Summary: | sci-libs/scipy-0.8 segfaults | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | Paulo da Silva <psdasilva> |
Component: | [OLD] Unspecified | Assignee: | Gentoo Science Related Packages <sci> |
Status: | RESOLVED WORKSFORME | ||
Severity: | normal | CC: | bugs |
Priority: | High | ||
Version: | unspecified | ||
Hardware: | AMD64 | ||
OS: | Linux | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- |
Description
Paulo da Silva
2010-11-09 05:51:49 UTC
Try python -v ; help("modules") and see if its a specific library that crashes it. (In reply to comment #1) > Try python -v ; help("modules") and see if its a specific library that crashes > it. > Below what I got. I reemerged scipy. No success. The sequence import scipy; help("modules") works. # /usr/lib64/python2.6/site-packages/scipy/fftpack/__init__.pyc matches /usr/lib64/python2.6/site-packages/scipy/fftpack/__init__.py import scipy.fftpack # precompiled from /usr/lib64/python2.6/site-packages/scipy/fftpack/__init__.pyc # /usr/lib64/python2.6/site-packages/scipy/fftpack/fftpack_version.pyc matches /usr/lib64/python2.6/site-packages/scipy/fftpack/fftpack_version.py import scipy.fftpack.fftpack_version # precompiled from /usr/lib64/python2.6/site-packages/scipy/fftpack/fftpack_version.pyc # /usr/lib64/python2.6/site-packages/scipy/fftpack/basic.pyc matches /usr/lib64/python2.6/site-packages/scipy/fftpack/basic.py import scipy.fftpack.basic # precompiled from /usr/lib64/python2.6/site-packages/scipy/fftpack/basic.pyc dlopen("/usr/lib64/python2.6/site-packages/scipy/fftpack/_fftpack.so", 102); Segmentation fault What version of scipy? Please try rebuilding it (In reply to comment #3) > What version of scipy? Please try rebuilding it > The version is 0.8.0. What do you mean by "rebuilding it"? I wrote I already reemerged it. Anything else I can do? (In reply to comment #4) > (In reply to comment #3) > > What version of scipy? Please try rebuilding it > > > > The version is 0.8.0. > > What do you mean by "rebuilding it"? I wrote I already reemerged it. Anything > else I can do? Sorry, I somehow missed that you actually reemerged it. I just tried building scipy and tried with python 2.6.6 and python 2.7 - weren't able to reproduce. Next step will be backporting to 2.6.5. If you could try updating to 2.6.6 (in unstable), that would be helpful. (In reply to comment #5) > (In reply to comment #4) > > (In reply to comment #3) > If you could > try updating to 2.6.6 (in unstable), that would be helpful. > I emerged python-2.6.6-r1 and scipy (again). The same symptoms! Seg fault. If I import scipy then help("modules") works as expected. Has anyone in the sci herd been able to reproduce this? I'm at a loss here :( i was able to replicate a while ago but it seems gone now. probably something fishy going on in the dlopen so i will close it for now, feel free to reopen |