Jeroen Demeyer wrote:
On 2014-05-09 10:55, Julien Puydt wrote:
But personally, I don't consider RAM usage by doctests an issue.
Perhaps because you don't use old or limited computers?
No, because I think that doctest coverage is more important than memory
usage of doctests. It is perfectly possible to *use* Sage on such old
and limited computers and that is still important to me.
As long as every Sage developer has access to a machine with sufficient
RAM to run all doctests (which is the case, given
boxen.math.washington.edu), I think it is fine.
Well, the point is not to test on just sage.math (or similarly equipped
machines). You'll currently hardly find an ARM box with 128GB RAM, say,
similar for 32-bit x86 or ppc.
Also, limited or limiting available memory /can/ help in detecting
regressions and space leaks (it doesn't necessarily do).
I wouldn't mind tagging *a few* doctests with '# optional - swap', along
with a definition of how much RAM is expected to be needed to run
doctests without that tag (i.e., all other tests, sequentially).
And probably some build- or patchbots should run the tests with 'ulimit
-v $SAGE_DOCTEST_MEMORY_CUTOFF' by default.
-leif
--
() The ASCII Ribbon Campaign
/\ Help Cure HTML E-Mail
--
You received this message because you are subscribed to the Google Groups
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.
