我想知道為什么子進程使這么多文件保持打開狀態(tài)。我有一個例子，其中某些文件似乎永遠保持打開狀態(tài)（在子進程完成后，甚至在程序崩潰之后）。請考慮以下代碼：import aiofilesimport tempfileasync def main():    return [await fds_test(i) for i in range(2000)]async def fds_test(index):    print(f"Writing {index}")    handle, temp_filename = tempfile.mkstemp(suffix='.dat', text=True)    async with aiofiles.open(temp_filename, mode='w') as fp:        await fp.write('stuff')        await fp.write('other stuff')        await fp.write('EOF\n')    print(f"Reading {index}")    bash_cmd = 'cat {}'.format(temp_filename)    process = await asyncio.create_subprocess_exec(*bash_cmd.split(), stdout=asyncio.subprocess.DEVNULL, close_fds=True)    await process.wait()    print(f"Process terminated {index}")if __name__ == "__main__":    import asyncio    asyncio.run(main())這將一個接一個地生成進程（按順序）。我希望同時打開的文件數(shù)量也是一個。但事實并非如此，在某些時候我得到以下錯誤：/Users/cglacet/.pyenv/versions/3.8.0/lib/python3.8/subprocess.py in _execute_child(self, args, executable, preexec_fn, close_fds, pass_fds, cwd, env, startupinfo, creationflags, shell, p2cread, p2cwrite, c2pread, c2pwrite, errread, errwrite, restore_signals, start_new_session)   1410             # Data format: "exception name:hex errno:description"   1411             # Pickle is not used; it is complex and involves memory allocation.-> 1412             errpipe_read, errpipe_write = os.pipe()   1413             # errpipe_write must not be in the standard io 0, 1, or 2 fd range.   1414             low_fds_to_close = []OSError: [Errno 24] Too many open files我嘗試在沒有該選項的情況下運行相同的代碼，但它仍然崩潰。這個答案表明，這可能是問題的根源，錯誤也指向線 。但這似乎不是問題所在（在沒有該選項的情況下運行會產(chǎn)生相同的錯誤）。stdout=asyncio.subprocess.DEVNULLerrpipe_read, errpipe_write = os.pipe()