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
|
/*
Copyright 2010-2011, D. E. Shaw Research.
All rights reserved.
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are
met:
* Redistributions of source code must retain the above copyright
notice, this list of conditions, and the following disclaimer.
* Redistributions in binary form must reproduce the above copyright
notice, this list of conditions, and the following disclaimer in the
documentation and/or other materials provided with the distribution.
* Neither the name of D. E. Shaw Research nor the names of its
contributors may be used to endorse or promote products derived from
this software without specific prior written permission.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*/
#ifndef UTIL_CUDA_H__
#define UTIL_CUDA_H__
#include "util.h"
#include <cuda.h>
#include <cuda_runtime_api.h>
// utility macros to check return codes and complain/exit on failure
#define CHECKLAST(MSG) do { cudaError_t e = cudaGetLastError(); if (e != cudaSuccess) {fprintf(stderr, "%s:%d: CUDA Error: %s: %s\n", __FILE__, __LINE__, (MSG), cudaGetErrorString(e)); exit(1); }} while(0)
#define CHECKCALL(RET) do { cudaError_t e = (RET); if (e != cudaSuccess) { fprintf(stderr, "%s:%d: CUDA Error: %s\n", __FILE__, __LINE__, cudaGetErrorString(e)); exit(1); } } while(0)
typedef struct cuda_info {
int devnum, cores, blocks_per_grid, threads_per_block;
double cycles;
struct cudaDeviceProp dev;
} CUDAInfo;
// If devstr is none, chooses device with most cores.
static CUDAInfo *cuda_init(const char *devstr)
{
CUDAInfo *tp;
int i, ndev, cores, devcores;
double cycles;
CHECKNOTZERO(tp = (CUDAInfo *) malloc(sizeof(CUDAInfo)));
CHECKCALL( cudaGetDeviceCount(&ndev) );
devcores = 0;
for (i = 0; i < ndev; i++) {
struct cudaDeviceProp cu;
CHECKCALL( cudaGetDeviceProperties (&cu, i) );
// Number of cores is not available from a query, have to hardwire
// some knowledge here!
cores = cu.multiProcessorCount;
if (cu.major == 1 && cu.minor >= 0 && cu.minor <= 3) {
// 1.0 (G80, G92, aka GTX880, Tesla [CSD]870) to 1.3 (GT200, aka GTX280, Tesla [CS]10xx) have 8 cores per MP
cores *= 8;
} else if (cu.major == 2 && cu.minor == 0) {
// 2.0 (G100, aka GTX480, Tesla/Fermi [CSM]20[567]0, and GF110, aka GTX580, M2090)
cores *= 32;
} else if (cu.major == 2 && cu.minor == 1) {
// 2.1 (GF104, GF114, GF116 aka GTX [45][56]0)
cores *= 48;
} else if (cu.major == 3 && (cu.minor == 0||cu.minor==5)) {
// 3.0 (Kepler GK104 aka GTX 680)
cores *= 192;
} else {
fprintf(stderr, "Unknown # of cores per MP for this device, assuming 1, so cpb will be wrong\n");
}
/* clockrate is in KHz */
cycles = 1e3 * cu.clockRate * cores;
printf(" %d: maj %d min %d %s%s ( %d units @ %g MHz ECC=%d %d cores %g Gcycles/s)\n",
i, cu.major, cu.minor, nameclean(cu.name), cu.integrated ? " integrated" : "",
cu.multiProcessorCount, cu.clockRate*1e-3, cu.ECCEnabled, cores, cycles*1e-9);
if (devstr && strstr(cu.name, devstr) == NULL) {
dprintf(("skipping device %s\n", cu.name));
continue;
}
if (cores > devcores) {
devcores = cores;
tp->devnum = i;
tp->cores = cores;
tp->cycles = cycles;
tp->dev = cu;
}
}
if (devcores == 0) {
fprintf(stderr, "could not find specified device\n");
exit(1);
}
tp->blocks_per_grid = tp->cores; /* seems like a good guess */
tp->threads_per_block = tp->dev.warpSize * 2;
dprintf(("Using CUDA device %d, %d cores, %g cycles, will try %d blocks/grid %d threads/block\n",
tp->devnum, tp->cores, tp->cycles, tp->blocks_per_grid, tp->threads_per_block));
CHECKCALL(cudaSetDevice(tp->devnum));
dprintf(("cuda_init done\n"));
return tp;
}
static void cuda_done(CUDAInfo *tp)
{
dprintf(("cuda_done\n"));
free(tp);
}
#endif /* UTIL_CUDA_H__ */
|
