Jaap Spies wrote: > Dr. David Kirkby wrote: >> Jaap Spies wrote: >> >>> Hi Dave, >> Hi Jaap >> >> >>> I tried to build gcc-4.4.2 from source, but failed. Got gcc-4.3.2 from the >>> repo. >>> Put gcc and friends in a directory $HOME/bin, changed my .bashrc so my path >>> is >> GCC can be difficult to build on Solaris. >> > > I now have: > j...@opensolaris:~/Downloads/build-gcc-4.4.2$ /usr/local/gcc-4.4.2/bin/gcc -v > Using built-in specs. > Target: i386-pc-solaris2.11 > Configured with: ../gcc-4.4.2/configure --prefix=/usr/local/gcc-4.4.2 > --with-gmp=/usr/local --with-mpfr=/usr/local --with-ld=/usr/ccs/bin/ld > --without-gnu-ld > --enable-languages=c,c++,fortran > Thread model: posix > gcc version 4.4.2 (GCC) > j...@opensolaris:~/Downloads/build-gcc-4.4.2$ which as > /usr/ccs/bin/as
You might find 4.4.2 is ok. I'm not 100% sure on that one. There was a bug which gave some odd messages from the assembler in earlier release. It was supposed to be fixed in the 4.3 series somewhere, but I still got it on 4.4.1. But perhaps it is fixed in 4.4.2. > So as you suggested I'll do it again with an other as from binutls-2.20. > >>> /usr/ccs/bin:/export/home/jaap/bin:/usr/gnu/bin:/usr/bin:/usr/X11/bin:/usr/sbin:/sbin >> That should not (will not) work!! >> > > It works because I move make out of the way! > mv make sunmake That's generally a bad idea to do that. It's much better to put a GNU version of make early in your path. I do something like mkdir $HOME/bins-for-sage cp /usr/sfw/bin/gtar $HOME/bins-for-sage/tar then after building GNU make (I don't think it is pre-installed), copy that to HOME/bins-for-sage, then finally export PATH=$HOME/bins-for-sage:$PATH now the gnu versions of make and tar are in your path first. You may find upgrades will fail if you move the system tools. One something like HP-UX, that would fail miserably, as sometimes the kernel has to be rebuilt. In which case things would no doubt go terribly wrong if the version of make was not as expected. >> http://trac.sagemath.org/sage_trac/ticket/7781 >> ([with patches ; needs review] Update prereq to version 0.6) >> > > I'll look into that. You wont need it, but it should stop people trying to build Sage in a way which will not work. >> Given the issue with 'make' I'm surprised you managed to get as far you di. >> Hopefully if #7781 gets reviewed and incorporated into Sage, such an error >> will >> be caught within the first minute of building Sage. >> > > There wasn't a make issue. OK >>>> Again, you can use the package manager to install that for you. >>>> >>> Built openssl-0.9.8l from source. Seems to work. >> Yes, agreed. Though you might note it configures in a 64-bit mode, and builds >> only 64-bit libraries. >> > > For the moment this is all I need as I'm building with SAGE64="yes" I think that is the most sensible way. See my post the other day. I'm reluctant to fight with OpenSSL to get it to build 32-bit binaries. I'd rather fight the bits of Sage that don't work in 64-bit mode. This machine has 12 GB RAM. I'm reluctant to run Sage in a way that can't use that 12 GB. >> It seems to be quite inevitable that someone will at some point be building >> Sage >> on mobile phones and similar devices (some have tried already). It also seems >> inevitable that someone will try on high performance systems like Cuda. It's >> quite likely different compilers may be used, needing different options. As >> much >> as possible needs to be in one place, not in 100 different .spkg files. >> > > As you stated before many spkgs are not in a good shape to handle this. > > I'm surprised I got so far > j...@opensolaris:~/Downloads/sage-4.3$ ls spkg/installed > bzip2-1.0.5 gnutils-2.2.1.p4 libpng-1.2.35 readline-6.0.p1 > termcap-1.3.1.p0 > cliquer-1.2.p2 gnutls-2.2.1.p4 opencdk-0.6.6.p2 > sage_scripts-4.3 zlib-1.2.3.p4 > conway_polynomials-0.2 libgcrypt-1.4.4.p1 prereq-0.5 > scons-1.2.0 > dir-0.1 libgpg_error-1.6.p2 python-2.6.2.p4 > sqlite-3.6.19.p0 > > All built in 64 bit mode. > > Cheers, > > Jaap Great, that is good news. I think William should be pleased someone else is working on OpenSolaris port too. He has someone asking him the other day about it, and I know someone at Sun is keen to know when Sage is running on the T5240, which it does now at: http://t2nb.math.washington.edu:8000/ I believe when 4.3.1 comes out, it should build on Solaris 10 (SPARC) irrespective of whether Sun Studio is installed or not (currently, it fails if Sun Studio is installed, but #6595 is marked as a 'blocker' for 4.3.1 and has positive review, so should sort out that issue). So hopefully Open Solaris wont be too difficult, especially now there is someone else working on it. 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
