为什么在循环中使用 gettimeofday，第二个 diff 与其他 diff 如此不同？

Question

这个功能对我来说比较陌生，所以写了个小程序来熟悉一下。这是我的程序（只是在多次调用 gettimeofday 之间打印圈数）。

#include <stdio.h>
#include <unistd.h>
#include <stdlib.h>
#include <sys/time.h>
#include <pthread.h>

typedef struct timeval t_timeval;

t_timeval    get_diff(t_timeval prev)
{
    t_timeval current_time;
    gettimeofday(&current_time, NULL);
    printf("diff : %ld seconds and %ld micro seconds\n",
        current_time.tv_sec - prev.tv_sec, current_time.tv_usec - prev.tv_usec);
    return (current_time);
}

int        get_time_laps()
{
    int            i = 0;
    t_timeval    prev;

    gettimeofday(&prev, NULL);
    while (i++ < 50)
        prev = get_diff(prev);
    return (0);
}

int        main()
{
    get_time_laps();
    return (0);
}

结果是：

diff : 0 seconds and 0 microseconds
diff : 0 seconds and 47 microseconds
diff : 0 seconds and 1 microseconds
diff : 0 seconds and 1 microseconds
diff : 0 seconds and 1 microseconds
diff : 0 seconds and 1 microseconds
diff : 0 seconds and 1 microseconds
diff : 0 seconds and 1 microseconds
diff : 0 seconds and 2 microseconds
diff : 0 seconds and 1 microseconds
diff : 0 seconds and 1 microseconds
diff : 0 seconds and 0 microseconds
[...]
diff : 0 seconds and 1 microseconds

所以我想知道为什么第二个差异与其他差异如此不同。仅供参考，我做了几次测试，每次我都得到这个模式。

Answer 1

发生这种情况的原因很有趣。由于 gettimeofday 会使用一个经常被调用的系统调用，linux 使用一个调用 vdso 的东西，它会自动映射到您的程序中，以便访问与时间信息相关的东西，而不需要系统调用（因为系统调用成本高昂）。

其工作方式是当您调用 gettimeofday 时，您的程序将在 vdso 中查找符号。当它找到它时，我会将它缓存在内存中以便更快地访问。这就是为什么第一次通话较慢的原因。你可以找到更好的解释方法here

Oftentimes the C library will do detection with the first call and then cache the result for subsequent calls.

还有一个很好的方法来对 vdso 与实际系统调用进行基准测试 here

为什么在循环中使用 gettimeofday，第二个 diff 与其他 diff 如此不同？

Why the 2nd diff is so different from the others, using gettimeofday in a loop?

c

gettimeofday