Роман Донченко <d...@corrigendum.ru> added the comment:
> What you do is split 'wait' into two parts: first it waits for me process to
> become reapable without actually reaping it. On Linux you can do this with
> waitid+WNOWAIT. On kqueue platforms you can do it with kqueue.
> Then, you take a lock, and use it to atomically call waitpid/waitid to reap
> the process + update self.returncode to record that you've done so.
> In send_signal, we use the same lock to atomically check whether the process
> has been reaped, and then send the signal if it hasn't.
It's a good idea, but it would break the scenario where two threads call wait()
concurrently. It would create this race condition:
1. Thread A reaps the process.
2. Thread B thinks the process is still running, so it calls waitid+WNOHANG on
a stale PID, with unpredictable results.
3. Thread A sets self.returncode.
What is needed here is a reader-writer lock. subprocess.wait would work like
this (pseudocode):
with lock taken for reading:
os.waitid(..., WNOHANG)
with lock taken for writing:
os.waitid(...)
self.returncode = ...
Whereas subprocess.send_signal would work like this:
with lock taken for reading:
os.kill(...)
Unfortunately, it doesn't seem like Python has reader-writer locks in the
library...
----------
nosy: +SpecLad
_______________________________________
Python tracker <rep...@bugs.python.org>
<https://bugs.python.org/issue38630>
_______________________________________
_______________________________________________
Python-bugs-list mailing list
Unsubscribe:
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
