aboutsummaryrefslogtreecommitdiff
path: root/src/core/CL/kernels/CLAccumulateKernel.cpp
blob: a7dfcdca9ba9649c3d09eb977cc2d5e22f2fddec (plain)
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
/*
 * Copyright (c) 2016-2020 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 "arm_compute/core/CL/kernels/CLAccumulateKernel.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/Error.h"
#include "arm_compute/core/IAccessWindow.h"
#include "arm_compute/core/Types.h"
#include "arm_compute/core/Validate.h"

namespace arm_compute
{
namespace
{
constexpr unsigned int num_elems_processed_per_iteration = 16;
} // namespace

void CLAccumulateKernel::configure(const ICLTensor *input, ICLTensor *accum)
{
    configure(CLKernelLibrary::get().get_compile_context(), input, accum);
}

void CLAccumulateKernel::configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *accum)
{
    ARM_COMPUTE_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(input, 1, DataType::U8);
    ARM_COMPUTE_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(accum, 1, DataType::S16);

    // Create kernel
    _kernel = create_kernel(compile_context, "accumulate");

    // Make sure _kernel is initialized before calling the parent's configure
    ICLSimple2DKernel::configure(input, accum, num_elems_processed_per_iteration);
}

void CLAccumulateWeightedKernel::configure(const ICLTensor *input, float alpha, ICLTensor *accum)
{
    configure(CLKernelLibrary::get().get_compile_context(), input, alpha, accum);
}

void CLAccumulateWeightedKernel::configure(const CLCompileContext &compile_context, const ICLTensor *input, float alpha, ICLTensor *accum)
{
    ARM_COMPUTE_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(input, 1, DataType::U8);
    ARM_COMPUTE_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(accum, 1, DataType::U8);
    ARM_COMPUTE_ERROR_ON(alpha < 0.0 || alpha > 1.0);

    // Create kernel
    _kernel = create_kernel(compile_context, "accumulate_weighted");

    // Set static kernel arguments
    unsigned int idx = 2 * num_arguments_per_2D_tensor(); //Skip the input and output parameters
    _kernel.setArg(idx++, alpha);

    // Configure kernel window
    ICLSimple2DKernel::configure(input, accum, num_elems_processed_per_iteration);
}

void CLAccumulateSquaredKernel::configure(const ICLTensor *input, uint32_t shift, ICLTensor *accum)
{
    configure(CLKernelLibrary::get().get_compile_context(), input, shift, accum);
}

void CLAccumulateSquaredKernel::configure(const CLCompileContext &compile_context, const ICLTensor *input, uint32_t shift, ICLTensor *accum)
{
    ARM_COMPUTE_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(input, 1, DataType::U8);
    ARM_COMPUTE_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(accum, 1, DataType::S16);
    ARM_COMPUTE_ERROR_ON(shift > 15);

    // Create kernel
    _kernel = create_kernel(compile_context, "accumulate_squared");

    // Set static kernel arguments
    unsigned int idx = 2 * num_arguments_per_2D_tensor(); //Skip the input and output parameters
    _kernel.setArg(idx++, shift);

    // Configure kernel window
    ICLSimple2DKernel::configure(input, accum, num_elems_processed_per_iteration);
}
} // namespace arm_compute