如何在python中使用多重处理？

Question

I'm trying to do a task where I would assign something to a list. I'll do it with multiprocessing. After all task is finish, I want to sum all of it. But, What I get is not what i expected. Do you know why do this happen? Below is just a sample code. The result should be 45, instead of 0.

from multiprocessing import Process

def a(b):
    for i in range(10):
        b[i] = i

b = [0 for _ in range(10)]

p = Process(target=a, args=(b,))
p.start()
p.join()

print sum(b)

Answer 1

The reason is because the list b is mutated in a separate process. When the process gets joined, the original process knows nothing about the mutations to b . The way to fix this is to have your list managed by a multiprocessing.Manager .

from multiprocessing import Process,Manager

def a(b):
    for i in range(10):
        b[i] = i

b = [0 for _ in range(10)]

#create a manager to manage access to `b`
manager = Manager()
b = manager.list(b)

p = Process(target=a, args=(b,))
p.start()
p.join()

print sum(b)