On Jan 14, 2008 6:39 AM, DuaneKaufman <[EMAIL PROTECTED]> wrote: > > Indeed, I finally got through the build with your suggestion, though > it took 1900 minutes. Is this normal?
No, that is not normal. On a 1.8Ghz Opteron under Linux using one processor, it takes 168 minutes (< 3 hours) real time to completely build Sage-2.10.alpha2. Thus there must have been an issue with ATLAS tuning still. On a 2.6Ghz core 2 duo it takes 109 minutes real time to build sage-2.9.3. -- William > Thanks for your help, > Duane > > On Jan 11, 7:55 am, mabshoff <[EMAIL PROTECTED] > dortmund.de> wrote: > > On Jan 11, 2:37 pm, DuaneKaufman <[EMAIL PROTECTED]> wrote: > > > > > Hi, > > > > Hello Duane, > > > > > I downloaded the sage 2.9.3 tarball to my notebook (Debian, Centrino > > > 1.5Ghz, 2Gb RAM), and started a build. > > > > > That was about 2 days ago, and it is _still_ grinding away at the > > > build (still seems to be progressing, plenty of memory free, no disk > > > thraashing) > > > > > Is this normal? Other than this, nothing else seems amiss with the > > > functioning of the notebook. Is there anything else I can check? > > > > You have hit an ATLAS build bug, namely that ATLAS mis-detects Pentium > > Ms as CoreDuos (seehttp://trac.sagemath.org/sage_trac/ticket/1740). > > That has been fixed for the next, i.e. Sage 2.10, release. > > > > To solve this kill the build, > > downloadhttp://sage.math.washington.edu/home/mabshoff/release-cycles-2.10/alp... > > and copy that file into spkg/standard. There should already be > > atlas-3.8.p6.spkg in that directory. Then restart the build via make > > from the base directory of Sage. If the build keeps running that long > > you have hit another bug. You should also disable power management > > while building Sage (you have a laptop after all). For instructions > > see the question "QUESTION: Sage 2.9 and higher fails compiling ATLAS > > on Linux. How can I fix this?" athttp://wiki.sagemath.org/faq > > > > > > According to top, there is a _lot_ of time spent running programs like > > > xctfc, and other x<nnnn> variants > > > > Yep, those are the tuning routines, which in your case do a complete > > tuning, which is very, very expensive. With a pretuned build ATLAS > > takes about 16 minutes to build on a Opteron 248. > > > > > Thanks, > > > Duane. > > > > Let me know if anything else odd happens or if you have any more > > questions. > > > > Cheers, > > > > Michael > > > -- William Stein Associate Professor of Mathematics University of Washington http://wstein.org --~--~---------~--~----~------------~-------~--~----~ To post to this group, send email to [email protected] To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/sage-forum URLs: http://sage.math.washington.edu/sage/ and http://sage.scipy.org/sage/ -~----------~----~----~----~------~----~------~--~---
