I would be interested in knowing if anyone has successfully built Sage on Solaris 10 (intel and/or sparc); due to timelines I've decided to go with another math package until I can clearly see stability across Solaris, Linux and Mac OSX; I am still testing with a select group of undergrad students but the masses (1000+ students) are using a package.
We are examining the following options for next year in line with some research a number of my students are doing: 1. start with R and extend it to handle some of the features resident in sage; we had not problems building and running R on solaris 10 sparc and intel; 2. remove unstable packages in Sage and re-write some of the core packages it uses that were actually based on experimental work publshe; I believe the major issues we encountered with sage centered around not the core packages but additional packages not in common use. 3. extend our current math package - vendor based with support for an extension api and distribute this back to the masses. Again I would be interested if there is any further progress on a Solaris Port on intel or sparc; I believe a few more months is needed in my research lab to get it to work on Solaris, although from what I am seeing our modifications would not represent the same software currently distributed. Thanks, Emanuel On Wednesday, January 22, 2014 12:20:38 PM UTC-8, Emanuel Koseos wrote: > > Hello, > > I've been trying to build Sage 5.13 and Sage 6.0 on a Solaris 10 x86 in > 32-bit mode, and planning to do the same on a Solaris Sparc; this is the > standard OS of our university. > > I'm using GNU Binutils 2.2.4 (AS) and Solaris LD (/usr/ccs/bin/ld) as this > combination is stable across many open source packages. > with GCC 4.1.2 and Sage 5.13 > > I receive compile errors on fpLLL to do with the NaN and isinfinte > defintions; for this fix I have e-mailed the developer or will look to fix > it myself > > Givaro results in link (ld) errors > with GCC 4.7.3 (built from within Sage) and 5.13 or 6.0 > > - fpLLL-4.0.4 seems to pass but I have not verified the package yet > - Givaro 3.7.1 fails with Sage 5.13 and 6.0 under GCC 4.7.3-sage > > > Here is the typicall error for Givaro: > > ld: elf error: file library/tools/.libs/libgivtools.a: elf_getarsym > ld: fatal: File processing errors. No output written to > .libs/libgivaro.so.0.5.1 > collect2: error: ld returned 1 exit status > > Is this a known bug; and wondering on how dependent Sage is on these two > packages; if so can remove them or is there a fix available. > > FYI: the other two packages I had issues with are ATLAS which I put in a > fix myself (builds/tests smoothly on Solaris x86, Sparc and GCC 4.1.2 or > 4.7.3) > and gf2x ( another fix I put in) > > Thank you very much for your help. > Emanuel Koseos (ekos...@gmail.com) > Dept. Mathematics/Computer Science > UFV (Canada) > > -- You received this message because you are subscribed to the Google Groups "sage-support" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-support+unsubscr...@googlegroups.com. To post to this group, send email to sage-support@googlegroups.com. Visit this group at http://groups.google.com/group/sage-support. For more options, visit https://groups.google.com/d/optout.
