在Python3.8中的多进程SharedMemory中共享字节对象

Question

我有一个很大的字节对象，我想把它放在SharedMemory中，这样我的多处理任务就可以访问它。我使用的是docs中描述的ShareableList。

from multiprocessing import shared_memory
s2v_a = Sense2Vec().from_disk(SENSE2VEC_FOLDER)
s2v_a_bytes = s2v_a.to_bytes()
print(sys.getsizeof(s2v_a_bytes)) #prints <class 'bytes'>
print(type(s2v_a_bytes)) #prints 4220733334 (4.2Gb)
memory = shared_memory.ShareableList([s2v_a_bytes])

但是，当我尝试创建ShareableList时，我得到一个格式不小于8的AssertionError。我可以看出这与struct packing format有关。

Traceback (most recent call last):
  File "/home/user/anaconda3/envs/uniqueness/lib/python3.8/runpy.py", line 193, in _run_module_as_main
    return _run_code(code, main_globals, None,
  File "/home/user/anaconda3/envs/uniqueness/lib/python3.8/runpy.py", line 86, in _run_code
    exec(code, run_globals)
  File "/home/user/dev/uniqueness/backend/app/coding.py", line 39, in <module>
    memory = shared_memory.ShareableList([s2v_a_bytes])
  File "/home/user/anaconda3/envs/uniqueness/lib/python3.8/multiprocessing/shared_memory.py", line 295, in __init__
    assert sum(len(fmt) <= 8 for fmt in _formats) == self._list_len
AssertionError

代码中的注释

Because values are packed into a memoryview as bytes, the struct
    packing format for any storable value must require no more than 8
    characters to describe its format."""

但据我所知，我没有对文档做任何不同的事情。

Answer 1

文档规定bytes和str在ShareableList中的最大大小是10M字节。4.2 GB远远超出了这个限制。