Reorganize the kernels into nhwc, nchw and common folders

The Following kernels have been split into nchw/nhwc kernels files:

- batchnormalization_layer
- batch_to_space
- channel_shuffle
- depth_to_space
- dequantization_layer
- im2col
- normalization_layer
- normalize_planar_yuv_layer
- normalize_planar_yuv_layer_quantized
- pooling_layer
- pooling_layer_quantized
- remap
- reorg_layer
- scale
- scale_quantized
- space_to_batch
- space_to_depth
- upsample_layer
- winograd_filter_transform
- winograd_input_transform
- winograd_output_transform

The following kernels have been moved to nchw folder:
- direct_convolution1x1
- direct_convolution3x3
- direct_convolution5x5
- direct_convolution_quantized
- prior_box_layer

The following kernels have been moved to nhwc folder:
- direct_convolution
- dwc_native_fp_nhwc
- dwc_native_quantized_nhwc

The following kernels have been removed:
- sobel_filter

While the rest kerenls have been moved to the common folder.

Partially resolves COMPMID-4453

Signed-off-by: Adnan AlSinan <adnan.alsinan@arm.com>
Change-Id: Ic327ac935687ec351c610c65a3c6357f364a5a58
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/5919
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>

1 files changed, 0 insertions, 111 deletions

diff --git a/src/core/CL/cl_kernels/depth_to_space.cl b/src/core/CL/cl_kernels/depth_to_space.cl
deleted file mode 100644
index f301e64d66..0000000000
--- a/src/core/CL/cl_kernels/depth_to_space.cl
+++ /dev/null
@@ -1,111 +0,0 @@
-/*
- * Copyright (c) 2019-2021 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(DATA_TYPE) && defined(BLOCK_SHAPE) && defined(CHANNEL_SIZE)
-/** Depth to space transformation. (NCHW)
- *
- * @note Datatype should be given as a preprocessor argument using -DDATA_TYPE=type. e.g. -DDATA_TYPE=float
- * @note The input tensor depth size must be passed at compile time using -DCHANNEL_SIZE. e.g. -DCHANNEL_SIZE=2
- * @note The block shape must be passed at compile time using -DBLOCK_SHAPE. e.g. -DBLOCK_SHAPE=2
- *
- * @param[in]  input_ptr                            Pointer to the source tensor. Supported data types: All.
- * @param[in]  input_stride_x                       Stride of the source tensor in X dimension (in bytes)
- * @param[in]  input_step_x                         input_stride_x * number of elements along X processed per workitem(in bytes)
- * @param[in]  input_stride_y                       Stride of the source tensor in Y dimension (in bytes)
- * @param[in]  input_step_y                         input_stride_y * number of elements along Y processed per workitem(in bytes)
- * @param[in]  input_stride_z                       Stride of the source tensor in Z dimension (in bytes)
- * @param[in]  input_step_z                         input_stride_z * number of elements along Z processed per workitem(in bytes)
- * @param[in]  input_offset_first_element_in_bytes  The offset of the first element in the first source tensor
- * @param[in]  batch_id                             The input tensor batch id
- * @param[out] output_ptr                           Pointer to the destination tensor. Supported data types: same as @p input_ptr
- * @param[in]  output_stride_x                      Stride of the destination tensor in X dimension (in bytes)
- * @param[in]  output_step_x                        output_stride_x * number of elements along X processed per workitem(in bytes)
- * @param[in]  output_stride_y                      Stride of the destination tensor in Y dimension (in bytes)
- * @param[in]  output_step_y                        output_stride_y * number of elements along Y processed per workitem(in bytes)
- * @param[in]  output_stride_z                      Stride of the source tensor in Z dimension (in bytes)
- * @param[in]  output_step_z                        output_stride_z * number of elements along Z processed per workitem(in bytes)
- * @param[in]  output_offset_first_element_in_bytes The offset of the first element in the destination tensor
- */
-__kernel void depth_to_space_nchw(
-    TENSOR3D_DECLARATION(input),
-    const int batch_id,
-    TENSOR4D_DECLARATION(output))
-{
-    Tensor3D in  = CONVERT_TO_TENSOR3D_STRUCT(input);
-    Tensor4D out = CONVERT_TO_TENSOR4D_STRUCT_NO_STEP(output, 0);
-
-    const int r = (CHANNEL_SIZE / (BLOCK_SHAPE * BLOCK_SHAPE));
-    const int x = get_global_id(0);
-    const int y = get_global_id(1);
-    const int z = get_global_id(2) % r;
-
-    const int out_x = x * BLOCK_SHAPE + (get_global_id(2) / r) % BLOCK_SHAPE;
-    const int out_y = y * BLOCK_SHAPE + (get_global_id(2) / r) / BLOCK_SHAPE;
-
-    *((__global DATA_TYPE *)tensor4D_offset(&out, out_x, out_y, z, batch_id)) = *((__global DATA_TYPE *)in.ptr);
-}
-/** Depth to space transformation. (NHWC)
- *
- * @note Datatype should be given as a preprocessor argument using -DDATA_TYPE=type. e.g. -DDATA_TYPE=float
- * @note The input tensor depth size must be passed at compile time using -DCHANNEL_SIZE. e.g. -DCHANNEL_SIZE=2
- * @note The block shape must be passed at compile time using -DBLOCK_SHAPE. e.g. -DBLOCK_SHAPE=2
- *
- * @param[in]  input_ptr                            Pointer to the source tensor. Supported data types: All.
- * @param[in]  input_stride_x                       Stride of the source tensor in X dimension (in bytes)
- * @param[in]  input_step_x                         input_stride_x * number of elements along X processed per workitem(in bytes)
- * @param[in]  input_stride_y                       Stride of the source tensor in Y dimension (in bytes)
- * @param[in]  input_step_y                         input_stride_y * number of elements along Y processed per workitem(in bytes)
- * @param[in]  input_stride_z                       Stride of the source tensor in Z dimension (in bytes)
- * @param[in]  input_step_z                         input_stride_z * number of elements along Z processed per workitem(in bytes)
- * @param[in]  input_offset_first_element_in_bytes  The offset of the first element in the first source tensor
- * @param[in]  batch_id                             The input tensor batch id
- * @param[out] output_ptr                           Pointer to the destination tensor. Supported data types: same as @p input_ptr
- * @param[in]  output_stride_x                      Stride of the destination tensor in X dimension (in bytes)
- * @param[in]  output_step_x                        output_stride_x * number of elements along X processed per workitem(in bytes)
- * @param[in]  output_stride_y                      Stride of the destination tensor in Y dimension (in bytes)
- * @param[in]  output_step_y                        output_stride_y * number of elements along Y processed per workitem(in bytes)
- * @param[in]  output_stride_z                      Stride of the source tensor in Z dimension (in bytes)
- * @param[in]  output_step_z                        output_stride_z * number of elements along Z processed per workitem(in bytes)
- * @param[in]  output_offset_first_element_in_bytes The offset of the first element in the destination tensor
- */
-__kernel void depth_to_space_nhwc(
-    TENSOR3D_DECLARATION(input),
-    const int batch_id,
-    TENSOR4D_DECLARATION(output))
-{
-    Tensor3D in  = CONVERT_TO_TENSOR3D_STRUCT(input);
-    Tensor4D out = CONVERT_TO_TENSOR4D_STRUCT_NO_STEP(output, 0);
-
-    const int r = (CHANNEL_SIZE / (BLOCK_SHAPE * BLOCK_SHAPE));
-    const int x = get_global_id(1);
-    const int y = get_global_id(2);
-    const int z = get_global_id(0) % r;
-
-    const int out_x = x * BLOCK_SHAPE + (get_global_id(0) / r) % BLOCK_SHAPE;
-    const int out_y = y * BLOCK_SHAPE + (get_global_id(0) / r) / BLOCK_SHAPE;
-
-    *((__global DATA_TYPE *)tensor4D_offset(&out, z, out_x, out_y, batch_id)) = *((__global DATA_TYPE *)in.ptr);
-}
-#endif // defined(DATA_TYPE) && defined(BLOCK_SHAPE) && defined(CHANNEL_SIZE)
\ No newline at end of file