Message 361867 - 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	methane
Recipients	jd, mark.dickinson, methane, vstinner
Date	2020-02-12.07:46:53
SpamBayes Score	-1.0
Marked as misclassified	Yes
Message-id	<1581493614.55.0.502307310834.issue39599@roundup.psfhosted.org>
In-reply-to

Content
https://github.com/python/cpython/blob/95905ce0f41fd42eb1ef60ddb83f057401c3d52f/Include/cpython/objimpl.h#L89 _Py_AS_GC and all APIs to get PyGC_Head from PyObject* are not only unstable, but also private. I thought PyGC_Head must not be used directly from extensions, so we changed its layout. Since Julien said "my code does not access PyGC_Head at all", I still think it was not real ABI breakage. I suppose other serious bug (e.g. dungling pointer, missing refcount increment, etc) caused the segfault.

https://github.com/python/cpython/blob/95905ce0f41fd42eb1ef60ddb83f057401c3d52f/Include/cpython/objimpl.h#L89

_Py_AS_GC and all APIs to get PyGC_Head from PyObject* are not only unstable, but also private.

I thought PyGC_Head must not be used directly from extensions, so we changed its layout.

Since Julien said "my code does not access PyGC_Head at all", I still think it was not real ABI breakage.

I suppose other serious bug (e.g. dungling pointer, missing refcount increment, etc) caused the segfault.

History
Date	User	Action	Args
2020-02-12 07:46:54	methane	set	recipients: + methane, mark.dickinson, vstinner, jd
2020-02-12 07:46:54	methane	set	messageid: <1581493614.55.0.502307310834.issue39599@roundup.psfhosted.org>
2020-02-12 07:46:54	methane	link	issue39599 messages
2020-02-12 07:46:53	methane	create