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
|
#include <assert.h>
#include <stdlib.h>
#include <stdio.h>
#include <CL/cl.h>
#include "cl_simple.h"
#include "cl_util.h"
int main (int argc, char ** argv)
{
int i,j;
struct cl_simple_context context;
cl_int error;
cl_mem out_buffer;
int * out_data;
size_t global_work_size = 5;
int iterations;
unsigned out_data_size;
if (argc != 2) {
fprintf(stderr, "Usage: loop iterations\n");
return EXIT_FAILURE;
}
iterations = atoi(argv[1]);
out_data_size = global_work_size * iterations * sizeof(int);
out_data = malloc(out_data_size);
if (!clSimpleSimpleInit(&context, "loop")) {
return EXIT_FAILURE;
}
out_buffer = clCreateBuffer(context.cl_ctx,
CL_MEM_WRITE_ONLY,
out_data_size,
NULL, &error);
assert(error == CL_SUCCESS);
if (!clSimpleKernelSetArg(context.kernel, 0, sizeof(cl_mem), &out_buffer)
|| !clSimpleKernelSetArg(context.kernel, 1, sizeof(int), &iterations)) {
return EXIT_FAILURE;
}
error = clEnqueueNDRangeKernel(context.command_queue,
context.kernel,
1,
NULL,
&global_work_size,
&global_work_size,
0, NULL, NULL);
assert(error == CL_SUCCESS);
error = clEnqueueReadBuffer(context.command_queue,
out_buffer,
CL_TRUE,
0,
out_data_size,
out_data,
0, NULL, NULL);
assert(error == CL_SUCCESS);
for (i = 0; i < global_work_size; i++) {
for (j = 0; j < iterations; j++) {
fprintf(stderr, "%2u ", out_data[(i * iterations) + j]);
}
fprintf(stderr, "\n");
}
}
|