JG points out that NIST...
> ...presently require a (faulty) blowout-preventer from
> PySparse when solving with Trilinos.

I could use a (functional) blowout-preventer for my code, just in general.
Please advise.

and DW says...
> ...I just fixed a memory leak yesterday in pysparse

Thanks, gentlemen.  I need to do a better job of monitoring and searching
the logs.  The fact that the problem does *not* show up with pysparse was a
trifle misleading.

> If you want to try valgrind, I can give you some
> pointers for using it with python.

A quick tour of what the web knows about valgrind makes it sound like a
useful tool, though a bit tricky to use with python -- pointers would be
appreciated.

> I'll do a minor release of the tarball either today or Monday

A grateful public anxiously awaits.  I am truly relieved to be spared the
rigors of a (pointless) Trilinos rebuild!   I shall now hang up the hair
shirt and put away the scourge...

+jtg+

Reply via email to