> On Apr 25, 4:33 pm, Francois Bissey <francois.bis...@canterbury.ac.nz> > > wrote: > > You could build numpy without lapack to check if it is the source of the > > problem. I gave someone instructions on how to do that a while ago. > > But that may have been on sage-release rather than sage-devel. > > It was me you sent them to. When calculus/riemann.pyx was not passing > tests (on sage-release). But you sent them to me off-list. But I > found them in my email store. I just posted them on the Trac ticket > for this. > > Thanks for reminding me about that. ;-) That's an experiment > somebody with the right configuration could try once they know they > have the failures. > > Meanwhile, another reply on the numpy-discussion list is casting > doubts on the blas+lapack combo on OSX. > > http://permalink.gmane.org/gmane.comp.python.numeric.general/43707 > Actually do we know what version of ATLAS/BLAS/LAPACK is shipped with OSX? I know there are some instabilities depending on hardware with some ATLAS-3.9.xx releases. There is some numerical noise and some errors that we have tracked down to it in sage-on-gentoo.
I have been contacted off-list for details but didn't realise it was not on the list. If you could repost instruction here for all to see it would be good. In fact I wonder if we should switch lapack off in numpy by default. It is not necessary for numpy to work and it triggers the occasional issue. Francois This email may be confidential and subject to legal privilege, it may not reflect the views of the University of Canterbury, and it is not guaranteed to be virus free. If you are not an intended recipient, please notify the sender immediately and erase all copies of the message and any attachments. Please refer to http://www.canterbury.ac.nz/emaildisclaimer for more information. -- To post to this group, send an email to sage-devel@googlegroups.com To unsubscribe from this group, send an email to sage-devel+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URL: http://www.sagemath.org
