#14754: Update ATLAS to stable version 3.10.1
-------------------------------------------------------------------+--------
Reporter: jdemeyer |
Owner: jdemeyer
Type: enhancement |
Status: needs_review
Priority: critical |
Milestone: sage-5.11
Component: packages: standard |
Resolution:
Keywords: | Work
issues:
Report Upstream: N/A |
Reviewers: Benjamin Jones, Karl-Dieter Crisman, Dmitrii Pasechnik, Georg
Weber, François Bissey, John Palmieri, Volker Braun
Authors: Volker Braun, Jeroen Demeyer, Jean-Pierre Flori | Merged
in:
Dependencies: |
Stopgaps:
-------------------------------------------------------------------+--------
Comment (by jpflori):
Replying to [comment:29 vbraun]:
> Replying to [comment:28 jpflori]:
> > And about 3DNow, we don't want to put it in base or fast anyway
because it's AMD only
>
> Kind of a moot point since we don't distribute binaries built on such an
old machine. Also, `SAGE_FAT_BINARY` is "base" so the 3DNow would be fair
game for the "fast" target. Except that, there, we already require SSE2.
It does not tackle the problem that 3DNow is AMD only...
I could imagine someone wanting to distribute a sage bdist with a "fast"
but still somehow portable ATLAS.
You can still argue that in that case he can very well remove 3DNow
himself if he wants ot play with fire...
I just don't like to add non-universal extensions to predifined targets
although I completely agree that if we consider that someone who builds
with fast should only expect it to work on the machine where it was built
then trying 3DNow makes sense.
Anyway, does using 3DNow in addition of SSE or SSE2 gives any speedup?
I guess not, so let's forget about 3DNow for "fast" because we have SSE2
and for "base" because we don't want people to end up distributing AMD
only binaries (though it's highly unlikely, I agree)...
--
Ticket URL: <http://trac.sagemath.org/sage_trac/ticket/14754#comment:37>
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.
For more options, visit https://groups.google.com/groups/opt_out.
