On Fri, Jun 6, 2008 at 10:35 PM, Dr. David Kirkby <[EMAIL PROTECTED]> wrote:
> > Hi all, > Hi Dave, > > I've got a bit further with building sage-3.0.3.alpha1. After a few > hacks to ntl and flint, I got problems when building polybori It is a known issue - see below. > > 1) One of the problems that stops this working is in C++ code, and I > don't know C++ - only C. Anyone who knows C++ feel they might be able > to fix polybori on Solaris? If they are a *regular* poster to sage- > devel, but do not have access to a Sun, I don't mind giving access to > a Sun Ultra 60 via SSH. I'd obviously ask my machine is not used as a > hacking platform and would want to know a narrowish range of IPs they > would use so I can restrict access in the firewall. > > Just be aware, the Sun Ultra 60 is not the fastest machine in the > world - 2 x 450 MHz, 1 GB RAM. It will take *hours* to compile Sage > from scratch, so you might want to set off a compilation and return > later to check on progress. > > 2) Also worth noting is that there appears to be made a call to 'CC', > which is the name of the Sun C++ compiler. Given I've been using all > GNU tools up to this point, any compilation of C should have used gcc > and C++ used g++. The command CC should not have been called. I'm > guessing a makefile or some script is a bit broken. I'll probably be > able to sort that out without too much hassle. > > 3) The compiler flag -KPIC, which is used on Sun compilers, is being > passed to gcc - usually one finds gcc specific flags passed to a Sun > compiler, but this is the opposite! I guess one might call it it > SUNism, which makes a change from GNUism! > > I think gcc will simply ignore this, and since the GNU equivalent (- > fPIC) is also passed, don't think the compiler flag problem should > stop this compiling. > > 4) gcc reports Anachronism. > > 5) Due to work commitments, I am going to get far less chance to work > on this over the next few weeks. I will be working almost 24/7 on real > (paid) work! Hence don't expect too much work from me. > > Just for the record, this is a Sun Blade 2000, 2 x 1.2 GHz UltraSPARC > III+ CPUs, 8 GB RAM, Solaris 10 update 4, gcc 4.0.3, gmake 3.81. > > Below is the errors recorded in install.log > <SNIP> > gcc -o Cudd/epd/so_epd.o -c -O3 -ansi -Wno-long-long -Wreturn-type -g - > fPIC -KPIC -DNDEBUG -DPACKED -DHAVE_M4RI -DHAVE_IEEE_754 -DBSD -I/ > export/home/drkirkby/sage-3.0.3.alpha1/spkg/build/polybori-0.3.1.p3/ > src/boost_1_34_1.cropped -I/export/home/drkirkby/sage-3.0.3.alpha1/ > local/include/python2.5 -Ipolybori/include -ICudd/obj -ICudd/util - > ICudd/cudd -ICudd/dddmp -ICudd/mtr -ICudd/st -ICudd/epd Cudd/epd/epd.c > gcc: unrecognized option '-KPIC' > /opt/SUNWspro/bin/CC -o polybori/src/so_BoolePolyRing.o -c -O3 -ansi - > Wno-long-long -Wreturn-type -g -fPIC -ftemplate-depth-100 -g -fPIC - > KPIC -DNDEBUG -DPACKED -DHAVE_M4RI -DHAVE_IEEE_754 -DBSD -I/export/ > home/drkirkby/sage-3.0.3.alpha1/spkg/build/polybori-0.3.1.p3/src/ > boost_1_34_1.cropped -I/export/home/drkirkby/sage-3.0.3.alpha1/local/ > include/python2.5 -Ipolybori/include -ICudd/obj -ICudd/util -ICudd/ > cudd -ICudd/dddmp -ICudd/mtr -ICudd/st -ICudd/epd polybori/src/ > BoolePolyRing.cc > CC: Warning: Option -ansi passed to ld, if ld is invoked, ignored > otherwise > CC: Warning: Option -Wno-long-long passed to ld, if ld is invoked, > ignored otherwise > CC: Warning: Option -Wreturn-type passed to ld, if ld is invoked, > ignored otherwise > CC: Warning: Option -fPIC passed to ld, if ld is invoked, ignored > otherwise > CC: Warning: Option -ftemplate-depth-100 passed to ld, if ld is > invoked, ignored otherwise > CC: Warning: Option -fPIC passed to ld, if ld is invoked, ignored > otherwise > <SNIP> Well, we are using CC here when the makefile should be using g++. The port of PolyBoRi to the Sun Forte compiler will likely be an adventure. I forced SCons for my Solaris builds to prefer g++ over CC on Solaris since the default is CC over g++. Cheers, Michael > > > > > > --~--~---------~--~----~------------~-------~--~----~ 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://www.sagemath.org -~----------~----~----~----~------~----~------~--~---
