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
|
#ifndef ELEMENT_OPENCL_HH
#define ELEMENT_OPENCL_HH
#include "test_opencl.hpp"
template <class T, class F, int number_of_tests, int max_dimension>
class bench_elementwise
{
public:
typedef test_opencl<T, F> test;
void run()
{
opencl::library lib;
int passedOperations = 0;
// get default device and setup context
compute::device device = compute::system::default_device();
compute::context context(device);
compute::command_queue queue(context, device);
std::srand(time(0));
ublas::matrix<T, F> a;
ublas::matrix<T, F> b;
//matrix-matrix operations of cpu
ublas::matrix<T, F> result_m_add;
ublas::matrix<T, F> result_m_sub;
ublas::matrix<T, F> result_m_mul;
//matrix-matrix operations of gpu
ublas::matrix<T, F> result_m_add_cl;
ublas::matrix<T, F> result_m_sub_cl;
ublas::matrix<T, F> result_m_mul_cl;
ublas::vector<T> va;
ublas::vector<T> vb;
//vector-vector operations of cpu
ublas::vector<T> result_v_add;
ublas::vector<T> result_v_sub;
ublas::vector<T> result_v_mul;
//vector-vector operations of gpu
ublas::vector<T> result_v_add_cl;
ublas::vector<T> result_v_sub_cl;
ublas::vector<T> result_v_mul_cl;
for (int i = 0; i<number_of_tests; i++)
{
int rows = std::rand() % max_dimension + 1;
int cols = std::rand() % max_dimension + 1;
a.resize(rows, cols);
b.resize(rows, cols);
va.resize(rows);
vb.resize(rows);
test::init_matrix(a, 200);
test::init_matrix(b, 200);
test::init_vector(va, 200);
test::init_vector(vb, 200);
result_m_add = a + b;
result_m_add_cl = opencl::element_add(a, b, queue);
result_m_sub = a - b;
result_m_sub_cl = opencl::element_sub(a, b, queue);
result_m_mul = ublas::element_prod(a, b);
result_m_mul_cl = opencl::element_prod(a, b, queue);
result_v_add = va + vb;
result_v_add_cl = opencl::element_add(va, vb, queue);
result_v_sub = va - vb;
result_v_sub_cl = opencl::element_sub(va, vb, queue);
result_v_mul = ublas::element_prod(va, vb);
result_v_mul_cl = opencl::element_prod(va, vb, queue);
if ((!test::compare(result_m_add, result_m_add_cl)) ||
(!test::compare(result_m_sub, result_m_sub_cl)) ||
(!test::compare(result_m_mul, result_m_mul_cl)) ||
(!test::compare(result_v_add, result_v_add_cl)) ||
(!test::compare(result_v_sub, result_v_sub_cl)) ||
(!test::compare(result_v_mul, result_v_mul_cl))
)
{
std::cout << "Error in calculations" << std::endl;
std::cout << "passed: " << passedOperations << std::endl;
return;
}
passedOperations++;
}
std::cout << "All is well (matrix opencl element wise operations) of " << typeid(T).name() << std::endl;
}
};
#endif
|
