aboutsummaryrefslogtreecommitdiff
path: root/arm_compute/core/CL/CLHelpers.h
blob: 26253e3f38cfaed90c4a1a89bc6564014c62964f (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
102
103
104
105
/*
 * Copyright (c) 2016, 2017 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.
 */
#ifndef __ARM_COMPUTE_CLHELPERS_H__
#define __ARM_COMPUTE_CLHELPERS_H__

#include "arm_compute/core/CL/OpenCL.h"
#include "arm_compute/core/Helpers.h"

#include <string>

namespace arm_compute
{
enum class DataType;
enum class GPUTarget;

/** Enable operation operations on GPUTarget enumerations */
template <>
struct enable_bitwise_ops<arm_compute::GPUTarget>
{
    static constexpr bool value = true;
};

/** Max vector width of an OpenCL vector */
static constexpr const unsigned int max_cl_vector_width = 16;

/** Translates a tensor data type to the appropriate OpenCL type.
 *
 * @param[in] dt @ref DataType to be translated to OpenCL type.
 *
 * @return The string specifying the OpenCL type to be used.
 */
std::string get_cl_type_from_data_type(const DataType &dt);

/** Translates a given gpu device target to string.
 *
 * @param[in] target Given gpu target.
 *
 * @return The string describing the target.
 */
const std::string &string_from_target(GPUTarget target);

/** Helper function to create and return a unique_ptr pointed to a CL kernel object
 *  It also calls the kernel's configuration.
 *
 * @param[in] args All the arguments that need pass to kernel's configuration.
 *
 * @return A unique pointer pointed to a CL kernel object
 */
template <typename Kernel, typename... T>
std::unique_ptr<Kernel> create_configure_kernel(T &&... args)
{
    std::unique_ptr<Kernel> k = arm_compute::cpp14::make_unique<Kernel>();
    k->configure(std::forward<T>(args)...);
    return k;
}

/** Helper function to create and return a unique_ptr pointed to a CL kernel object
 *
 * @return A unique pointer pointed to a CL kernel object
 */
template <typename Kernel>
std::unique_ptr<Kernel> create_kernel()
{
    std::unique_ptr<Kernel> k = arm_compute::cpp14::make_unique<Kernel>();
    return k;
}

/** Helper function to get the GPU target from CL device
 *
 * @param[in] device A CL device
 *
 * @return the GPU target
 */
GPUTarget get_target_from_device(cl::Device &device);

/** Helper function to get the GPU arch
 *
 * @param[in] target GPU target
 *
 * @return the GPU target which shows the arch
 */
GPUTarget get_arch_from_target(GPUTarget target);
}
#endif