Re: [deal.II] erorr: recompile with -fPIC

You should build Trilinos as shared library. Best, Daniel On Thu, Aug 29, 2024 at 2:26 AM ztdep...@gmail.com wrote: > > /usr/bin/ld: > /home/ztdep/2T/lib/Trilinos-13.4/lib/libnox.a(NOX_Abstract_Vector.C.o): > relocation R_X86_64_32S against symbol > `_ZTVN7Teuchos11RCPNodeTmplIN3NOX11MultiVec

Re: [deal.II] Adding AVX support libdeal.ii debian 12 package

I am seeing something like -- Include /tmp/dealii_new/cmake/checks/check_01_cpu_features.cmake -- Performing Test DEAL_II_HAVE_SSE2 -- Performing Test DEAL_II_HAVE_SSE2 - Failed -- Performing Test DEAL_II_HAVE_AVX -- Performing Test DEAL_II_HAVE_AVX - Failed -- Performing Test DEAL_II_HAVE_AVX512

Re: [deal.II] Adding AVX support libdeal.ii debian 12 package

Did you actually reconfigure and recompile deal.II itself or only step-76? You should see in the CMake configuration output why AVX-support wasn't detected.\ Best, Daniel On Wed, Aug 14, 2024 at 8:43 AM Gregory Christian wrote: > > Hi Daniel, > > The output is Vectorization over 1 doubles = 64 b

Re: [deal.II] Adding AVX support libdeal.ii debian 12 package

Greg, What does your detailed.log look like? The "Vectorization level" line should tell you the detected capabilities. Best, Daniel On Wed, Aug 14, 2024 at 7:16 AM Gregory Christian wrote: > > Hi all, > > I'm trying to add AVX2 support for step-76 but have had no luck. > Steps taken: > sudo apt

Re: [deal.II] Help Needed with Implementing Thickness Inhomogeneity in deal.II Mesh

Tarik, Have a look at GridTools::transform(https://www.dealii.org/current/doxygen/deal.II/namespaceGridTools.html#a212e99cf0d923cebfa04f1d23fa60b04). Best, Daniel On Tue, Jul 30, 2024 at 2:35 PM Tarik Nadeem wrote: > > Dear All, > > I am new to deal.II and need help with mesh creation. Specifi

Re: [deal.II] unknown DEAL_II_HOST_DEVICE

Are you sure that you are using the same deal.II version for the examples as you installed? That macro should be defined in deal.II/base/numbers.h. Best, Daniel On Sun, Jul 28, 2024 at 7:02 PM H wrote: > > Hi Everyone, > > I am new to the dealii. I use Mac M1 and installed dealii from candi. I t

Re: [deal.II] Parallized BEM (Step-34)

Max, the problem is that GridGenerator::extract_boundary_mesh doesn't support the volume mesh to be an object of type parallel::distributed::Triangulation. This just isn't implemented yet. Do you actually need the input volume mesh to be distributed? Best, Daniel On Mon, May 6, 2024 at 12:06 PM

Re: [deal.II] Output domain id for different parts of the domain

Chen, Have a look at how the subdomain id is added to the output results in step-40 (https://www.dealii.org/current/doxygen/deal.II/step_40.html#LaplaceProblemoutput_results). Best, Daniel On Mon, Apr 29, 2024 at 9:23 AM 陈敏 wrote: > > Dear all > > I'm dealing with a fluid-structure interaction

Re: [deal.II] There are some mistake about the function ”GridTools::collect_periodic_faces()"

I meant #include Best, Daniel Daniel Arndt 8:43 AM (0 minutes ago) to dealii On Wed, Apr 17, 2024 at 8:43 AM Daniel Arndt wrote: > > You will have to include > > #include > > Best, > Daniel > > On Wed, Apr 17, 2024 at 7:21 AM Xue Yu wrote: > > > >

Re: [deal.II] There are some mistake about the function ”GridTools::collect_periodic_faces()"

You will have to include #include Best, Daniel On Wed, Apr 17, 2024 at 7:21 AM Xue Yu wrote: > > Dear deal.ii community, > > first of all, a big thank you for all the effort going into this > software. > With reference to step-45, I am solving the homogenization problem with > dealii software

Re: [deal.II] gridin.read_exodusii to import exodusii mesh in dealii 9.5.1 fails

Sina, You'll have to provide the path directly as a function argument, i.e., gridin.read_exodusii(case_path + "mesh.exo"); Best, Daniel On Fri, Mar 29, 2024 at 7:32 AM Sina Tajfirooz wrote: > > Dear developers/users, > > How can I import an exodusii mesh in dealii version 9.5.1, which is > c

Re: [deal.II] Possible typo in step 26 tutorial

Kyle, Why don't you just submit a pull request for this fix? Best, Daniel On Fri, Mar 22, 2024 at 9:10 AM Kyle Schwiebert wrote: > Hello, > > I'm sorry for not just mentioning this on github. If I'm honest I wasn't > sure of the procedure for such a small change. I think there is a typo in: Th

Re: [deal.II] [Assigning boundary condition to node]

Can specify what you mean by "assign a certain BC [boundary condition] to a node"? Typically boundary conditions only make sense on the boundary of a mesh with codimension 1 (embedded in the same physical space). Thus, you would only consider setting boundary conditions on a node/point (0-dimension

Re: [deal.II] Re: Can't configure with cmake

Sean, that problem doesn't sound specific to compiling with Kokkos+Cuda. Can you make sure to start from a clean work directory and again select `nvcc_wrapper` as compiler? Best, Daniel On Tue, Feb 6, 2024 at 6:26 PM Sean Johnson wrote: > Bruno, > > I apparently spoke to soon. It passed "make

Re: [deal.II] cmake configure does not find cgal

What happens if you configure with CGAL_DIR= /home/spack/opt/spack/linux-opensuse15-broadwell/gcc-12.3.0/cgal-5.5.2-6umjskepmayihkexoscjiovjmtgo5toy/lib64/cmake/CGAL ? Best, Daniel On Sun, Jan 14, 2024 at 4:14 AM Praveen C wrote: > CGALConfig.cmake is present > > *$ ls -l `spack location -i cga

Re: [deal.II] Shape gradients of (f(x)*\phi_i)

Abbas, \dfrac{\partial f(x) \phi_i}{\partial x} is just \dfrac{\partial f(x)}{\partial x} \phi_i+f(x)\dfrac{f(x)}{\partial x} which I would use. There is no need to integrate by parts. Best, Daniel On Tue, Jan 23, 2024 at 2:37 PM Abbas Ballout wrote: > I know I can query the standard shape gra

Re: [deal.II] cmake configure does not find cgal

What about CGALConfig.cmake? A CMake project installation should provide that. Best, Daniel On Fri, Jan 5, 2024 at 11:09 AM Praveen C wrote: > I think deal.II is supposed to provide FindCGAL.cmake, there is no such > file inside cgal. > > Thanks > praveen > > On 05-Jan-202

Re: [deal.II] cmake configure does not find cgal

It seems like the spark installation doesn't provide FindCGAL.cmake or at least not in the path that you provide as CGAL_DIR. Can you try finding that file and (if it exists) setting the respective directory as CGAL_DIR? Best, Daniel On Fri, Jan 5, 2024 at 9:37 AM timo Hyvärinen wrote: > hi, pr

Re: [deal.II] Reporting issue with repo

Alex, grid_reordering,h was removed in https://github.com/dealii/dealii/commit/fd0c356263d7115ebf58fcea15d0412a32da199e (which isn't part of any release yet). Best, Daniel On Tue, Dec 5, 2023 at 7:47 AM Alex Quinlan wrote: > Dear deal.ii devs, > > I cloned the deal.ii repo yesterday and found

Re: [deal.II] How to find the cell index by using global dof index

Lance, Have a look at GridTools::vertex_to_cell_map( https://www.dealii.org/current/doxygen/deal.II/namespaceGridTools.html#a9b7e2ca8ecd26a472e5225ba91a58acb ). Best, Daniel On Sun, Dec 3, 2023 at 11:38 PM Lance Zhang wrote: > Hello team, > > may I know how to find the cell index with the info

Re: [deal.II] How to find the neighbor of one cell

Lance, You would check for the cell being at a boundary first before asking for its neighbor: // Loop over all neighboring cells for (unsigned int f = 0; f < cell->n_faces(); ++f) { if (cell->at_boundary(f)) continue; auto neighbor = cell->nei

Re: [deal.II] Re: merge_triangulation error

Vinayak, Would you have a minimal reproducer? Best, Daniel On Mon, Nov 20, 2023 at 4:13 AM Vinayak Vijay wrote: > Hello, > > I also tried to do the same using the > GridGenerator::replicate_triangulation(triangulation1,{2},triangulation2) > function. This should result in a triangulation (tri

Re: [deal.II] how to treat the intersection point at the first and second kined bcs

Dirichlet boundary conditions are normally enforced strongly while Neumann boundary conditions are enforced weakly. Thus, when interpolating the Dirichlet boundary conditions on the respective boundary you would set the degrees of freedom accordingly while the Neumann boundary conditions are part o

Re: [deal.II] how to use const std::function< T(const T &, const T &)> & combiner

It's simply a function that takes two arguments of the same type by reference and returns a new value of that type. For a sum reduction using doubles, you could pass in in a lambda, for example: [](const double& a, const double& b) {return a+b;} Best, Daniel On Thu, Nov 2, 2023 at 9:03 AM ztdep.

Re: [deal.II] Mixed codimensionality

Alex, If you can formulate your problem/approach in a way (iteratively?) that you could solve separately on the three triangulations (using input from the solution on the other triangulations), that might be easier. On the other hand, if you only want one system matrix you would identify matching

Re: [deal.II] Mixed codimensionality

Alex, FiniteElements in deal.II are used with a Triangulation and DoFHandler of the same dimension and space dimension. >From your sketch, it seems like the Triangulation for your 3 types of finite elements is (expectedly) different. Thus, it doesn't make much sense to combine those in a FESystem.

Re: [deal.II] Compilation of deal.II fails due to conversion error from std::complex to double

Roland, Using PETSc with complex values is pretty niche and there isn't much CI in place for this case. It looks like boost::serialization::array_wrapper wrapper( should be replaced with boost::serialization::array_wrapper wrapper( and similarly in VectorBase::load. Best, Daniel On Sat,

Re: [deal.II] An error occurred in line <1467> of file in function dealii::IndexSet dealii::TrilinosWrappers::MPI::Vector::locally_owned_elements() const

Yi, How are you initializing your TrilinosWrappers::MPI::BlockVector? It looks like you are calling the constructor with only one std::vector but some of the index sets are overlapping. You should use the constructor taking two std::vector objects with the first one containing the locally owned in

Re: [deal.II] I am experiencing varying execution speeds when running my code with different configurations of Ubuntu and Deal.II. Here are the observed speeds for the same code on different systems

What is your code doing and how do you configure deal.II in all of these cases? Are you running in Release mode when comparing the run time of your program with different configurations? I would be very surprised to find that vastly different runtimes are a result of using different Ubuntu versions

Re: [deal.II] Vector-valued matrix-free operatior with CUDA

Justin, As far as I know, we haven't made any significant progress with vector-valued support for CudaWrappers::MatrixFree (apart from porting it to Kokkos). What's your specific use case? What functionality do you need specifically (same finite elements? FESystsem?)? Best, Daniel On Mon, Aug 7,

Re: [deal.II] Question about compiling with CUDA

I'm glad you found something that is working for you! Best, Daniel On Mon, Aug 7, 2023 at 9:52 AM Justin O'Connor wrote: > For posterity, ere's a dockerfile I got working that installs deal.ii with > cuda capability. :) > > Thanks, > Justin > On Friday, July 28, 2023 at 4:33:46 PM UTC-6 Justin

Re: [deal.II] Integrate difference isn't picking on the

Abbas, Try following the steps at https://github.com/dealii/dealii/wiki/Contributing. Best, Daniel On Thu, Jul 27, 2023 at 1:24 PM Abbas Ballout wrote: > Lucas > It's not the linker but I am guessing it's when I edit something deep > within the library.h files. > Since everything is linked to

Re: [deal.II] Question about compiling with CUDA

Justin, a docker file that we are using in a different project can be found at https://github.com/arborx/ArborX/blob/master/docker/Dockerfile in case that helps. In general, I would advise against using nvhpc/nvc++ which proves to be pretty buggy as compared to nvcc/nvcc_wrapper or even (clang++ w

Re: [deal.II] Q: Question about extracting part of a vector

Najwa, the relevant code reads something like const unsigned int n_components = dof.get_fe(0).n_components(); AssertDimension(exact_solution.n_components, n_components); if (weight != nullptr) { Assert((weight->n_components == 1) || (weight->

Re: [deal.II] Q: Question about extracting part of a vector

Najwa, The documentation of VectorTools::integrate_difference( https://www.dealii.org/current/doxygen/deal.II/namespaceVectorTools.html#a676190d2c897ac5da68a9c460fa95832) says The additional argument weight allows to evaluate weighted norms. The weight function may be scalar, establishing a spat

Re: [deal.II] create reference simplex in 2d (triangle) based on three vertices

Math, You can use ReferenceCells::get_simplex() https://www.dealii.org/current/doxygen/deal.II/namespaceReferenceCells.html#ace7fb0fa0c1a98b17ae9c6f6d2eaa650 to create a ReferenceCell object describing the reference simplex. Best, Daniel On Tue, Jul 18, 2023 at 1:49 PM Mathieu wrote: > Thank y

Re: [deal.II] create reference simplex in 2d (triangle) based on three vertices

Mathieu, You should be able to use GridGenerator::reference_cell ( https://www.dealii.org/current/doxygen/deal.II/namespaceGridGenerator.html#aeb9a83e353f8d69ce49ebdd191a3a51f ), and then move the vertices to your liking, e.g. by using GridTools::transform ( https://www.dealii.org/current/doxygen/

Re: [deal.II] how to define periodic bc in delal.II

Have a look at step-45. On Tue, Jul 11, 2023 at 11:48 PM ztdep...@gmail.com wrote: > I waant to know how to define periodic bc in delal.II. > > -- > 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 > -

Re: [deal.II] Re: Finding unconnected regions in a triangulation

Kaushik, The implementation for those functions is pretty simple so it shouldn't be too hard to modify them to meet your needs, see https://dealii.org/current/doxygen/deal.II/grid__tools_8cc_source.html#l03828 . Best, Daniel On Mon, Jul 10, 2023 at 1:01 PM Kaushik Das wrote: > Hi Bruno, than

Re: [deal.II] Unable to compile in ubuntu 22.04

Sabyasachi, What is the boost version of the system-provided boost installation? You should be able to work around this issue by forcing deal.II to use its bundled boost package via using `-DDEAL_II_FORCE_BUNDLED_BOOST=ON` when configuring deal.II. Best, Daniel On Sun, Jun 18, 2023 at 2:01 AM sa

Re: [deal.II] Re: Mean Value Constraints

Corbin, - Using ARPACK might indeed be an option. Do you really need to know the whole spectrum, though? - Yes, you would just call AffineConstraints::add_line() to add constrain a given DoF to zero. Note that it's a no-op if the DoF is already constrained. This doesn't change the size of the line

Re: [deal.II] determine global dof index of a point

There is also DoFTools::map_support_points_to_dofs, see https://www.dealii.org/current/doxygen/deal.II/namespaceDoFTools.html#a3540ceb577e65414bde1b6b14808da2c Best, Daniel On Thu, May 18, 2023 at 11:15 AM Wolfgang Bangerth wrote: > On 5/17/23 12:00, Mathieu wrote: > > > > I apologize for my si

Re: [deal.II] Error in using FESystem usage

Vinayak, an alternative to initializing an object of type FESystem in the constructor is to use a smart pointer like std::unique_ptr> fe_system_ptr; as a member variable since that one would have a default constructor. Best, Daniel On Tue, May 16, 2023 at 10:26 AM Lucas Myers wrote: > Hi Vin

Re: [deal.II] Re: Could not find partitioner that fits vector

Wayne, > I have another suggestion that the mg_constrained_dofs::make_zero_boundary() seems only accept one > boundary_ids set + ComponentMask combination. So if there are 2 Dirichlet BC that mask different components > in vector problems, more make_zero_boundary functions might be added. Not a bi

Re: [deal.II] Unable to refresh spack module

es? > > ==> Error: Cannot use invalid module set default.Valid module set > names are [] > > Thanks again for your help. > Vinayak > > On Fri, 12 May 2023 at 3:12 PM, Daniel Arndt > wrote: > >> Yes, our HDF5 integration only works well with a raw install (

Re: [deal.II] Unable to refresh spack module

Yes, our HDF5 integration only works well with a raw install (like a system package manager provides) that doesn't expose CMake files. https://github.com/dealii/dealii/pull/15199 offers a workaround for that issue but otherwise, it's best to just use the system installation obtainable through apt-

Re: [deal.II] Two-component reaction-diffusion equations

In the end, you have decoupled heat equations so step-26 ( https://www.dealii.org/current/doxygen/deal.II/step_26.html) and tutorials leading up to it would be good places to start from. Best, Daniel On Tue, May 9, 2023 at 4:26 AM D. Sarkar wrote: > Hello, deal.II users, > > I am new to deal.II

Re: [deal.II] Problem compiling deal.II with CUDA

Sean, we removed the implementation for atomic add for doubles for nvcc < 8 and compute capability < 6 in https://github.com/dealii/dealii/pull/10298 (maybe in error). I would expect that you could the code in that pull request as a patch. Best, Daniel On Sat, Apr 29, 2023 at 1:46 PM Sean Johnso

Re: [deal.II] Applying boundary conditions to a point on the boundary in MP framework

Wasim, What does the backtrace in your debugger (gdb?) look like? Best, Daniel On Wed, Apr 19, 2023 at 7:27 AM Wasim Niyaz Munshi ce21d400 < ce21d...@smail.iitm.ac.in> wrote: > Hello everyone. > I am trying to apply a particular boundary condition at one single point > in the domain for a probl

Re: [deal.II] Trouble installing p4est from candi M1 mac

Matteo, what happens if you just use p4est CMake installation interface instead of our scripts? You probably only need to make sure to enable MPI support. Best, Daniel On Mon, Apr 10, 2023 at 8:25 AM Matteo Malvestiti wrote: > Hi, thanks for your response! > Could you please tell me how to com

Re: [deal.II] Extracting element solution in step-40

Wasim, step-40 uses Q2 elements which means that you have 9 dofs/cell in 2D and for 32 cells in each direction (33+32)^2=65^2=4225 dofs. Best, Daniel On Fri, Apr 7, 2023 at 5:47 AM Wasim Niyaz Munshi ce21d400 < ce21d...@smail.iitm.ac.in> wrote: > Thank You, Prof. Bangerth. > I have one more d

Re: [deal.II] Extracting element solution in step-40

Wasim, The answer depends very much on what you actually want to do with that solution vector. Do you want a representation of the solution (assuming you are using Q1? nodal elements) on a single process/all processes or are you just interested in the partial solution on every process separately?

Re: [deal.II] Compile deal.ii with metis

Li Xiangyue, You need to tell deal.II to enable METIS support using a CMake option. Have a look at https://dealii.org/9.1.1/users/cmake.html#configurefeature. Best, Daniel On Wed, Apr 5, 2023 at 3:59 AM 李相越 wrote: > Dear deal.ii developers, > > I’m Li Xiangyue, a novice programmer. I initiall

dealii@googlegroups.com

Laryssa, The bundled boost package only includes the files that are used within deal.II. If you want to use additional boost features, you should build boost yourself (and tell deal.II where to find it). Best, Daniel On Tue, Mar 21, 2023 at 1:19 PM Laryssa Abdala wrote: > > Great, that solved

Re: [deal.II] Using affine constraints object to apply boundary condition at a vertex

Wasim, Have a look at step-11 ( https://www.dealii.org/current/doxygen/deal.II/step_11.html). Best, Daniel On Sun, Mar 12, 2023 at 9:17 AM Wasim Niyaz Munshi ce21d400 < ce21d...@smail.iitm.ac.in> wrote: > I am solving an elasticity problem wherein I have a pin support at one > vertex. Earlier,

Re: [deal.II] How to apply Spherical Manifold

Deepika, https://www.dealii.org/current/doxygen/deal.II/namespaceGridGenerator.html#a533c4778cbc9bcbed365dcab42ca4418 says By default, the manifold_id is set to 0 on the boundary faces, 1 on the boundary cells, and numbers::flat_manifold_id

Re: [deal.II] Discontinous Galerkin with periodic boundary conditions

Nils, Using DG with periodic boundary conditions is not very common to my knowledge. It seems you are hitting a bug in deal.II in Debug mode and it should work with diff --git a/include/deal.II/meshworker/mesh_loop.h b/include/deal.II/meshworker/mesh_loop.h index f46c737c2a..699264551e 100644 ---

Re: [deal.II] VectorTools::interpolate_boundary_values doesn't work for DG

Abbas, Yes, FE_DGQ only defines dofs in the interior of cells. Thus, trying to interpolate the boundary has no effect. You would normally use penalty terms to prescribe boundary values for DG methods. Best, Daniel On Wed, Jan 4, 2023 at 6:04 PM Abbas wrote: > Hello, > > The VectorTools::interp

Re: [deal.II] Linking warning: could not create compact unwind for ___emutls_get_address: registers 23 and 24 not saved contiguously in frame

Praveen, I'm also seeing this warning and from what I found on the internet it's harmless. Exception handling might be a bit slower but that's about it. Best, Daniel On Tue, Dec 13, 2022 at 11:22 PM Praveen C wrote: > Hello > > When compiling deal.ii on macos M1 with clang I get these messages

Re: [deal.II] Hoe to merge hyper_cube_with_cylindrical_hole with hyper_ball()

Deepika, You will need to appropriate manifolds ids and manifolds after merging the triangulations. Also, you need to make sure that the coarse triangulations share the same vertices and edges at the interior boundary. Best, Daniel On Wed, Dec 7, 2022 at 2:24 AM Deepika Kushwah wrote: > Hello

Re: [deal.II] Re: Getting global DoF indices of a face

Alexander, Yes, we are identifying DoFs on faces and thus FE_DGQ doesn't have any DoFs on faces. Best, Daniel On Tue, Nov 22, 2022 at 6:48 AM Alexander Kiselyov < aleksandr.kisel...@gmail.com> wrote: > Dear Peter, > > Thank you for the reply! > > The thing is that the input data is a FE_DGQ DoF

Re: [deal.II] Tutorial 40 , error: use of deleted function

You would get this compile-time error if you haven't configured deal.II with `DEAL_II_WITH_MPI=ON`. Best, Daniel On Mon, Nov 7, 2022 at 9:51 AM ztdep...@gmail.com wrote: > when i compile the tutorial 40, i met the following error. It semms that > the function has been deleted. > > error: use o

Re: [deal.II] Applying a zero constraints to cell based on material_id

Bruno, I don't think we have a function that would do that already. I would just loop over all (locally relevant) cells, and add the respective (locally relevant) dofs to the AffineConstraints object with add_line(). Best, Daniel On Mon, Oct 31, 2022 at 8:36 AM blais...@gmail.com wrote: > Dear

Re: [deal.II] C++ language in step16

cell_worker and copier are lambda functions. You can think of them as in-place structs with a call operator (operator()) with the specified signature. We define a new cell_worker object because the place we are passing it into expects the object to have a call operator that does what this->cell_wor

Re: [deal.II] Issue linking GMSH when installing & building dealii 9.4

Matthew, I don't know if the missing references are due to missing components in your gmsh installation (but I wouldn't be surprised). In any way, *"netgen, tetgen and blas support has to be enabled in Gmsh" * relates to CMake options when building gmsh from source, e.g., https://gmsh.info/src/gms

Re: [deal.II] compiling deal.ii 9.4 on wsl error

Qiong, It's not sufficient to include deal.II's include files but you need to link the library target. Have a look at https://www.dealii.org/current/users/cmake_user.html. Best, Daniel On Wed, Sep 7, 2022 at 9:38 PM jim wo wrote: > and at the same time, I write the codes and it comes "the > d

Re: [deal.II] compiling deal.ii 9.4 on wsl error

Qiong, What CMake settings did you use to configure deal.II? Does your system indeed provide an MPI installation? Bets, Daniel On Wed, Sep 7, 2022 at 10:22 AM jim wo wrote: > Hello, > I want to develop my own project on wsl with deal.ii 9.4. after make I > got this message: > > In file inclu

Re: [deal.II] visualizing a finite element field: different piecewise constant values within cells

Simon, You could also refine once globally and then manually assign values. Implementing a new finite element class just for the sake of visualization sounds like much more work. Best, Daniel On Fri, Aug 26, 2022 at 12:31 PM Simon wrote: > Dear all, > > > my triangulation is two-dimensional an

Re: [deal.II] Re: Solving the linear system of equations using PETSc BlockSparseMatrix

Adding to what Bruno is saying, If you are not taking advantage of the block structure, you should be able to treat the block matrix as "regular", i.e., non-block, matrix possibly by copying it to a corresponding non-block type. Best, Daniel On Mon, Aug 22, 2022 at 10:51 AM Bruno Turcksin wrote

Re: [deal.II] Issue encountered while solving Step-40 in 1 dimension

Fri, 19 Aug 2022, 8:18 pm Daniel Arndt, wrote: > >> Syed, >> >> parallel::distributed::Triangulation is just not implemented for dim==1 >> so you can't run step-40 for the one-dimensional case. >> >> Best, >> Daniel >> >> On Fri, Aug 19,

Re: [deal.II] Issue encountered while solving Step-40 in 1 dimension

Syed, parallel::distributed::Triangulation is just not implemented for dim==1 so you can't run step-40 for the one-dimensional case. Best, Daniel On Fri, Aug 19, 2022 at 7:07 AM syed ansari wrote: > Dear all, > I was trying to run step-40 in 1 dimension and encountered the > error

Re: [deal.II] Refine per direction

Uclus, Use GridOut::write_vtu or GridOut::write_vtu_with_pvtu_record as demonstrated in step-40 instead. Best, Daniel On Wed, Aug 17, 2022 at 3:06 AM Uclus Heis wrote: > Dear Wolfgang, > > Thank you very much I could solve that. > > I would like to ask another question if it is ok. When I try

Re: [deal.II] Installation problem using spack

Marco, It appears that -- Insufficient hdf5 installation found: hdf5 has to be configured with MPI support. is the problem. HDF5 has not been compiled with MPI support. Thus, you can either try to make sure that HDF5 is built with MPI support or, in case you are not going to use HDF5 anyway, jus

Re: [deal.II] trilinos error with candi

My best bet would be that your compiler is just too old, what are you using? Also, what does the referenced line in packages/kokkos/core/src/impl/Kokkos_Core.cpp look like? Best, Daniel On Wed, Jul 20, 2022 at 11:12 AM erkin yildiz wrote: > I am getting[image: Screenshot from 2022-07-20 17-08-0

Re: [deal.II] 3D Periodic BCs

Jack, step-45 https://www.dealii.org/current/doxygen/deal.II/step_45.html provides an example for dealing with periodic boundary conditions. Even though it's a 2D example, the interface is the same in 3D. Assuming that the arguments for your set_periodic_boundary_condition function are similar to

Re: [deal.II] discretization of a analytical function on a second triangulation

triangulation based on the > formula implemented in the just derived class. > During the assembly routine, I access some interpolated values at > quadrature points,... > > With this in mind, is this an appropriate way? > > > Best > Simon > > Am Di., 5. Juli 2022

Re: [deal.II] error: installation of dealii using spack

Simon, It looks like all the errors are related to the Sundials dependency (release 6.0.0). Looking at https://github.com/dealii/dealii/pull/13918, it seems that you need the latest deal.II release (9.4.0) to support that Sundials version. If you don't anticipate needing it I would just turn the d

Re: [deal.II] discretization of a analytical function on a second triangulation

Simon, defining a class derived from dealii::Function sounds sensible. Whether calling VectorTools::interpolate makes sense, depends on what you want to do with the interpolated function. In may cases, it's enough to just evaluate it locally during assembly using functionality like FEValues::get_f

Re: [deal.II] Re: Trouble getting UMFPACK to work

What does the log for running cmake -DDEAL_II_WITH_UMFPACK=ON . look like? The summary printed at the end should tell you if `UMFPACK` really was found. N=Any problems in finding a compatible library should be contained in the output printed when running the CMake command. Best, Daniel On Mon, Ju

Re: [deal.II] Compilation fails for deal.II when compiling with PETSc+Kokkos

Roland, Instead of debugging where thighs go wring in choosing the correct compiler when using the launch_compiler, you could either use Kokkos' nvcc_wrapper (in the Kokkos installation directory in the bin directory) or use clang++. The former one forces all compilation to go through nvcc which y

Re: [deal.II] Compilation fails for deal.II when compiling with PETSc+Kokkos

Somewhere the compiler flag -DKOKKOS_DEPENDENCE got lost, you might try if adding it fixes your problem. Kokkos allows specifying an arbitrary CMake C++ compiler to be used as host compiler for nvcc under the hood. As long as the compiler is not clang++ (which can compile CUDA code on its own), eve

Re: [deal.II] No printing of TimerOutput after catch of exception in MPI based code

Maurice, You could try to wrap the calls that might throw in a try-catch block to deal with the exception yourself (including ignoring it). Best, Daniel On Mon, May 9, 2022 at 11:14 AM 'maurice rohracker' via deal.II User Group < dealii@googlegroups.com> wrote: > Dear all, > > In the attached m

Re: [deal.II] simple application of anti-periodic boundary conditions

Paras, You can also use add_periodicity with regular Triangulation objects, see https://www.dealii.org/current/doxygen/deal.II/classTriangulation.html#adbf04756b28dae69194870812acaf941. This should make sure that periodic neighbors differ at most by one refinement. Best, Daniel On Wed, Apr 20, 2

Re: [deal.II] Issue with deal.ii installation on Cori

Vishwas, A deal.II installation should not use that much disk space or are you talking about the memory being used while compiling? Every compilation unit uses a couple GB so running with 64 threads might be too much. If it's the disk space, do you have any idea which directories are particularly

Re: [deal.II] FunctionManifold class

Giuseppe, Searching for FunctionMnaifold in the tests folder should give you a couple of examples on how to use FunctionManifold with an analytical function. Most of them use std::strings but tests/fe/fe_values_function_manifold.cc and tests/mappings/mapping_q_manifold_02.cc actually use Function

Re: [deal.II] Is any faster version of VectorTools::point_gradient?

Hello, Have a look at VectorTools::point_gradients ( https://www.dealii.org/developer/doxygen/deal.II/namespaceVectorTools.html#aada37fd09d8be1b9ed9bb12d698f8ee4 ). Best, Daniel Am Di., 14. Dez. 2021 um 10:16 Uhr schrieb 沈键 : > Hello, > I want to compute streamlines in deal.II after solving a

Re: [deal.II] evaluating basis functions at random points

Andrea, have a look at GridTools::compute_point_locations( https://www.dealii.org/developer/doxygen/deal.II/namespaceGridTools.html#a8e8bb9211264d2106758ac4d7184117e ). Step-60 demonstrates its use. Best, Daniel Am Mo., 13. Dez. 2021 um 16:14 Uhr schrieb Andrea Bonito : > Howdy: > > this might

Re: [deal.II] Building error

Naren, What compiler are you using and how are you configuring deal.II (What does detailed.log look like)? Best, Daniel Am Di., 9. Nov. 2021 um 15:44 Uhr schrieb Naren Vohra < voh...@oregonstate.edu>: > Hello, > > I am trying to install dealii on a Linux system and I get the following > error:

Re: [deal.II] Centroids of a distributed triangulation

Vachan, There is TriaAccessor::barycenter( https://www.dealii.org/current/doxygen/deal.II/classTriaAccessor.html#a56cc737f4b28f8d28a576d2e3d32ef9f) and TriaAccessor::measure( https://www.dealii.org/current/doxygen/deal.II/classTriaAccessor.html#a9cecb2b7c9a1644fb5fd44bbba40ab0c) for a single cell

Re: [deal.II] Missing (block_sparse_matrix)

Ahmed, The file and the respective classes were deleted in https://github.com/dealii/dealii/pull/9761 and deprecated in https://github.com/dealii/dealii/pull/5352. If you need to replace BlockMatrixArray, you should have a look at the documentation in the latter pull request to use BlockLinearOper

Re: [deal.II] Direct solver matrix.free

Joss, The MatrixFree framework relies on the operator to be only defined by its action. Accessing individual elements (as direct solvers typically require) falls out of this concept. That is to say that the MatrixFree framework only works well with an iterative solver. Best, Daniel Am Do., 21. O

Re: [deal.II] Time dependent boundary conditions

Kyle, Function is derived from FunctionTime which implements get_time()/set_time(). In particular, step-26 uses get_time() to access a previously stored time via set_time(). Best, Daniel Am Fr., 1. Okt. 2021 um 16:09 Uhr schrieb Kyle Schwiebert : > I am having trouble understanding the class Fu

Re: [deal.II] Element-wise DOF numbering and shape function numbering

Lixing, Yes, using FiniteElemnt::component_to_system_index() seems to be reasonable to use to identify the overall DoF number for copies of the same finite element. The FiniteElement class doesn't need class accessor argument and the mapping is thus generic. Also note that there is DoFTools::map_d

Re: [deal.II] Calculating an integral over the boundary

Kyle, Have a look at VectorTools::point_value( https://www.dealii.org/current/doxygen/deal.II/namespaceVectorTools.html#a7be5c7eed52308898dfaad91c4cff204 ). Best, Daniel Am Di., 28. Sept. 2021 um 13:04 Uhr schrieb Kyle Schwiebert < kjsch...@mtu.edu>: > Thank you for the help. I think that has t

Re: [deal.II] MatrixTools::apply_boundary_values for BlockSparseMatrix> and BlockVector> entries

Mariia, I also tried fixing this within deal.II and discovered that some more changes are necessary, see https://github.com/dealii/dealii/pull/12761. Would you mind checking if the changes proposed there work for you? Best, Daniel Am Fr., 17. Sept. 2021 um 06:25 Uhr schrieb Мария Бронзова < masi

Re: [deal.II] MatrixTools::apply_boundary_values for BlockSparseMatrix> and BlockVector> entries

Mariia, That's a problem a little more complicated to solve. Basically, we are not instantiating the template for complex and BlockSparseMatrix. If you are OK with recompiling deal.II, you should modify source/numerics/ matrix_tools.inst.in so that all overloads are also compiled for complex numbe

Re: [deal.II] MatrixTools::apply_boundary_values for BlockSparseMatrix> and BlockVector> entries

Mariia, You need to make sure that the std::map also uses values of type std::complex. The compile error you shared shows that you use doubles. Best, Daniel Am Di., 14. Sept. 2021 um 07:30 Uhr schrieb Мария Бронзова < masianic...@gmail.com>: > Hi everyone, > > I ran into the following difficult

Re: [deal.II] Massively parallel & inverse matrix solver

Hermes, The SparseDirectUMFPACK solver doesn't work with MPI parallel matrices. You will need to find another solver based on the matrix class you are using. Standard choices would be GMRES or CG but I am assuming that your linear system might also be non-symmetric or indefinite. Best, Daniel

Re: [deal.II] Equation implementation

Toddy, I would start with thinking about the time discretization scheme you want to use and in particular which terms you want to treat explicitly and which implicitly. After you have done that, you might still end up with a nonlinear problem in which case you should have a look at https://www.dea

Re: [deal.II] point_value, Real&Imaginary parts, step-29

Hermes, Use another overload. The one returning the solution as a parameter should work: https://www.dealii.org/current/doxygen/deal.II/namespaceVectorTools.html#acd358e9b110ccbf4a7f76796d206b9c7 Best, Daniel Am Di., 10. Aug. 2021 um 09:41 Uhr schrieb Hermes Sampedro < hermesampe...@gmail.com>:

  1   2   3   4   5   6   >