AliceO2/GPU/Common/GPUCommonAlgorithmThrust.h at 68bca25aa1284b12eaeaf444cf3cc451fb08ca40 · alibuild/AliceO2 · GitHub

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
// Copyright 2019-2020 CERN and copyright holders of ALICE O2.
// See https://alice-o2.web.cern.ch/copyright for details of the copyright holders.
// All rights not expressly granted are reserved.
//
// This software is distributed under the terms of the GNU General Public
// License v3 (GPL Version 3), copied verbatim in the file "COPYING".
//
// In applying this license CERN does not waive the privileges and immunities
// granted to it by virtue of its status as an Intergovernmental Organization
// or submit itself to any jurisdiction.

/// \file GPUCommonAlgorithmThrust.h
/// \author Michael Lettrich

#ifndef GPUCOMMONALGORITHMTHRUST_H
#define GPUCOMMONALGORITHMTHRUST_H

#pragma GCC diagnostic push
#pragma GCC diagnostic ignored "-Wshadow"
#include <thrust/sort.h>
#include <thrust/execution_policy.h>
#include <thrust/device_ptr.h>
#pragma GCC diagnostic pop

#include "GPUCommonDef.h"

#ifdef __CUDACC__
#define GPUCA_THRUST_NAMESPACE thrust::cuda
#else
#define GPUCA_THRUST_NAMESPACE thrust::hip
#endif

namespace o2
{
namespace gpu
{

// - Our quicksort and bubble sort implementations are faster
/*
template <class T>
GPUdi() void GPUCommonAlgorithm::sort(T* begin, T* end)
{
  thrust::device_ptr<T> thrustBegin(begin);
  thrust::device_ptr<T> thrustEnd(end);
  thrust::sort(thrust::seq, thrustBegin, thrustEnd);
}

template <class T, class S>
GPUdi() void GPUCommonAlgorithm::sort(T* begin, T* end, const S& comp)
{
  thrust::device_ptr<T> thrustBegin(begin);
  thrust::device_ptr<T> thrustEnd(end);
  thrust::sort(thrust::seq, thrustBegin, thrustEnd, comp);
}

template <class T>
GPUdi() void GPUCommonAlgorithm::sortInBlock(T* begin, T* end)
{
  if (get_local_id(0) == 0) {
    sortDeviceDynamic(begin, end);
  }
}

template <class T, class S>
GPUdi() void GPUCommonAlgorithm::sortInBlock(T* begin, T* end, const S& comp)
{
  if (get_local_id(0) == 0) {
    sortDeviceDynamic(begin, end, comp);
  }
}

*/

template <class T>
GPUdi() void GPUCommonAlgorithm::sortDeviceDynamic(T* begin, T* end)
{
  thrust::device_ptr<T> thrustBegin(begin);
  thrust::device_ptr<T> thrustEnd(end);
  thrust::sort(GPUCA_THRUST_NAMESPACE::par, thrustBegin, thrustEnd);
}

template <class T, class S>
GPUdi() void GPUCommonAlgorithm::sortDeviceDynamic(T* begin, T* end, const S& comp)
{
  thrust::device_ptr<T> thrustBegin(begin);
  thrust::device_ptr<T> thrustEnd(end);
  thrust::sort(GPUCA_THRUST_NAMESPACE::par, thrustBegin, thrustEnd, comp);
}

} // namespace gpu
} // namespace o2

#endif