On 03/10/2010 08:40 AM, Johann Cohen-Tanugi wrote: > Pauli, isn't it hopeless to follow the execution of the source code when > the crash actually occurs when I exit, and not when I execute. I would > have to understand enough of this umath_tests.c.src to spot a refcount > error or things like that???? > > On 03/10/2010 03:11 PM, Pauli Virtanen wrote: > >> Wed, 10 Mar 2010 10:28:07 +0100, Johann Cohen-Tanugi wrote: >> >> >> >>> On 03/10/2010 01:55 AM, Pauli Virtanen wrote: >>> >>> >>>>> more fun : >>>>> [co...@jarrett tests]$ pwd >>>>> /home/cohen/sources/python/numpy/numpy/core/tests [co...@jarrett >>>>> tests]$ python -c 'import test_ufunc' python: Modules/gcmodule.c:277: >>>>> visit_decref: Assertion `gc->gc.gc_refs != 0' failed. >>>>> Aborted (core dumped) >>>>> >>>>> >>>>> >>>> What happens if you only import the umath_tests module (or something, >>>> it's a .so under numpy.core)? >>>> >>>> >>>> >>> [co...@jarrett core]$ export >>> PYTHONPATH=/home/cohen/.local/lib/python2.6/site-packages/numpy/core: >>> >>> >> $PYTHONPATH >> >> >>> [co...@jarrett core]$ python >>> Python 2.6.2 (r262:71600, Jan 25 2010, 18:46:45) [GCC 4.4.2 20091222 >>> (Red Hat 4.4.2-20)] on linux2 Type "help", "copyright", "credits" or >>> "license" for more information. >>> >>> import umath_tests >>> >>> >>> python: Modules/gcmodule.c:277: visit_decref: Assertion `gc->gc.gc_refs >>> != 0' failed. >>> Aborted (core dumped) >>> >>> so this import also trigger the crash at exit... >>> >>> >> Then it is clear that the umath_tests module does something that is not >> permitted. It's possible that there is a some sort of a refcount error >> somewhere in the generalized ufuncs mechanisms -- that part of Numpy is >> not heavily used. >> >> Bug spotting challenge: Start from umath_tests.c.src:initumath_tests, >> follow the execution, and spot the bug (if any). >> >> Pauli >> >> PS. it might be a good idea to file a bug ticket now >> >> _______________________________________________ >> NumPy-Discussion mailing list >> NumPy-Discussion@scipy.org >> http://mail.scipy.org/mailman/listinfo/numpy-discussion >> >> >> > _______________________________________________ > NumPy-Discussion mailing list > NumPy-Discussion@scipy.org > http://mail.scipy.org/mailman/listinfo/numpy-discussion > Hi, I can say that I have the same problem with Fedora 11 and Python 2.6 (both Fedora version and Unladen Swallow versions) with numpy 2.0.0.dev8272. However, I first noticed it with scipy but I see it with numpy.
It does not appear to be with Python 2.5 with an svn version and I did not see it with numpy 1.3 or the removed numpy 1.4. Bruce _______________________________________________ NumPy-Discussion mailing list NumPy-Discussion@scipy.org http://mail.scipy.org/mailman/listinfo/numpy-discussion
