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
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
|
/*
* Copyright 2008-2013 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.
*/
#pragma once
#include <thrust/detail/config.h>
#include <thrust/detail/cstdint.h>
#include <thrust/system/detail/generic/scan_by_key.h>
#include <thrust/functional.h>
#include <thrust/transform.h>
#include <thrust/replace.h>
#include <thrust/iterator/zip_iterator.h>
#include <thrust/iterator/iterator_traits.h>
#include <thrust/detail/temporary_array.h>
#include <thrust/detail/internal_functional.h>
#include <thrust/scan.h>
THRUST_NAMESPACE_BEGIN
namespace system
{
namespace detail
{
namespace generic
{
namespace detail
{
template <typename OutputType, typename HeadFlagType, typename AssociativeOperator>
struct segmented_scan_functor
{
AssociativeOperator binary_op;
typedef typename thrust::tuple<OutputType, HeadFlagType> result_type;
__host__ __device__
segmented_scan_functor(AssociativeOperator _binary_op) : binary_op(_binary_op) {}
__host__ __device__
result_type operator()(result_type a, result_type b)
{
return result_type(thrust::get<1>(b) ? thrust::get<0>(b) : binary_op(thrust::get<0>(a), thrust::get<0>(b)),
thrust::get<1>(a) | thrust::get<1>(b));
}
};
} // end namespace detail
template<typename DerivedPolicy,
typename InputIterator1,
typename InputIterator2,
typename OutputIterator>
__host__ __device__
OutputIterator inclusive_scan_by_key(thrust::execution_policy<DerivedPolicy> &exec,
InputIterator1 first1,
InputIterator1 last1,
InputIterator2 first2,
OutputIterator result)
{
return thrust::inclusive_scan_by_key(exec, first1, last1, first2, result, thrust::equal_to<>());
}
template<typename DerivedPolicy,
typename InputIterator1,
typename InputIterator2,
typename OutputIterator,
typename BinaryPredicate>
__host__ __device__
OutputIterator inclusive_scan_by_key(thrust::execution_policy<DerivedPolicy> &exec,
InputIterator1 first1,
InputIterator1 last1,
InputIterator2 first2,
OutputIterator result,
BinaryPredicate binary_pred)
{
return thrust::inclusive_scan_by_key(exec, first1, last1, first2, result, binary_pred, thrust::plus<>());
}
template<typename DerivedPolicy,
typename InputIterator1,
typename InputIterator2,
typename OutputIterator,
typename BinaryPredicate,
typename AssociativeOperator>
__host__ __device__
OutputIterator inclusive_scan_by_key(thrust::execution_policy<DerivedPolicy> &exec,
InputIterator1 first1,
InputIterator1 last1,
InputIterator2 first2,
OutputIterator result,
BinaryPredicate binary_pred,
AssociativeOperator binary_op)
{
using OutputType = typename thrust::iterator_traits<InputIterator2>::value_type;
using HeadFlagType = thrust::detail::uint8_t;
const size_t n = last1 - first1;
if(n != 0)
{
// compute head flags
thrust::detail::temporary_array<HeadFlagType,DerivedPolicy> flags(exec, n);
flags[0] = 1; thrust::transform(exec, first1, last1 - 1, first1 + 1, flags.begin() + 1, thrust::detail::not2(binary_pred));
// scan key-flag tuples,
// For additional details refer to Section 2 of the following paper
// S. Sengupta, M. Harris, and M. Garland. "Efficient parallel scan algorithms for GPUs"
// NVIDIA Technical Report NVR-2008-003, December 2008
// http://mgarland.org/files/papers/nvr-2008-003.pdf
thrust::inclusive_scan(exec,
thrust::make_zip_iterator(thrust::make_tuple(first2, flags.begin())),
thrust::make_zip_iterator(thrust::make_tuple(first2, flags.begin())) + n,
thrust::make_zip_iterator(thrust::make_tuple(result, flags.begin())),
detail::segmented_scan_functor<OutputType, HeadFlagType, AssociativeOperator>(binary_op));
}
return result + n;
}
template<typename DerivedPolicy,
typename InputIterator1,
typename InputIterator2,
typename OutputIterator>
__host__ __device__
OutputIterator exclusive_scan_by_key(thrust::execution_policy<DerivedPolicy> &exec,
InputIterator1 first1,
InputIterator1 last1,
InputIterator2 first2,
OutputIterator result)
{
typedef typename thrust::iterator_traits<InputIterator2>::value_type InitType;
return thrust::exclusive_scan_by_key(exec, first1, last1, first2, result, InitType{});
}
template<typename DerivedPolicy,
typename InputIterator1,
typename InputIterator2,
typename OutputIterator,
typename T>
__host__ __device__
OutputIterator exclusive_scan_by_key(thrust::execution_policy<DerivedPolicy> &exec,
InputIterator1 first1,
InputIterator1 last1,
InputIterator2 first2,
OutputIterator result,
T init)
{
return thrust::exclusive_scan_by_key(exec, first1, last1, first2, result, init, thrust::equal_to<>());
}
template<typename DerivedPolicy,
typename InputIterator1,
typename InputIterator2,
typename OutputIterator,
typename T,
typename BinaryPredicate>
__host__ __device__
OutputIterator exclusive_scan_by_key(thrust::execution_policy<DerivedPolicy> &exec,
InputIterator1 first1,
InputIterator1 last1,
InputIterator2 first2,
OutputIterator result,
T init,
BinaryPredicate binary_pred)
{
return thrust::exclusive_scan_by_key(exec, first1, last1, first2, result, init, binary_pred, thrust::plus<>());
}
template<typename DerivedPolicy,
typename InputIterator1,
typename InputIterator2,
typename OutputIterator,
typename T,
typename BinaryPredicate,
typename AssociativeOperator>
__host__ __device__
OutputIterator exclusive_scan_by_key(thrust::execution_policy<DerivedPolicy> &exec,
InputIterator1 first1,
InputIterator1 last1,
InputIterator2 first2,
OutputIterator result,
T init,
BinaryPredicate binary_pred,
AssociativeOperator binary_op)
{
using OutputType = T;
using HeadFlagType = thrust::detail::uint8_t;
const size_t n = last1 - first1;
if(n != 0)
{
InputIterator2 last2 = first2 + n;
// compute head flags
thrust::detail::temporary_array<HeadFlagType,DerivedPolicy> flags(exec, n);
flags[0] = 1; thrust::transform(exec, first1, last1 - 1, first1 + 1, flags.begin() + 1, thrust::detail::not2(binary_pred));
// shift input one to the right and initialize segments with init
thrust::detail::temporary_array<OutputType,DerivedPolicy> temp(exec, n);
thrust::replace_copy_if(exec, first2, last2 - 1, flags.begin() + 1, temp.begin() + 1, thrust::negate<HeadFlagType>(), init);
temp[0] = init;
// scan key-flag tuples,
// For additional details refer to Section 2 of the following paper
// S. Sengupta, M. Harris, and M. Garland. "Efficient parallel scan algorithms for GPUs"
// NVIDIA Technical Report NVR-2008-003, December 2008
// http://mgarland.org/files/papers/nvr-2008-003.pdf
thrust::inclusive_scan(exec,
thrust::make_zip_iterator(thrust::make_tuple(temp.begin(), flags.begin())),
thrust::make_zip_iterator(thrust::make_tuple(temp.begin(), flags.begin())) + n,
thrust::make_zip_iterator(thrust::make_tuple(result, flags.begin())),
detail::segmented_scan_functor<OutputType, HeadFlagType, AssociativeOperator>(binary_op));
}
return result + n;
}
} // end namespace generic
} // end namespace detail
} // end namespace system
THRUST_NAMESPACE_END
|
