blob: 2b04ac97cd3544aad9ad609fc839c597cb3997f6 (
plain)
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
|
#include <string>
#include <cuda.h>
#include <iostream>
#ifdef _WIN32
#define EXPORT __declspec(dllexport)
#else
#define EXPORT
#endif
int dynamic_base_func(int);
EXPORT int __host__ cuda_dynamic_host_func(int x)
{
return dynamic_base_func(x);
}
static
__global__
void DetermineIfValidCudaDevice()
{
}
EXPORT int choose_cuda_device()
{
int nDevices = 0;
cudaError_t err = cudaGetDeviceCount(&nDevices);
if (err != cudaSuccess)
{
std::cerr << "Failed to retrieve the number of CUDA enabled devices"
<< std::endl;
return 1;
}
for (int i = 0; i < nDevices; ++i)
{
cudaDeviceProp prop;
cudaError_t err = cudaGetDeviceProperties(&prop, i);
if (err != cudaSuccess)
{
std::cerr << "Could not retrieve properties from CUDA device " << i
<< std::endl;
return 1;
}
if (prop.major >= 4)
{
err = cudaSetDevice(i);
if (err != cudaSuccess)
{
std::cout << "Could not select CUDA device " << i << std::endl;
}
else
{
return 0;
}
}
}
std::cout << "Could not find a CUDA enabled card supporting compute >=3.0"
<< std::endl;
return 1;
}
EXPORT void cuda_dynamic_lib_func()
{
DetermineIfValidCudaDevice <<<1,1>>> ();
cudaError_t err = cudaGetLastError();
if(err != cudaSuccess)
{
std::cerr << "DetermineIfValidCudaDevice [SYNC] failed: "
<< cudaGetErrorString(err) << std::endl;
}
err = cudaDeviceSynchronize();
if(err != cudaSuccess)
{
std::cerr << "DetermineIfValidCudaDevice [ASYNC] failed: "
<< cudaGetErrorString(cudaGetLastError()) << std::endl;
}
}
|
