ParallelLinAl/timer.h at master · bhuvan002/ParallelLinAl · 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
#ifndef TIMER_H_
#define TIMER_H_

#include <sys/time.h>
#include <cuda_runtime.h>

struct CPUTimer
{
  int err;
  double start;
  double stop;
  struct timeval Tv;

  CPUTimer()
  {
    start = 0.0;
    stop = 0.0;
    err = 0;

  }

  void Start()
  {
    err = gettimeofday(&Tv, NULL);
    if (!err)
      start = Tv.tv_sec + Tv.tv_usec * 1.0e-6;
  }

  void Stop()
  {
    err = gettimeofday(&Tv, NULL);
    if (!err)
      stop = Tv.tv_sec + Tv.tv_usec * 1.0e-6;
  }

  double Elapsed()
  {
    double elapsed;
    elapsed = stop - start;
    return elapsed;
  }
};


struct GPUTimer
{
	cudaEvent_t start;
	cudaEvent_t stop;

	GPUTimer()
	{
		cudaEventCreate(&start);
		cudaEventCreate(&stop);
	}

	~GPUTimer()
	{
		cudaEventDestroy(start);
		cudaEventDestroy(stop);
	}

	void Start()
	{
		cudaEventRecord(start, 0);
	}

	void Stop()
	{
		cudaEventRecord(stop, 0);
	}

	float Elapsed()
	{
		float elapsed;
		cudaEventSynchronize(stop);
		cudaEventElapsedTime(&elapsed, start, stop); // returns elapsed time in milli second
		return elapsed;
	}
};

#endif


/* Note on usage */

/*
 * CPUTimer cputimer;
 * cputimer.Start();
 * < piece of code to be timed >
 * cputimer.Stop();
 * printf("time taken for piece of code to run = %f", cputimer.Elapsed());

 **************************************************************************
 * GPUTimer gputimer;
 * gputimer.Start();
 * kernelToTime<<<...>>>(...);
 * gputimer.Stop();
 * cudaDeviceSynchronize(); // or some blocking call eg. cudaMemcpy(...)
 * printf("time taken for kernel code to run = %f", gputimer.Elapsed());

*/