Thomas Heller thel...@ctypes.org added the comment:
Tested myself, on a mipsel debian qemu instance.
--
resolution: accepted - fixed
status: open - closed
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue4305
Thomas Heller thel...@ctypes.org added the comment:
I've applied py-issue-4305.patch and reran configure 2.61. Committed as
rev 72081 (trunk), 72082 (py3k branch), 72083 (release26-maint branch),
72084 (release30-maint branch).
Can someone please confirm that it works now, so that I can close
Roumen Petrov bugtr...@roumenpetrov.info added the comment:
Thomas, the report is for:
ffi_sources += ffi_platforms['MIPS']
KeyError: 'MIPS'
you may test patch without to cross-compile. As example create two
directories.
In first one run:
Thomas Heller thel...@ctypes.org added the comment:
I wonder if fixes for issues like these (ctypes build errors on some
plaforms) could be verified by using cross compilers on linux?
--
___
Python tracker rep...@bugs.python.org
Joshua Kinard ku...@gentoo.org added the comment:
Is there any movement on this perchance? Just bumped into this on my
MIPS platform and discovered this bug.
--
nosy: +kumba
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue4305
Thomas Heller thel...@ctypes.org added the comment:
Joshua Kinard schrieb:
Joshua Kinard ku...@gentoo.org added the comment:
Is there any movement on this perchance? Just bumped into this on my
MIPS platform and discovered this bug.
There is no MIPS buildbot at the moment, and I'm unclear
Roumen Petrov bugtr...@roumenpetrov.info added the comment:
May I propose a simple py-issue-4305.patch. Let me know if patch work
for you.
--
nosy: +rpetrov
Added file: http://bugs.python.org/file13083/py-issue-4305.patch
___
Python tracker
Mark Miller m...@mirell.org added the comment:
Not entirely. I applied what you had for the autoconf section to my
configure, since I do not have autoconf installed in my environment.
Specifically, this part of configure is what's messing with the build:
if test x$TARGET = xMIPS; then
Roumen Petrov bugtr...@roumenpetrov.info added the comment:
So I attach patch with changes in Modules/_ctypes/libffi/configure for
the tests.
Added file: http://bugs.python.org/file13084/py-issue-4305+configure.patch
___
Python tracker rep...@bugs.python.org
Mark Miller m...@mirell.org added the comment:
Would it be possible to get some feedback as to whether this is a
candidate for Python 2.6 on this patch?
___
Python tracker rep...@bugs.python.org
http://bugs.python.org/issue4305
Mark Miller m...@mirell.org added the comment:
It's actually due to the merge of the libffi3 branch back in March,
which removed the logic for configure to deal with MIPS in
fficonfig.py.in. I've attached a patch which allows the build to work
correctly on the QEMU instance of the MIPSEL machine
Mark Miller m...@mirell.org added the comment:
Adding this as affecting the 2.6 branch as well, since I initially
tested on a 2.6.1 tarball with this problem.
--
versions: +Python 2.6
___
Python tracker rep...@bugs.python.org
New submission from Matthias Klose [EMAIL PROTECTED]:
seen on the mipsel buildbot
running build
running build_ext
INFO: Can't locate Tcl/Tk libs and/or headers
Traceback (most recent call last):
File ./setup.py, line 1880, in module
main()
File ./setup.py, line 1875, in main
13 matches
Mail list logo
