21 * src_stride_z));
+    float d26 = *((__global float *)(src_addr + 22 * src_stride_z));
+    float d27 = *((__global float *)(src_addr + 23 * src_stride_z));
+
+    float d30 = *((__global float *)(src_addr + 24 * src_stride_z));
+    float d31 = *((__global float *)(src_addr + 25 * src_stride_z));
+    float d32 = *((__global float *)(src_addr + 26 * src_stride_z));
+    float d33 = *((__global float *)(src_addr + 27 * src_stride_z));
+    float d34 = *((__global float *)(src_addr + 28 * src_stride_z));
+    float d35 = *((__global float *)(src_addr + 29 * src_stride_z));
+    float d36 = *((__global float *)(src_addr + 30 * src_stride_z));
+    float d37 = *((__global float *)(src_addr + 31 * src_stride_z));
+
+    float d40 = *((__global float *)(src_addr + 32 * src_stride_z));
+    float d41 = *((__global float *)(src_addr + 33 * src_stride_z));
+    float d42 = *((__global float *)(src_addr + 34 * src_stride_z));
+    float d43 = *((__global float *)(src_addr + 35 * src_stride_z));
+    float d44 = *((__global float *)(src_addr + 36 * src_stride_z));
+    float d45 = *((__global float *)(src_addr + 37 * src_stride_z));
+    float d46 = *((__global float *)(src_addr + 38 * src_stride_z));
+    float d47 = *((__global float *)(src_addr + 39 * src_stride_z));
+
+    float d50 = *((__global float *)(src_addr + 40 * src_stride_z));
+    float d51 = *((__global float *)(src_addr + 41 * src_stride_z));
+    float d52 = *((__global float *)(src_addr + 42 * src_stride_z));
+    float d53 = *((__global float *)(src_addr + 43 * src_stride_z));
+    float d54 = *((__global float *)(src_addr + 44 * src_stride_z));
+    float d55 = *((__global float *)(src_addr + 45 * src_stride_z));
+    float d56 = *((__global float *)(src_addr + 46 * src_stride_z));
+    float d57 = *((__global float *)(src_addr + 47 * src_stride_z));
+
+    float d60 = *((__global float *)(src_addr + 48 * src_stride_z));
+    float d61 = *((__global float *)(src_addr + 49 * src_stride_z));
+    float d62 = *((__global float *)(src_addr + 50 * src_stride_z));
+    float d63 = *((__global float *)(src_addr + 51 * src_stride_z));
+    float d64 = *((__global float *)(src_addr + 52 * src_stride_z));
+    float d65 = *((__global float *)(src_addr + 53 * src_stride_z));
+    float d66 = *((__global float *)(src_addr + 54 * src_stride_z));
+    float d67 = *((__global float *)(src_addr + 55 * src_stride_z));
+
+    float d70 = *((__global float *)(src_addr + 56 * src_stride_z));
+    float d71 = *((__global float *)(src_addr + 57 * src_stride_z));
+    float d72 = *((__global float *)(src_addr + 58 * src_stride_z));
+    float d73 = *((__global float *)(src_addr + 59 * src_stride_z));
+    float d74 = *((__global float *)(src_addr + 60 * src_stride_z));
+    float d75 = *((__global float *)(src_addr + 61 * src_stride_z));
+    float d76 = *((__global float *)(src_addr + 62 * src_stride_z));
+    float d77 = *((__global float *)(src_addr + 63 * src_stride_z));
+
+    // Compute the 8x4 intermediate tensor
+    float4 comm_fact0, comm_fact1, comm_fact2;
+    float4 tmp_col0, tmp_col1, tmp_col2, tmp_col3, tmp_col4, tmp_col5, tmp_col6, tmp_col7;
+
+    COMPUTE_TMP_COL(tmp_col0, d00, d10, d20, d30, d40, d50, d60, d70, comm_fact0);
+    COMPUTE_TMP_COL(tmp_col1, d01, d11, d21, d31, d41, d51, d61, d71, comm_fact0);
+    COMPUTE_TMP_COL(tmp_col2, d02, d12, d22, d32, d42, d52, d62, d72, comm_fact0);
+    COMPUTE_TMP_COL(tmp_col3, d03, d13, d23, d33, d43, d53, d63, d73, comm_fact0);
+    COMPUTE_TMP_COL(tmp_col4, d04, d14, d24, d34, d44, d54, d64, d74, comm_fact0);
+    COMPUTE_TMP_COL(tmp_col5, d05, d15, d25, d35, d45, d55, d65, d75, comm_fact0);
+    COMPUTE_TMP_COL(tmp_col6, d06, d16, d26, d36, d46, d56, d66, d76, comm_fact0);
+    COMPUTE_TMP_COL(tmp_col7, d07, d17, d27, d37, d47, d57, d67, d77, comm_fact0);
+
+    // Compute the 4x4 output tile
+    comm_fact0 = tmp_col1 + tmp_col2;
+    comm_fact1 = tmp_col3 + tmp_col4;
+    comm_fact2 = tmp_col5 + tmp_col6;
+
+    float4 out_col0 = comm_fact0 + comm_fact1 + 8.f * comm_fact2 + tmp_col0;
+    float4 out_col2 = comm_fact0 + 4.f * comm_fact1 + 2.f * comm_fact2;
+
+    comm_fact0 = tmp_col1 - tmp_col2;
+    comm_fact1 = tmp_col3 - tmp_col4;
+    comm_fact2 = tmp_col5 - tmp_col6;
+
+    float4 out_col1 = comm_fact0 + 2.f * comm_fact1 + 4.f * comm_fact2;
+    float4 out_col3 = comm_fact0 + 8.f * comm_fact1 + comm_fact2 + tmp_col7;
+
+    int y_in  = get_global_id(1);
+    int x_out = (y_in % NUM_TILES_X) * 4;
+    int y_out = (y_in / NUM_TILES_X) * 4;
+    int z_out = get_global_id(0);
+
+#if defined(HAS_BIAS)
+    // Add bias
+    Vector bias = CONVERT_TO_VECTOR_STRUCT_NO_STEP(bias);
+
+    float b = (float) * ((__global float *)(vector_offset(&bias, z_out)));
+
+    out_col0 += (float4)b;
+    out_col1 += (float4)b;
+    out_col2 += (float4)b;
+    out_col3 += (float4)b;
+#endif // defined(HAS_BIAS)
+
+    // Get output address
+    __global uchar *dst_addr = dst_ptr + dst_offset_first_element_in_bytes + x_out * dst_stride_x + y_out * dst_stride_y + z_out * dst_stride_z;
+
+    // Store the 4x4 output tile
+    *(__global float *)(dst_addr + 0 * dst_stride_x + 0 * dst_stride_y) = out_col0.s0;
+    *(__global float *)(dst_addr + 1 * dst_stride_x + 0 * dst_stride_y) = out_col1.s0;
+    *(__global float *)(dst_addr + 2 * dst_stride_x + 0 * dst_stride_y) = out_col2.s0;
+    *(__global float *)(dst_addr + 3 * dst_stride_x + 0 * dst_stride_y) = out_col3.s0;
+    *(__global float *)(dst_addr + 0 * dst_stride_x + 1 * dst_stride_y) = out_col0.s1;
+    *(__global float *)(dst_addr + 1 * dst_stride_x + 1 * dst_stride_y) = out_col1.s1;
+    *(__global float *)(dst_addr + 2 * dst_stride_x + 1 * dst_stride_y) = out_col2.s1;
+    *(__global float *)(dst_addr + 3 * dst_stride_x + 1 * dst_stride_y) = out_col3.s1;
+    *(__global float *)(dst_addr + 0 * dst_stride_x + 2 * dst_stride_y) = out_col0.s2;
+    *(__global float *)(dst_addr + 1 * dst_stride_x + 2 * dst_stride_y) = out_col1.s2;
+    *(__global float *)(dst_addr + 2 * dst_stride_x + 2 * dst_stride_y) = out_col2.s2;
+    *(__global float *)(dst_addr + 3 * dst_stride_x + 2 * dst_stride_y) = out_col3.s2;
+    *(__global float *)(dst_addr + 0 * dst_stride_x + 3 * dst_stride_y) = out_col0.s3;
+    *(__global float *)(dst_addr + 1 * dst_stride_x + 3 * dst_stride_y) = out_col1.s3;
+    *(__global float *)(dst_addr + 2 * dst_stride_x + 3 * dst_stride_y) = out_col2.s3;
+    *(__global float *)(dst_addr + 3 * dst_stride_x + 3 * dst_stride_y) = out_col3.s3;
+}
+
+/** This OpenCL kernel performs Winograd output transform when the output tile is 4x4, the filter size 5x5 and the data layout is NHWC
+ *
+ * @note The number of tiles along the X direction must be passed at compile time using -DNUM_TILES_X: e.g. -DNUM_TILES_X=16
+ *
+ * @param[in]  src_ptr                           Pointer to the source tensor. Supported data types: 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]  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]  dst_offset_first_element_in_bytes The offset of the first element in the destination tensor
+ */
+__kernel void winograd_output_transform_4x4_5x5_nhwc(
+    TENSOR3D_DECLARATION(src),
+    TENSOR3D_DECLARATION(dst),
+#if defined(HAS_BIAS)
+    VECTOR_DECLARATION(bias),
+#endif // defined(HAS_BIAS)
+    int dst_size)
+{
+    // Each thread stores a 4x4 tile
+    Tensor3D src = CONVERT_TO_TENSOR3D_STRUCT(src);
+
+    const __global uchar *src_addr = tensor3D_offset(&src, 0, 0, 0);
+
+    // Load the values across the 64 channels to compose the 8x8 input tile
+    float d00 = *((__global float *)(src_addr + 0 * src_stride_z));
+    float d01 = *((__global float *)(src_addr + 1 * src_stride_z));
+    float d02 = *((__global float *)(src_addr + 2 * src_stride_z));
+    float d03 = *((__global float *)(src_addr + 3 * src_stride_z));
+    float d04 = *((__global float *)(src_addr + 4 * src_stride_z));
+    float d05 = *((__global float *)(src_addr + 5 * src_stride_z));
+    float d06 = *((__global float *)(src_addr + 6 * src_stride_z));
+    float d07 = *((__global float *)(src_addr + 7 * src_stride_z));
+
+    float d10 = *((__global float *)(src_addr + 8 * src_stride_z));
+    float d11 = *((__global float *)(src_addr + 9 * src_stride_z));
+    float d12 = *((__global float *)(src_addr + 10 * src_stride_z));
+    float d13 = *((__global float *)(src_addr + 11 * src_stride_z));
+    float d14 = *((__global float *)(src_addr + 12 * src_stride_z));
+    float d15 = *((__global float *)(src_addr + 13 * src_stride_z));
+    float d16 = *((__global float *)(src_addr + 14 * src_stride_z));
+    float d17 = *((__global float *)(src_addr + 15 * src_stride_z));
+
+    float d20 = *((__global float *)(src_addr + 16 * src_stride_z));
+    float d21 = *((__global float *)(src_addr + 17 * src_stride_z));
+    float d22 = *((__global float *)(src_addr + 18 * src_stride_z));
+    float d23 = *((__global float *)(src_addr + 19 * src_stride_z));
+    float d24 = *((__global float *)(src_addr + 20 * src_stride_z));
+    float d25 = *((__global float *)(src_addr + 21 * src_stride_z));
+    float d26 = *((__global float *)(src_addr + 22 * src_stride_z));
+    float d27 = *((__global float *)(src_addr + 23 * src_stride_z));
+
+    float d30 = *((__global float *)(src_addr + 24 * src_stride_z));
+    float d31 = *((__global float *)(src_addr + 25 * src_stride_z));
+    float d32 = *((__global float *)(src_addr + 26 * src_stride_z));
+    float d33 = *((__global float *)(src_addr + 27 * src_stride_z));
+    float d34 = *((__global float *)(src_addr + 28 * src_stride_z));
+    float d35 = *((__global float *)(src_addr + 29 * src_stride_z));
+    float d36 = *((__global float *)(src_addr + 30 * src_stride_z));
+    float d37 = *((__global float *)(src_addr + 31 * src_stride_z));
+
+    float d40 = *((__global float *)(src_addr + 32 * src_stride_z));
+    float d41 = *((__global float *)(src_addr + 33 * src_stride_z));
+    float d42 = *((__global float *)(src_addr + 34 * src_stride_z));
+    float d43 = *((__global float *)(src_addr + 35 * src_stride_z));
+    float d44 = *((__global float *)(src_addr + 36 * src_stride_z));
+    float d45 = *((__global float *)(src_addr + 37 * src_stride_z));
+    float d46 = *((__global float *)(src_addr + 38 * src_stride_z));
+    float d47 = *((__global float *)(src_addr + 39 * src_stride_z));
+
+    float d50 = *((__global float *)(src_addr + 40 * src_stride_z));
+    float d51 = *((__global float *)(src_addr + 41 * src_stride_z));
+    float d52 = *((__global float *)(src_addr + 42 * src_stride_z));
+    float d53 = *((__global float *)(src_addr + 43 * src_stride_z));
+    float d54 = *((__global float *)(src_addr + 44 * src_stride_z));
+    float d55 = *((__global float *)(src_addr + 45 * src_stride_z));
+    float d56 = *((__global float *)(src_addr + 46 * src_stride_z));
+    float d57 = *((__global float *)(src_addr + 47 * src_stride_z));
+
+    float d60 = *((__global float *)(src_addr + 48 * src_stride_z));
+    float d61 = *((__global float *)(src_addr + 49 * src_stride_z));
+    float d62 = *((__global float *)(src_addr + 50 * src_stride_z));
+    float d63 = *((__global float *)(src_addr + 51 * src_stride_z));
+    float d64 = *((__global float *)(src_addr + 52 * src_stride_z));
+    float d65 = *((__global float *)(src_addr + 53 * src_stride_z));
+    float d66 = *((__global float *)(src_addr + 54 * src_stride_z));
+    float d67 = *((__global float *)(src_addr + 55 * src_stride_z));
+
+    float d70 = *((__global float *)(src_addr + 56 * src_stride_z));
+    float d71 = *((__global float *)(src_addr + 57 * src_stride_z));
+    float d72 = *((__global float *)(src_addr + 58 * src_stride_z));
+    float d73 = *((__global float *)(src_addr + 59 * src_stride_z));
+    float d74 = *((__global float *)(src_addr + 60 * src_stride_z));
+    float d75 = *((__global float *)(src_addr + 61 * src_stride_z));
+    float d76 = *((__global float *)(src_addr + 62 * src_stride_z));
+    float d77 = *((__global float *)(src_addr + 63 * src_stride_z));
+
+    // Compute the 8x4 intermediate tensor
+    float4 comm_fact0, comm_fact1, comm_fact2;
+    float4 tmp_col0, tmp_col1, tmp_col2, tmp_col3, tmp_col4, tmp_col5, tmp_col6, tmp_col7;
+
+    COMPUTE_TMP_COL(tmp_col0, d00, d10, d20, d30, d40, d50, d60, d70, comm_fact0);
+    COMPUTE_TMP_COL(tmp_col1, d01, d11, d21, d31, d41, d51, d61, d71, comm_fact0);
+    COMPUTE_TMP_COL(tmp_col2, d02, d12, d22, d32, d42, d52, d62, d72, comm_fact0);
+    COMPUTE_TMP_COL(tmp_col3, d03, d13, d23, d33, d43, d53, d63, d73, comm_fact0);
+    COMPUTE_TMP_COL(tmp_col4, d04, d14, d24, d34, d44, d54, d64, d74, comm_fact0);
+    COMPUTE_TMP_COL(tmp_col5, d05, d15, d25, d35, d45, d55, d65, d75, comm_fact0);
+    COMPUTE_TMP_COL(tmp_col6, d06, d16, d26, d36, d46, d56, d66, d76, comm_fact0);
+    COMPUTE_TMP_COL(tmp_col7, d07, d17, d27, d37, d47, d57, d67, d77, comm_fact0);
+
+    // Compute the 4x4 output tile
+    comm_fact0 = tmp_col1 + tmp_col2;
+    comm_fact1 = tmp_col3 + tmp_col4;
+    comm_fact2 = tmp_col5 + tmp_col6;
+
+    float4 out_col0 = comm_fact0 + comm_fact1 + 8.f * comm_fact2 + tmp_col0;
+    float4 out_col2 = comm_fact0 + 4.f * comm_fact1 + 2.f * comm_fact2;
+
+    comm_fact0 = tmp_col1 - tmp_col2;
+    comm_fact1 = tmp_col3 - tmp_col4;
+    comm_fact2 = tmp_col5 - tmp_col6;
+
+    float4 out_col1 = comm_fact0 + 2.f * comm_fact1 + 4.f * comm_fact2;
+    float4 out_col3 = comm_fact0 + 8.f * comm_fact1 + comm_fact2 + tmp_col7;
+
+    int y_in  = get_global_id(1);
+    int x_out = get_global_id(0);
+    int y_out = (y_in % NUM_TILES_X) * 4;
+    int z_out = (y_in / NUM_TILES_X) * 4;
+
+#if defined(HAS_BIAS)
+    // Add bias
+    Vector bias = CONVERT_TO_VECTOR_STRUCT_NO_STEP(bias);
+
+    float b = (float) * ((__global float *)(vector_offset(&bias, z_out)));
+
+    out_col0 += (float4)b;
+    out_col1 += (float4)b;
+    out_col2 += (float4)b;
+    out_col3 += (float4)b;
+#endif // defined(HAS_BIAS)
+
+    // Get output address
+    int4 offset = (int4)(dst_offset_first_element_in_bytes + x_out * sizeof(float) + y_out * dst_stride_y + z_out * dst_stride_z);
+    offset      = min(offset + (int4)(0, 1, 2, 3) * (int4)dst_stride_z, dst_size); // If address is beyond the last plane, clamp it to dst_size (which points to the last padding).
+    int4 mult_y = min(dst_size - offset, 1);                                       // If out of bound, we don't want to increase dst_stride_y, so we set the multiplier to 0. It will be 1 otherwise.
+
+    // Store the 4x4 output tile
+    *(__global float *)(dst_ptr + mult_y.s0 * 0 * dst_stride_y + offset.s0) = out_col0.s0;
+    *(__global float *)(dst_ptr + mult_y.s0 * 1 * dst_stride_y + offset.s0) = out_col1.s0;
+    *(__global float *)(dst_ptr + mult_y.s0 * 2 * dst_stride_y + offset.s0) = out_col2.s0;
+    *(__global float *)(dst_ptr + mult_y.s0 * 3 * dst_stride_y + offset.s0) = out_col3.s0;
+    *(__global float *)(dst_ptr + mult_y.s0 * 0 * dst_stride_y + offset.s1) = out_col0.s1;
+    *(__global float *)(dst_ptr + mult_y.s0 * 1 * dst_stride_y + offset.s1) = out_col1.s1;
+    *(__global float *)(dst_ptr + mult_y.s0 * 2 * dst_stride_y + offset.s1) = out_col2.s1;
+    *(__global float *)(dst_ptr + mult_y.s0 * 3 * dst_stride_y + offset.s1) = out_col3.s1;
+    *(__global float *)(dst_ptr + mult_y.s0 * 0 * dst_stride_y + offset.s2) = out_col0.s2;
+    *(__global float *)(dst_ptr + mult_y.s0 * 1 * dst_stride_y + offset.s2) = out_col1.s2;
+    *(__global float *)(dst_ptr + mult_y.s0 * 2 * dst_stride_y + offset.s2) = out_col2.s2;
+    *(__global float *)(dst_ptr + mult_y.s0 * 3 * dst_stride_y + offset.s2) = out_col3.s2;
+    *(__global float *)(dst_ptr + mult_y.s0 * 0 * dst_stride_y + offset.s3) = out_col0.s3;
+    *(__global float *)(dst_ptr + mult_y.s0 * 1 * dst_stride_y + offset.s3) = out_col1.s3;
+    *(__global float *)(dst_ptr + mult_y.s0 * 2 * dst_stride_y + offset.s3) = out_col2.s3;
+    *(__global float *)(dst_ptr + mult_y.s0 * 3 * dst_stride_y + offset.s3) = out_col3.s3;
+}
+#endif // defined(NUM_TILES_X) && defined(OUTPUT_TILE_W) && defined(OUTPUT_TILE_H)
\ No newline at end of file