➜
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.
Created on 2002-05-11 00:08 by mattbehrens, last changed 2022-04-10 16:05 by admin. This issue is now closed.
| Files | ||||
|---|---|---|---|---|
| File name | Uploaded | Description | Edit | |
| va_copy.patch | mattbehrens, 2002-05-11 00:08 | __va_copy patches | ||
| Messages (2) | |||
|---|---|---|---|
| msg39973 - (view) | Author: Matt Behrens (mattbehrens) | Date: 2002-05-11 00:08 | |
This issue was discovered when preparing for OpenBSD 3.1, and compiling on our non-i386 arches. Let me quote a mail from drahn@openbsd.org: > [Tell the Python guys] the vararg handling is poor, and that this is possible solution, but not a great solution. If possible It would be best to not parse the varargs argument twice. > Different architectures deal with varargs differently, __va_copy is a way that some architectures use do a deep copy. __va_copy is present in solaris and powerpc (*BSD and Linux) as far as I know. Attached is the patches we are using to build our Python package; without it we cannot build Python 2.2 on arches like powerpc as the built python cores. Python 2.1 does not need these patches. |
|||
| msg39974 - (view) | Author: Martin v. Löwis (loewis) *  |
Date: 2002-07-28 10:24 | |
Logged In: YES user_id=21627 Thanks for the patch, applied as abstract.c 2.93.6.5 stringobject.c 2.147.6.6 getargs.c 2.90.6.1 modsupport.c 2.58.16.2 abstract.c 2.104 stringobject.c 2.171 getargs.c 2.93 modsupport.c 2.61 |
|||
| History | |||
|---|---|---|---|
| Date | User | Action | Args |
| 2022-04-10 16:05:18 | admin | set | github: 36585 |
| 2002-05-11 00:08:34 | mattbehrens | create | |