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
|
/*
* Copyright (c) 2016-2018 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_CLGEMM_H__
#define __ARM_COMPUTE_CLGEMM_H__
#include "arm_compute/core/CL/kernels/CLGEMMInterleave4x4Kernel.h"
#include "arm_compute/core/CL/kernels/CLGEMMMatrixAdditionKernel.h"
#include "arm_compute/core/CL/kernels/CLGEMMMatrixMultiplyKernel.h"
#include "arm_compute/core/CL/kernels/CLGEMMTranspose1xWKernel.h"
#include "arm_compute/runtime/CL/CLMemoryGroup.h"
#include "arm_compute/runtime/CL/CLTensor.h"
#include "arm_compute/runtime/IFunction.h"
#include "arm_compute/runtime/IMemoryManager.h"
namespace arm_compute
{
class ICLTensor;
/** Basic function to execute GEMM on OpenCL. This function calls the following OpenCL kernels:
*
* -# @ref CLGEMMInterleave4x4Kernel (only if the reshaped GEMM is selected by the heuristic model)
* -# @ref CLGEMMTranspose1xWKernel (only if the reshaped GEMM is selected by the heuristic model)
* -# @ref CLGEMMMatrixMultiplyKernel
* -# @ref CLGEMMMatrixAdditionKernel (if c != nullptr and beta != 0.0)
*
*/
class CLGEMM : public IFunction
{
public:
/** Default constructor.
*
* @param[in] memory_manager (Optional) Memory manager.
*/
CLGEMM(std::shared_ptr<IMemoryManager> memory_manager = nullptr);
/** Prevent instances of this class from being copied (As this class contains pointers) */
CLGEMM(const CLGEMM &) = delete;
/** Default move constructor */
CLGEMM(CLGEMM &&) = default;
/** Prevent instances of this class from being copied (As this class contains pointers) */
CLGEMM &operator=(const CLGEMM &) = delete;
/** Default move assignment operator */
CLGEMM &operator=(CLGEMM &&) = default;
/** Initialise the kernel's inputs and output
*
* @note GEMM: General Matrix Multiply - [alpha * A * B + beta * C].
*
* @note All tensors must have the same data type.
*
* @note Whilst the first input tensor can be a vector, the second input tensor must be at least a matrix
*
* @param[in] a First input tensor (Matrix or Vector A). Data types supported: QS8/QS16/F16/F32
* @param[in] b Second input tensor (Matrix B). Data type supported: same as @p a.
* @param[in] c Third input tensor (Matrix C). It can be a nullptr if just the multiplication between @p a and @p b is needed. Data type supported: same as @p a.
* @param[out] output Output tensor. Data type supported: same as @p a
* @param[in] alpha Weight of the matrix product
* @param[in] beta Weight of matrix C
* @param[in] gemm_info (Optional) Specifies if the matrix A and/or matrix B have been reshaped and
* if the reshape of matrix B should happen only for the first run. GEMMInfo also contains information about the reshaping
* in case matrix A and matrix B have been already transformed.
*/
void configure(const ICLTensor *a, const ICLTensor *b, const ICLTensor *c, ICLTensor *output, float alpha, float beta, const GEMMInfo &gemm_info = GEMMInfo());
/** Static function to check if given info will lead to a valid configuration of @ref CLGEMM.
*
* @param[in] a First input tensor info (Matrix or Vector A). Data types supported: QS8/QS16/F16/F32
* @param[in] b Second input tensor info (Matrix B). Data type supported: same as @p a.
* @param[in] c Third input tensor info (Matrix C). It can be a nullptr if just the multiplication between @p a and @p b is needed. Data type supported: same as @p a.
* @param[out] output Output tensor info. Data type supported: same as @p a
* @param[in] alpha Weight of the matrix product
* @param[in] beta Weight of matrix C
* @param[in] gemm_info (Optional) Specifies if the matrix A and/or matrix B have been reshaped and
* if the reshape of matrix B should happen only for the first run
*
* @return a status
*/
static Status validate(const ITensorInfo *a, const ITensorInfo *b, const ITensorInfo *c, const ITensorInfo *output, float alpha, float beta, const GEMMInfo &gemm_info = GEMMInfo());
// Inherited methods overridden:
void run() override;
void prepare() override;
private:
CLMemoryGroup _memory_group;
CLGEMMInterleave4x4Kernel _interleave_kernel;
CLGEMMTranspose1xWKernel _transpose_kernel;
CLGEMMMatrixMultiplyKernel _mm_kernel;
CLGEMMMatrixAdditionKernel _ma_kernel;
CLTensor _tmp_a;
CLTensor _tmp_b;
const ICLTensor *_original_b;
bool _is_interleaved_transposed;
bool _run_addition;
bool _reshape_b_only_on_first_run;
bool _is_prepared;
};
}
#endif /* __ARM_COMPUTE_CLGEMM_H__ */
|
