On Sep 21, 5:41 pm, "William Stein" <[EMAIL PROTECTED]> wrote:
> On Sun, Sep 21, 2008 at 4:44 PM, mabshoff <[EMAIL PROTECTED]> wrote:
<SNIP>
> >> What is a problem with it and
> >> how should it be fixed, or rewritten from scratch?
>
> I did a little work on rewriting it from scratch, but using
> pyprocessing:
>
> http://trac.sagemath.org/sage_trac/ticket/3765
>
> The above is just done by adding a couple of lines to setup.py, so
> it's really trivial.
>
> To make it usable for most people, I think I'll change it to
> use a command line option, e.g.,
>
> sage -br 10
>
> will build sage in parallel using at most 10 cores.
> (If the number is >= 2 * the number of processors, say,
> then it'll stop with an error, since it is easy to
> accidently do "sage -br 10000" oops... boom.)
>
> What do you think?
Yeah, the above was the one thing about the patch that really bothered
me since on sage.math the old version would just grab 16 cores
regardless of the load for example. Using -br #n works for me.
I am not sure if if I mentioned this on the ticket, but in that case
we should make the Sage library depend on pyprocessing and also use
some env variable to pass the number of threads to the build process
when building the Sage library for the first time.
> William
Cheers,
Michael
--~--~---------~--~----~------------~-------~--~----~
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-devel
URLs: http://www.sagemath.org
-~----------~----~----~----~------~----~------~--~---
