p.s. I can confirm that using reinit() which takes vectors of DoFHandlers and Constraints solved the problem.
@Martin: nevermind about documentation, there is also a paragraph above which discuss similar things. Maybe eventually we need to have another matrix-free tutorial to demonstrate all those options. Regards, Denis. On Monday, January 29, 2018 at 5:13:58 PM UTC+1, Denis Davydov wrote: > > Hi Martin, > > Thanks for confirming that it's indeed possible. > > On Monday, January 29, 2018 at 4:34:07 PM UTC+1, Martin Kronbichler wrote: >> >> Dear Denis, >> >> p.s. the mapping is the same for both cases. >> >> On Monday, January 29, 2018 at 4:16:48 PM UTC+1, Denis Davydov wrote: >>> >>> Dear all, >>> >>> I am playing around with something in the following setup: >>> 1. I have a single p::d::Tria >>> 2. two different DoFHandler's using scalar FE_Q<dim> of different degrees >>> 3. The same quadrature rule (QGauss) used for both cases >>> 4. Two different MatrixFree objects >>> >>> I am not 100% sure, but it looks like the iteration over SIMD chunks of >>> cells via two different matrix-free >>> objects actually leads to the different sequence of cells. I see that >>> when I try to calculate some integral quantities >>> of my quadrature point data. >>> >> >> I guess that's possible. Once there are different DoFHandler objects, the >> algorithm that collects cells and puts them into batches might change. The >> documentation does state the following (in FEEvaluation): >> https://www.dealii.org/developer/doxygen/deal.II/classFEEvaluation.html >> section on "Handling several integration tasks and data storage in >> quadrature points" towards the end of the general class documentation. It >> says "For the combination of different fields, including different solution >> vectors that come from different time steps, it is mandatory that all >> FEEvaluation >> <https://www.dealii.org/developer/doxygen/deal.II/classFEEvaluation.html> >> objects share the same MatrixFree >> <https://www.dealii.org/developer/doxygen/deal.II/classMatrixFree.html> >> object." >> >> Did you expect anything in addition? >> > > I guess i just missed this part. I always forget that FEEvaluation has > much more info than MatrixFree class. > > I suppose the right approach here is to use `reinit()` which takes a > vector of constraints and DoFHandlers and use it > in combination with different operators. I guess this might be the only > thing to mention in that paragraph. > > Regards, > Denis > -- The deal.II project is located at http://www.dealii.org/ For mailing list/forum options, see https://groups.google.com/d/forum/dealii?hl=en --- You received this message because you are subscribed to the Google Groups "deal.II User Group" group. To unsubscribe from this group and stop receiving emails from it, send an email to dealii+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
