<div dir="ltr"><br><div class="gmail_quote"><div dir="ltr">On Fri, Sep 8, 2017 at 12:52 PM Antoine Pitrou <<a href="mailto:solipsis@pitrou.net">solipsis@pitrou.net</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">On Fri, 8 Sep 2017 12:40:34 -0700<br>
Nathaniel Smith <<a href="mailto:njs@pobox.com" target="_blank">njs@pobox.com</a>> wrote:<br>
><br>
> PyPy just abandons everything when shutting down, instead of running<br>
> finalizers. See the last paragraph of :<br>
> <a href="http://doc.pypy.org/en/latest/cpython_differences.html#differences-related-to-garbage-collection-strategies" rel="noreferrer" target="_blank">http://doc.pypy.org/en/latest/cpython_differences.html#differences-related-to-garbage-collection-strategies</a><br>
><br>
> So that might be a useful source of experience.<br>
<br>
CPython can be embedded in applications, though, and that is why we try<br>
to be a bit more thorough during the interpreter cleanup phase.<br></blockquote><div><br></div><div>Indeed. My gut feeling is that proposing to not run finalizers on interpreter shutdown is a non-starter and would get the pep rejected. We've previously guaranteed that they were run unless the process dies via an unhandled signal or calls os._exit() in CPython.</div><div><br></div><div>-gps</div></div></div>