1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
|
/*
* Copyright (c) 2017-2021 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 "src/core/gpu/cl/kernels/ClReshapeKernel.h"
#include "arm_compute/core/CL/CLHelpers.h"
#include "arm_compute/core/CL/CLKernelLibrary.h"
#include "arm_compute/core/CL/ICLTensor.h"
#include "arm_compute/core/CL/OpenCL.h"
#include "arm_compute/core/Helpers.h"
#include "arm_compute/core/IAccessWindow.h"
#include "arm_compute/core/TensorInfo.h"
#include "arm_compute/core/Utils.h"
#include "src/core/AccessWindowStatic.h"
#include "src/core/CL/CLValidate.h"
#include "src/core/helpers/WindowHelpers.h"
#include "support/Cast.h"
#include <string>
/** [ClReshapeKernel Kernel] **/
namespace arm_compute
{
namespace opencl
{
namespace kernels
{
namespace
{
Status validate_arguments(const ITensorInfo *src, const ITensorInfo *dst)
{
ARM_COMPUTE_RETURN_ERROR_ON_NULLPTR(src, dst);
ARM_COMPUTE_RETURN_ERROR_ON_F16_UNSUPPORTED(src);
ARM_COMPUTE_RETURN_ERROR_ON(src->data_type() == DataType::UNKNOWN);
ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_DATA_TYPES(src, dst);
ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_QUANTIZATION_INFO(src, dst);
ARM_COMPUTE_RETURN_ERROR_ON(src->tensor_shape().total_size() != dst->tensor_shape().total_size());
return Status{};
}
} // namespace
void ClReshapeKernel::configure(const CLCompileContext &compile_context, const ITensorInfo *src, ITensorInfo *dst)
{
ARM_COMPUTE_ERROR_ON_NULLPTR(src, dst);
ARM_COMPUTE_ERROR_THROW_ON(validate_arguments(src, dst));
auto padding_info = get_padding_info({ src, dst });
// Create kernel
std::set<std::string> build_opts = { "-DDATA_TYPE=" + get_cl_unsigned_type_from_element_size(src->element_size()) };
_kernel = create_kernel(compile_context, "reshape_layer", build_opts);
// Add static arguments
const cl_int2 src_shape =
{
{
static_cast<cl_int>(src->tensor_shape()[0]),
static_cast<cl_int>(src->tensor_shape()[1])
}
};
const cl_int2 dst_shape =
{
{
static_cast<cl_int>(dst->tensor_shape()[0]),
static_cast<cl_int>(dst->tensor_shape()[1])
}
};
unsigned int idx = 2 * num_arguments_per_3D_tensor(); // Skip the src and dst parameters
_kernel.setArg<cl_int2>(idx++, src_shape);
_kernel.setArg<cl_int2>(idx++, dst_shape);
// Configure kernel window
Window win = calculate_max_window(*src);
// Set the dst valid region
dst->set_valid_region(ValidRegion(Coordinates(), dst->tensor_shape()));
ICLKernel::configure_internal(win);
ARM_COMPUTE_ERROR_ON(has_padding_changed(padding_info));
}
Status ClReshapeKernel::validate(const ITensorInfo *src, const ITensorInfo *dst)
{
ARM_COMPUTE_RETURN_ON_ERROR(validate_arguments(src, dst));
return Status{};
}
void ClReshapeKernel::run_op(ITensorPack &tensors, const Window &window, cl::CommandQueue &queue)
{
ARM_COMPUTE_ERROR_ON_UNCONFIGURED_KERNEL(this);
ARM_COMPUTE_ERROR_ON_INVALID_SUBWINDOW(IKernel::window(), window);
Window window_collapsed = window.collapse_if_possible(ICLKernel::window(), Window::DimZ);
Window slice = window_collapsed.first_slice_window_3D();
const auto src = utils::cast::polymorphic_downcast<const ICLTensor *>(tensors.get_const_tensor(TensorType::ACL_SRC));
auto dst = utils::cast::polymorphic_downcast<ICLTensor *>(tensors.get_tensor(TensorType::ACL_DST));
// Set srcs
unsigned int idx = 0;
add_3D_tensor_argument(idx, src, window_collapsed);
add_3D_tensor_argument(idx, dst, window_collapsed);
enqueue(queue, *this, slice, lws_hint());
}
} // namespace kernels
} // namespace opencl
} // namespace arm_compute
/** [ClReshapeKernel Kernel] **/
|
