#6743: cygwin metaticket: port Sage to Microsoft Windows (via Cygwin): stage 1
--
make building Sage automatic
-------------------------------+--------------------------------------------
Reporter: was | Owner: was
Type: enhancement | Status: needs_info
Priority: major | Milestone: sage-5.9
Component: cygwin | Resolution:
Keywords: sd31 sd32 | Work issues:
Report Upstream: N/A | Reviewers: Jean-Pierre Flori, Dmitrii
Pasechnik, Karl-Dieter Crisman, Mike Hansen, William Stein, Luis Tabera
Authors: | Merged in:
Dependencies: | Stopgaps:
-------------------------------+--------------------------------------------
Comment (by jpflori):
Replying to [comment:209 leif]:
> Replying to [comment:206 kcrisman]:
> > In the prereq spkg log, at the end:
> > {{{
> > ...
> > checking for sqrtl in -lm... no
> > configure: You have an outdated and/or broken math library.
> > configure: error: Exiting, since the library function 'sqrtl()' was
not found.
> > ...
> > }}}
> > Supposedly this was (I think?) provided by Cephes, but of course that
never actually built on Cygwin, as JP discovered.
>
> WHAT?! The only reason for including Cephes as a standard package was
Cygwin, because it was said to lack some `long double` versions ('`l`'
suffix) of standard math functions (such as `sqrt()`).
I don't really know, but Cygwin libm was really lacking a LOT of
functions, some of them were pulled from Cephes at some point, but not al
of what Cephes provides, and the libm Cygwin now ships seems mostly
sufficient to build Sage.
At least it was until a few realease and an update of R which requires one
"long double"function, but that got easily patched (rather than requiring
again to build Cephes which I would be happy with as well, but seems
harder to maintain).
>
> > Grepping through the entire Sage root directory shows that sqrtl
mostly comes in play with Pynac and !Scipy/Numpy, and we seem to even (?)
define our own sqrtl...
>
> I recall somebody somewhere replacing (`#define`ing?) `sqrtl()` by
`sqrt()`, arguing `double` precision was sufficient there, for the sake of
Cygwin, and that AFAIK didn't get removed even after the inclusion of
Cephes. (Not 100% sure whether it was really `sqrt()`; probably some
other function, but Cygwin was lacking its `long double` implementation.)
>
> > oh, also in the partitions code in combinat this is needed. The Pari
references I think are to sqrtlift, something else. Maybe we don't need
this any more? I presume this was added by David Kirkby for Solaris or
other platforms with weird base libraries.
>
> Certainly not for Solaris... ;-)
>
--
Ticket URL: <http://trac.sagemath.org/sage_trac/ticket/6743#comment:213>
Sage <http://www.sagemath.org>
Sage: Creating a Viable Open Source Alternative to Magma, Maple, Mathematica,
and MATLAB
--
You received this message because you are subscribed to the Google Groups
"sage-trac" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/sage-trac?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.
