Question

I am using the following code to compute execution time in milli-secs.

struct timespec tp;
 if (clock_gettime (CLOCK_REALTIME, &tp) == 0)
  return ((tp.tv_sec * 1000000000) + tp.tv_nsec);
 else
  return ;

Can you please tell me whether this is correct?
Let’s name this function comptime_nano().

Now, I write the following code in main() to check execution times of following operations.

 unsigned long int a, b, s1, s3;
 a = (unsigned long int)(1) << 63;
 b = (unsigned long int)(1) << 63;
 btime = comptime_nano();
 s1 = b >> 30;
 atime = comptime_nano();
 printf ("Time =%ld for %lu\n", (atime - btime), s1);
 btime = comptime_nano();
 s3 = a >> 1;
 atime = comptime_nano();
 printf ("Time =%ld for %lu\n", (atime - btime), s3);

To my surprise, the first operation takes about roughly 4 times more time than the second. Again, if I change the relative ordering of these operations, the respective timings change drastically.

Please comment…

Answer 1

clock_gettime is not accurate enough for that kind of measurement. If you need to measure operations like that, do the operation several thousand (or several million times) in a loop before comparison. The two operations above should take the same amount of time but the second in your example code does not have the overhead of loading a, b, s1, and s3 into the processor’s cache.

Also, what’s going on here?

struct timespec tp;
 if (clock_gettime (CLOCK_REALTIME, &tp) == 0)
  return ((tp.tv_sec * 1000000000) + tp.tv_nsec);
 else
  return ;

The first return is illegal if the function returns void, and the second is illegal if it does not return void….

EDIT: 1000000000 also overflows the range of int.