On 7/23/07, Justin C. Walker <[EMAIL PROTECTED]> wrote: > Tried this on both the PowerMac (2x2.7GHz G5) and MacBook Pro (2.33 > Mhz Intel Core 2 Duo). SAGE built without an apparent problem on > both systems.
Excellent. This is great news. I also posted a slightly newer alpha5 tarball tonight here http://sage.math.washington.edu/home/was/sage2.7.1/, and build-tested it on 32-bit debian, 64-bit debian, 32-bit fedora, 32-bit mandriva, OSX PowerPC, and OS X intel, and it built fine on all, and most doctests pass. (A few doctest failed, since I upgrade pari to the latest stable version, but neglected to include the Galois groups database in the upgraded package.) So I think we've succeeded at creating an easy-to-distribute and build scipy. > FWIW, it seems to take at least 40-45 processes > running simultaneously to do it! Interesting. At what point (or points) in the make processes does this occur? Building each spkg is a separate event, so it should be the case that the process usage should depend entirely on which package you're building. Maybe there is one particularly strange package that tries to build using lots of processes at once? > In any case, the results from "-testall" are: Thanks. The failures are fine and I've subsequently fixed them. They had to do with some interesting new code that Christian Wuthrich wrote for doing computations involving the p-adic BSD conjecture. > sage -t matrix/matrix2.pyx > sage -t schemes/elliptic_curves/ell_rational_field.py > sage -t schemes/elliptic_curves/padic_lseries.py > Total time for all tests: 1684.8 seconds > Testing of examples currently not implemented. > ---------------------------------------------------------------------- > > sage -t functions/transcendental.py > sage -t matrix/matrix2.pyx > sage -t schemes/elliptic_curves/ell_rational_field.py > sage -t schemes/elliptic_curves/padic_lseries.py > Total time for all tests: 3277.8 seconds > Testing of examples currently not implemented. > ---------------------------------------------------------------------- > > Why the tests take 2x on the G5 is a puzzle, but it is more heavily > used than the MacBook Pro. I've noticed that sort of performance difference for many years; it's why I didn't like Macs until last year when Apple switched from PowerPC to Intel. Now I really really like Macs. > Are the failures expected? Yes, and understood. -- William Stein Associate Professor of Mathematics University of Washington http://www.williamstein.org --~--~---------~--~----~------------~-------~--~----~ To post to this group, send email to sage-devel@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/sage-devel URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/ -~----------~----~----~----~------~----~------~--~---
