On 29 February 2012 13:58, deSitter <antimatter3...@gmail.com> wrote:
> Well there's no doubt that it's happening. The compiler had no problem > at all building and installing mpc, mpfr, gmp, ppl, and cloog, so it's > not a problem with how it handles complex numbers as such. It's > something in the interaction of Sage, gcc, and Sun Studio. > > -drl > If you are compiling with gcc, I don't understand why the presence (or absence) or Sun Studio should be relevant. That said, I do recall one bug, fixed some time ago, where Sage would build with gcc on Solaris if Sun Studio was NOT installed. But if Sun Studio was installed, the build would fail. That was due to SCons picking up the wrong C++ compiler. A lot of Sage would build with this obscure bug related to complex numbers in older versions of gcc. So just because mpc, mpfr and gmp all build, I would not assume there is no issues with complex numbers. What's the output of gcc -v (or if that is not useful, gcc -V)? I'm not on a Unix/Linux box now, so can't say what the right switch to gcc is. If you are trying to build with the Sun compiler, then things wont work very well. There is too much code in Sage that is not valid C/C++, so it wont compile with the Sun compiler. There are just too many GNU-specific extensions. William said he was going to switch t2.math on, but has not to my knowledge done so yet. If he does, I'll try to build Sage on Solaris SPARC. 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
