Compute Library
 21.02
ClReshapeKernel.cpp
Go to the documentation of this file.
1 /*
2  * Copyright (c) 2017-2021 Arm Limited.
3  *
4  * SPDX-License-Identifier: MIT
5  *
6  * Permission is hereby granted, free of charge, to any person obtaining a copy
7  * of this software and associated documentation files (the "Software"), to
8  * deal in the Software without restriction, including without limitation the
9  * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
10  * sell copies of the Software, and to permit persons to whom the Software is
11  * furnished to do so, subject to the following conditions:
12  *
13  * The above copyright notice and this permission notice shall be included in all
14  * copies or substantial portions of the Software.
15  *
16  * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
17  * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
18  * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
19  * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
20  * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
21  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
22  * SOFTWARE.
23  */
25 
33 #include "arm_compute/core/Utils.h"
35 #include "src/core/CL/CLValidate.h"
37 #include "support/Cast.h"
38 
39 #include <string>
40 
41 /** [ClReshapeKernel Kernel] **/
42 namespace arm_compute
43 {
44 namespace opencl
45 {
46 namespace kernels
47 {
48 namespace
49 {
50 Status validate_arguments(const ITensorInfo *src, const ITensorInfo *dst)
51 {
55 
58  ARM_COMPUTE_RETURN_ERROR_ON(src->tensor_shape().total_size() != dst->tensor_shape().total_size());
59 
60  return Status{};
61 }
62 } // namespace
63 
64 void ClReshapeKernel::configure(const CLCompileContext &compile_context, const ITensorInfo *src, ITensorInfo *dst)
65 {
68 
69  auto padding_info = get_padding_info({ src, dst });
70 
71  // Create kernel
72  std::set<std::string> build_opts = { "-DDATA_TYPE=" + get_cl_unsigned_type_from_element_size(src->element_size()) };
73  _kernel = create_kernel(compile_context, "reshape_layer", build_opts);
74 
75  // Add static arguments
76  const cl_int2 src_shape =
77  {
78  {
79  static_cast<cl_int>(src->tensor_shape()[0]),
80  static_cast<cl_int>(src->tensor_shape()[1])
81  }
82  };
83  const cl_int2 dst_shape =
84  {
85  {
86  static_cast<cl_int>(dst->tensor_shape()[0]),
87  static_cast<cl_int>(dst->tensor_shape()[1])
88  }
89  };
90  unsigned int idx = 2 * num_arguments_per_3D_tensor(); // Skip the src and dst parameters
91  _kernel.setArg<cl_int2>(idx++, src_shape);
92  _kernel.setArg<cl_int2>(idx++, dst_shape);
93 
94  // Configure kernel window
95  Window win = calculate_max_window(*src);
96 
97  // Set the dst valid region
99  ICLKernel::configure_internal(win);
100 
102 }
103 
105 {
107 
108  return Status{};
109 }
110 
111 void ClReshapeKernel::run_op(ITensorPack &tensors, const Window &window, cl::CommandQueue &queue)
112 {
115 
116  Window window_collapsed = window.collapse_if_possible(ICLKernel::window(), Window::DimZ);
117  Window slice = window_collapsed.first_slice_window_3D();
118 
119  const auto src = utils::cast::polymorphic_downcast<const ICLTensor *>(tensors.get_const_tensor(TensorType::ACL_SRC));
120  auto dst = utils::cast::polymorphic_downcast<ICLTensor *>(tensors.get_tensor(TensorType::ACL_DST));
121 
122  // Set srcs
123  unsigned int idx = 0;
124  add_3D_tensor_argument(idx, src, window_collapsed);
125  add_3D_tensor_argument(idx, dst, window_collapsed);
126  enqueue(queue, *this, slice, lws_hint());
127 }
128 } // namespace kernels
129 } // namespace opencl
130 } // namespace arm_compute
131 /** [ClReshapeKernel Kernel] **/
static Status validate(const ITensorInfo *src, const ITensorInfo *dst)
Static function to check if given info will lead to a valid configuration of ClReshapeKernel.
Window calculate_max_window(const ValidRegion &valid_region, const Steps &steps, bool skip_border, BorderSize border_size)
#define ARM_COMPUTE_RETURN_ERROR_ON_F16_UNSUPPORTED(tensor)
Definition: CLValidate.h:35
const Window & window() const
The maximum window the kernel can be executed on.
Definition: IKernel.cpp:28
void enqueue(IGCKernel &kernel, const Window &window, const gles::NDRange &lws=gles::NDRange(1U, 1U, 1U))
Add the kernel to the command queue with the given window.
Definition: IGCKernel.cpp:41
#define ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_QUANTIZATION_INFO(...)
Definition: Validate.h:610
cl::NDRange lws_hint() const
Return the Local-Workgroup-Size hint.
Definition: ICLKernel.h:276
#define ARM_COMPUTE_RETURN_ON_ERROR(status)
Checks if a status contains an error and returns it.
Definition: Error.h:204
#define ARM_COMPUTE_ERROR_ON(cond)
If the condition is true then an error message is printed and an exception thrown.
Definition: Error.h:466
Store the tensor&#39;s metadata.
Definition: ITensorInfo.h:40
#define ARM_COMPUTE_ERROR_THROW_ON(status)
Definition: Error.h:455
Status class.
Definition: Error.h:52
#define ARM_COMPUTE_RETURN_ERROR_ON(cond)
If the condition is true, an error is returned.
Definition: Error.h:296
void add_3D_tensor_argument(unsigned int &idx, const ICLTensor *tensor, const Window &window)
Add the passed 3D tensor&#39;s parameters to the object&#39;s kernel&#39;s arguments starting from the index idx...
Definition: ICLKernel.h:172
SimpleTensor< float > src
Definition: DFT.cpp:155
Copyright (c) 2017-2021 Arm Limited.
virtual void set_valid_region(const ValidRegion &valid_region)=0
Set the valid region of the tensor.
void run_op(ITensorPack &tensors, const Window &window, cl::CommandQueue &queue) override
Enqueue the OpenCL kernel to process the given window on the passed OpenCL command queue...
#define ARM_COMPUTE_RETURN_ERROR_ON_NULLPTR(...)
Definition: Validate.h:163
const ITensor * get_const_tensor(int id) const
Get constant tensor of a given id.
Definition: ITensorPack.cpp:40
cl::Kernel create_kernel(const CLCompileContext &ctx, const std::string &kernel_name, const std::set< std::string > &build_opts=std::set< std::string >())
Creates an opencl kernel using a compile context.
Definition: CLHelpers.cpp:403
static constexpr unsigned int num_arguments_per_3D_tensor()
Returns the number of arguments enqueued per 3D tensor object.
Definition: ICLKernel.h:214
Window collapse_if_possible(const Window &full_window, size_t first, size_t last, bool *has_collapsed=nullptr) const
Collapse the dimensions between first and last if possible.
Definition: Window.inl:68
virtual const TensorShape & tensor_shape() const =0
Size for each dimension of the tensor.
Coordinates of an item.
Definition: Coordinates.h:37
void configure(const CLCompileContext &compile_context, const ITensorInfo *src, ITensorInfo *dst)
Set the src and dst of the kernel.
virtual size_t element_size() const =0
Element size in bytes calculated as data_size() * num_channels()
#define ARM_COMPUTE_ERROR_ON_UNCONFIGURED_KERNEL(k)
Definition: Validate.h:941
bool has_padding_changed(const std::unordered_map< const ITensorInfo *, PaddingSize > &padding_map)
Check if the previously stored padding info has changed after configuring a kernel.
Definition: Utils.cpp:528
CLCompileContext class.
ITensor * get_tensor(int id)
Get tensor of a given id from the pac.
Definition: ITensorPack.cpp:50
static constexpr size_t DimZ
Alias for dimension 2 also known as Z dimension.
Definition: Window.h:47
Manages all the OpenCL kernels compilation and caching, provides accessors for the OpenCL Context...
#define ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_DATA_TYPES(...)
Definition: Validate.h:545
std::string get_cl_unsigned_type_from_element_size(size_t element_size)
Translates the element size to an unsigned integer data type.
Definition: CLHelpers.cpp:103
std::unordered_map< const ITensorInfo *, PaddingSize > get_padding_info(std::initializer_list< const ITensorInfo *> infos)
Stores padding information before configuring a kernel.
Definition: Utils.cpp:513
Status validate_arguments(const ITensorInfo *input, const ITensorInfo *bias, const ITensorInfo *output, const GEMMLowpOutputStageInfo *output_stage)
Wrapper to configure the Khronos OpenCL C++ header.
Tensor packing service.
Definition: ITensorPack.h:37
#define ARM_COMPUTE_ERROR_ON_NULLPTR(...)
Definition: Validate.h:161
Container for valid region of a window.
Definition: Types.h:188
Window first_slice_window_3D() const
First 3D slice of the window.
Definition: Window.h:291
Describe a multidimensional execution window.
Definition: Window.h:39
#define ARM_COMPUTE_ERROR_ON_INVALID_SUBWINDOW(f, s)
Definition: Validate.h:205
SimpleTensor< T > slice(const SimpleTensor< T > &src, Coordinates starts, Coordinates ends)