/home/docs/checkouts/readthedocs.org/user_builds/advanced-micro-devices-hipcub/checkouts/docs-5.4.1/hipcub/include/hipcub/backend/rocprim/block/block_discontinuity.hpp Source File

/home/docs/checkouts/readthedocs.org/user_builds/advanced-micro-devices-hipcub/checkouts/docs-5.4.1/hipcub/include/hipcub/backend/rocprim/block/block_discontinuity.hpp Source File#

hipCUB: /home/docs/checkouts/readthedocs.org/user_builds/advanced-micro-devices-hipcub/checkouts/docs-5.4.1/hipcub/include/hipcub/backend/rocprim/block/block_discontinuity.hpp Source File
block_discontinuity.hpp
1 /******************************************************************************
2  * Copyright (c) 2010-2011, Duane Merrill. All rights reserved.
3  * Copyright (c) 2011-2018, NVIDIA CORPORATION. All rights reserved.
4  * Modifications Copyright (c) 2017-2020, Advanced Micro Devices, Inc. All rights reserved.
5  *
6  * Redistribution and use in source and binary forms, with or without
7  * modification, are permitted provided that the following conditions are met:
8  * * Redistributions of source code must retain the above copyright
9  * notice, this list of conditions and the following disclaimer.
10  * * Redistributions in binary form must reproduce the above copyright
11  * notice, this list of conditions and the following disclaimer in the
12  * documentation and/or other materials provided with the distribution.
13  * * Neither the name of the NVIDIA CORPORATION nor the
14  * names of its contributors may be used to endorse or promote products
15  * derived from this software without specific prior written permission.
16  *
17  * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
18  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
19  * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
20  * DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE FOR ANY
21  * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
22  * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
23  * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
24  * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
25  * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
26  * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
27  *
28  ******************************************************************************/
29 
30 #ifndef HIPCUB_ROCPRIM_BLOCK_BLOCK_DISCONTINUITY_HPP_
31 #define HIPCUB_ROCPRIM_BLOCK_BLOCK_DISCONTINUITY_HPP_
32 
33 #include "../../../config.hpp"
34 
35 #include <rocprim/block/block_discontinuity.hpp>
36 
37 BEGIN_HIPCUB_NAMESPACE
38 
39 template<
40  typename T,
41  int BLOCK_DIM_X,
42  int BLOCK_DIM_Y = 1,
43  int BLOCK_DIM_Z = 1,
44  int ARCH = HIPCUB_ARCH /* ignored */
45 >
47  : private ::rocprim::block_discontinuity<
48  T,
49  BLOCK_DIM_X,
50  BLOCK_DIM_Y,
51  BLOCK_DIM_Z
52  >
53 {
54  static_assert(
55  BLOCK_DIM_X * BLOCK_DIM_Y * BLOCK_DIM_Z > 0,
56  "BLOCK_DIM_X * BLOCK_DIM_Y * BLOCK_DIM_Z must be greater than 0"
57  );
58 
59  using base_type =
60  typename ::rocprim::block_discontinuity<
61  T,
62  BLOCK_DIM_X,
63  BLOCK_DIM_Y,
64  BLOCK_DIM_Z
65  >;
66 
67  // Reference to temporary storage (usually shared memory)
68  typename base_type::storage_type& temp_storage_;
69 
70 public:
71  using TempStorage = typename base_type::storage_type;
72 
73  HIPCUB_DEVICE inline
74  BlockDiscontinuity() : temp_storage_(private_storage())
75  {
76  }
77 
78  HIPCUB_DEVICE inline
79  BlockDiscontinuity(TempStorage& temp_storage) : temp_storage_(temp_storage)
80  {
81  }
82 
83  template<int ITEMS_PER_THREAD, typename FlagT, typename FlagOp>
84  HIPCUB_DEVICE inline
85  void FlagHeads(FlagT (&head_flags)[ITEMS_PER_THREAD],
86  T (&input)[ITEMS_PER_THREAD],
87  FlagOp flag_op)
88  {
89  base_type::flag_heads(head_flags, input, flag_op, temp_storage_);
90  }
91 
92  template<int ITEMS_PER_THREAD, typename FlagT, typename FlagOp>
93  HIPCUB_DEVICE inline
94  void FlagHeads(FlagT (&head_flags)[ITEMS_PER_THREAD],
95  T (&input)[ITEMS_PER_THREAD],
96  FlagOp flag_op,
97  T tile_predecessor_item)
98  {
99  base_type::flag_heads(head_flags, tile_predecessor_item, input, flag_op, temp_storage_);
100  }
101 
102  template<int ITEMS_PER_THREAD, typename FlagT, typename FlagOp>
103  HIPCUB_DEVICE inline
104  void FlagTails(FlagT (&tail_flags)[ITEMS_PER_THREAD],
105  T (&input)[ITEMS_PER_THREAD],
106  FlagOp flag_op)
107  {
108  base_type::flag_tails(tail_flags, input, flag_op, temp_storage_);
109  }
110 
111  template<int ITEMS_PER_THREAD, typename FlagT, typename FlagOp>
112  HIPCUB_DEVICE inline
113  void FlagTails(FlagT (&tail_flags)[ITEMS_PER_THREAD],
114  T (&input)[ITEMS_PER_THREAD],
115  FlagOp flag_op,
116  T tile_successor_item)
117  {
118  base_type::flag_tails(tail_flags, tile_successor_item, input, flag_op, temp_storage_);
119  }
120 
121  template<int ITEMS_PER_THREAD, typename FlagT, typename FlagOp>
122  HIPCUB_DEVICE inline
123  void FlagHeadsAndTails(FlagT (&head_flags)[ITEMS_PER_THREAD],
124  FlagT (&tail_flags)[ITEMS_PER_THREAD],
125  T (&input)[ITEMS_PER_THREAD],
126  FlagOp flag_op)
127  {
128  base_type::flag_heads_and_tails(
129  head_flags, tail_flags, input,
130  flag_op, temp_storage_
131  );
132  }
133 
134  template<int ITEMS_PER_THREAD, typename FlagT, typename FlagOp>
135  HIPCUB_DEVICE inline
136  void FlagHeadsAndTails(FlagT (&head_flags)[ITEMS_PER_THREAD],
137  FlagT (&tail_flags)[ITEMS_PER_THREAD],
138  T tile_successor_item,
139  T (&input)[ITEMS_PER_THREAD],
140  FlagOp flag_op)
141  {
142  base_type::flag_heads_and_tails(
143  head_flags, tail_flags, tile_successor_item, input,
144  flag_op, temp_storage_
145  );
146  }
147 
148  template<int ITEMS_PER_THREAD, typename FlagT, typename FlagOp>
149  HIPCUB_DEVICE inline
150  void FlagHeadsAndTails(FlagT (&head_flags)[ITEMS_PER_THREAD],
151  T tile_predecessor_item,
152  FlagT (&tail_flags)[ITEMS_PER_THREAD],
153  T (&input)[ITEMS_PER_THREAD],
154  FlagOp flag_op)
155  {
156  base_type::flag_heads_and_tails(
157  head_flags, tile_predecessor_item, tail_flags, input,
158  flag_op, temp_storage_
159  );
160  }
161 
162  template<int ITEMS_PER_THREAD, typename FlagT, typename FlagOp>
163  HIPCUB_DEVICE inline
164  void FlagHeadsAndTails(FlagT (&head_flags)[ITEMS_PER_THREAD],
165  T tile_predecessor_item,
166  FlagT (&tail_flags)[ITEMS_PER_THREAD],
167  T tile_successor_item,
168  T (&input)[ITEMS_PER_THREAD],
169  FlagOp flag_op)
170  {
171  base_type::flag_heads_and_tails(
172  head_flags, tile_predecessor_item, tail_flags, tile_successor_item, input,
173  flag_op, temp_storage_
174  );
175  }
176 
177 private:
178  HIPCUB_DEVICE inline
179  TempStorage& private_storage()
180  {
181  HIPCUB_SHARED_MEMORY TempStorage private_storage;
182  return private_storage;
183  }
184 };
185 
186 END_HIPCUB_NAMESPACE
187 
188 #endif // HIPCUB_ROCPRIM_BLOCK_BLOCK_DISCONTINUITY_HPP_
Definition: block_discontinuity.hpp:53