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
|
/*
* Copyright 2008-2009 NVIDIA Corporation
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
/*
* Modifications to this file:
* Copyright (c) 2014-2015, The University of Queensland
* Licensed under the Apache License, Version 2.0.
*
*/
#include <cusp/array1d.h>
//MW: if device is not cuda then do not include cuda stuff and redirect device multiply to host
#include <cusp/detail/host/multiply.h>
#include <cusp/detail/host/transposed_multiply.h>
#if THRUST_DEVICE_SYSTEM == THRUST_DEVICE_SYSTEM_CUDA
#include <cusp/detail/device/multiply.h>
#include <cusp/detail/device/transposed_multiply.h>
#endif
namespace cusp
{
namespace detail
{
namespace dispatch
{
////////////////
// Host Paths //
////////////////
template <typename LinearOperator,
typename MatrixOrVector1,
typename MatrixOrVector2>
void multiply(const LinearOperator& A,
const MatrixOrVector1& B,
MatrixOrVector2& C,
cusp::host_memory,
cusp::host_memory,
cusp::host_memory)
{
cusp::detail::host::multiply(A, B, C);
}
template <typename LinearOperator,
typename MatrixOrVector1,
typename MatrixOrVector2>
void transposed_multiply(const LinearOperator& A,
const MatrixOrVector1& B,
MatrixOrVector2& C,
cusp::host_memory,
cusp::host_memory,
cusp::host_memory)
{
cusp::detail::host::transposed_multiply(A, B, C);
}
//////////////////
// Device Paths //
//////////////////
template <typename LinearOperator,
typename MatrixOrVector1,
typename MatrixOrVector2>
void multiply(const LinearOperator& A,
const MatrixOrVector1& B,
MatrixOrVector2& C,
cusp::device_memory,
cusp::device_memory,
cusp::device_memory)
{
#if THRUST_DEVICE_SYSTEM == THRUST_DEVICE_SYSTEM_CUDA
cusp::detail::device::multiply(A, B, C);
#else
cusp::detail::host::multiply(A, B, C);
#endif
}
template <typename LinearOperator,
typename MatrixOrVector1,
typename MatrixOrVector2>
void transposed_multiply(const LinearOperator& A,
const MatrixOrVector1& B,
MatrixOrVector2& C,
cusp::device_memory,
cusp::device_memory,
cusp::device_memory)
{
#if THRUST_DEVICE_SYSTEM == THRUST_DEVICE_SYSTEM_CUDA
cusp::detail::device::transposed_multiply(A, B, C);
#else
cusp::detail::host::transposed_multiply(A, B, C);
#endif
}
} // end namespace dispatch
} // end namespace detail
} // end namespace cusp
|
