On 04/17/2015 03:24 PM, Peter Bruin wrote: > John Cremona <[email protected]> writes: > >> Testing 6.7.beta1 after building from source. Running "make >> testlong", it has been on >> >> sage -t --long src/sage/schemes/elliptic_curves/lseries_ell.py >> >> for a while. But the underlying gp process has been running for 17 >> minutes and is using 0.108tb (terabytes!) of RAM, so I had to kill it >> as it was causing inconvenience (the machine has 128gb but still). > > This has been happening to me regularly as well (on a different > machine). For me the doctest times out eventually after consuming about > 4 GB of memory.
I've now also seen this two times (in ptestlong) on a machine (running Ubuntu 10.04.4, kernel 2.6.32-73, FWIW) I haven't built Sage many times; not sure whether the compiler version also matters, but I don't think so. Interestingly, the doctest(s) pass(es) for me when the greedy gp instance gets killed (in time). Testing the file alone (so far) hasn't shown that behavior. Presumably (the pretty useless) sorting of doctests by previous runtime has an influence, too, but I suspect once again pexpect is to blame, in conjunction with heavy system load, maybe PARI as well... John said he'd run 'make testlong' (not ptestlong), but I guess his machine was already sufficiently busy. -leif -- You received this message because you are subscribed to the Google Groups "sage-release" 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-release. For more options, visit https://groups.google.com/d/optout.
