[deal.II] Re: Problem with AffineConstraints for periodicity and MeshWorker in parallel

was calling copier on all cells. Using a filtered range on locally owned cells fixed the problem. Thanks praveen > On 4 May 2024, at 12:33 PM, Praveen C wrote: > > Hello > > In a 2d problem with all periodic bc, I was able to use WorkStream for > assembly. >

[deal.II] Problem with AffineConstraints for periodicity and MeshWorker in parallel

Hello In a 2d problem with all periodic bc, I was able to use WorkStream for assembly. But when I switch to MeshWorker, I get error about constraints (when running in parallel, serial is ok) An error occurred in line <2248> of file in function types::global_dof_index

Re: [deal.II] Assembling gradient jump at faces with FESystem

Hello Wolfgang FEFaceValues has two types of functions voidget_function_values <> (const InputVector _function, std::vector< typename InputVector::value_type > ) const voidget_function_values <> (const InputVector _function, std::vector< Vector <>< typename InputVector::value_type > >

Re: [deal.II] Assembling gradient jump at faces with FESystem

to implement a face_worker without FEInterfaceValues >> can be found in Step 12b, i.e. on >> https://github.com/dealii/dealii/issues/15839. >> >> Notice, it is probably best to focus on the face_worker only, because >> the tutorial seem to contain outdated code, see &

[deal.II] Assembling gradient jump at faces with FESystem

Hello In a continuous Galerkin method, I want to assemble terms involving jump in solution gradient and shape function gradient. For this, FEInterfacesValues seems like a good choice. But the function get_jump_in_function_gradients

Re: [deal.II] Growing memory usage with multiple threads

ote: > > > On 4/21/24 05:41, Praveen C wrote: >> I am using WorkStream in a code, and if I run with multiple threads, the >> memory usage grows continuously, as reported by "top". >> This does not happen with single thread. >> Can you give any tips on how to find

[deal.II] Growing memory usage with multiple threads

Hello I am using WorkStream in a code, and if I run with multiple threads, the memory usage grows continuously, as reported by "top". This does not happen with single thread. Can you give any tips on how to find the source of the problem ? Thanks praveen -- The deal.II project is located at

Re: [deal.II] Using some delta_t in embedded RK

variable which can be used to control when solution output is done. best praveen > On 20 Apr 2024, at 9:28 PM, Wolfgang Bangerth wrote: > > On 4/20/24 06:05, Praveen C wrote: >> I am using embedded RK scheme >> https://www.dealii.org/cu

[deal.II] Using some delta_t in embedded RK

Hello I am using embedded RK scheme https://www.dealii.org/current/doxygen/deal.II/classTimeStepping_1_1EmbeddedExplicitRungeKutta.html which chooses the time step in some adaptive manner. I also want to save solution to file at some intervals, e.g., at t = 1.0, 2.0, etc for which sometimes

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

CGALConfig.cmake is present *$ ls -l `spack location -i cgal`/lib64/cmake/CGAL/CGALConfig** -rw-r--r-- 1 praveen 8.5K Jun 13 2022 /home/spack/opt/spack/linux-opensuse15-broadwell/gcc-12.3.0/cgal-5.5.2-6umjskepmayihkexoscjiovjmtgo5toy/lib64/cmake/CGAL/CGALConfig_binary.cmake.in -rw-r--r-- 1

Re: [deal.II] Compilation Error with the Trilinos Package in deal.II

I also got this error recently when I used gcc-13.2 on opensuse 15.5. /home/ubuntu/dealii-candi/tmp/unpack/Trilinos-trilinos-release-13-2-0/packages/teuchos/core/src/Teuchos_BigUIntDecl.hpp:67:8: error: ‘uint32_t’ in namespace ‘std’ does not name a type; did you mean ‘wint_t’? Using gcc-12.3

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

I think deal.II is supposed to provide FindCGAL.cmake, there is no such file inside cgal. Thanks praveen > On 05-Jan-2024, at 8:16 PM, Daniel Arndt wrote: > > 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

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

So this error > On 05-Jan-2024, at 7:28 PM, Praveen C wrote: > > [100%] Built target step-1 > $ ./step-1 > step-1: malloc.c:2382: sysmalloc: Assertion `(old_top == initial_top (av) && > old_size == 0) || ((unsigned long) (old_size) >= MINSIZE && prev_

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

Hello Tim I download cgal-5.6 and use that, it seems there is no need to compile it as it is include only. And cmake does succeed to configure, I had to add -DGMP_INCLUDE_DIR=`spack location -i gmp` \ -DGMP_LIBRARIES=`spack location -i gmp` \ -DMPFR_INCLUDE_DIR=`spack location -i mpfr` \

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

Boost is external, from spack. That works fine, I can compile deal.II with this external boost if I disable cgal. Thanks praveen > On 05-Jan-2024, at 3:11 PM, timo Hyvärinen wrote: > > hi, praveen > > Did you check which Boost deal.ii was built from? The external one or bundled > one? If

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

In cmake/modules/FindDEAL_II_CGAL.cmake I print CGAL_DIR and remove QUIET # # CGAL requires C++17 and an externally configured Boost, otherwise the # call to find_package(CGAL) will fail. Guard the call to FIND_PACKAGE to # fail gracefully: # if(DEAL_II_HAVE_CXX17 AND NOT

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

5 gen 2024, alle ore 03:48, Praveen C >> ha scritto: >> >> spack location -i cgal >> /home/spack/opt/spack/linux-opensuse15-broadwell/gcc-12.3.0/cgal-5.5.2-6umjskepmayihkexoscjiovjmtgo5toy >> >> does return the correct directory, which contains >

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

spack location -i cgal /home/spack/opt/spack/linux-opensuse15-broadwell/gcc-12.3.0/cgal-5.5.2-6umjskepmayihkexoscjiovjmtgo5toy does return the correct directory, which contains CGAL/include, etc., and they do have the include files. I have tried directly putting the cgal directory path also,

[deal.II] cmake configure does not find cgal

Hello I have cgal installed via spack $ spack location -i cgal /home/spack/opt/spack/linux-opensuse15-broadwell/gcc-12.3.0/cgal-5.5.2-6umjskepmayihkexoscjiovjmtgo5toy and I add -DDEAL_II_CXX_FLAGS="-march=native -std=c++17" \ -DCGAL_DIR=`spack location -i cgal` \ -DDEAL_II_WITH_CGAL:BOOL=ON

Re: [deal.II] Overriden Function::value not being called

Hello You have specified boundary_indicator=1 when you call interpolate_boundary_values, but all boundaries are marked with indicator 0. You did not specify colorize=True option when creating the mesh, so all boundaries are marked as 0. So your function is fine, you just never called it.

Re: [deal.II] FullMatrix::print_formatted() issue

The manual page (assuming cell_matrix is a FullMatrix) https://dealii.org/developer/doxygen/deal.II/classFullMatrix.html#abdd86a102fefa37a3916a612d8e329bd says there is an option “zero_string”, may be set it to “0” and try it. best praveen > On 17-May-2023, at 8:28 PM, Vinayak Vijay wrote: >

Re: [deal.II] Re: dealii and hdf5 compile problem

't exist in my installation. I don't know why spack didn't install the > preferred version of hdf5 (1.12.2). I'm wondering whether I should install > the preferred version or the most recent version (1.13.2) of hdf5 and try > reinstalling deal.II? Could this solve the problem? Any suggest

[deal.II] Finding cut cells

Dear all I am looking for available methods to find cells cut by some given curve. I found such a case in step-60 and made this example https://bitbucket.org/cpraveen/deal_ii/src/master/embedded_curve/ As shown in the figures, some cells are not identified though they are cut by the curve.

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

Hello When compiling deal.ii on macos M1 with clang I get these messages while linking [ 90%] Linking CXX shared library ../lib/libdeal_II.g.dylib ld: warning: could not create compact unwind for ___emutls_get_address: registers 23 and 24 not saved contiguously in frame Compiling succeeds and

Re: [deal.II] MUMPS solver with Trilinos

openmpi is what I have now, I remember having compilation problems with mpich, but I have not tried it recently on M1 machines, so dont know if it works or not. Best praveen > On 17-Oct-2022, at 7:29 PM, Rahul Gopalan Ramachandran > wrote: > > Hello Praveen, > > Excellent! Thank you for

Re: [deal.II] MUMPS solver with Trilinos

I have compiled and used mumps from trilinos, using spack, on M1 machines. trilinos@13.0.1 mumps@5.4.1 best praveen > On 16-Oct-2022, at 9:12 PM, Rahul Gopalan Ramachandran > wrote: > > Hello everyone, > > I am not able to get Trilinos working with MUMPS direct solver, so I am > reaching

Re: [deal.II] automatic differentiation of pde solution wrt design variables in dealii

It depends on what kind of problem you have, linear or nonlinear, steady-state or transient. For example, in case of a linear BVP you may have A(E) U(E) = b(E) then A(E) U’(E) = b’(E) - A’(E) U(E) Use AD to compute b’(E), A’(E) which involves applying AD to the assemble process, that is

[deal.II] Re: dealii and hdf5 compile problem

-gxrwbuzg3xom562obmqaqtu5forevio5/include #HDF5_LIBRARIES = hdf5-shared and last line above seems wrong. Thanks praveen > On 17-Aug-2022, at 11:49 AM, Praveen C wrote: > > Hello > > I am compiling dealii@9.4.0 myself using dependent packages installed via > spack

[deal.II] dealii and hdf5 compile problem

Hello I am compiling dealii@9.4.0 myself using dependent packages installed via spack. I am facing the issue reported here https://github.com/spack/spack/issues/32023 and is fixed here https://github.com/spack/spack/pull/32079

Re: [deal.II] Accessing and modifying modal properties of a FE_DGQ solution

FE_DGQ are tensor product polynomials but FE_DGP is not. FE_DGP is a smaller space. So you cannot exactly transfer the solution between them. There is also a tensor product modal version, see FE_DGQLegendre, maybe thats what you want. Vandermonde matrix is good idea to go from modal —> nodal

Re: [deal.II] SparseDirectMUMPS before deal.ii 8.3.0

You need this https://dealii.org/current/doxygen/deal.II/classConditionalOStream.html > On 21-Apr-2022, at 1:30 PM, hkch...@gmail.com wrote: > > Hi, everyone. > > I have installed the deal.ii version 8.2.1 to use SparseDirectMUMPS replace > SparseDirectUMFPACK. It seems that I can run it in

[deal.II] Re: Clemson Computational Math Seminar: Bruno Blais - Feb 4

Is the video of this talk archived somewhere ? Thanks praveen On Friday, March 25, 2022 at 10:42:11 PM UTC+5:30 Timo Heister wrote: > Hi all, > > I would like to announce the following seminar talk in our Clemson > Computational Math seminar that is related to deal.II. If you are > interested,

Re: [deal.II] Website down?

You can download the offline documentation here which contains the tutorials https://github.com/dealii/dealii/releases best praveen > On 15-Mar-2022, at 3:36 PM, divyaprak...@gmail.com > wrote: > > Any information on when it'll be restored? I need

Re: [deal.II] Periodic bc for dim=1

? In hyper.deal > (https://github.com/hyperdeal/hyperdeal), we also have PBC in "1D". > > Peter > > On Tuesday, 8 February 2022 at 06:34:42 UTC+1 Praveen C wrote: > Dear all > > I was trying to write a code that works for 1/2/3d navier stokes. >

[deal.II] Periodic bc for dim=1

Dear all I was trying to write a code that works for 1/2/3d navier stokes. But it seems like collect_periodic_faces does not work for 1d ? Is this the case, and is there any other way to deal with this situation ? Thanks praveen -- The deal.II project is located at http://www.dealii.org/ For

Re: [deal.II] Wrong vertices at the boundary

There is a small error in this code. > On 16-Dec-2021, at 5:43 PM, Masoud Ahmadi wrote: > > for (const auto : dof_handler.active_cell_iterators()){ > for (const auto : cell->face_iterators()){ > if (face->at_boundary()){ > for (const auto v : face->vertex_indices()) >

[deal.II] v9.3.2 compilation error: sundials

Dear all While compiling v9.3.2, I get failure at this point. It seems to be this problem https://github.com/dealii/dealii/issues/12914 but I am still facing it. Is it not fixed in v9.3.2 ? Thanks praveen [ 58%] Building CXX object

Re: [deal.II] PETSc autodetect

There is a DEAL_II_ALLOW_AUTODETECTION = ON|OFF option, which probably disables all auto detection. best praveen On Thu, Dec 9, 2021 at 11:43 PM Michał Wichrowski wrote: > Dear all, > While I was setting up deal.II compilation on another machine I got stuck > for weeks: it was crashing when

Re: [deal.II] Are there known issue between Trilinos 13.0.1 and deal.II?

I have compiled dealii@9.3 with trilinos@13.0.1 on various systems with gcc 8/9/10 via spack. On Fri, Oct 29, 2021 at 10:04 AM blais...@gmail.com wrote: > Dear All, > I hope you are well. > I am trying to compile deal.II on a new cluster in Canada called Narval > (named after another cute

[deal.II] Re: Strange malloc error on step-1

The gmsh sdk I had linked to seems to be responsible for this. After removing it and recompiling deal.II its fine. Thanks praveen > On 18-Oct-2021, at 6:17 PM, Praveen C wrote: > > I am getting this error when I run step-1 on my Debian-11 with gcc@10.2.1 > > $ make > [

[deal.II] Strange malloc error on step-1

I am getting this error when I run step-1 on my Debian-11 with gcc@10.2.1 $ make [ 50%] Building CXX object CMakeFiles/step-1.dir/step-1.cc.o [100%] Linking CXX executable step-1 [100%] Built target step-1 $ ./step-1 step-1: malloc.c:2379: sysmalloc: Assertion `(old_top == initial_top (av) &&

[deal.II] Using mapping as manifold

Dear all I want to transform a cartesian grid with my own transformation which is defined by a formula. I have done this and created a MappingFEField as in this example https://bitbucket.org/cpraveen/deal_ii/src/master/mapping_fe_field/ Saving to vtu format I can see it is working.

Re: [deal.II] Issues related to usage of parallel::fullydistributed::triangulation

In your check periodicity function you must replace p:f:t with Triangulation. Attached code has those changes and compiles.bestpraveen -- 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

Re: [deal.II] The assignment of initial values by user-defined initialValues

But VectorTools::project needs a vector without ghosts, right ? And thats what the error is saying. In step-40 solve function, we already use a vector without ghosts when we solve the linear system. Similar thing has to be done for project. best praveen > On 04-Oct-2021, at 10:06 AM, Wolfgang

Re: [deal.II] Interlacing update ghost and assemble

levant code is here: >> https://github.com/dealii/dealii/blob/master/include/deal.II/matrix_free/cuda_matrix_free.templates.h#L1012-L1057 >> >> <https://github.com/dealii/dealii/blob/master/include/deal.II/matrix_free/cuda_matrix_free.templates.h#L1012-L1057> >&g

[deal.II] Interlacing update ghost and assemble

Dear all I use la::d::Vector in a continuous nodal FE code. I want to mix update ghost and assemble so that I dont wait for update ghost to finish solution.update_ghost_values_start() Assemble on all locally owned cells that do not need ghost values solution.update_ghost_values_finish() Assemble

Re: [deal.II] dealii on Apple M1

It depends on what all other packages you want along with deal.II Using spack I could install almost all dependencies except oce trilinos It needed some tweaks to some packages though. If your needs are not too broad, you may be able to get all needed dependencies from homebrew itself. best

Re: [deal.II] Question on TimeStepping methods

28-Aug-2021, at 9:16 AM, Praveen C wrote: > > Thank you, I could pass NULL in place of id_minus_tau_J_inverse and it works. > > best > praveen -- The deal.II project is located at http://www.dealii.org/ For mailing list/forum options, see https://groups.google.com/d/forum/dea

Re: [deal.II] Question on TimeStepping methods

Thank you, I could pass NULL in place of id_minus_tau_J_inverse and it works. best praveen -- 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

[deal.II] Question on TimeStepping methods

Dear all In TimeStepping methods We have functions taking these arguments https://www.dealii.org/current/doxygen/deal.II/classTimeStepping_1_1ExplicitRungeKutta.html#ad43855dc35abf17e9c707444c0901790

[deal.II] Re: Problem using EmbeddedRungeKutta

I modified step-52 in following way Change this line TimeStepping::EmbeddedExplicitRungeKutta> embedded_explicit_runge_kutta(method, coarsen_param, refine_param, min_delta,

[deal.II] Problem using EmbeddedRungeKutta

Dear all I am using TimeStepper with classical RK4 which works fine. I am facing some issue with using embedded RK schemes. My usage is like this. typedef LinearAlgebra::distributed::Vector PVector; const TimeStepping::runge_kutta_method method = TimeStepping::DOPRI; If I do

Re: [deal.II] Computing DFT

Thank you. Related to this question. Can I save a parallel distributed solution using serialize and read it back as a serial triangulation/solution ? This may help me to sample the solution on a Cartesian grid and do FFT. Thanks praveen > On 23-Jul-2021, at 8:30 PM, Wolfgang Bangerth

[deal.II] Computing DFT

Dear all I want to compute the DFT of a PDE solution. I have seen some Fourier Series in deal.II https://www.dealii.org/developer/doxygen/deal.II/classFESeries_1_1Fourier.html and it says it is used to compute the DFT on the reference cell. I have seen examples of its use to estimate local

Re: [deal.II] Printing git version from my deal.ii-based code

Thank you, this is exactly what I wanted. > On 01-Jul-2021, at 3:00 PM, 'peterrum' via deal.II User Group > wrote: > > I am doing in a project like this: > https://github.com/hyperdeal/hyperdeal/blob/0d1529484306b034e442d79ab85ae7dbe8815e63/CMakeLists.txt#L61-L67 > >

Re: [deal.II] Printing git version from my deal.ii-based code

Thanks a lot. That gives me the deal.II git version. Is there a way to put the git version of my application code also ? I suppose this has to be done to cmake stage I suppose. Thanks praveen > On 01-Jul-2021, at 2:52 PM, 'peterrum' via deal.II User Group > wrote: > > There are some macros:

[deal.II] Printing git version from my deal.ii-based code

Dear all I have my deal.II based application code under git. Is there a way to print the git version from my deal.II code ? Something like this https://github.com/smessmer/gitversion but I couldnt get that to work. My knowledge of cmake almost zero. Any suggestions ? Best praveen -- The

Re: [deal.II] order of dofs within a cell in a FESystem

> On 02-Jun-2021, at 8:44 PM, Wolfgang Bangerth wrote: > > For DG, it probably wouldn't be very difficult to implement something like > this, but it doesn't exist at the moment. > > Can I ask why you want this particular property? Why is it not enough to have > all DoFs associated with a

Re: [deal.II] SparseDirectUMFPACK - member function "initialize" takes very long

The initialize function does the LU factorization which is the most expensive part https://www.dealii.org/developer/doxygen/deal.II/classSparseDirectUMFPACK.html#af23a74dd9cac006c9d87c18e5d638076

Re: [deal.II] order of dofs within a cell in a FESystem

Thanks David for the clarification. I am using nodal DG basis. I would like all components located at a support point to be consecutive in memory. I could achieve this by implementing a renumber_node_wise function. However, when I get local dofs on a cell

[deal.II] order of dofs within a cell in a FESystem

Is it possible to change the order of dofs within a cell in a FESystem ? I have situation like this: FESystem fe; degree = 1 nvar = 4 // number of unknowns at each support point fe(FE_DGQArbitrarynodes(QGauss<1>(degree+1), nvar) I print for i = 0; ihttp://www.dealii.org/ For mailing

Re: [deal.II] Looping over cells and faces: MeshWorker vs WorkStream

it. >> >>> I know MeshWorker can be used for this, but it seems to be not used by many >>> people and I suspect it may not be maintained well. >> >> Yes, it seems not to be used terribly often (at least, the questions on the >> mailing list are infrequent, and

[deal.II] Looping over cells and faces: MeshWorker vs WorkStream

Dear all I have a DG-type scheme which is totally quadrature-free and matrix-free. I need to first loop over cells and then loop over faces. I know MeshWorker can be used for this, but it seems to be not used by many people and I suspect it may not be maintained well. Is it possible to use

Re: [deal.II] Regarding installation of dealii in MacOS Big Sur

Using the m1 docker version, I was able to run the docker ubuntu image on my macmini-m1 like this docker pull dealii/dealii:latest cd /path/of/your/deal.II/code docker run -ti -v $(pwd):/home/dealii/shared -w /home/dealii/shared dealii/dealii:latest Compile and run your code inside docker.

Re: [deal.II] Regarding installation of dealii in MacOS Big Sur

Do you have intel cpu or m1 ? I could not run the dmg on my m1. Get same error as you. best praveen > On 19-May-2021, at 8:37 PM, Debabrata Auddya wrote: > > Hello, > > I am trying to install dealii v 9.2.0 v3 on MacOS Big Sur after following > this thread. > >

Re: [deal.II] dealii/spack on apple m1

somewhat slow, though. best praveen > On 17-Apr-2021, at 9:25 AM, Praveen C wrote: > > Thanks for pointing this out. I think these recent fixes should be the > solution to compiling on M1. I will give it a try. > > best > praveen > >> On 17-Apr-2021, at 12:26 AM, Jean-P

Re: [deal.II] dealii/spack on apple m1

Thanks for pointing this out. I think these recent fixes should be the solution to compiling on M1. I will give it a try. best praveen > On 17-Apr-2021, at 12:26 AM, Jean-Paul Pelteret wrote: > > Dear Praveen, > > Even though you’re trying to install deal.II, it would seem that the issue >

Re: [deal.II] dealii/spack on apple m1

Thank you for sharing your experience. So it looks like we cannot yet install spack natively, right ? > On 15-Apr-2021, at 12:08 PM, Alberto Salvadori > wrote: > > 1 - After opening a terminal rosetta, installing gcc, installing deal.ii via > spack and using gfortran from gcc installation,

[deal.II] dealii/spack on apple m1

Dear all Has anyone succeeded in installing dealii via spack on an apple m1 ? What works/does not work ? spack gives me a spec but also gives some warnings $ spack spec dealii Input spec dealii Concretized sysctl: unknown oid

Re: [deal.II] Query about external meshing software

In your geo file, you can check orientation of boundary normals by adding Geometry.Normals = 100; // adjust 100 if needed Then check that all normals are pointing outward. If some surface has opposite normal, then flip sign for example, if Line Loop(53) = {17, 18, -1, -16}; Ruled Surface(54)

[deal.II] Re: Compilation problem on macos bigsur intel with xcode 12.3

Dear all I could compile the latest dealii master branch, which is good for me. You can ignore my previous message. Thanks praveen > On 06-Jan-2021, at 8:40 AM, Praveen C wrote: > > Dear all > > I am running into lot of possibly clang errors when installing dealii-9.

Re: [deal.II] dealii installation Mac OS 10.15.5 (Catalina)

I think this should work arch -x86_64 ./step-1 > On 08-Dec-2020, at 7:33 PM, Wells, David wrote: > > Hi Deepak, > > You may be the first person to ever run deal.II on the new M1 Mac, > congratulations :) > > I don't have an M1 machine so I will have to make some educated guesses on > how

Re: [deal.II] building with PETSc and METIS

You are installing to /usr/local which needs root permissions Use sudo make install best praveen > On 14-Oct-2020, at 5:44 PM, Christos Karliampas > wrote: > > Thanks for the answer, the problem as I understood appeared when building > with documentation , so I just turned it to >

Re: [deal.II] install problems with clang@12.0.0 and Xcode12.0 from spack

Here is an observation. I set DYLD_FALLBACK_LIBRARY_PATH=/path/to/spack/view/lib and run as ./step-40 then it runs fine. But if I use “make run”, then it cannot find libumfpack !!! Best praveen > On 02-Oct-2020, at 6:56 PM, 'Alexander Greiner' via deal.II User Group > wrote: > > Hi

Re: [deal.II] install problems with clang@12.0.0 and Xcode12.0 from spack

I notice that you are using cmake from /opt/local/bin, but spack would have installed its own cmake. Could you try using cmake from spack to compile and then run the example ? best praveen > On 29-Sep-2020, at 7:35 PM, 'Alexander Greiner' via deal.II User Group > wrote: > > Hi Luca, > > I

Re: [deal.II] install problems with clang@12.0.0 and Xcode12.0 from spack

find libmuelu… ? Best praveen > On 28-Sep-2020, at 2:32 PM, Praveen C wrote: > > Try adding path to you spack lib, like this > > export DYLD_FALLBACK_LIBRARY_PATH=/opt/spack/lib > > I add this to my $HOME/.profile file. > > Best > praveen > >> On 28-Sep-20

Re: [deal.II] install problems with clang@12.0.0 and Xcode12.0 from spack

Try adding path to you spack lib, like this export DYLD_FALLBACK_LIBRARY_PATH=/opt/spack/lib I add this to my $HOME/.profile file. Best praveen > On 28-Sep-2020, at 2:21 PM, 'Alexander Greiner' via deal.II User Group > wrote: > > Hi Wolfgang, > > at least I could not find a log file, but

Re: [deal.II] step-7 remarks on superconvergence and quadrature

I did this quick test in 2d for a Poisson problem https://github.com/cpraveen/fembook/blob/master/deal.II/ex04/demo.cc (Change quadrature rules in this code as indicated below) degree=1 assembly using QGauss(2) error

[deal.II] step-7 remarks on superconvergence and quadrature

Dear all In step-7 https://www.dealii.org/current/doxygen/deal.II/step_7.html under section on "Verification of correctness”, there is this statement (e.g., for linear elements, do not use the QGauss(2) quadrature formula) because solution may exhibit superconvergence at the QGauss(2) points.

Re: [deal.II] Step-4 1D problem

In 1-D, GridGenerator::hyper_cube(triangulation, -1, 1); will assign different boundary indicators to left and right side, even though colorize=false is default, see https://dealii.org/developer/doxygen/deal.II/namespaceGridGenerator.html#acea0cbcd68e52ce8113d1134b87de403

Re: [deal.II] Small step-22 modification gives unexpected result

The step-22 code uses the stress form of the NS equations. If you use the form -mu*Laplace(u) + grad(p) = 0 and use the same dirichlet+neumann bc, then it seems to give the correct solution. Please check it. These are the lines to be changed std::vector> symgrad_phi_u(dofs_per_cell);

Re: [deal.II] Small step-22 modification gives unexpected result

AM, Praveen C wrote: > > Some things I noticed in your code. > > Your velocity seems to be from right to left since you put a minus sign > > if (component == 0) > return -umax*p[1]*(1.0 + p[1]); > > but you say you put neumann bc on right boundary. > > You

Re: [deal.II] Small step-22 modification gives unexpected result

Some things I noticed in your code. Your velocity seems to be from right to left since you put a minus sign if (component == 0) return -umax*p[1]*(1.0 + p[1]); but you say you put neumann bc on right boundary. You are using equality test to identify boundary faces if

Re: [deal.II] How to visualise the simulation output in vtk format on a one dimensional mesh?

For 1d, you can save data in gnuplot format https://www.dealii.org/current/doxygen/deal.II/classDataOutInterface.html#a85407e870a68179ebe62410d9efc153f and use gnuplot to visualize it. Best praveen > On 16-Mar-2020, at 8:33 PM, Krishnakumar Gopalakrishnan > wrote: > > Same as before -- have

Re: [deal.II] discontinous contour over elements

Is your inflow conditions uniform ? If not then large elements at inflow could introduce errors. If you compute vorticity as curl of velocity pointwise, then the vorticity would be discontinuous at element boundaries, since you are using only C0 elements for the velocity. If you are averaging

Re: [deal.II] circl e obstacle in step-35

Maybe this helps https://bitbucket.org/cpraveen/deal_ii/src/master/step-35/ I used gmsh to generate the mesh. Best Praveen On Tue, 26 Nov 2019 at 3:21 PM, hussan zeb wrote: > How to change the geometry of the Navier stokes problem from around a > square obstacle to a circular obstacle in

Re: [deal.II] vmult function

Here is an example of vmult int n = 10; FullMatrix A(n,n); // Fill A Vector v(n), w(n); // Fill v A.vmult(w, v); This implements w = A*v Best praveen > On 08-Nov-2019, at 11:48 AM, Muhammad Adil wrote: > > Hi Everyone, > If someone used the function of vmult, will you please tell me what

Re: [deal.II] Loading a .msh mesh of big size into parallel::distributed::triangulation using p4est

Just an observation running your code with deal.II-9.1.1 In debug mode, it runs at 65 sec. In release mode, I get segmentation error. Best praveen $ make Scanning dependencies of target step-40 [ 50%] Building CXX object CMakeFiles/step-40.dir/step-40.cc.o [100%] Linking CXX executable step-40

Re: [deal.II] Doubt regarding largrange element gradient value in step-3

On 17-Oct-2019, at 11:19 PM, het patel wrote:That is a nice piece of information. Yes, it is doing integration on the real cell. I didn't knew that there was something like FE.shape_grad for integration on the reference cell. I thought it did by default isoparametric mapping

Re: [deal.II] Doubt regarding largrange element gradient value in step-3

You may be comparing the gradient on reference and real cells. They will differ by some factor due to the mapping. FE.shape_grad gives gradient on reference cell https://dealii.org/current/doxygen/deal.II/classFiniteElement.html#fefdf45fc2a5e7db18acb64063f5d

Re: [deal.II] How to get the coordinates of dof with FE_Raviart_Thomas?

If you are going to use Cartesian meshes only, then you may want to check out https://dealii.org/developer/doxygen/deal.II/classFE__RaviartThomasNodal.html Best praveen > On 24-Sep-2019, at 10:42 AM, Phạm Ngọc Kiên

Re: [deal.II] Re: DG explicit time integration for linear advection equation with MeshWorker (suggestions)

> On 18-Sep-2019, at 11:40 AM, Doug wrote: > > Praveen, > > I did inspire my loops out of your dflo ;). Note that if you use a > distributed grid, the user_index() will not be consistent across MPI > boundaries. Therefore, you might end up with doubly-counted faces. Here is a >

Re: [deal.II] Re: DG explicit time integration for linear advection equation with MeshWorker (suggestions)

> I now have one more question. In assemble_face_term function of step-33, the > normal numerical flux is calculated. This function is called for every > cell-neighbor pair from the assemble_system function. This means, if I am not > wrong, at every interior face quadrature point, the

Re: [deal.II] DG explicit time integration for linear advection equation with MeshWorker (suggestions)

If you are working only on Cartesian grids, these matrices will be same for all cells, except maybe for scalar factors. In this case, you can first compute them on a reference cell. For assembly, you loop over cells and perform a small matrix-vector product on each cell. Similar things can be

Re: [deal.II] Step 26 with advection term

For an advection-diffusion equation you have to specify boundary conditions everywhere on the boundary. You have to decide what boundary conditions you want, dirichlet, neuman, or some mixture of these. Also you have just one boundary condition for the pde, there is no separate advection boundary

[deal.II] Saving and retrieving solution vectors for adjoint based optimization

Dear all I am doing an adjoint based optimization for which I solve a forward problem and an adjoint. I need to save the forward solution at different time instants and read it back to compute a gradient. What methods are available to do this in deal.II ? Thanks praveen -- The deal.II

Re: [deal.II] Alternative to local_range in LinearAlgebra::distributed::Vector

> On 02-Feb-2019, at 8:41 PM, Wolfgang Bangerth wrote: > > No :-) The reason why it was scheduled for removal is that it only works in > case the locally owned range is really just a single interval. But that's not > always the case. For example, you may have reordered degrees of freedom in

Re: [deal.II] Non-unique solution values for FE_Q_DG0

Hello Jinhyun This is not an error. Since you have discontinuous solution, at a face in 2d you have two different values and at a vertex in 2d you may have 4 different values. When deal.II saves solution files for visualization, it does not do any averaging etc. at face/vertex, hence you will

Re: [deal.II] Alternative to local_range in LinearAlgebra::distributed::Vector

Thank you for this tip. For my need, I only require the starting global index on each partition since I am using local_element to do some operations like multiplying by a local time step which is different for each cell. The local_range function gave this info. Here is a sample usage in my code

[deal.II] Alternative to local_range in LinearAlgebra::distributed::Vector

Dear all This is marked deprecated https://dealii.org/developer/doxygen/deal.II/classLinearAlgebra_1_1distributed_1_1Vector.html#aa3749d01e5c34fa01a5753b1998f9716 What is the way to get local_range.first ? Thanks praveen -- The deal.II project is located at http://www.dealii.org/ For mailing

  1   2   3   >