[petsc-dev] testing: with parmetis, without 64-bit, without complex

Matthew Knepley knepley at gmail.com
Mon May 13 12:35:09 CDT 2013


On Mon, May 13, 2013 at 11:42 AM, Jed Brown <jedbrown at mcs.anl.gov> wrote:

> src/vec/is/ao/examples/tutorials/ex2.c fails because it was not written
> to use the typedefs.
>
>
> http://ftp.mcs.anl.gov/pub/petsc/nightlylogs/examples_master_arch-linux-pkgs-64idx_thrash.log
>
> Expressing this set of configuration constraints is a total mess in the
> current system.  Should we just remove it from the nightlies entirely?
> Or have it #error or compilation when attempting to use a broken
> configuration (so the test error is understandable)?
>

I think we remove it.


> The example is also in the wrong place: it depends on Mat, so it cannot
> be in src/vec/.
>

We need to

  a) Switch the nightlies from make to Python

  b) Change the categories to tags

  c) Simply parse parts of the output for numerical comparison

   Matt


-- 
What most experimenters take for granted before they begin their
experiments is infinitely more interesting than any results to which their
experiments lead.
-- Norbert Wiener
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20130513/68bbc6cc/attachment.html>


More information about the petsc-dev mailing list