1 /*
2 * Copyright (c) 2018-2021 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 */
24 #include "src/core/NEON/kernels/NEStackLayerKernel.h"
25
26 #include "arm_compute/core/Error.h"
27 #include "arm_compute/core/Helpers.h"
28 #include "arm_compute/core/Helpers.h"
29 #include "arm_compute/core/ITensor.h"
30 #include "arm_compute/core/TensorInfo.h"
31 #include "arm_compute/core/Utils.h"
32 #include "arm_compute/core/Validate.h"
33 #include "arm_compute/core/Window.h"
34 #include "arm_compute/core/utils/misc/ShapeCalculator.h"
35 #include "src/core/helpers/AutoConfiguration.h"
36 #include "src/core/helpers/WindowHelpers.h"
37
38 namespace arm_compute
39 {
40 using namespace arm_compute::misc::shape_calculator;
41
42 namespace
43 {
validate_arguments(const ITensorInfo * input,unsigned int axis,unsigned int idx_input,unsigned int num_tensors,const ITensorInfo * output)44 Status validate_arguments(const ITensorInfo *input, unsigned int axis, unsigned int idx_input, unsigned int num_tensors, const ITensorInfo *output)
45 {
46 ARM_COMPUTE_RETURN_ERROR_ON_NULLPTR(input, output);
47 // Note: ARM_COMPUTE_RETURN_ERROR_ON_CPU_F16_UNSUPPORTED(input) is not needed here as this kernel doesn't use CPU FP16 instructions.
48 ARM_COMPUTE_RETURN_ERROR_ON(input->data_type() == DataType::UNKNOWN);
49 ARM_COMPUTE_RETURN_ERROR_ON(idx_input >= num_tensors);
50 ARM_COMPUTE_RETURN_ERROR_ON(axis > input->num_dimensions());
51 ARM_COMPUTE_RETURN_ERROR_ON(input->num_dimensions() > 4);
52
53 if(output->total_size() != 0)
54 {
55 ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_DIMENSIONS(output->tensor_shape(), compute_stack_shape(*input, axis, num_tensors));
56 ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_DATA_TYPES(input, output);
57 ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_QUANTIZATION_INFO(input, output);
58 }
59
60 return Status{};
61 }
62
validate_and_configure_window(ITensorInfo * input,unsigned int axis,unsigned int num_tensors,ITensorInfo * output)63 std::pair<Status, Window> validate_and_configure_window(ITensorInfo *input, unsigned int axis, unsigned int num_tensors, ITensorInfo *output)
64 {
65 // Output auto inizialitation if not yet initialized
66 auto_init_if_empty(*output, input->clone()->set_tensor_shape(compute_stack_shape(*input, axis, num_tensors)));
67
68 // Configure kernel window
69 Window win = calculate_max_window(*input);
70
71 return std::make_pair(Status{}, win);
72 }
73
shift_from_axis_and_replace_coordinate(const Coordinates & id,unsigned int axis,unsigned int idx_input)74 inline Coordinates shift_from_axis_and_replace_coordinate(const Coordinates &id, unsigned int axis, unsigned int idx_input)
75 {
76 constexpr int max_out_coord = 5; // Input shape is max a 4D shape, output is max 5D
77 Coordinates id_out = id;
78 for(unsigned int i = max_out_coord - 1; i > axis; --i)
79 {
80 id_out.set(i, id[i - 1]);
81 }
82 id_out.set(axis, idx_input);
83 return id_out;
84 }
85 } // namespace
86
NEStackLayerKernel()87 NEStackLayerKernel::NEStackLayerKernel()
88 : _input(nullptr), _output(nullptr), _axis(), _idx_input()
89 {
90 }
91
configure(const ITensor * input,unsigned int axis,unsigned int idx_input,unsigned int num_tensors,ITensor * output)92 void NEStackLayerKernel::configure(const ITensor *input, unsigned int axis, unsigned int idx_input, unsigned int num_tensors, ITensor *output)
93 {
94 ARM_COMPUTE_ERROR_ON_NULLPTR(input, output);
95 ARM_COMPUTE_ERROR_THROW_ON(validate_arguments(input->info(), axis, idx_input, num_tensors, output->info()));
96
97 _input = input;
98 _output = output;
99 _axis = axis;
100 _idx_input = idx_input;
101
102 // Configure kernel window
103 auto win_config = validate_and_configure_window(input->info(), axis, num_tensors, output->info());
104
105 ARM_COMPUTE_ERROR_THROW_ON(win_config.first);
106 INEKernel::configure(win_config.second);
107 }
108
validate(const ITensorInfo * input,unsigned int axis,unsigned int idx_input,unsigned int num_tensors,const ITensorInfo * output)109 Status NEStackLayerKernel::validate(const ITensorInfo *input, unsigned int axis, unsigned int idx_input, unsigned int num_tensors, const ITensorInfo *output)
110 {
111 ARM_COMPUTE_RETURN_ON_ERROR(validate_arguments(input, axis, idx_input, num_tensors, output));
112 ARM_COMPUTE_RETURN_ON_ERROR(validate_and_configure_window(input->clone().get(), axis, num_tensors, output->clone().get()).first);
113 return Status{};
114 }
115
run(const Window & window,const ThreadInfo & info)116 void NEStackLayerKernel::run(const Window &window, const ThreadInfo &info)
117 {
118 ARM_COMPUTE_UNUSED(info);
119 ARM_COMPUTE_ERROR_ON_UNCONFIGURED_KERNEL(this);
120 ARM_COMPUTE_ERROR_ON_INVALID_SUBWINDOW(INEKernel::window(), window);
121
122 Window window_out;
123 window_out.use_tensor_dimensions(_output->info()->tensor_shape());
124
125 Iterator input(_input, window);
126 Iterator output(_output, window_out);
127
128 const int stride_x = _output->info()->strides_in_bytes()[0];
129 const int stride_y = _output->info()->num_dimensions() >= 1 ? _output->info()->strides_in_bytes()[1] : 0;
130 const int stride_z = _output->info()->num_dimensions() >= 2 ? _output->info()->strides_in_bytes()[2] : 0;
131 const int stride_w = _output->info()->num_dimensions() >= 3 ? _output->info()->strides_in_bytes()[3] : 0;
132 const int stride_k = _output->info()->num_dimensions() >= 4 ? _output->info()->strides_in_bytes()[4] : 0;
133
134 execute_window_loop(window, [&](const Coordinates & id)
135 {
136 Coordinates id_out = shift_from_axis_and_replace_coordinate(id, _axis, _idx_input);
137 const int idx = id_out[0] * stride_x + id_out[1] * stride_y + id_out[2] * stride_z + id_out[3] * stride_w + id_out[4] * stride_k;
138 std::memcpy(output.ptr() + idx, input.ptr(), _input->info()->element_size());
139 },
140 input);
141 }
142 } // namespace arm_compute
143