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
|
#include "utest_helper.hpp"
static void cpu(int global_id, int *src, int *dst) {
int tmp[32];
for (int i = 0; i < 16; ++i) {
for (int j = 0; j < 16; ++j)
tmp[j] = global_id;
for (int j = 0; j < src[0]; ++j)
tmp[j] = 1+src[j];
tmp[16+i] = tmp[i];
}
dst[global_id] = tmp[16+global_id];
}
void compiler_array3(void)
{
const size_t n = 16;
int cpu_dst[16], cpu_src[16];
// Setup kernel and buffers
OCL_CREATE_KERNEL("compiler_array3");
OCL_CREATE_BUFFER(buf[0], 0, n * sizeof(uint32_t), NULL);
OCL_CREATE_BUFFER(buf[1], 0, n * sizeof(uint32_t), NULL);
OCL_SET_ARG(0, sizeof(cl_mem), &buf[0]);
OCL_SET_ARG(1, sizeof(cl_mem), &buf[1]);
globals[0] = 16;
locals[0] = 16;
// Run random tests
for (uint32_t pass = 0; pass < 8; ++pass) {
OCL_MAP_BUFFER(0);
for (int32_t i = 0; i < (int32_t) n; ++i)
cpu_src[i] = ((int32_t*)buf_data[0])[i] = rand() % 16;
OCL_UNMAP_BUFFER(0);
// Run the kernel on GPU
OCL_NDRANGE(1);
// Run on CPU
for (int32_t i = 0; i <(int32_t) n; ++i) cpu(i, cpu_src, cpu_dst);
// Compare
OCL_MAP_BUFFER(1);
for (int32_t i = 0; i < 11; ++i)
OCL_ASSERT(((int32_t*)buf_data[1])[i] == cpu_dst[i]);
OCL_UNMAP_BUFFER(1);
}
}
MAKE_UTEST_FROM_FUNCTION(compiler_array3);
|
