Newer
Older
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
/***************************************************************************
WriteOne.h - description
-------------------
begin : Dec 19, 2018
copyright : (C) 2018 by oberhuber
email : tomas.oberhuber@fjfi.cvut.cz
***************************************************************************/
/* See Copyright Notice in tnl/Copyright */
// Implemented by: Tomas Oberhuber
#pragma once
#include <TNL/ParallelFor.h>
#include <TNL/Devices/Host.h>
#include <TNL/Devices/Cuda.h>
#include <TNL/Containers/Vector.h>
namespace TNL {
namespace Benchmarks {
template< int Dimension,
typename Device,
typename Real,
typename Index >
class GridTraversersBenchmark{};
template< typename Device,
typename Real,
typename Index >
class GridTraversersBenchmark< 1, Device, Real, Index >
{
public:
using Vector = Containers::Vector< Real, Device, Index >;
GridTraversersBenchmark( Index size )
:v( size ), size( size )
{}
void writeOne()
{
auto f = [] __cuda_callable__ ( Index i, Real* data )
{
data[ i ] = i;
};
ParallelFor< Device >::exec( ( Index ) 0, size, f, v.getData() );
}
protected:
Index size;
Vector v;
};
template< typename Device,
typename Real,
typename Index >
class GridTraversersBenchmark< 2, Device, Real, Index >
{
public:
using Vector = Containers::Vector< Real, Device, Index >;
GridTraversersBenchmark( Index size )
:size( size ), v( size * size ) { }
void writeOne()
{
Index _size = this->size;
auto f = [=] __cuda_callable__ ( Index i, Index j, Real* data )
{
data[ i * _size + j ] = i + j;
};
ParallelFor2D< Device >::exec( ( Index ) 0,
( Index ) 0,
this->size,
this->size,
f, v.getData() );
}
protected:
Index size;
Vector v;
};
template< typename Device,
typename Real,
typename Index >
class GridTraversersBenchmark< 3, Device, Real, Index >
{
public:
using Vector = Containers::Vector< Real, Device, Index >;
GridTraversersBenchmark( Index size )
: size( size ), v( size * size * size ) {}
void writeOne()
{
Index _size = this->size;
auto f = [=] __cuda_callable__ ( Index i, Index j, Index k, Real* data )
{
data[ ( i * _size + j ) * _size + k ] = i + j + k;
};
ParallelFor3D< Device >::exec( ( Index ) 0,
( Index ) 0,
( Index ) 0,
this->size,
this->size,
this->size,
f, v.getData() );
}
protected:
Index size;
Vector v;
};
} // namespace Benchmarks
} // namespace TNL