#9385: Building ATLAS goes into an infinite loop
--------------------+-------------------------------------------------------
Reporter: olazo | Owner: olazo
Type: defect | Status: new
Priority: major | Milestone: sage-4.6.1
Component: build | Resolution:
Keywords: | Author:
Upstream: N/A | Reviewer:
Merged: | Work_issues:
--------------------+-------------------------------------------------------
Changes (by vbraun):
* cc: was (added)
* milestone: sage-duplicate/invalid/wontfix => sage-4.6.1
Comment:
I think it is unavoidable that that ATLAS sometimes fails to get accurate
timings. Nor is it always desirable to tune it to precisely your CPU, for
example if you are building a binary distribution. Therefore, I propose we
add a new variable `SAGE_ATLAS_ARCH` with values
1. `auto` - run through the tuning process
2. `fast` - reasonably modern ~2005 cpu: sse3 on x86, Niagara SPARC, ...
3. `base` - really old cpus
4. A particular architecture from `ATLAS/CONFIG/ARCHS/*.tgz`, e.g.
`AMD64K10h32SSE3`
The ATLAS spkg then should then build a configuration according to
`SAGE_ATLAS_ARCH`. If it it is not set, try 2x`auto`, if that fails
`fast`, and if that fails `base`.
The Fedora rpm package has pre-built configurations for various cpus and
shows how to patch them into the ATLAS build system.
--
Ticket URL: <http://trac.sagemath.org/sage_trac/ticket/9385#comment:44>
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.
