1 /*
2 * Copyright (c) 2018-2022 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/gpu/cl/kernels/ClWidthConcatenateKernel.h"
25
26 #include "arm_compute/core/CL/CLHelpers.h"
27 #include "arm_compute/core/CL/CLKernelLibrary.h"
28 #include "arm_compute/core/CL/ICLTensor.h"
29 #include "arm_compute/core/Helpers.h"
30 #include "arm_compute/core/Utils.h"
31 #include "src/core/CL/CLValidate.h"
32 #include "src/core/helpers/WindowHelpers.h"
33 #include "support/Cast.h"
34
35 #include "support/StringSupport.h"
36
37 namespace arm_compute
38 {
39 namespace opencl
40 {
41 namespace kernels
42 {
43 namespace
44 {
validate_arguments(const ITensorInfo * src,unsigned int width_offset,const ITensorInfo * dst)45 Status validate_arguments(const ITensorInfo *src, unsigned int width_offset, const ITensorInfo *dst)
46 {
47 ARM_COMPUTE_RETURN_ERROR_ON_NULLPTR(src, dst);
48 ARM_COMPUTE_RETURN_ERROR_ON_F16_UNSUPPORTED(src);
49 ARM_COMPUTE_RETURN_ERROR_ON(src->data_type() == DataType::UNKNOWN);
50
51 ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_DATA_TYPES(src, dst);
52 ARM_COMPUTE_RETURN_ERROR_ON(src->dimension(0) + width_offset > dst->dimension(0));
53
54 for(size_t i = 1; i < Coordinates::num_max_dimensions; ++i)
55 {
56 ARM_COMPUTE_RETURN_ERROR_ON(src->dimension(i) != dst->dimension(i));
57 }
58 ARM_COMPUTE_RETURN_ERROR_ON(src->num_dimensions() > 4);
59
60 return Status{};
61 }
62 } // namespace
63
ClWidthConcatenateKernel()64 ClWidthConcatenateKernel::ClWidthConcatenateKernel()
65 {
66 _type = CLKernelType::ELEMENTWISE;
67 }
68
validate(const ITensorInfo * src,unsigned int width_offset,const ITensorInfo * dst)69 Status ClWidthConcatenateKernel::validate(const ITensorInfo *src, unsigned int width_offset, const ITensorInfo *dst)
70 {
71 ARM_COMPUTE_RETURN_ON_ERROR(validate_arguments(src, width_offset, dst));
72 return Status{};
73 }
74
configure(const CLCompileContext & compile_context,ITensorInfo * src,unsigned int width_offset,ITensorInfo * dst)75 void ClWidthConcatenateKernel::configure(const CLCompileContext &compile_context, ITensorInfo *src, unsigned int width_offset, ITensorInfo *dst)
76 {
77 ARM_COMPUTE_ERROR_ON_NULLPTR(src, dst);
78 ARM_COMPUTE_ERROR_THROW_ON(validate_arguments(src, width_offset, dst));
79
80 auto padding_info = get_padding_info({ src, dst });
81
82 const unsigned int num_elems_processed_per_iteration = adjust_vec_size(16, src->dimension(0));
83
84 // Add build options
85 CLBuildOptions build_opts;
86 build_opts.add_option("-DDATA_TYPE=" + get_cl_type_from_data_type(src->data_type()));
87 build_opts.add_option("-DVEC_SIZE=" + support::cpp11::to_string(num_elems_processed_per_iteration));
88 build_opts.add_option("-DVEC_SIZE_LEFTOVER=" + support::cpp11::to_string(src->dimension(0) % num_elems_processed_per_iteration));
89 build_opts.add_option("-DWIDTH_OFFSET=" + support::cpp11::to_string(width_offset));
90
91 if(is_data_type_quantized_asymmetric(src->data_type()) && src->quantization_info() != dst->quantization_info())
92 {
93 const UniformQuantizationInfo iqinfo = src->quantization_info().uniform();
94 const UniformQuantizationInfo oqinfo = dst->quantization_info().uniform();
95
96 build_opts.add_option("-DOFFSET_IN1=" + float_to_string_with_full_precision(iqinfo.offset));
97 build_opts.add_option("-DOFFSET_OUT=" + float_to_string_with_full_precision(oqinfo.offset));
98 build_opts.add_option("-DSCALE_IN1=" + float_to_string_with_full_precision(iqinfo.scale));
99 build_opts.add_option("-DSCALE_OUT=" + float_to_string_with_full_precision(oqinfo.scale));
100 }
101 _depth = src->dimension(2);
102 std::string kernel_name = "concatenate_width";
103
104 // A macro guard to compile ONLY the kernel of interest
105 build_opts.add_option("-D" + upper_string(kernel_name));
106
107 // Create kernel
108 _kernel = create_kernel(compile_context, kernel_name, build_opts.options());
109
110 // Configure kernel window
111 Window win = calculate_max_window(*src, Steps(num_elems_processed_per_iteration));
112 ICLKernel::configure_internal(win.collapse(win, Window::DimZ));
113
114 ARM_COMPUTE_ERROR_ON(has_padding_changed(padding_info));
115 }
116
run_op(ITensorPack & tensors,const Window & window,::cl::CommandQueue & queue)117 void ClWidthConcatenateKernel::run_op(ITensorPack &tensors, const Window &window, ::cl::CommandQueue &queue)
118 {
119 ARM_COMPUTE_ERROR_ON_UNCONFIGURED_KERNEL(this);
120 ARM_COMPUTE_ERROR_ON_INVALID_SUBWINDOW(ICLKernel::window(), window);
121
122 const auto src = utils::cast::polymorphic_downcast<const ICLTensor *>(tensors.get_const_tensor(TensorType::ACL_SRC));
123 auto dst = utils::cast::polymorphic_downcast<ICLTensor *>(tensors.get_tensor(TensorType::ACL_DST));
124
125 unsigned int idx = 0;
126 add_4D_tensor_argument(idx, src, window);
127 add_4D_tensor_argument(idx, dst, window);
128 _kernel.setArg<cl_uint>(idx++, _depth);
129 enqueue(queue, *this, window, lws_hint());
130 }
131 } // namespace kernels
132 } // namespace opencl
133 } // namespace arm_compute
134