MPI_Accumulate后得到不同的答案

在下面的代码中，每次执行后我都会得到不同的答案，例如 1、2、3 mpirun -n 3 ./a.out而正确的答案是 3。为什么会这样？

#include <mpi.h>
#include <iostream>
int main()
{
    MPI_Init(NULL, NULL);
    int rank, size;
    MPI_Comm_rank(MPI_COMM_WORLD, &rank);
    MPI_Comm_size(MPI_COMM_WORLD, &size);
    int* a;
    MPI_Win win;
    MPI_Win_allocate(size*sizeof(int), sizeof(int), MPI_INFO_NULL, MPI_COMM_WORLD, &a, &win);
    int one = 1;
    int target_proc = 0;
    int shift = target_proc;
    MPI_Win_lock(MPI_LOCK_EXCLUSIVE, target_proc, 0, win);
    MPI_Accumulate(&one, 1, MPI_INT, target_proc, shift, 1, MPI_INT, MPI_SUM, win);
    MPI_Win_unlock(target_proc, win);
    if (rank == target_proc)
        std::cout << a[shift] << std::endl;
    MPI_Win_free(&win);
    MPI_Finalize();
    return 0;
}