sdcuda/cmemory.cpp at master · oxford-pcs/sdcuda · GitHub

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
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
#include "cmemory.h"

#include <cuda_runtime.h>
#include <stdio.h>
#include <valarray>

#include "ccomplex.h"
#include "cdevice.cuh"
#include "errors.h"

using std::valarray;

template <class T>
void memory<T>::memcpydd(T* dst, T* src, long size) {
	/*
	Copy block of memory of size [size] from device location [src] to device location [dst].
	*/
	cudaMemcpy(dst, src, size, cudaMemcpyDeviceToDevice);
	if (cudaGetLastError() != cudaSuccess) {
		throw_error(CUDA_FAIL_MEMCPY_DD);
	}
}

template <class T>
void memory<T>::memcpydh(T* dst, T* src, long size) {
	/*
	Copy block of memory of size [size] from device location [src] to host location [dst].
	*/
	cudaMemcpy(dst, src, size, cudaMemcpyDeviceToHost);
	if (cudaGetLastError() != cudaSuccess) {
		throw_error(CUDA_FAIL_MEMCPY_DH);
	}
}

template <class T>
void memory<T>::memcpyhd(T* dst, T* src, long size) {
	/*
	Copy block of memory of size [size] from host location [src] to device location [dst].
	*/
	cudaMemcpy(dst, src, size, cudaMemcpyHostToDevice);
	if (cudaGetLastError() != cudaSuccess) {
		throw_error(CUDA_FAIL_MEMCPY_HD);
	}
}

template <class T>
void memory<T>::memcpyhh(T* dst, T* src, long size) {
	/*
	Copy block of memory of size [size] from host location [src] to host location [dst].
	*/
	cudaMemcpy(dst, src, size, cudaMemcpyHostToHost);
	if (cudaGetLastError() != cudaSuccess) {
		throw_error(CUDA_FAIL_MEMCPY_HH);
	}
}


template <class T>
void hmemory<T>::free(T* data) {
	/*
	Free host memory block.
	*/
	if (data != NULL) {
		cudaFreeHost(data);
		if (cudaGetLastError() != cudaSuccess) {
			throw_error(CUDA_FAIL_FREE_MEMORY_H);
		}
	}
}

template <class T>
T* hmemory<T>::malloc(long size, bool zero_initialise) {
	/*
	Allocate a piece of memory on the host of size [size].
	*/
	T* data = NULL;
	cudaMallocHost(&(data), size);
	if (cudaGetLastError() != cudaSuccess) {
		throw_error(CUDA_FAIL_ALLOCATE_MEMORY_H);
	}
	if (zero_initialise) {
		memset(data, 0, size);
	}
	return data;
}

template <class T>
T* hmemory<T>::realloc(T* old_data, long new_size, long old_size, bool zero_initialise_if_grow) {
	/*
	Reallocate a piece of memory of size [old_size] on the host of size [new_size].
	*/
	T* new_data = NULL;
	if (new_size > old_size) {
		if (zero_initialise_if_grow) {
			new_data = hmemory::malloc(new_size, true);
		} else {
			new_data = hmemory::malloc(new_size, false);
		}
		hmemory::memcpyhh(new_data, old_data, old_size);
	} else {
		new_data = hmemory::malloc(new_size, false);
		hmemory::memcpyhh(new_data, old_data, new_size);
	}
	hmemory::free(old_data);
	return new_data;
}


template <class T>
void dmemory<T>::free(T* data) {
	/*
	Free device memory block.
	*/
	if (data != NULL) {
		cudaFree(data);
		if (cudaGetLastError() != cudaSuccess) {
			throw_error(CUDA_FAIL_FREE_MEMORY_D);
		}
	}
}

template <class T>
T* dmemory<T>::malloc(long size, bool zero_initialise) {
	/*
	Allocate a piece of memory on the device of size [size].
	*/
	T* data = NULL;
	cudaMalloc((void**)&(data), size);
	if (cudaGetLastError() != cudaSuccess) {
		throw_error(CUDA_FAIL_ALLOCATE_MEMORY_D);
	}
	if (zero_initialise) {
		cudaMemset(data, 0, size);
		if (cudaGetLastError() != cudaSuccess) {
			throw_error(CUDA_FAIL_SET_MEMORY_D);
		}
	}
	return data;
}

template <class T>
T* dmemory<T>::realloc(T* old_data, long new_size, long old_size, bool zero_initialise_if_grow) {
	/*
	Reallocate a piece of memory of size [old_size] on the device of size [new_size].
	*/
	T* new_data = NULL;
	if (new_size > old_size) {
		if (zero_initialise_if_grow) {
			new_data = dmemory::malloc(new_size, true);
		} else {
			new_data = dmemory::malloc(new_size, true);
		}
		dmemory::memcpydd(new_data, old_data, old_size);
	} else {
		new_data = dmemory::malloc(new_size, false);
		dmemory::memcpydd(new_data, old_data, new_size);
	}
	dmemory::free(old_data);
	return new_data;
}

// explicit declarations required to construct classes of required types
template class memory<Complex>;
template class hmemory<Complex>;
template class dmemory<Complex>;
template class memory<Complex*>;
template class hmemory<Complex*>;
template class dmemory<Complex*>;
template class memory<Complex**>;
template class hmemory<Complex**>;
template class dmemory<Complex**>;
template class memory<int>;
template class hmemory<int>;
template class dmemory<int>;
template class memory<int*>;
template class hmemory<int*>;
template class dmemory<int*>;