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
178
179
180
181
182
183
184
|
/*
* (C) Copyright 2024- ECMWF.
*
* This software is licensed under the terms of the Apache Licence Version 2.0
* which can be obtained at http://www.apache.org/licenses/LICENSE-2.0.
* In applying this licence, ECMWF does not waive the privileges and immunities
* granted to it by virtue of its status as an intergovernmental organisation
* nor does it submit to any jurisdiction.
*/
#include <algorithm>
#include <chrono>
#include <iostream>
#include <string_view>
#include <vector>
#include "hic/hic.h"
#include "pluto/pluto.h"
// ---------------------------------------------------------------------------------------------------
// Kernel to add +1 to device array. This could be in a separate CUDA/HIP source file
template <typename T>
void plus_one_on_host(T* d, int n) {
std::cout << " ~ plus_one_on_host" << std::endl;
for (int i = 0; i < n; ++i) {
d[i] += 1.;
}
}
#if PLUTO_HAVE_HIC
template <typename T>
HIC_GLOBAL void kernel_plus_one_on_device(T* d, int n) {
const int idx{int(blockDim.x) * int(blockIdx.x) + int(threadIdx.x)};
const int stride{int(blockDim.x) * int(gridDim.x)};
for (int i{idx}; i < n; i += stride) {
d[i] += 1.;
}
}
template <typename T>
void plus_one_on_device(T* d, int n) {
std::cout << " ~ plus_one_on_device" << std::endl;
const int threads_per_block{1024};
const int blocks_per_grid{32};
HIC_CHECK_KERNEL_LAUNCH();
kernel_plus_one_on_device<<<blocks_per_grid, threads_per_block>>>(d, n);
// HIC_CHECK_KERNEL_LAUNCH();
}
#else
template <typename T>
void plus_one_on_device(T* d, int n) {
plus_one_on_host(d, n);
}
#endif
// ---------------------------------------------------------------------------------------------------
template <typename T>
void print_array(std::string_view symbol, T* d, int n) {
std::cout << '\n' << symbol << "[0..size] = ";
for (std::size_t i = 0; i < 5; ++i) {
std::cout << d[i] << " ";
}
std::cout << "... ";
for (int i = n - 2; i < n; ++i) {
std::cout << d[i] << " ";
}
std::cout << '\n' << std::endl;
}
void print_info(const void* ptr, bool debug = false) {
std::cout << "Pointer introspection [" << ptr << "]" << std::endl;
std::cout << " pluto::is_managed: " << pluto::is_managed(ptr) << std::endl;
std::cout << " pluto::is_host: " << pluto::is_host(ptr) << std::endl;
std::cout << " pluto::is_device: " << pluto::is_device(ptr) << std::endl;
std::cout << " pluto::is_pinned: " << pluto::is_pinned(ptr) << std::endl;
std::cout << " pluto::is_device_accessible: " << pluto::is_device_accessible(ptr) << std::endl;
std::cout << " pluto::is_host_accessible: " << pluto::is_host_accessible(ptr) << std::endl;
if (debug) {
#if HIC_BACKEND_HIP && HIP_VERSION_MAJOR < 6
{
// Debugging HIP backend pointer attributes because it did not map one-to-one with CUDA
hipPointerAttribute_t attr_;
auto err = hipPointerGetAttributes(&attr_, ptr);
if (err != hipSuccess) {
std::cout << "Warning: could not access hipPointerGetAttributes" << std::endl;
return;
}
std::cout << "\nhipPointerAttribute_t: " << std::endl;
std::cout << " memoryType: " << attr_.memoryType << std::endl;
std::cout << " device: " << attr_.device << std::endl;
std::cout << " devicePointer: " << attr_.devicePointer << std::endl;
std::cout << " hostPointer: " << attr_.hostPointer << std::endl;
std::cout << " isManaged: " << attr_.isManaged << std::endl;
std::cout << "Legend(hipMemoryType):" << std::endl;
std::cout << " hipMemoryTypeHost: " << hipMemoryTypeHost << std::endl;
std::cout << " hipMemoryTypeDevice: " << hipMemoryTypeDevice << std::endl;
std::cout << " hipMemoryTypeUnified: " << hipMemoryTypeUnified << std::endl;
}
#endif
#if PLUTO_HAVE_HIC
{
// Debugging HIC
hicPointerAttributes attr_;
auto err = hicPointerGetAttributes(&attr_, ptr);
if (err != hicSuccess) {
std::cout << "Warning: could not access hicPointerGetAttributes" << std::endl;
return;
}
std::cout << "\nhicPointerAttributes: " << std::endl;
std::cout << " type: " << attr_.type << std::endl;
std::cout << " device: " << attr_.device << std::endl;
std::cout << " devicePointer: " << attr_.devicePointer << std::endl;
std::cout << " hostPointer: " << attr_.hostPointer << std::endl;
std::cout << "Legend(hicMemoryType):" << std::endl;
std::cout << " hicMemoryTypeUnregistered: " << hicMemoryTypeUnregistered << std::endl;
std::cout << " hicMemoryTypeHost: " << hicMemoryTypeHost << std::endl;
std::cout << " hicMemoryTypeDevice: " << hicMemoryTypeDevice << std::endl;
std::cout << " hicMemoryTypeManaged: " << hicMemoryTypeManaged << std::endl;
}
#endif
}
}
// ---------------------------------------------------------------------------------------------------
enum class Memory
{
Managed,
Pinned
};
int main([[maybe_unused]] int argc, [[maybe_unused]] char* argv[]) {
// Choose data type
using value_type = float;
// Total problem size
std::size_t size = 1'000;
// Choose Memory allocation strategy
Memory strategy = Memory::Managed;
std::cout << "pluto::devices: " << pluto::devices() << std::endl;
pluto::host::allocator<value_type> allocator{strategy == Memory::Managed ? pluto::managed_resource()
: strategy == Memory::Pinned ? pluto::pinned_resource()
: nullptr};
value_type* h_array = allocator.allocate(size);
value_type* d_array = pluto::get_registered_device_pointer(h_array);
print_info(h_array);
print_info(d_array);
// Fill array with zero
if (pluto::is_host_accessible(h_array)) {
std::fill(h_array, h_array + size, 0.);
}
print_array("array", h_array, size);
// Add +1 to each element of array
if (pluto::is_device_accessible(d_array)) {
plus_one_on_device(d_array, size);
pluto::wait();
}
else {
plus_one_on_host(d_array, size);
}
print_array("array", h_array, size);
allocator.deallocate(h_array, size);
// No need to deallocate d_array !
std::cout << pluto::memory::report() << std::endl;
}
|
