On Jan 21, 2012, at 16:10 , Jeroen Demeyer wrote:
> On 2012-01-21 19:30, Justin C. Walker wrote:
>> I tried this on my system (Mac Pro, Dual 6-core Xeon, 10.6.8, 24 GB RAM)
>> with 4.8.rc0, using "-j24". For 'make build', I got
>>
>> real23m34.692s
>> user159m17.675s
>> sys 22m54.6
On 2012-01-21 19:30, Justin C. Walker wrote:
> I tried this on my system (Mac Pro, Dual 6-core Xeon, 10.6.8, 24 GB RAM) with
> 4.8.rc0, using "-j24". For 'make build', I got
>
> real23m34.692s
> user159m17.675s
> sys 22m54.627s
>
> Lots of parallelism. For 'make doc', I got
>
> re
On 2012-01-21 23:56, Justin C. Walker wrote:
> Jeroen: did you apply these patches for your tests, or did you apply other
> magic?
No, I didn't use anything special (apart from #12329 which has nothing
to do with doc).
--
To post to this group, send an email to sage-devel@googlegroups.com
To uns
On Jan 21, 2012, at 11:10 , Volker Braun wrote:
> Sphinx works serial only. The patch at
> http://trac.sagemath.org/sage_trac/ticket/6495 aims to break the reference
> manual into bits that can be processed in parallel.
Thanks for the pointer, but the patches don't apply cleanly to 4.8.rc0. P
Sphinx works serial only. The patch at
http://trac.sagemath.org/sage_trac/ticket/6495 aims to break the reference
manual into bits that can be processed in parallel.
--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to
sage-dev
On Jan 20, 2012, at 05:25 , Jeroen Demeyer wrote:
> Some observations from building sage-4.8 on a very fast system (32
> cores, 512G RAM) using "make -j64 -l32".
>
> First of all, apply the patch from #12329 to prune many unneeded
> dependencies of the Sage library (needs_review by the way...):
On 2012-01-20 14:39, Martin Albrecht wrote:
> Seems I should improve my tuning code: building M4RI is very quick, but the
> find-the-cache-size-via-tuning stuff takes a lot of time.
I don't think it's a big deal really. libm4ri isn't in the critical
path, and it's only the 10th slowest package.
> The 10 slowest packages to build (slowest first):
> r
> atlas
> scipy
> polybori
> maxima
> ntl
> singular
> ecl
> sage
> libm4ri
Seems I should improve my tuning code: building M4RI is very quick, but the
find-the-cache-size-via-tuning stuff takes a lot of time.
Cheers,
Martin
--
name: Mart
