[petsc-dev] synchronized printing, singletons, and jumbled GAMG/BJacobi output

[Matthew Knepley]

On Sat, May 25, 2013 at 8:38 AM, Jed Brown <jedbrown at mcs.anl.gov> wrote:

> Matthew Knepley <knepley at gmail.com> writes:
>
> > Okay, I thought as the docs stand today, this is misleading. If we
> declare
> > that Singletons are
> > for synchronized printing, then that is fine.
>
> If you don't synchronize the printing, it always produces jumbled
> output.  I would document that it is indeed synchronizing, but pretty
> much all use cases are wrong, and are making noise in the test suite.
>

I should have been more clear. The part I thought was misleading was a user
getting
out this Singleton and printing, but not seeing anything immediately, even
though its
on SELF. We just need to tell them nothing will come out until its
destroyed.

   Matt


> For example:
>
> $ cd src/vec/vec/examples/tests
> $ make ex24
> $ make runex24
> 31d30
> < ----
> 41d39
> < ----
> 48a47,48
> > ----
> > ----
> /home/jed/petsc/src/vec/vec/examples/tests
> Possible problem with with ex24_1, diffs above
> \n=========================================
> $ make runex24
> 27c27
> < 3
> ---
> > Vector Object:Vector Object:3
> 28a29,31
> >  1 MPI processes
> >  1 MPI processes
> >   type: seq
> 30,32d32
> < 6
> < ----
> < Vector Object: 1 MPI processes
> 33a34
> > 6
> 35a37
> > 10
> 41,44d42
> < ----
> < Vector Object: 1 MPI processes
> <   type: seq
> < 10
> 48a47,48
> > ----
> > ----
> /home/jed/petsc/src/vec/vec/examples/tests
> Possible problem with with ex24_1, diffs above
> \n=========================================
>



-- 
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/20130525/324b96c1/attachment.html>