Did the build work? (Maybe I need to wait 7 more hours for an answer.) I use Ubuntu 11.10, 64 bit, and I have a core i7-2600. I think building Sage using many processes usually takes me about 2 hours + 15 minutes, where the 2 hours is all ATLAS, and the 15 minutes is everything else. I'm building 4.8 now with the SAGE_ATLAS_LIB environment variable set, so maybe it will be faster (but I have a feeling that it doesn't make a difference for me). If you're still having trouble, I can post a binary. Maybe I should do that anyway.
When ATLAS is trying to tune, it can get really messed up if there are other programs using the cpu too much, because that can make its timings inconsistent, and it can even get messed up if nothing else is running but cpu frequency scaling is turned on (and it is on my default, as it should be), and it can also messed up by the "turboboost" stuff that all processors do nowadays. So you might try turning off cpu frequency scaling, and if you get desperate you can probably go into the bios to turn off turboboost. (These clever tricks really make CPUs better, but they make it so much harder to get reliable benchmarks...) Also, some things that make the build go faster are: export SAGE_PARALLEL_SPKG_BUILD="yes" export MAKE="make -j8" (For some reason, you can't just type "make -j8" to make things build in parallel.) On Thu, Feb 9, 2012 at 6:02 AM, firebird <graham.gerr...@gmail.com> wrote: > Thanks Jason/Volker > > Will persevere... Type of computer is I5-2320 Quadcore... Not on your > list of known processors. > Will try and find how to save tuning information after next test. > I am getting the message that compilation will take many hours. So I > will give it 12-24 hours. > I guess that I am compiling many of the packages to which there is a > SAGE interface as well. > > For me and others, how long is a typical compiliation and how big > should the install.log file be? > Are there any sample install.logs that I could check out? > > Finally, is there a Ubuntu 11.10 version pending? > If not, I would be happy to go through a few extra loops and try to > publish mine! > > FireBird > > > On Feb 8, 6:27 pm, Jason Grout <jason-s...@creativetrax.com> wrote: > > On 2/8/12 10:39 AM, firebird wrote: > > > > > Thanks for the prompt response, but it is barely credible... > > > > Atlas will sometimes do tuning, which means that it will run the same > > program lots of times with different values in order to ascertain > > exactly what parameters to use to get the fastest linear algebra. > > Sometimes this tuning takes hours and hours, if atlas doesn't already > > know about your CPU and computer configuration. The important thing > > then is to save that tuning information for the next time you compile > > atlas, or even better, submit that tuning information back upstream to > > the atlas project. > > > > How long did you let it go before terminating? What kind of computer > > are you compiling on? > > > > Thanks, > > > > Jason > > -- > To post to this group, send email to sage-support@googlegroups.com > To unsubscribe from this group, send email to > sage-support+unsubscr...@googlegroups.com > For more options, visit this group at > http://groups.google.com/group/sage-support > URL: http://www.sagemath.org > -- To post to this group, send email to sage-support@googlegroups.com To unsubscribe from this group, send email to sage-support+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-support URL: http://www.sagemath.org
