/*
AS/400 High-Resolution Clock Reader

Choose your browser's option to 'Save As...' to download this code
example.  Send the program to your AS/400 using FTP or similar method
and compile it using the development facilities supplied there.  This
program was developed on a V4R3 system.

This small program that is furnished by IBM is a simple example to
provide an illustration.  This example has not been thoroughly tested
under all conditions.  IBM, therefore, cannot guarantee or imply
reliability, serviceability, or function of this program.  All programs
contained herein are provided to you "AS IS".  THE IMPLIED WARRANTIES OF
MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE EXPRESSLY
DISCLAIMED.

-----------------------------------------------------------------------

You can use the Materialize Time-of-Day (mattod) function to get the
current tick on the system clock.  Currently, the system clock is really
only 49 bits long, but it is returned nicely in a 64-bit unsigned integer
as the first (high-order) 49 bits.  The low-order 15 bits are not clocked.
They are used, though.  No 2 succeeding calls to the mattod() function
will return the same number.  The low-order 15 bits are incremented to
cover the situation of 2 calls within the same clock tick.  Naturally, the
2 calls ought not return the same time.

There are 8 microseconds to a single clock tick.  This is the finest
resolution presented by the clock above the MI.  This translates to a
precision of 0.008 ms.  (Actually, the potenial for sub-nanosecod
precision is built in, but the low 15 bits are not clocked as of V4R3.)
Simply right-shift the 64-bit int by 15 bits to get the number of ticks.
Interestingly enough, this number will be 0x0FFFFFFFFFFFF at the end of
1999, and 0x1000000000000 in the first instant of the year 2000.  How's
that for century-checking efficiency, required by Y2K fixes?  (This was
done way before 1990!)

This sample program ignores the last 15 bits and displays 10 tick counts.
Note that this uses 64-bit integer support, which is first implemented by
ILE C at V4R3.  For earlier releases, use a structure of 2 32-bit unsigned
integers or the _MI_Time type, which would be syntactically correct.

*/

#include <stdio.h>
#include <unistd.h>
#include <QSYSINC/MIH/MATTOD>

void main()
{
   unsigned long long t[10];
   int i;

   /* get 10 consecutive time readings */
   for (i=0; i<10; i++){
      mattod((char*)&(t[i]));
      printf(" "); /* slight delay */
   }
   printf("\n");

   for (i=0; i<10; i++)
   {
      printf("Original Hex:0x%llX " , t[i]);
      /* right-shift the clock bits from high to low for number of ticks */
      t[i] = t[i] >> 15;
      printf("ticks:%lld \n", t[i]);
   }
}