On May 1, 1:53 am, "Dr. David Kirkby" <david.kir...@onetel.net> wrote:
> Somewhat confused by the way Sage appeared to have failed at two
> different points before, I decided to delete all the sage directory
> and start again. This time, the Solaris tool chain had the right paths
> in before I started.
<SNIP>
> clisp fails to build though.
This is a known issue on Sparc with this toolchain. I have a special
gcc 3.2.3, but it is not included in the binary for now :(
> However, but I can't work out what the error message about clisp even
> means. It's not clear that the compiler could not compile something,
> or the linker can't link something. I'm just confused. There is a
> message that the output from make goes to a file
Any clisp release I ever tried to build on Sparc is basically broken
with anything later than gcc 3.2.3. There was a thread on clisp-devel
about this and it boils down to tthe following:
* the clisp people know which file miscompiles
* the file is *huge*, ergo the gcc folks want a simpler test case
* RMS was consulted and his conclusion was: The gcc folks need to fix
this anyway
Result: clisp is broken on Solaris with any reasonable compiler. It
also frequently miscompiles with -O2 on any platform with *any* gcc >=
4.0.0. Guess why we are switching to ecl next week :)
I have a precompiled clisp binary I use. Another option is to use ecl.
Since it is 2:20 am I am about to go to sleep, but in the morning I
should post a prebuild clisp.spkg.
Cheers,
Michael
--~--~---------~--~----~------------~-------~--~----~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~----------~----~----~----~------~----~------~--~---
