File: alignment.h

package info (click to toggle)
libthrust 1.17.2-2
  • links: PTS, VCS
  • area: main
  • in suites: bookworm
  • size: 10,900 kB
  • sloc: ansic: 29,519; cpp: 23,989; python: 1,421; sh: 811; perl: 460; makefile: 112
file content (250 lines) | stat: -rw-r--r-- 4,227 bytes parent folder | download | duplicates (6)
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
246
247
248
249
250
/*
 *  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.
 */

// TODO: This can probably be removed.

#pragma once

#include <thrust/detail/config.h>

#include <thrust/system/cuda/detail/util.h>

THRUST_NAMESPACE_BEGIN
namespace cuda_cub {
namespace alignment_of_detail {


  template <typename T>
  class alignment_of_impl;

  template <typename T, std::size_t size_diff>
  struct helper
  {
    static const std::size_t value = size_diff;
  };

  template <typename T>
  class helper<T, 0>
  {
  public:
    static const std::size_t value = alignment_of_impl<T>::value;
  };

  template <typename T>
  class alignment_of_impl
  {
  private:
    struct big
    {
      T    x;
      char c;
    };

  public:
    static const std::size_t value = helper<big, sizeof(big) - sizeof(T)>::value;
  };


}    // end alignment_of_detail


template <typename T>
struct alignment_of
    : alignment_of_detail::alignment_of_impl<T>
{
};


template <std::size_t Align>
struct aligned_type;

// __align__ is CUDA-specific, so guard it
#if THRUST_DEVICE_COMPILER == THRUST_DEVICE_COMPILER_NVCC

// implementing aligned_type portably is tricky:

#if THRUST_HOST_COMPILER == THRUST_HOST_COMPILER_MSVC
// implement aligned_type with specialization because MSVC
// requires literals as arguments to declspec(align(n))
template <>
struct aligned_type<1>
{
  struct __align__(1) type{};
};

template <>
struct aligned_type<2>
{
  struct __align__(2) type{};
};

template <>
struct aligned_type<4>
{
  struct __align__(4) type{};
};

template <>
struct aligned_type<8>
{
  struct __align__(8) type{};
};

template <>
struct aligned_type<16>
{
  struct __align__(16) type{};
};

template <>
struct aligned_type<32>
{
  struct __align__(32) type{};
};

template <>
struct aligned_type<64>
{
  struct __align__(64) type{};
};

template <>
struct aligned_type<128>
{
  struct __align__(128) type{};
};

template <>
struct aligned_type<256>
{
  struct __align__(256) type{};
};

template <>
struct aligned_type<512>
{
  struct __align__(512) type{};
};

template <>
struct aligned_type<1024>
{
  struct __align__(1024) type{};
};

template <>
struct aligned_type<2048>
{
  struct __align__(2048) type{};
};

template <>
struct aligned_type<4096>
{
  struct __align__(4096) type{};
};

template <>
struct aligned_type<8192>
{
  struct __align__(8192) type{};
};
#elif (THRUST_HOST_COMPILER == THRUST_HOST_COMPILER_GCC) && (THRUST_GCC_VERSION < 40300)
// implement aligned_type with specialization because gcc 4.2
// requires literals as arguments to __attribute__(aligned(n))
template <>
struct aligned_type<1>
{
  struct __align__(1) type{};
};

template <>
struct aligned_type<2>
{
  struct __align__(2) type{};
};

template <>
struct aligned_type<4>
{
  struct __align__(4) type{};
};

template <>
struct aligned_type<8>
{
  struct __align__(8) type{};
};

template <>
struct aligned_type<16>
{
  struct __align__(16) type{};
};

template <>
struct aligned_type<32>
{
  struct __align__(32) type{};
};

template <>
struct aligned_type<64>
{
  struct __align__(64) type{};
};

template <>
struct aligned_type<128>
{
  struct __align__(128) type{};
};

#else
// assume the compiler allows template parameters as
// arguments to __align__
template <std::size_t Align>
struct aligned_type
{
  struct __align__(Align) type{};
};
#endif    // THRUST_HOST_COMPILER
#else
template <std::size_t Align>
struct aligned_type
{
  struct type
  {
  };
};
#endif    // THRUST_DEVICE_COMPILER


template <std::size_t Len, std::size_t Align>
struct aligned_storage
{
  union type
  {
    unsigned char data[Len];

    typename aligned_type<Align>::type align;
  };
};


}    // end cuda_

THRUST_NAMESPACE_END