#15738: upgrade eclib to version 20140128
-------------------------------------+-------------------------------------
Reporter: cremona | Owner: cremona
Type: enhancement | Status: needs_review
Priority: major | Milestone: sage-6.1
Component: packages: | Resolution:
standard | Merged in:
Keywords: eclib | Reviewers: Jeroen Demeyer
Authors: John Cremona | Work issues:
Report Upstream: N/A | Commit:
Branch: | bc292bace3dadf71ada95342098453d5f82b9f5e
u/cremona/ticket/15738 | Stopgaps:
Dependencies: |
-------------------------------------+-------------------------------------
Comment (by cremona):
Replying to [comment:23 jdemeyer]:
> Replying to [comment:22 cremona]:
> > The self-test which eclib runs does not work perfectly when done with
a lot of parallel threads owing (I presume) to race conditions.
> Are you open to bug reports about that? I have had a quick look and the
problem seems to be that most programs are actually built ''twice'' in
`make check` (check the logs, you will see for example `tmrank` is built
twice, which can obviously be a race condition if that happens twice at
the same time).
Of course I am open to bug reports! It's up to you whether you think
they are reasons not to include this version in Sage. But if easily
fixed, now would be a good time to fix them.
I know of no reason ahy any test program should be built twice. If I
delete etsts/tmrank and type "make tmrank" I see this:
{{{
/bin/bash ../libtool --tag=CXX --mode=link g++ -g -O2 -L/usr/local/lib
-L/usr/lib -o tmrank tmrank.o ../libsrc/libec.la -lflint -lpari -lntl
-lboost_system-mt -lboost_thread
libtool: link: g++ -g -O2 -o .libs/tmrank tmrank.o -L/usr/local/lib
-L/usr/lib ../libsrc/.libs/libec.so -lflint -lpari
/usr/local/lib/libntl.so -lboost_system-mt -lboost_thread -Wl,-rpath
-Wl,/home/jec/eclib-parallel/lib
}}}
but surely that is only one build?
Apart from building, some of the tests will only work properly if the are
run in the correct sequence since one writes a temporary files which
another needs to read. So far this has worked ok, but it would be
possible to change the target {{{check_g0n}}} in tests/Makefile.am to
forece them to run in order. That would be easy, in fact. None of the
other check targets have this defect.
--
Ticket URL: <http://trac.sagemath.org/ticket/15738#comment:24>
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 unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/sage-trac.
For more options, visit https://groups.google.com/groups/opt_out.
