On Wed, Feb 8, 2012 at 10:27 AM, Jason Grout
<[email protected]> wrote:
> On 2/8/12 10:39 AM, firebird wrote:
>>
>> Thanks for the prompt response, but it is barely credible...
>
>
> Atlas will sometimes do tuning, which means that it will run the same
> program lots of times with different values in order to ascertain exactly
> what parameters to use to get the fastest linear algebra. Sometimes this
> tuning takes hours and hours, if atlas doesn't already know about your CPU
> and computer configuration. The important thing then is to save that tuning
> information for the next time you compile atlas, or even better, submit that
> tuning information back upstream to the atlas project.
Also, the spkg-install attempts to build ATLAS in four different ways
before giving up:
print 'First attempt: automatic tuning.'
rc = build()
if rc!=0:
print 'Waiting 5 minutes...'
time.sleep(5*60)
print 'Second attempt: Re-running make.'
rc = make_core()
if rc!=0:
print 'Waiting 5 minutes...'
time.sleep(5*60)
try:
print 'Third attempt: use "fast" options.'
rc = build(arch='fast')
if rc!=0:
print 'Fourth attempt: use "base" options.'
rc = build(arch='base')
except NotImplementedError:
pass
assert rc==0, 'Failed to build ATLAS.'
>
> How long did you let it go before terminating? What kind of computer are
> you compiling on?
>
> Thanks,
>
> Jason
>
>
>
> --
> 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-support
> URL: http://www.sagemath.org
--
William Stein
Professor of Mathematics
University of Washington
http://wstein.org
--
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-support
URL: http://www.sagemath.org
