boost_compute 0.1.0

Boost C++ library boost_compute packaged using Zanbil
Documentation 
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
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156

//---------------------------------------------------------------------------//
// Copyright (c) 2013-2014 Kyle Lutz <kyle.r.lutz@gmail.com>
//
// Distributed under the Boost Software License, Version 1.0
// See accompanying file LICENSE_1_0.txt or copy at
// http://www.boost.org/LICENSE_1_0.txt
//
// See http://boostorg.github.com/compute for more information.
//---------------------------------------------------------------------------//

#define BOOST_TEST_MODULE TestSvmPtr
#include <boost/test/unit_test.hpp>

#include <iostream>

#include <boost/compute/core.hpp>
#include <boost/compute/svm.hpp>
#include <boost/compute/container/vector.hpp>
#include <boost/compute/utility/source.hpp>

#include "quirks.hpp"
#include "check_macros.hpp"
#include "context_setup.hpp"

namespace compute = boost::compute;

BOOST_AUTO_TEST_CASE(empty)
{
}

#ifdef BOOST_COMPUTE_CL_VERSION_2_0
BOOST_AUTO_TEST_CASE(alloc)
{
    REQUIRES_OPENCL_VERSION(2, 0);

    compute::svm_ptr<cl_int> ptr = compute::svm_alloc<cl_int>(context, 8);
    compute::svm_free(context, ptr);
}

BOOST_AUTO_TEST_CASE(svmmemcpy)
{
    REQUIRES_OPENCL_VERSION(2, 0);

    if(bug_in_svmmemcpy(device)){
        std::cerr << "skipping svmmemcpy test case" << std::endl;
        return;
    }

    cl_int input[] = { 1, 2, 3, 4, 5, 6, 7, 8 };
    cl_int output[] = { 0, 0, 0, 0, 0, 0, 0, 0 };
    compute::svm_ptr<cl_int> ptr = compute::svm_alloc<cl_int>(context, 8);
    compute::svm_ptr<cl_int> ptr2 = compute::svm_alloc<cl_int>(context, 8);

    // copying from and to host mem
    queue.enqueue_svm_memcpy(ptr.get(), input, 8 * sizeof(cl_int));
    queue.enqueue_svm_memcpy(output, ptr.get(), 8 * sizeof(cl_int));
    queue.finish();

    CHECK_HOST_RANGE_EQUAL(cl_int, 8, output, (1, 2, 3, 4, 5, 6, 7, 8));

    // copying between svm mem
    queue.enqueue_svm_memcpy(ptr2.get(), ptr.get(), 8 * sizeof(cl_int));
    queue.enqueue_svm_memcpy(output, ptr2.get(), 8 * sizeof(cl_int));
    queue.finish();

    CHECK_HOST_RANGE_EQUAL(cl_int, 8, output, (1, 2, 3, 4, 5, 6, 7, 8));

    compute::svm_free(context, ptr);
    compute::svm_free(context, ptr2);
}

BOOST_AUTO_TEST_CASE(sum_svm_kernel)
{
    REQUIRES_OPENCL_VERSION(2, 0);

    const char source[] = BOOST_COMPUTE_STRINGIZE_SOURCE(
        __kernel void sum_svm_mem(__global const int *ptr, __global int *result)
        {
            int sum = 0;
            for(uint i = 0; i < 8; i++){
                sum += ptr[i];
            }
            *result = sum;
        }
    );

    compute::program program =
        compute::program::build_with_source(source, context, "-cl-std=CL2.0");

    compute::kernel sum_svm_mem_kernel = program.create_kernel("sum_svm_mem");

    cl_int data[] = { 1, 2, 3, 4, 5, 6, 7, 8 };
    compute::svm_ptr<cl_int> ptr = compute::svm_alloc<cl_int>(context, 8);
    queue.enqueue_svm_map(ptr.get(), 8 * sizeof(cl_int), CL_MAP_WRITE);
    for(size_t i = 0; i < 8; i ++) {
        static_cast<cl_int*>(ptr.get())[i] = data[i];
    }
    queue.enqueue_svm_unmap(ptr.get());

    compute::vector<cl_int> result(1, context);

    sum_svm_mem_kernel.set_arg(0, ptr);
    sum_svm_mem_kernel.set_arg(1, result);
    queue.enqueue_task(sum_svm_mem_kernel);

    queue.finish();
    BOOST_CHECK_EQUAL(result[0], (36));

    compute::svm_free(context, ptr);
}
#endif // BOOST_COMPUTE_CL_VERSION_2_0

#ifdef BOOST_COMPUTE_CL_VERSION_2_1
BOOST_AUTO_TEST_CASE(migrate)
{
    REQUIRES_OPENCL_VERSION(2, 1);

    compute::svm_ptr<cl_int> ptr =
        compute::svm_alloc<cl_int>(context, 8);

    // Migrate to device
    std::vector<const void*> ptrs(1, ptr.get());
    std::vector<size_t> sizes(1, 8 * sizeof(cl_int));
    queue.enqueue_svm_migrate_memory(ptrs, sizes).wait();

    // Set on device
    const char source[] = BOOST_COMPUTE_STRINGIZE_SOURCE(
        __kernel void foo(__global int *ptr)
        {
            for(int i = 0; i < 8; i++){
                ptr[i] = i;
            }
        }
    );
    compute::program program =
        compute::program::build_with_source(source, context, "-cl-std=CL2.0");
    compute::kernel foo_kernel = program.create_kernel("foo");
    foo_kernel.set_arg(0, ptr);
    queue.enqueue_task(foo_kernel).wait();

    // Migrate to host
    queue.enqueue_svm_migrate_memory(
        ptr.get(), 0, boost::compute::command_queue::migrate_to_host
    ).wait();

    // Check
    CHECK_HOST_RANGE_EQUAL(
        cl_int, 8,
        static_cast<cl_int*>(ptr.get()),
        (0, 1, 2, 3, 4, 5, 6, 7)
    );
    compute::svm_free(context, ptr);
}
#endif // BOOST_COMPUTE_CL_VERSION_2_1

BOOST_AUTO_TEST_SUITE_END()