#9356: make SAGE_ATLAS_LIB work on Solaris
-------------------------------+--------------------------------------------
Reporter: jhpalmieri | Owner: drkirkby
Type: defect | Status: closed
Priority: minor | Milestone: sage-4.5.3
Component: solaris | Resolution: fixed
Keywords: | Author: John Palmieri
Upstream: N/A | Reviewer: David Kirkby
Merged: sage-4.5.3.alpha0 | Work_issues:
-------------------------------+--------------------------------------------
Comment(by drkirkby):
Replying to [comment:19 fbissey]:
> Replying to [comment:15 drkirkby]:
>
> > Yes, you are probably right. That's something I can address at #9780.
In fact, I don't think there's any need to have the lapack package in
Sage.
>
> The sage spkg doesn't use it. However:
>
> * numpy can use it.
> * scipy '''requires''' it.
> * linbox pretends to want it. - the reality is more subtle than that.
Which is why even so liblinbox*.so are compiled against lapack sage
components compiled against these do not need to be linked to lapack as
well.
> * R can use it.
What's confusing me is that in the ATLAS package, in the file
{{{make_correct_shared.sh}}} there is this line:
{{{
lapack_command="ld -L"$f95_dir" -L"$SAGE_LOCAL"/lib -shared
-soname liblapack.so -o liblapack.so --whole-archive liblapack.a --no-
whole-archive -lc -lm -lf95"
}}}
which makes a shared library {{{liblapack.so}}} from a static library
{{{liblapack.a}}}. But the LAPACK package also creates {{{liblapack.a}}}
Since LAPACK is listed as a dependency of ATLAS, it would appear that
LAPACK gets built first, creates {{{liblapack.a}}}, then ATLAS overwrites
{{{liblapack.a}}}. (Overwriting files like this is not supposed to happen
in Sage, but if ATLAS creates a more optimised version, I can see there
may be the logical thing to do).
So it seems if you want to make use of a pre-existing {{{liblapack.a}}}
from ATLAS, which means creating a symbolic link, would need to remove the
{{{liblapack.a}}} created by LAPACK first, then create the link.
(Otherwise, I think creating the link might fail, though I'm not sure.) I
need to rush out, so dont have time to test that.
Dave
--
Ticket URL: <http://trac.sagemath.org/sage_trac/ticket/9356#comment:20>
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 post to this group, send email to [email protected].
To unsubscribe from this group, send email to
[email protected].
For more options, visit this group at
http://groups.google.com/group/sage-trac?hl=en.
