mpi_parallel_shit

// R-B For (i+j) % 2 == 1:

      for(i=id + 1; i<=N-2; i += num_procs)
      for(j= 1 + i % 2; j<=N-2; j+= 2)
      if ((i + j) % 2 == 1)
      {
        double b;
        b = w*((A[i-1][j]+A[i+1][j]+A[i][j-1]+A[i][j+1])/4. - A[i][j]);
        eps =  Max(fabs(b),eps);
        A[i][j] = A[i][j] + b;
      }


      for(i = id + 1; i<=N-2; i += num_procs)
      for(j= 1 + (i + 1) % 2; j <= N-2; j+= 2)
      if ((i + j) % 2 == 1)
      {
        double b;
        b = w*((A[i-1][j]+A[i+1][j]+A[i][j-1]+A[i][j+1])/4. - A[i][j]);
        eps =  Max(fabs(b),eps);
        A[i][j] = A[i][j] + b;
      }

    // matrix sync for all the programs:
    for(i = id + 1; i<=N-2; i += num_procs)
      for(j= 1 + (i + 1) % 2; j <= N-2; j+= 2)
          MPI_Bcast(&A[i][j], 1, MPI_DOUBLE, id, MPI_COMM_WORLD);

    for(i = id + 1; i<=N-2; i += num_procs)
      for(j= 1 + i % 2; j <= N-2; j+= 2)
          MPI_Bcast(&A[i][j], 1, MPI_DOUBLE, id, MPI_COMM_WORLD);

    // Barrier:
    MPI_Barrier(MPI_COMM_WORLD);
    // R-B For (i+j) % 2 == 0:

      for(i = id + 1; i<=N-2; i += num_procs)
      for(j=1 + i % 2; j <= N-2; j += 2)
      if ((i + j) % 2 == 0)
      {
        double b;
        b = w*((A[i-1][j]+A[i+1][j]+A[i][j-1]+A[i][j+1])/4. - A[i][j]);
        A[i][j] = A[i][j] + b;
      }


      for(i = id + 1; i<=N-2; i += num_procs)
      for(j=1 + (i + 1) % 2; j <= N-2; j += 2)
      if ((i + j) % 2 == 0)
      {
        double b;
        b = w*((A[i-1][j]+A[i+1][j]+A[i][j-1]+A[i][j+1])/4. - A[i][j]);
        A[i][j] = A[i][j] + b;
      }
}