[petsc-dev] ts_monitor_solution_vtk does not work when used with DMComposite

[Matthew Knepley]

On Fri, Feb 7, 2014 at 4:40 PM, Jed Brown <jed at jedbrown.org> wrote:

> Mani Chandra <mc0710 at gmail.com> writes:
>
> > I have a heavy residual evaluation function which I cannot seem to
> > vectorize at all, so then I tried changing the global data layout putting
> > each variable in a contiguous chunk and it finally does vectorize.
>
> I suggest locally extracting from interlaced to fill vector registers at
> the cell level (e.g., group of a 4 cells).  The memory performance
> (number of streams and cache reuse) is better with interlaced and if you
> have enough flops that local vectorization matters, packing will be
> cheap.
>

This is EXACTLY what you should do, and what PetscFE does now.

   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/20140207/2cde54cd/attachment.html>