/******************************************************************************
* FILE: omp_reduction.c
* DESCRIPTION:
*   OpenMP Example - Combined Parallel Loop Reduction - C/C++ Version
*   This example demonstrates a sum reduction within a combined parallel loop
*   construct.  Notice that default data element scoping is assumed - there
*   are no clauses specifying shared or private variables.  OpenMP will 
*   automatically make loop index variables private within team threads, and
*   global variables shared.
* AUTHOR: Blaise Barney  5/99
* LAST REVISED: 04/06/05
******************************************************************************/
/**
* This program computes the dot product of two vectors.
* Online source:
* https://computing.llnl.gov/tutorials/openMP/samples/C/omp_reduction.c
**/
#include <civlc.h>
#include <stdio.h>
#include <stdlib.h>
#include "omp.cvh"
$input int N;

int main (int argc, char *argv[]) {
  int   i;
  double a[N], b[N], sum; 

  /* Some initializations */
  for (i=0; i < N; i++)
    a[i] = b[i] = i * 1.0;
  sum = 0.0;
  
  /* The procedure of each thread for the for loop */
  void __for_0(int __tid, int __start, int __end, int __extra){	
//	if(__extra > 0) 
//      printf("   __for_1 thread %d starts from index %d to %d, %d\n", __tid, __start, __end, __extra);
//	else
//	  printf("   __for_1 thread %d starts from index %d to %d\n", __tid, __start, __end);
    for(int __i = __start; __i < __end; __i++) {
  	  sum = sum + (a[__i] * b[__i]);
//      printf("loop %d\n", __i);
    }
    if(__extra > 0) {
      sum = sum + (a[__extra] * b[__extra]);
      printf("loop %d\n", __extra);
    }
  }
  
  /* Helper variales introduced for the parallel for loop */
  $proc __for_0_procs[OMP_NUM_THREADS];
  
  /* Create processes to conduct the parallel for loop  */
  for(i = 0; i < OMP_NUM_THREADS; i++) {
	  int __start = __for_start(i, N);
	  int __end = __for_end(i, N);
	  int __extra = __for_extra(i, N);
	
	  __for_0_procs[i] = $spawn __for_0(i, __start, __end, __extra);
  }
  
  /** The orginal OpenMP parallel for loop
  #pragma omp parallel for reduction(+:sum)
    for (i=0; i < n; i++){
      sum = sum + (a[i] * b[i]);
      printf("loop %d\n", i);
    }
  **/
  
  /* Join parallel for processes  */
  for(i = 0; i < OMP_NUM_THREADS; i++) {
	  $wait(__for_0_procs[i]);
  }
  /* Print result  */
  printf("   Sum = %f\n",sum);
}