Does anyone know off hand if F14 is going to get Python 2.7.1/2.7.2? I'm trying to help move miro from Fedora to RPM Fusion due to the new ffmpeg requirement and it's having issues with the version of python in F14 but not F15.
I figure there's a reason but I went ahead and tried a mock build of the F15 source in F14 and ran into a build problem[1]. I'm assuming that's the reason?
Thanks, Richard
[1] build.log excerpt:
gcc -fno-strict-aliasing -O2 -g -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector --param=ssp-buffer-size=4 -m64 -mtune=generic -D_GNU_SOURCE -fPIC -fwrapv -O2 -g -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector --param=ssp-buffer-size=4 -m64 -mtune=generic -D_GNU_SOURCE -fPIC -fwrapv -O2 -g -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector --param=ssp-buffer-size=4 -m64 -mtune=generic -D_GNU_SOURCE -fPIC -fwrapv -Xlinker -export-dynamic -o python-debug \ Modules/python.o \ -L. -lpython2.7_d -lpthread -ldl -lutil -lm ./libpython2.7_d.so: undefined reference to `python_function__return_semaphore' ./libpython2.7_d.so: undefined reference to `python_function__entry_semaphore' collect2: ld returned 1 exit status
On 08/01/2011 12:21 AM, Richard Shaw wrote:
Does anyone know off hand if F14 is going to get Python 2.7.1/2.7.2? I'm trying to help move miro from Fedora to RPM Fusion due to the new ffmpeg requirement and it's having issues with the version of python in F14 but not F15.
Here's the review ticket on RPMfusion, if anyone's interested:
https://bugzilla.rpmfusion.org/show_bug.cgi?id=1845
I figure there's a reason but I went ahead and tried a mock build of the F15 source in F14 and ran into a build problem[1]. I'm assuming that's the reason?
That build problem is similar to a recent problem affecting F-16 Python builds:
https://bugzilla.redhat.com/show_bug.cgi?id=711427
While the systemtap in F-14 is of a higher revision than the one which fixes one Python build problem (comment #3) on that report, it appears that the Python Makefile fix (comment #6) introduced in 2.7.2-3.fc16 hasn't been backported to F-15 yet.
Getting late here so this is pure conjecture at this point; I tried a scratch build of F-16's python for F-14 to see if the miro problem on F-14 is solely due to python 2.7.0, but it's failing with the same error:
http://koji.fedoraproject.org/koji/getfile?taskID=3242306&name=build.log
On 07/31/2011 05:07 PM, Michel Alexandre Salim wrote:
On 08/01/2011 12:21 AM, Richard Shaw wrote:
I figure there's a reason but I went ahead and tried a mock build of the F15 source in F14 and ran into a build problem[1]. I'm assuming that's the reason?
That build problem is similar to a recent problem affecting F-16 Python builds:
https://bugzilla.redhat.com/show_bug.cgi?id=711427
While the systemtap in F-14 is of a higher revision than the one which fixes one Python build problem (comment #3) on that report, it appears that the Python Makefile fix (comment #6) introduced in 2.7.2-3.fc16 hasn't been backported to F-15 yet.
Despite the apparent sync of revision numbers, systemtap in F14 doesn't actually have all the same patches applied. I just ran a few mock builds to confirm python's FTBFS, and I also confirmed that this is fixed by systemtap 1.6. That update is now waiting in bodhi...
Josh
devel@lists.stg.fedoraproject.org