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
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
|
/*
* 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.
*/
#include <cusp/copy.h>
#include <cusp/dia_matrix.h>
#include <thrust/tuple.h>
#include <thrust/reduce.h>
#include <thrust/scan.h>
#include <thrust/count.h>
#include <thrust/functional.h>
namespace cusp
{
namespace gallery
{
namespace detail
{
template <typename StencilPoint, typename GridDimension, typename IndexType, int i, int n>
struct inside_grid_helper
{
__host__ __device__
static bool inside_grid(StencilPoint point, GridDimension grid, IndexType index)
{
IndexType x = index % thrust::get<i>(grid) + thrust::get<i>(thrust::get<0>(point));
if (x < 0 || x >= thrust::get<i>(grid))
return false;
else
return inside_grid_helper<StencilPoint,GridDimension,IndexType,i + 1,n>::inside_grid(point, grid, index / thrust::get<i>(grid));
}
};
template <typename StencilPoint, typename GridDimension, typename IndexType, int n>
struct inside_grid_helper<StencilPoint,GridDimension,IndexType,n,n>
{
__host__ __device__
static bool inside_grid(StencilPoint point, GridDimension grid, IndexType index)
{
return true;
}
};
template <typename StencilPoint, typename GridDimension, typename IndexType>
__host__ __device__
bool inside_grid(StencilPoint point, GridDimension grid, IndexType index)
{
return inside_grid_helper<StencilPoint,GridDimension,IndexType, 0, thrust::tuple_size<GridDimension>::value >::inside_grid(point, grid, index);
}
template <typename Tuple, typename UnaryFunction, int i, int size>
struct tuple_for_each_helper
{
static UnaryFunction for_each(Tuple& t, UnaryFunction f)
{
f(thrust::get<i>(t));
return tuple_for_each_helper<Tuple,UnaryFunction,i + 1,size>::for_each(t, f);
}
};
template <typename Tuple, typename UnaryFunction, int size>
struct tuple_for_each_helper<Tuple,UnaryFunction,size,size>
{
static UnaryFunction for_each(Tuple& t, UnaryFunction f)
{
return f;
}
};
template <typename Tuple, typename UnaryFunction>
UnaryFunction tuple_for_each(Tuple& t, UnaryFunction f)
{
return tuple_for_each_helper<Tuple, UnaryFunction, 0, thrust::tuple_size<Tuple>::value>::for_each(t, f);
}
template <typename Tuple, typename OutputIterator>
struct unpack_tuple_functor
{
OutputIterator output;
unpack_tuple_functor(OutputIterator output) : output(output) {}
template <typename T>
void operator()(T v)
{
*output++ = v;
}
};
template <typename Tuple, typename OutputIterator>
OutputIterator unpack_tuple(const Tuple & t, OutputIterator output)
{
return tuple_for_each(t, unpack_tuple_functor<Tuple,OutputIterator>(output)).output;
}
template <typename IndexType,
typename ValueType,
typename StencilPoint,
typename GridDimension>
struct fill_diagonal_entries
{
StencilPoint point;
GridDimension grid;
fill_diagonal_entries(StencilPoint point, GridDimension grid)
: point(point), grid(grid) {}
__host__ __device__
ValueType operator()(IndexType index)
{
if (inside_grid(point, grid, index))
return thrust::get<1>(point);
else
return ValueType(0);
}
};
} // end namespace detail
template <typename IndexType,
typename ValueType,
typename MemorySpace,
typename StencilPoint,
typename GridDimension>
void generate_matrix_from_stencil( cusp::dia_matrix<IndexType,ValueType,MemorySpace>& matrix,
const cusp::array1d<StencilPoint,cusp::host_memory>& stencil,
const GridDimension& grid)
{
IndexType num_dimensions = thrust::tuple_size<GridDimension>::value;
cusp::array1d<IndexType,cusp::host_memory> grid_indices(num_dimensions);
detail::unpack_tuple(grid, grid_indices.begin());
IndexType num_rows = thrust::reduce(grid_indices.begin(), grid_indices.end(), IndexType(1), thrust::multiplies<IndexType>());
IndexType num_diagonals = stencil.size();
cusp::array1d<IndexType,cusp::host_memory> strides(grid_indices.size());
thrust::exclusive_scan(grid_indices.begin(), grid_indices.end(), strides.begin(), IndexType(1), thrust::multiplies<IndexType>());
cusp::array1d<IndexType,cusp::host_memory> offsets(stencil.size(), 0);
for(size_t i = 0; i < offsets.size(); i++)
{
cusp::array1d<IndexType,cusp::host_memory> stencil_indices(num_dimensions);
detail::unpack_tuple(thrust::get<0>(stencil[i]), stencil_indices.begin());
for(IndexType j = 0; j < num_dimensions; j++)
{
offsets[i] += strides[j] * stencil_indices[j];
}
}
// TODO compute num_entries directly from stencil
matrix.resize(num_rows, num_rows, 0, num_diagonals); // XXX we set NNZ to zero for now
cusp::copy(offsets, matrix.diagonal_offsets);
// ideally we'd have row views and column views here
for(IndexType i = 0; i < num_diagonals; i++)
{
thrust::transform(thrust::counting_iterator<IndexType>(0),
thrust::counting_iterator<IndexType>(num_rows),
matrix.values.values.begin() + matrix.values.pitch * i,
detail::fill_diagonal_entries<IndexType,ValueType,StencilPoint,GridDimension>(stencil[i], grid));
}
matrix.num_entries = matrix.values.values.size() - thrust::count(matrix.values.values.begin(), matrix.values.values.end(), ValueType(0));
}
// TODO add an entry point and make this the default path
template <typename MatrixType,
typename StencilPoint,
typename GridDimension>
void generate_matrix_from_stencil( MatrixType& matrix,
const cusp::array1d<StencilPoint,cusp::host_memory>& stencil,
const GridDimension& grid)
{
CUSP_PROFILE_SCOPED();
typedef typename MatrixType::index_type IndexType;
typedef typename MatrixType::value_type ValueType;
typedef typename MatrixType::memory_space MemorySpace;
cusp::dia_matrix<IndexType,ValueType,MemorySpace> dia;
generate_matrix_from_stencil(dia, stencil, grid);
cusp::convert(dia, matrix);
}
} // end namespace gallery
} // end namespace cusp
|
