On Sep 7, 3:09 am, Jeroen Demeyer <jdeme...@cage.ugent.be> wrote:
> Hello sage-devel,
>
> As far as we know, there are no more remaining issues for the PARI
> update (#9343). We haven't had any doctest failures for a while now.
> The main issues recently have been with PARI not compiling properly on
> various machines, but all those seem to be fixed now.
>
> I have made a complete Sage distribution sage-4.6.prealpha4, based on
> sage-4.5.3.rc0 with the new PARI and some more updates. You can
> download it
> fromhttp://sage.math.washington.edu/home/jdemeyer/dist/sage-4.6.prealpha4...
>
> In order to test this thoroughly, extract the tarball and then:
> cd sage-4.6.prealpha4
> make
> env SAGE_CHECK=yes SAGE_TUNE_pari=yes ./sage -f
> pari-2.4.3.svn-12577.p5.spkg
> make ptestlong
>
> On sage.math.washington.edu, this worked without problems (apart from a
> sometimes-reproducible error in devel/sage/sage/tests/startup.py)
>
> The following tickets have been merged in sage-4.6.prealpha4 (w.r.t.
> sage-4.5.3.rc0):
> #9343: PARI upgrade
> #9860: Port changes in PARI 2.3.5.p4 (#9722) to current 2.4.3
> #9591: Upgrade genus2reduction due to Pari upgrade to svn snapshot 12577
> - a pre-release of Pari 2.4.3
> #9592: Upgrade lcalc to work with Pari svn snapshot 12577 - a
> pre-release of Pari 2.4.3
> #9845: lcalc doesn't build on cygwin due to missing time.h include
> #9750: Document that PARI no longer assumes more than GRH.
> #9636: Catch output from PARI in Sage.
> #9400: Modify the NumberField constructor to pass in optional integer B
> such that all the internal PARI routines will replace the discriminant
> by its gcd with B, making some things massively faster.
>
> Jeroen.
All tests passed on skynet machines taurus (x86_64-Linux-nehalem-fc),
menas (x86_64-Linux-core2-suse), and fulvia (x86_64-SunOS-core2).
Tests are still in progress for mark2 (sparc-SunOS-ultrasparc3).
These all use gcc 4.5.1.
On t2.math, just running "make ptestlong" (without setting SAGE_CHECK
or SAGE_TUNE_pari) worked fine: all tests passed. However, setting
SAGE_CHECK produced an error:
+++ [BUG] Total bench for gp-sta is 3288122
+++ [BUG] Total bench for gp-dyn is 3363785
PROBLEMS WERE NOTED. The following files list them in diff format:
Directory: /scratch/palmieri/sage-4.6.prealpha4/spkg/build/
pari-2.4.3.svn-12577.p5/src/Osolaris-sparcv9
rnfkummer-sta.dif
rnfkummer-dyn.dif
I'm now trying with SAGE_TUNE_pari=yes (but SAGE_CHECK unset).
Everything worked fine on an Intel Mac OS X 10.6 box.
--
John
--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to
sage-devel+unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org
