/home/docs/checkouts/readthedocs.org/user_builds/advanced-micro-devices-hipcub/checkouts/docs-5.5.1/hipcub/include/hipcub/backend/rocprim/warp/warp_merge_sort.hpp Source File

/home/docs/checkouts/readthedocs.org/user_builds/advanced-micro-devices-hipcub/checkouts/docs-5.5.1/hipcub/include/hipcub/backend/rocprim/warp/warp_merge_sort.hpp Source File#

hipCUB: /home/docs/checkouts/readthedocs.org/user_builds/advanced-micro-devices-hipcub/checkouts/docs-5.5.1/hipcub/include/hipcub/backend/rocprim/warp/warp_merge_sort.hpp Source File
warp_merge_sort.hpp
1 /******************************************************************************
2  * Copyright (c) 2011-2021, NVIDIA CORPORATION. All rights reserved.
3  * Modifications Copyright (c) 2017-2021, Advanced Micro Devices, Inc. All
4  * 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_WARP_WARP_MERGE_SORT_
31 #define HIPCUB_ROCPRIM_WARP_WARP_MERGE_SORT_
32 
33 #include "../../../config.hpp"
34 
35 #include "../block/block_merge_sort.hpp"
36 #include "../util_ptx.hpp"
37 #include "../util_type.hpp"
38 
39 #include <rocprim/functional.hpp>
40 #include <rocprim/intrinsics/thread.hpp>
41 
42 BEGIN_HIPCUB_NAMESPACE
43 
120 template <
121  typename KeyT,
122  int ITEMS_PER_THREAD,
123  int LOGICAL_WARP_THREADS = HIPCUB_DEVICE_WARP_THREADS,
124  typename ValueT = NullType,
125  int PTX_ARCH = HIPCUB_ARCH>
126 class WarpMergeSort
127  : public BlockMergeSortStrategy<
128  KeyT,
129  ValueT,
130  LOGICAL_WARP_THREADS,
131  ITEMS_PER_THREAD,
132  WarpMergeSort<KeyT, ITEMS_PER_THREAD, LOGICAL_WARP_THREADS, ValueT, PTX_ARCH>>
133 {
134 private:
135  constexpr static bool IS_ARCH_WARP = LOGICAL_WARP_THREADS == HIPCUB_DEVICE_WARP_THREADS;
136  constexpr static bool KEYS_ONLY = ::rocprim::Equals<ValueT, NullType>::VALUE;
137  constexpr static int TILE_SIZE = ITEMS_PER_THREAD * LOGICAL_WARP_THREADS;
138 
139  using BlockMergeSortStrategyT = BlockMergeSortStrategy<KeyT,
140  ValueT,
141  LOGICAL_WARP_THREADS,
142  ITEMS_PER_THREAD,
143  WarpMergeSort>;
144 
145  const unsigned int warp_id;
146  const uint64_t member_mask;
147 
148 public:
149  WarpMergeSort() = delete;
150 
151  HIPCUB_DEVICE __forceinline__
152  WarpMergeSort(typename BlockMergeSortStrategyT::TempStorage &temp_storage)
153  : BlockMergeSortStrategyT(temp_storage,
154  IS_ARCH_WARP
155  ? LaneId()
156  : (LaneId() % LOGICAL_WARP_THREADS))
157  , warp_id(IS_ARCH_WARP ? 0 : (LaneId() / LOGICAL_WARP_THREADS))
158  , member_mask(WarpMask<LOGICAL_WARP_THREADS>(warp_id))
159  {
160  }
161 
162  HIPCUB_DEVICE __forceinline__ uint64_t get_member_mask() const
163  {
164  return member_mask;
165  }
166 
167 private:
168  HIPCUB_DEVICE __forceinline__ void SyncImplementation() const
169  {
170  WARP_SYNC(member_mask);
171  }
172 
173  friend BlockMergeSortStrategyT;
174 };
175 
176 
177 END_HIPCUB_NAMESPACE
178 
179 #endif // HIPCUB_ROCPRIM_WARP_WARP_MERGE_SORT_
hipcub::BlockMergeSortStrategy
Generalized merge sort algorithm.
Definition: block_merge_sort.hpp:175
hipcub::WarpMergeSort
The WarpMergeSort class provides methods for sorting items partitioned across a CUDA warp using a mer...
Definition: warp_merge_sort.hpp:133
hipcub::BlockMergeSortStrategy::TempStorage
\smemstorage{BlockMergeSort}
Definition: block_merge_sort.hpp:207