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
|
/*
* Copyright 2020 Patrick Stotko
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
#include <iostream>
#include <thrust/copy.h>
#include <thrust/reduce.h>
#include <thrust/sequence.h>
#include <stdgpu/deque.cuh> // stdgpu::deque
#include <stdgpu/iterator.h> // device_begin, device_end
#include <stdgpu/memory.h> // createDeviceArray, destroyDeviceArray
#include <stdgpu/platform.h> // STDGPU_HOST_DEVICE
struct is_odd
{
STDGPU_HOST_DEVICE bool
operator()(const int x) const
{
return x % 2 == 1;
}
};
__global__ void
insert_neighbors_with_duplicates(const int* d_input, const stdgpu::index_t n, stdgpu::deque<int> deq)
{
stdgpu::index_t i = static_cast<stdgpu::index_t>(blockIdx.x * blockDim.x + threadIdx.x);
if (i >= n)
return;
int num = d_input[i];
int num_neighborhood[3] = { num - 1, num, num + 1 };
is_odd odd;
for (int num_neighbor : num_neighborhood)
{
if (odd(num_neighbor))
{
deq.push_back(num_neighbor);
}
else
{
deq.push_front(num_neighbor);
}
}
}
int
main()
{
//
// EXAMPLE DESCRIPTION
// -------------------
// This example demonstrates how stdgpu::deque is used to compute a set of duplicated numbers.
// Every number is contained 3 times, except for the first and last one.
// Furthermore, even numbers are put into the front, whereas odd number are put into the back.
//
const stdgpu::index_t n = 100;
int* d_input = createDeviceArray<int>(n);
stdgpu::deque<int> deq = stdgpu::deque<int>::createDeviceObject(3 * n);
thrust::sequence(stdgpu::device_begin(d_input), stdgpu::device_end(d_input), 1);
// d_input : 1, 2, 3, ..., 100
stdgpu::index_t threads = 32;
stdgpu::index_t blocks = (n + threads - 1) / threads;
insert_neighbors_with_duplicates<<<static_cast<unsigned int>(blocks), static_cast<unsigned int>(threads)>>>(d_input,
n,
deq);
cudaDeviceSynchronize();
// deq : 0, 1, 1, 2, 2, 2, 3, 3, 3, ..., 99, 99, 99, 100, 100, 101
auto range_deq = deq.device_range();
int sum = thrust::reduce(range_deq.begin(), range_deq.end(), 0, thrust::plus<int>());
const int sum_closed_form = 3 * (n * (n + 1) / 2);
std::cout << "The set of duplicated numbers contains " << deq.size() << " elements (" << 3 * n
<< " expected) and the computed sum is " << sum << " (" << sum_closed_form << " expected)" << std::endl;
destroyDeviceArray<int>(d_input);
stdgpu::deque<int>::destroyDeviceObject(deq);
}
|
