Burcin Erocal wrote: > On Sun, 27 Sep 2009 13:47:49 -0700 > William Stein <wst...@gmail.com> wrote: > >> On Sun, Sep 27, 2009 at 6:38 AM, Dr. David Kirkby >> <david.kir...@onetel.net> wrote: >>> Using >>> >>> * Solaris 10 update 7 on SPARC >>> * sage-4.1.2.alpha2 >>> * Sun Studio 12.1 >>> * An updated configure script to allow the Sun compiler to be >>> used >>> >>> I found that pynac-0.1.8.p2 will not build. It appears the Sun C++ >>> compiler simply does not like the C++ code in pynac 0.1.8.p2 >> Does ginac build? Maybe this is something you can report to the >> ginac list too. > > I guess a recent version of ginac would build successfully. AFAIR, at > least some of the problems below were also seen while trying to compile > with MSVC, and there were patches to fix them. > > I will try to grab the relevant patches from ginac, and see if they are > enough to fix this for pynac. I would appreciate some help testing my > changes though. > > I have access to t2, but I suppose this message didn't come from that > machine. How can I reproduce it? Would I get the same error on t2? Is > there a (perhaps half built) Sage installation on t2 that I could copy > to my home directory and do "sage -sh" to get going? > > Thanks. > > Burcin
I don't have a recent build on t2 which will work for you. I would suggest you start from scratch with the following steps. 1) Type $ export CC=/opt/SUNWspro/bin/cc $ export CXX=/opt/SUNWspro/bin/CC $ export SAGE_FORTRAN=/opt/SUNWspro/bin/f95 This will tell Sage to use all the Sun compilers, though in practice several parts of Sage ignore the settings of CC and CXX. 2) extract the tar file 3) You *MUST* apply the fix at http://sagetrac.org/sage_trac/ticket/7021 Do not take the files from the attachments on the ticket, but from here. http://sage.math.washington.edu/home/kirkby/Solaris-fixes/prereq-0.4/ Download the two files to the directory $SAGE_HOME/spkg/base/ That fix is awaiting review, so you might want to comment on it later. You should see a warning that you are not using GCC, and that Sage has only every been built with gcc. Without that fix, any attempt to use the Sun compiler would fail, as the configure script in Sage would say your gcc is too old. 4) Type 'make' You should soon see that prereq-0.4 is being built, rather than the prereq-0.3 which is usual. 5) You will get a few failures. I got the following fail before 'pynac-0.1.9.p0' was built. It would be interesting if you get any more/less on 't2', as I have not tried recently. ntl-5.4.2.p9 eclib-20080310.p7 flint-1.5.0.p2 lapack-20071123.p0 atlas-3.8.3.p9 givaro-3.2.13rc2 linbox-1.1.6.p2 matplotlib-0.99.0 To get past the failures, I just used touch spkg/installed/ntl-5.4.2.p9 or whatever package would not build. I documented all the failures I observed on my own machine at: http://sagetrac.org/sage_trac/ticket/7056 6) I'm not sure what prerequisites pynac might have. But at some point you can try: $ ./sage -f pynac-xxx I actually used a slightly newer version of the Sun tools than what is on 't2', but that should not matter. It would be interesting if you get the same error as I did, at http://sagetrac.org/sage_trac/ticket/7029 Dave --~--~---------~--~----~------------~-------~--~----~ 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 -~----------~----~----~----~------~----~------~--~---
