This is a timer that times in clock cycles. It uses __rdtscp(). It only workes on unix. I will not port it to windows as i dont think an application like this would make much use there. You can port it easily.

Timer.c

/*** PRINTING STUFF ***/
#include <stdio.h>  /* printf() */
#include <stdlib.h> /* system() */
#include <stdint.h> /* fixed width ints */

/*** UNIX STUFF ***/
#include <x86intrin.h> /* __rdtscp() */
#include <unistd.h>    /* sleep()    */

#define SHOW_CPUID 0      /* True/False */
#define GET_TSC_PER_SEC 1 /* True/False */
#define TSC_WAIT_TIME 2   /* Seconds    */

uint64_t tscPerSecond()
{
  uint64_t start, end; /* Start/End Variables */
  uint32_t dummy[4]; /* Dummy to pass for CPUID */

  /* Record Sleep Function */
  start = __rdtscp(dummy);

  /* Sleep has an error width of */
  /* around 1ms, which is fine */
  sleep(TSC_WAIT_TIME);

  /* Record End Time */
  end = __rdtscp(dummy);

  /* Return Ticks elapsed in a Second */
  return (end - start) / TSC_WAIT_TIME;
}

int main(int argc, char** argv)
{
  uint64_t start, end; /* Start/End Variables */
  uint32_t scpu[4], ecpu[4]; /* CPU Information Gathering */

  /* CPUID Prevents Out Of Order Execution */
  /* Get start time / CPU information */
  start = __rdtscp(scpu);

  /* Run Specify Information */
  for(int i = 1; i < argc; ++i)
    system(argv[i]);

  /* Get end time / CPU information */
  end = __rdtscp(ecpu);

  /* Make Space */
  printf("\n");

  /* Used if you want to show CPUID */
  /* Mainly here to prevent CPUID,  */
  /* from getting optimized away    */
  if(SHOW_CPUID)
  {
    printf("+--------- STARTING  CPUID ---------+\n");
    printf("| EAX: 0x%08x | EBX: 0x%08x |\n", scpu[0], scpu[1]);
    printf("| ECX: 0x%08x | EDX: 0x%08x |\n", scpu[2], scpu[3]);
    printf("+-----------------------------------+\n\n");

    printf("+---------- ENDING  CPUID ----------+\n");
    printf("| EAX: 0x%08x | EBX: 0x%08x |\n", ecpu[0], ecpu[1]);
    printf("| ECX: 0x%08x | EDX: 0x%08x |\n", ecpu[2], ecpu[3]);
    printf("+-----------------------------------+\n\n");
  }

  /* Get Cycles */
  const uint64_t pTime = end - start; // Get overall time
  printf("Execution Cycles:  %lu ticks\n", pTime);

  if(GET_TSC_PER_SEC)
  {
    /* Get Cycle Speed */
    printf("Calculating Clock Speed... (etc=%is)\r", TSC_WAIT_TIME);
    fflush(stdout);

    const uint64_t CPS = tscPerSecond();
    if(CPS > 100000)
      if(CPS > 100000000)
        if(CPS > 100000000000)
          printf("Cycles Per Second: %luGHz", CPS/1000000);
        else printf("Cycles Per Second: %luMHz", CPS/1000000);
      else printf("Cycles Per Second: %luKHz", CPS/1000);
    else printf("Cycles Per Second: %luHz", CPS);
    printf(" (ESTIMATE)                                  \n");

    /* Get Nano Time */
    const uint64_t nano = (pTime * 1000000000) / CPS;

    /* Unit Simplification */
    if(nano > 100000)
      if(nano > 100000000)
        if(nano > 100000000000)
          printf("Execution Time:    %lus", nano/1000000);
        else printf("Execution Time:    %lums", nano/1000000);
      else printf("Execution Time:    %luus", nano/1000);
    else printf("Execution Time:    %luns", nano);
    printf(" (ESTIMATE)\n");
  }
}