I don't really need to play with it. Your new code path didn't cause any
diffs, it just caused slowness which if you've fixed, I'm satisfied.
It's on my list to improve our test timing so we can capture this kind of
thing automatically. RIght now our timed tests are all way to small to be
of any r
On Tue, 27 Oct 2015, Cody Permann wrote:
> So it looks like nearly all of the time for this problem is spent in
> System::project_vector() as you already guessed. The most expensive
> call being in "OldSolutionValue".
Looks like the problem was that I put the "project" back into
"project_vector"
Yep - all sounds right to me. Like Cody says we have to do another
reinit() because of adding extra ghosted degrees of freedom.
Derek
On Tue, Oct 27, 2015 at 11:05 AM Roy Stogner
wrote:
>
> On Tue, 27 Oct 2015, Cody Permann wrote:
>
> > So it looks like nearly all of the time for this problem
On Tue, 27 Oct 2015, Cody Permann wrote:
So it looks like nearly all of the time for this problem is spent in
System::project_vector() as you already guessed. The most expensive
call being in "OldSolutionValue".
Yup. I can replicate this now; it's definitely my fault. :-P
So how are we ge
So it looks like nearly all of the time for this problem is spent in
System::project_vector() as you already guessed. The most expensive call
being in "OldSolutionValue".
So how are we getting on this code path? The PenetrationAux kernels in
MOOSE turn on various machinery for GeometricSearch whic
Fortunately we did our libMesh update before that merge so none of our
users are using it. We probably won't do another for several weeks. This
may or may not affect certain libMesh users though so I'll let you make
that decision.
On Mon, Oct 26, 2015 at 3:05 PM Roy Stogner
wrote:
>
> On Mon, 26
On Mon, 26 Oct 2015, Cody Permann wrote:
I just pulled the latest libMesh head and saw a serious performance penalty
that I narrowed down to 445b1c3 with John's help. We've had this test
for several years and it's been relatively stable. It's two overlapping
structured 3D grids, one slightly
