Message 180833 - Python tracker

➜

This issue tracker has been migrated to GitHub, and is currently read-only.
For more information, see the GitHub FAQs in the Python's Developer Guide.

Author	cowwoc
Recipients	brian.curtin, cowwoc, r.david.murray
Date	2013-01-28.06:53:46
SpamBayes Score	-1.0
Marked as misclassified	Yes
Message-id	<1359356027.43.0.714476711247.issue17058@psf.upfronthosting.co.za>
In-reply-to

Content
I read http://mail.python.org/pipermail/python-dev/2012-August/121460.html and I believe they are wrong. I have personally run into these problems (each library maintaining its own CRT with separate heaps, file handles, etc) when static linking was used, but when dynamic linking is used there is only one CRT instance and therefore you only end up with one heap, set of file handles, etc. I've never seen the kind of problems you're referring to if dynamic linking is used. My guess is that someone (Python or an extension) is using static linking which is causing these problems. Is that a reasonable assumption?

I read http://mail.python.org/pipermail/python-dev/2012-August/121460.html and I believe they are wrong.

I have personally run into these problems (each library maintaining its own CRT with separate heaps, file handles, etc) when static linking was used, but when dynamic linking is used there is only one CRT instance and therefore you only end up with one heap, set of file handles, etc. I've never seen the kind of problems you're referring to if dynamic linking is used.

My guess is that someone (Python or an extension) is using static linking which is causing these problems. Is that a reasonable assumption?

History
Date	User	Action	Args
2013-01-28 06:53:47	cowwoc	set	recipients: + cowwoc, r.david.murray, brian.curtin
2013-01-28 06:53:47	cowwoc	set	messageid: <1359356027.43.0.714476711247.issue17058@psf.upfronthosting.co.za>
2013-01-28 06:53:47	cowwoc	link	issue17058 messages
2013-01-28 06:53:46	cowwoc	create