mabshoff wrote: > > > On Feb 5, 12:28 pm, Jaap Spies <j.sp...@hccnet.nl> wrote: >> mabshoff wrote: >>> something deeper inside libSingular. I have posted some speculation >>> why some runs fail and other runs do not, so I am not going to repeat >>> it here and bore 99.5% of Sage devs to death again :) >> I'm one of the other 0.5%! This will not stop me :) > > Well, feel free to fix the problem and to submit a patch :) >
This is really boring :-)! >> I have news from my Fedora 9 install. I built sage-3.2.3 again and tested: >> No mysterious failures! >> ---------------------------------------------------------------------- >> All tests passed! >> Total time for all tests: 5741.2 seconds >> Please see /home/jaap/downloads/sage-3.2.3/tmp/test.log for the complete log >> from this test. >> [j...@paix sage-3.2.3]$ >> >> Upgrading to sage-3.2.alpha0 gives: >> The following packages will be upgraded: >> atlas-3.8.2.p1 doc-3.3.alpha0 ecm-6.2.1.p0 examples-3.3.alpha0 >> extcode-3.3.alpha0 gmp-mpir-svn1555.p0 ipython-0.8.4.p1 jmol-11.6.rc8.p0 >> jquery-1.2.6.p0 >> jqueryui-1.6r807svn.p0 jsmath-3.6a.p0 libgcrypt-1.4.3 matplotlib-0.98.3.p5 >> mercurial-1.1.2 sage-3.3.alpha0 sage_scripts-3.3.alpha0 tinyMCE-3.2.0.2.p0 >> >> Now running ./sage -check: >> >> First occurence: >> sage -t "devel/sage/sage/categories/functor.pyx" >> A mysterious error (perphaps a memory error?) occurred, which may have >> crashed doctest. >> [0.8 s] >> etcetera, >> >> ---------------------------------------------------------------------- >> The following tests failed: >> >> sage -t "devel/sage/sage/categories/functor.pyx" >> sage -t "devel/sage/sage/combinat/combinat.py" >> sage -t "devel/sage/sage/rings/polynomial/toy_d_basis.py" >> sage -t "devel/sage/sage/interfaces/gap.py" >> sage -t "devel/sage/sage/schemes/elliptic_curves/ec_database.py" >> Total time for all tests: 5908.4 seconds >> Please see /home/jaap/downloads/sage-3.2.3/tmp/test.log for the complete log >> from this test. >> >> All but the toy_d_basis.py they are of type: >> A mysterious error (perphaps a memory error?) occurred, which may have >> crashed doctest. >> >> You probably know very well what changed from 3.2.3 -> 3.3.alpha0 > > Yes, but nothing touched libSingular and there is no obvious culprit > which touched the memory subsystem. > >> Thoughts? > > This bug has been around and detected ever since we tested on FC 9 64 > bit, but on the affected machine we had the kernel downgraded to FC8 > since we blamed somewhat incorrectly the FC9 kernel/glibc. But now we > cannot ignore the problem any more, so now that we know what goes > wrong we need to fix it :) > I remember I noticed this problem as early as 20090119 on Fedora 9, 32 bits. You seem to know more! I'm clueless to be honnest. For one thing I know for sure(?) on this very system testing sage-3.2.3 has no failures. Testing 3.3.alpha0+ has mysterious errors! We need to fix this :-)! Jaap >> Jaap > > Cheers, > > Michael > > > --~--~---------~--~----~------------~-------~--~----~ To post to this group, send email to sage-devel@googlegroups.com To unsubscribe from this group, send email to sage-devel-unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URLs: http://www.sagemath.org -~----------~----~----~----~------~----~------~--~---
