Message 366344 - Python tracker

Message366344

Author	pablogsal
Recipients	corona10, eelizondo, gregory.p.smith, nascheme, pablogsal, pitrou, steve.dower, tim.peters, vstinner
Date	2020-04-13.22:51:54
SpamBayes Score	-1.0
Marked as misclassified	Yes
Message-id	<1586818315.02.0.315524751787.issue40255@roundup.psfhosted.org>
In-reply-to

Content
> gc.freeze() has a similar issue, no? This function also comes from Instagram specific use case ;-) Not really because this patch is much more impacting. gc.freeze() moves objects into a permanent generation making them not participate in the GC so the only leaks are cycle-based. With this patch immortal object will leak every strong reference that they have even if they don't participate in cycles (but also if they participate in cycles).

Content

> gc.freeze() has a similar issue, no? This function also comes from Instagram specific use case ;-)

Not really because this patch is much more impacting. gc.freeze() moves objects into a permanent generation making them not participate in the GC so the only leaks are cycle-based. With this patch immortal object will leak every strong reference that they have even if they don't participate in cycles (but also if they participate in cycles).

History
Date	User	Action	Args
2020-04-13 22:51:55	pablogsal	set	recipients: + pablogsal, tim.peters, nascheme, gregory.p.smith, pitrou, vstinner, steve.dower, corona10, eelizondo
2020-04-13 22:51:55	pablogsal	set	messageid: <1586818315.02.0.315524751787.issue40255@roundup.psfhosted.org>
2020-04-13 22:51:55	pablogsal	link	issue40255 messages
2020-04-13 22:51:54	pablogsal	create