将马随机放在国际象棋的8*8棋盘的某个方格中，马按走棋规则进行移动。要求每个方格上只进入一次，走遍棋盘上全部64个方格。编制非递归程序，求出马的行走路线，并按求出的行走路线，将数字1，2，…，64依次填入一个8*8的方阵，输出之。

这里给出一个非递归的深度优先搜索算法，使用栈来模拟递归过程：

class Pos:
    def __init__(self, x, y):
        self.x = x
        self.y = y

def knight_tour():
    board = [[0]*8 for _ in range(8)]
    moves = [(-2, -1), (-2, 1), (-1, -2), (-1, 2), (1, -2), (1, 2), (2, -1), (2, 1)]
    stack = [Pos(0, 0)]
    board[0][0] = 1
    idx = 2

    while len(stack) > 0:
        pos = stack[-1]
        found = False
        for move in moves:
            next_pos = Pos(pos.x + move[0], pos.y + move[1])
            if next_pos.x < 0 or next_pos.x > 7 or next_pos.y < 0 or next_pos.y > 7:
                continue
            if board[next_pos.x][next_pos.y] == 0:
                stack.append(next_pos)
                board[next_pos.x][next_pos.y] = idx
                idx += 1
                found = True
                break
        if not found:
            board[pos.x][pos.y] = 0
            stack.pop()
            idx -= 1

    for row in board:
        print(row)

运行该函数即可得到马的行走路线并填入方阵中。