How Can I Update Global Variables from Subprocesses in Multiprocessing?

Multiprocessing | Global Variable Updates Not Returned to Parent

In multiprocessing, each subprocess runs as a separate Python instance with its own memory and global state. Therefore, any updates made to global variables within a subprocess are not visible to the parent process. Additionally, data passed between processes using shared proxies or queues must be pickleable.

To address the specific issue regarding the non-pickleability of a LORR object, consider the following approaches:

1. Create a pickleable subclass of LORR:
   This involves creating a new class that inherits from LORR and implements __getstate__ and __setstate__ methods to enable pickling.
2. Pass a reference to a ctype object:
   Find a way to store the relevant data in a ctype object and pass a reference to that object to the child process. However, this approach may not be reliable.
3. Use queues to send messages:
   Establish a communication system between the parent and child processes using queues. The parent can send method calls to the child, and the child can return the results via the queue.

Remember, the most suitable approach will depend on the specific implementation and capabilities of the LORR object.

The above is the detailed content of How Can I Update Global Variables from Subprocesses in Multiprocessing?. For more information, please follow other related articles on the PHP Chinese website!