Implement a reduction in mpi_manual_reduce.c to compute the sum of all the processor IDs using MPI send/recv routines ra

[answerhappygod]

Implement a reduction in mpi_manual_reduce.c to compute the sum of all the processor IDs using MPI send/recv routines rather than the built-in MPI reduction. Your code should achieve a time conplexity of O(log n). As an example, 5 number of processors used should output a 10 (10 = 0+1+2+3+4).
#include <stdio.h>
#include <stdlib.h>
#include "mpi.h"

int main(int argc, char* argv[]) {
int numProcs, myProcID;

MPI_Init(&argc, &argv);
MPI_Comm_size(MPI_COMM_WORLD, &numProcs);
MPI_Comm_rank(MPI_COMM_WORLD, &myProcID);

int local_val = myProcID;//local value

/* TODO: Implement a reduction to compute the sum of all the proc
IDs using MPI send/recv routines rather than the built-in MPI
reduction. Your implementation should achieve a time complecity of O(log n). You should not use the built-in reduce method. */


if(myProcID == 0) {//check result
int expected = 0;
for(int i = 0; i < numProcs; ++i)
expected += i;
printf("The final value is %d, expected %d\n", local_val, expected);
}

MPI_Finalize();
return 0;
}