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
|
/// \file Interface.cu
/// \author Johannes de Fine Licht (johannes.definelicht@cern.ch)
#include "VecGeom/backend/cuda/Interface.h"
#include <stdio.h>
#include <cuda.h>
// includes CUDA Runtime
#include <cuda_runtime.h>
// maybe you need also helpers
//#include <helper_cuda.h>
//#include <helper_functions.h> // helper utility functions
namespace vecgeom {
#ifdef VECCORE_CUDA
namespace cxx {
#else
inline namespace cxx {
#endif
cudaError_t CudaCheckError(const cudaError_t err)
{
if (err != cudaSuccess) {
printf("CUDA reported error with message: \"%s\"\n", cudaGetErrorString(err));
}
return err;
}
/**
* Retrieve the last cuda error and print it in case of problems.
* This clears the error state of the cuda API.
* @return The cuda error state.
*/
cudaError_t CudaCheckError()
{
return CudaCheckError(cudaGetLastError());
}
/**
* Assert that `err == cudaSuccess` in debug builds.
* Print the error and terminate the program if it's not.
* The function is a no-op in release builds.
* @return The cuda error state.
*/
void CudaAssertError(const cudaError_t err)
{
assert(CudaCheckError(err) == cudaSuccess);
}
/**
* In debug builds, retrieve the last cuda error and assert that it is cudaSuccess.
* The error will not be cleared in release builds. Use CudaCheckError() for this.
*/
void CudaAssertError()
{
assert(CudaCheckError(cudaGetLastError()) == cudaSuccess);
}
cudaError_t CudaMalloc(void **ptr, unsigned size)
{
return cudaMalloc(ptr, size);
}
cudaError_t CudaCopyToDevice(void *tgt, void const *src, unsigned size)
{
return cudaMemcpy(tgt, src, size, cudaMemcpyHostToDevice);
}
cudaError_t CudaCopyFromDevice(void *tgt, void const *src, unsigned size)
{
return cudaMemcpy(tgt, src, size, cudaMemcpyDeviceToHost);
}
cudaError_t CudaCopyFromDeviceAsync(void *dst, void const * src, unsigned size, cudaStream_t stream)
{
return cudaMemcpyAsync(dst, src, size, cudaMemcpyDeviceToHost, stream);
}
cudaError_t CudaFree(void *ptr)
{
return cudaFree(ptr);
}
cudaError_t CudaDeviceSetStackLimit(unsigned size)
{
return cudaDeviceSetLimit(cudaLimitStackSize, size);
}
cudaError_t CudaDeviceSetHeapLimit(unsigned size)
{
return cudaDeviceSetLimit(cudaLimitMallocHeapSize, size);
}
} // End namespace cuda
} // End namespace vecgeom
|
