[deal.II] Re: test failing with message /usr/lib/openmpi/lib/libmpi_cxx.so: error: undefined reference to 'opal_list_item_t_class'

2016-06-08 Thread Bruno Turcksin
Giovanna, can you run a simple "hello world" program with MPI? Best, Bruno On Wednesday, June 8, 2016 at 4:48:24 PM UTC-4, Giovanna Bucci wrote: > > Hello, > > I have compiled deal II without problems (see attached detailed.log file > for settings) on a new machine with Ubuntu 16.04. However

Re: [deal.II] test failing with message /usr/lib/openmpi/lib/libmpi_cxx.so: error: undefined reference to 'opal_list_item_t_class'

2016-06-09 Thread Bruno Turcksin
having both mpich and openmpi may cause problems but this is not > the case. > > Best regards, > Giovanna > > > > On Jun 9, 2016, at 10:02 AM, Bruno Turcksin <bruno.turck...@gmail.com> wrote: > >> Giovanna >> >> 2016-06-09 9:25 GMT-04:00 Gi

[deal.II] Re: Compilation problem on Power8 system

2016-06-09 Thread Bruno Turcksin
Todd, On Thursday, June 9, 2016 at 11:02:20 AM UTC-4, Todd Raeker wrote: > > > /gpfs/gpfs0/software/src/dealii-8.4.1/bundled/tbb41_20130401oss/src/tbb/tools_api/ittnotify_static.c:570:72: > > warning: ‘cdecl’ attribute directive ignored [-Wattributes] > static void ITTAPI

Re: [deal.II] test failing with message /usr/lib/openmpi/lib/libmpi_cxx.so: error: undefined reference to 'opal_list_item_t_class'

2016-06-09 Thread Bruno Turcksin
Giovanna, 2016-06-09 8:50 GMT-04:00 giovanna bucci : > Yes, a simple program like that runs ok. I have never seen something like that :-( I haven't had any problem to install deal.II on Ubuntu 16.04. Can you go where you have installed deal (probably something like

Re: [deal.II] Segmentation fault in WorkStream::run() for colored iterators in parallel(MPI)

2016-05-25 Thread Bruno Turcksin
Julius, On Wednesday, May 25, 2016 at 8:42:41 AM UTC-4, juliuswit...@gmail.com wrote: > Here is the simplified code (also attached): > This code won't work. The colored_iterators contains every active cells. You need to check that they are locally owned. Best, Bruno -- The deal.II project

Re: [deal.II] Read and write PETScWrappers::MPI::Vector to binary file

2016-06-02 Thread Bruno Turcksin
Shiva, take a look here for an example of checkpoint-restart: https://github.com/geodynamics/aspect/blob/master/source/simulator/checkpoint_restart.cc You need to use SolutionTransfer to do the serialization/deserialization. Best, Bruno On Thursday, June 2, 2016 at 9:40:12 AM UTC-4, Shiva

[deal.II] Re: How to group elements for highly heterogeneous material

2016-06-21 Thread Bruno Turcksin
Chenchen, is there a reason why you can't use material_id? If you can't, you could use user_flag or user_pointer https://dealii.org/developer/doxygen/deal.II/DEALGlossary.html#GlossUserFlags Best, Bruno On Tuesday, June 21, 2016 at 11:07:48 AM UTC-4, Chenchen Liu wrote: > > Hi all, > > I

Re: [deal.II] Re: Segmentation fault (core dumped) error

2016-05-20 Thread Bruno Turcksin
2016-05-20 9:12 GMT-04:00 Ehsan : > #2 0x75cb614b in dealii::SparsityPattern::reinit > (this=this@entry=0x7fffb498, m=829992, n=829992, max_per_row=8032) > at /deal_II/dealii-8.3.0/source/lac/sparsity_pattern.cc:251 It could be a memory problem. You try

Re: [deal.II] Re: Segmentation fault (core dumped) error

2016-05-20 Thread Bruno Turcksin
Martin, 2016-05-20 9:38 GMT-04:00 Martin Kronbichler : > If I look at the current source code of SparsityPattern, I think there is > indeed a bug: We define the variable max_vec_len to be of type size_type (= > types::global_dof_index) whereas it actually should be

Re: [deal.II] Re: "Make test" failure following Trilinos installation with Mesquite

2016-05-20 Thread Bruno Turcksin
Oded, 2016-05-20 13:25 GMT-04:00 Oded Yaakobi : > (ii) I downloaded the most recent version of Spack (version 0.8.17), but its > installation failed. You may see below the error message that I got: You need to use the development version on github. Best, Bruno -- The

Re: [deal.II] Re: Segmentation fault (core dumped) error

2016-05-20 Thread Bruno Turcksin
Ehsan, 2016-05-20 9:41 GMT-04:00 Ehsan : > it starts to allocate and when it faces the memory shortage it returns the > error This is what he does but you may have a hit a bug in deal.II. You have less than 4 billions degrees of freedom but you try to create a matrix

Re: [deal.II] Re: Post-processing of solution (computing stress distribution)

2016-05-24 Thread Bruno Turcksin
David, 2016-05-24 8:12 GMT-04:00 David : > And the error message on this part of code is: > > error: undefined reference to 'void > dealii::DataOut_DoFData, 2, > 2>::add_data_vector >>(std::vector

Re: [deal.II] Re: Post-processing of solution (computing stress distribution)

2016-05-24 Thread Bruno Turcksin
David, 2016-05-24 9:24 GMT-04:00 David : > I only declared it as an independent class and then instantiate it in my > problem class, is that OK? Yes, that's fine. I would need to see the code to see what's going on but first, you should create the default constructor

Re: [deal.II] Re: Segmentation fault (core dumped) error

2016-05-20 Thread Bruno Turcksin
-20 9:56 GMT-04:00 Bruno Turcksin <bruno.turck...@gmail.com>: > Ehsan, > > 2016-05-20 9:41 GMT-04:00 Ehsan <rabizadeh.eh...@gmail.com>: >> it starts to allocate and when it faces the memory shortage it returns the >> error > This is what he does but you may have

[deal.II] Re: Memory usage using MPI/PETSc on single processor

2016-07-25 Thread Bruno Turcksin
Pete, you need to make sure that the codes do exactly the same thing. I haven't compared the codes too much but they use different preconditioners. This will impact how much memory you will use. Best, Bruno On Monday, July 25, 2016 at 1:59:53 PM UTC-4, Pete Griffin wrote: > > I found that

[deal.II] Re: merging a mesh

2016-07-28 Thread Bruno Turcksin
Javier, On Thursday, July 28, 2016 at 9:34:15 AM UTC-4, Jose Javier Muñoz Criollo wrote: > > > I have a problem that involves the consideration of a mesh that varies in > size over time. In particular it requires the creation and removal of a > layer of elements on top of a cubic domain. This

Re: [deal.II] Re: Memory usage using MPI/PETSc on single processor

2016-07-27 Thread Bruno Turcksin
ion of reply check the > original thread. > > Pete Griffin > > On Monday, July 25, 2016 at 2:31:09 PM UTC-4, Bruno Turcksin wrote: >> >> Pete, >> >> you need to make sure that the codes do exactly the same thing. I haven't >> compared the codes too much bu

[deal.II] Re: Extra layer around mesh

2016-08-11 Thread Bruno Turcksin
Joel, On Thursday, August 11, 2016 at 11:29:35 AM UTC-4, Joel Davidsson wrote: > > > I just had a question about the third option. If one used FE_Nothing, what > is the boundary that is used when doing calculations? Is it the boundary of > the big box or the green box? Is this handle

[deal.II] Re: Observations on CPU usage on a single processor

2016-08-03 Thread Bruno Turcksin
Pete, On Wednesday, August 3, 2016 at 2:06:12 PM UTC-4, Pete Griffin wrote: > > I was surprised to see all 4 cores (8 threads) being used in step-8 which > does not have PETSc/MPI support. In the solver, the main thread was using > 100% of it's resources. The others were using very consistently

[deal.II] Re: Possibly wrong format of msh file

2016-07-15 Thread Bruno Turcksin
Marek, On Friday, July 15, 2016 at 9:27:02 AM UTC-4, Marek Čapek wrote: > P.S.: I have tried finer meshes (refined in gmsh) and merged meshes of > the geometries in the files in the attachment. > Here is the picture of the whole chamber >

Re: [deal.II] Re: Possibly wrong format of msh file

2016-07-15 Thread Bruno Turcksin
Marek 2016-07-15 11:27 GMT-04:00 Vinetou Incucuna : > i have redesigned my *.geo files and got > much coarser meshes, which work in dealii GridIn Great. > I dont know much about adaptive refinement in deal.ii. > (Till now I used only the uniform refinement featur). > Are

[deal.II] Re: External point forces on elastic bodies

2016-07-12 Thread Bruno Turcksin
Hamed, On Tuesday, July 12, 2016 at 12:40:35 PM UTC-4, Hamed Babaei wrote: > I want to solve a very simple elastic problem to obtain stress and > displacement field, applying a point force on one of the external surfaces > of a rectangular 2D domain. > Although I found some sort of external

Re: [deal.II] test failing with message /usr/lib/openmpi/lib/libmpi_cxx.so: error: undefined reference to 'opal_list_item_t_class'

2016-07-13 Thread Bruno Turcksin
in case someone else run in this issue. I fixed > it by using mpich, instead of openmpi. Whatever I tried with openmpi was > giving me linking problems. > > Thank you, > Giovanna > > > On Jun 9, 2016, at 11:14 AM, Bruno Turcksin <bruno.turck...@gmail.com> wrote: >

Re: [deal.II] Re: External point forces on elastic bodies

2016-07-12 Thread Bruno Turcksin
Hamed 2016-07-12 15:10 GMT-04:00 Hamed Babaei : > > I was wondering if I should use the rhs_vector when assembling the > cell_rhs as follows: > > cell_rhs(i) += (rhs_vector(i) * fe_values.JxW (q)); > > or do something else. would you be able to give me part of a sample

Re: [deal.II] Re: introduce random initial condition

2016-07-05 Thread Bruno Turcksin
Anup, On 07/05/2016 08:17 PM, Anup Basak wrote: But my question is that is there any function in dealii through which I can generate random numbers within a range at the nodal points. No, you will need to do it yourself using rand (http://www.cplusplus.com/reference/cstdlib/rand/). Best,

Re: [deal.II] Re: Extracting a component from Block FESystem and DofHandler

2016-08-16 Thread Bruno Turcksin
gust 16, 2016 at 12:35:05 PM UTC-5, Spencer Patty wrote: >> >> Bruno, >> >> Thanks for responding! >> >> >> >> On Tuesday, August 16, 2016 at 11:16:41 AM UTC-5, Bruno Turcksin wrote: >>> >>> Spencer, >>> >>> I

[deal.II] Re: Re_Error during installation of dealii.8.4.1

2017-02-02 Thread Bruno Turcksin
Sathish, which version of the intel compiler are you using? Older versions are not tested. Best, Bruno On Thursday, February 2, 2017 at 1:55:07 PM UTC-5, Sathish Kumar wrote: > > Hi , > > I am trying to install dealii in a Linux environment and when I run the > command make install after the

Re: [deal.II] Re: Re_Error during installation of dealii.8.4.1

2017-02-02 Thread Bruno Turcksin
rently I am using icc/2013.5.192. Also the available ones in the cluster > are : > > icc/11.1.080, icc/2013.5.192, icc/2013_sp1.1.106 and icc/2015.0.090. > > Is any of the above supported? > > Regards > Sathish > > > > On Thursday, 2 February 2017 20:00:18 UTC+1,

Re: [deal.II] Re: WorkStream + hp : resizing members of Copy/Scratch data

2017-02-03 Thread Bruno Turcksin
2017-02-03 8:45 GMT-05:00 Denis Davydov : > Thanks for confirming this, Bruno. > Do you think we should mention in the documentation? Sure. Since you asked the question, there are probably others with the same question. Bruno -- The deal.II project is located at

[deal.II] Re: WorkStream + hp : resizing members of Copy/Scratch data

2017-02-03 Thread Bruno Turcksin
Denis, On Friday, February 3, 2017 at 8:36:03 AM UTC-5, Denis Davydov wrote: > > So far i don't have any indication that such resizing is not allowed, but > just wanted to double check. > It should work. I can't think of a reason why it wouldn't. Best, Bruno -- The deal.II project is located

Re: [deal.II] cannot call member function ‘void tbb::task::spawn(tbb::task&)’ without object

2017-01-31 Thread Bruno Turcksin
Lev, On Tuesday, January 31, 2017 at 4:40:14 PM UTC-5, Lev Karatun wrote: > yes, the latest version. > I'm not really sure what tbb is.. I'm using the bundled version I guess? > Since I didn't install it separately and found task.h > under /bundled/tbb41_20130401oss/include/tbb/ > According to

[deal.II] Re: question about a lage deformation model based on the updated Lagrangian foemolation

2017-01-25 Thread Bruno Turcksin
H.B., On Wednesday, January 25, 2017 at 3:02:01 PM UTC-5, ha.badn...@gmail.com wrote: > > > The image of the mapping applied to cell with center [0.295666 0.850622] > is distorted. The cell geometry or the mapping are invalid, giving a > non-positive volume fraction of -0.000119973 in

Re: [deal.II] A bug of visualization with VISIT

2017-02-22 Thread Bruno Turcksin
Hi, while it looks strange I don't think there is any problem. In plottingBug-0002.png, Visit only knows the value at the four vertices of the domain and then it tries to "fill" the domain with a possible solution. In figure1.png, you give Visit the values on 9 points and thus, it has more

Re: [deal.II] Re: Who is owning the instance in an exception raised by Subscriptor?

2017-02-14 Thread Bruno Turcksin
2017-02-14 11:30 GMT-05:00 'Uwe Köcher' via deal.II User Group : > I'm wondering, if it is impossible to call dof.clear(), > which should remove the link to triangulation object, > in python or the c++ code beyond? > > I've included such a line in a destructor within my

Re: [deal.II] Re: Renumbering dofs with petsc + block + MPI + Direct solver work around

2017-02-09 Thread Bruno Turcksin
2017-02-09 14:33 GMT-05:00 Spencer Patty : > Interesting, I wondered if SuperLU_dist might be parallel but I hadn't > looked into it yet. If it does work, then that certainly makes things much > simpler since I have trilinos integrated well. I will look into installing > it

Re: [deal.II] Re: Who is owning the instance in an exception raised by Subscriptor?

2017-02-15 Thread Bruno Turcksin
2017-02-15 6:53 GMT-05:00 Sebastian.Gonzalez-Pintor : > I have been thinking, and the problem is that I do not know when to do the > `dof.clear()`. In my case `dof` belongs to an instance (state) that still > exists when python try to remove the instance containing the

[deal.II] Re: Who is owning the instance in an exception raised by Subscriptor?

2017-02-14 Thread Bruno Turcksin
Hi, most of time this problem happens because the DoFHandler was declared before the Triangulation. In the destructor, the Triangulation will be destroyed before the DoFHandler. Best, Bruno On Tuesday, February 14, 2017 at 8:35:10 AM UTC-5, Sebastian.Gonzalez-Pintor wrote: > > > > Hello! I

Re: [deal.II] Re: Who is owning the instance in an exception raised by Subscriptor?

2017-02-14 Thread Bruno Turcksin
Sebas, 2017-02-14 9:29 GMT-05:00 Sebastian.Gonzalez-Pintor : > I do not know why this happens, because I do not wrap the interface for > Subscriptor (it is called internally inside the c++ classes, whithout python > knowing about its existence). I will try to ask for this

Re: [deal.II] Re: Recommendation: BCs in config

2017-02-13 Thread Bruno Turcksin
2017-02-13 9:46 GMT-05:00 Franco Milicchio : > Yes, I have thought of that but I cannot see right now how I could express > vector functions. Using a custom separator isn't really nice here... Why not? You can split the string several times using different separators:

Re: [deal.II] Re: Recommendation: BCs in config

2017-02-13 Thread Bruno Turcksin
2017-02-13 10:34 GMT-05:00 Franco Milicchio : > Because I can imagine that having several BCs with some magic char there is > simply a recipe for disaster :) > > Just think about this example: > > set dirichlet = x^2+y^2 : if(x < 0, 1, 3) : 0; 0 : 0 : 0 > > Since I

[deal.II] Re: Renumbering dofs with petsc + block + MPI + Direct solver work around

2017-02-09 Thread Bruno Turcksin
Hi, this is not the answer to your question but if I understand correctly, everything works fine with Trilinos and the only reason why you need PETSc is to use MUMPS. If that's the case, instead of using Amesos_KLU with Trilinos, you can use SuperLU_dist

[deal.II] Re: Matrix tools in MatrixCreator namespace

2017-02-16 Thread Bruno Turcksin
Sumedh, On Thursday, February 16, 2017 at 6:58:48 AM UTC-5, Sumedh Yadav wrote: > > Hey Daniel, > Thanks for the reply! > I tried doing what you suggested. The code snippet: > void * ptr = nullptr; //I tried with NULL and 0 synonyms as well > Why are you using void* You should use const

[deal.II] Re: Approximate Derivative Tensor functionality

2017-02-15 Thread Bruno Turcksin
Sumedh, use DerivativeApproximation::approximate_derivative_tensor(dof_handler, solution, cell, cell_grad) or DerivativeApproximation::approximate_derivative_tensor(dof_handler, solution, cell, cell_grad, 0) Best, Bruno On Wednesday, February 15, 2017 at 10:03:28 AM UTC-5, Sumedh Yadav

[deal.II] Re: Assemble system with variational right hand side

2017-01-23 Thread Bruno Turcksin
Hi, On Monday, January 23, 2017 at 2:06:17 PM UTC-5, Jaekwang Kim wrote: > > I just wanted to make sure that whether I implemented it right, > but I still cannot sure whether I am doing correctly. Let me be with > more specific questions. > This is not the right way to do it. You cannot

[deal.II] Re: Extracting a component from Block FESystem and DofHandler

2016-08-16 Thread Bruno Turcksin
Spencer, I think that passing around the entire system is the way to go but what I would do is to use component mask (http://dealii.org/developer/doxygen/deal.II/DEALGlossary.html#GlossComponentMask) and block mask

[deal.II] Re: Reference article for modified Kelly Estimator

2016-08-24 Thread Bruno Turcksin
Deepak, On Wednesday, August 24, 2016 at 7:08:34 AM UTC-4, Deepak Gupta wrote: > > In addition, I am looking for a reference where it is > discussed that the KellyEstimator works well as an indicator for a wide > variety of problems. > I don't think that such a reference exists see this thread

[deal.II] Re: Installation/Tutorial 1 Issue

2016-08-25 Thread Bruno Turcksin
Hi, On Thursday, August 25, 2016 at 8:59:11 AM UTC-4, peter.woer...@gmail.com wrote: > > So I installed dealII and everything presumably went according to plan. 0 > out of the 4 tests failed. Then when I copy and pasted tutorial 1 and > tried to run it, I ended up with 792 errors. Here are

Re: [deal.II] Question concerning BlockSparsityPattern.copy_from() member function

2016-09-07 Thread Bruno Turcksin
Dustin, 2016-09-07 11:07 GMT-04:00 Dustin Kumor : > I'm using the function "add_entries" only in combination with the > DynamicSparsityPattern and not with the SparsityPattern class. Moreover, it > seems that the DynamicSparsityPattern class doesn't even have a "copy_form"

[deal.II] Re: applying 3D mesh in step-20

2016-08-31 Thread Bruno Turcksin
Kyusik, On Wednesday, August 31, 2016 at 10:45:08 AM UTC-4, Jean-Paul Pelteret wrote: > > > You welcome :-) No, deal.II only supports quads/hexes > > so > you cannot use prisms.

Re: [deal.II] Help with Vector template instantiations

2016-09-14 Thread Bruno Turcksin
Dragan, On Wednesday, September 14, 2016 at 7:34:46 AM UTC-4, dragniko...@gmail.com wrote: > I am finishing what I started doing and will write you about an > interesting application of deal.II. It is mostly useful for multi scale > modelling in chemical engineering but is general in nature.

[deal.II] Re: Help with the weak form

2016-09-09 Thread Bruno Turcksin
Dragon, On Friday, September 9, 2016 at 7:02:15 AM UTC-4, dragniko...@gmail.com wrote: > > Thank you Denis. > > I tried that. The function log(c/(1-c)) looks like: > > > >

[deal.II] Re: Parallel hp::DoFHandler

2016-09-30 Thread Bruno Turcksin
Andreas, On Friday, September 30, 2016 at 5:40:43 AM UTC-4, Andreas Krämer wrote: > > can you already foresee when there will be a version of hp::DoFHandler > that works with parallel::distributed::Triangulation? > It is in our todo list but nobody has time to implement it right now. You will

Re: [deal.II] Re: Running deal.ii with METIS

2016-10-05 Thread Bruno Turcksin
Chenchen, 2016-10-04 18:15 GMT-04:00 Chenchen Liu : > Second, for "#PBS -l nodes=1:ppn=x", no matter what the value x is, the > error always exists. But this is not running across multiple nodes. You are only using one node, so this the same as running the code on your

[deal.II] Re: Nonlinear Material

2016-10-05 Thread Bruno Turcksin
Lisa On Wednesday, October 5, 2016 at 10:17:17 AM UTC-4, Lisa Collins wrote: > I would like to solve a problem with material non-linearity. > Can you please tell me which tutorial steps could be the base of my code. > Take a look at step 15

Re: [deal.II] Re: Running deal.ii with METIS

2016-10-05 Thread Bruno Turcksin
2016-10-05 10:19 GMT-04:00 Chenchen Liu : > For "#PBS -l nodes=1:ppn=x", > when x=1, and use "mpirun -np 1 ./step-17", it works well as a single > processor. > when x=1, and use "mpirun -np 4 ./step-17", memory error happens. > For "#PBS -l nodes=4:ppn=1", > when I use

[deal.II] Re: Running deal.ii with METIS

2016-10-04 Thread Bruno Turcksin
Chenchen, On Tuesday, October 4, 2016 at 5:16:12 PM UTC-4, Chenchen Liu wrote: > > > ***Memory allocation failed for SetupCtrl: maxvwgt. Requested size: > 18446744073709551608 bytes > This line says that you are asking for 18446744073 GB of memory... > It is said that there is a bottleneck for

[deal.II] Re: Issue with Trilinos installation (issue with blas?)

2016-10-04 Thread Bruno Turcksin
Matt, On Tuesday, October 4, 2016 at 4:49:28 PM UTC-4, Matt Weller wrote: > > packages/teuchos/numerics/src/CMakeFiles/teuchosnumerics.dir/build.make:280: > recipe for target > 'packages/teuchos/numerics/src/libteuchosnumerics.so.11.4.2' failed > Why are you using such an old version of

[deal.II] Re: Can we say that the higher order method, the more accurate?

2016-09-19 Thread Bruno Turcksin
On Monday, September 19, 2016 at 2:12:31 AM UTC-4, Martin Kronbichler wrote: > > > In addition to what Daniel and Wolfgang said: One does definitely benefit > from going to higher degrees and deal.II is able to handle this (the > accurate boundary representation by a mapping is one thing that

[deal.II] Re: Input Q2 element generated by Gmsh using Gridin

2016-09-21 Thread Bruno Turcksin
Chenchen, On Wednesday, September 21, 2016 at 11:56:31 AM UTC-4, Daniel Arndt wrote: > > We only support to read in meshes that are not refined. Translated to your > situation this means that you are only able to use Q1 meshes as initial > grid. > Of course, you can then use

[deal.II] Re: Help with time stepping methods

2016-09-15 Thread Bruno Turcksin
Vaibhav On Thursday, September 15, 2016 at 5:52:41 AM UTC-4, Vaibhav Palkar wrote: > 1. What is wrong with the embedded method? Does the problem lie in my > understanding? > It's hard to say what is wrong from here. Is coarsen_param greater than one? Have you reached max_delta? Finally, just

[deal.II] Re: How can I have access to the all element of previous solution?

2016-09-23 Thread Bruno Turcksin
Kyusik, On Friday, September 23, 2016 at 2:09:26 AM UTC-4, hanks0...@gmail.com wrote: > > In the lecture, previous_solution_gradients[] is used. But, I also want to > know How I can use just previous_solution[]. > > At first, I used previous_solution[q_index] in assemble_system as in >

Re: [deal.II] Re: How can I have access to the all element of previous solution?

2016-09-23 Thread Bruno Turcksin
Kyusik, 2016-09-23 9:11 GMT-04:00 : > In the lecture 31.65, the professor change the step-6.cc a little to apply > Picard iteration. > > Doing this, he uses previous_solution_gradients[q_index]. > > So, like this, I just used previous_solution[q_index] > > But, he also say

Re: [deal.II] Re: Input Q2 element generated by Gmsh using Gridin

2016-09-22 Thread Bruno Turcksin
Chenchen, 2016-09-22 10:54 GMT-04:00 Chenchen Liu : > Thank you for your sharing. To my mind, after we generate Q2 mesh in Gmsh, > we have all the information about vertices and cells. We can read these > information to deal.ii as point list and cell list, and then generate

Re: [deal.II] Re: Help with time stepping methods

2016-09-16 Thread Bruno Turcksin
2016-09-16 9:22 GMT-04:00 Vaibhav Palkar : > > The error being exactly zero is indeed intriguing. I'm still looking out for > the reason why that is happening. I didn't understand what you meant by a > "ramp in time". Could you please elaborate? Since you are using a

Re: [deal.II] Re: Help with time stepping methods

2016-09-16 Thread Bruno Turcksin
I have created a pull request here: https://github.com/dealii/dealii/pull/3123 Thanks again Bruno 2016-09-16 10:01 GMT-04:00 Bruno Turcksin <bruno.turck...@gmail.com>: > 2016-09-16 9:22 GMT-04:00 Vaibhav Palkar <vaibhav_pal...@iitgn.ac.in>: >> >> The error b

[deal.II] Re: 1D version of step3

2016-09-16 Thread Bruno Turcksin
Weixong, On Friday, September 16, 2016 at 11:35:09 AM UTC-4, Weixiong Zheng wrote: > > When running the program, the solution is zero on the left side as > expected. However, the right side solution is something nonzero. Since I am > using the same apply_boundary_condition settings, I am

[deal.II] Re: Untraceable "not implemented" error in GDB

2016-09-16 Thread Bruno Turcksin
Alex, before the run command, you need to use: catch throw Best, Bruno On Friday, September 16, 2016 at 1:21:41 PM UTC-4, Alex Zimmerman wrote: > > A couple of times now I have encountered GDB output similar to my attached > snapshot. > > The previous time I had only made a very small change

[deal.II] Re: PETScWrappers::MPI::Vector Vs parallel::distributed::Vector

2016-08-26 Thread Bruno Turcksin
Hi, I guess it's more a question of preference. What I do is using the same vector type as the matrix type: PETSc matrix -> PETSc vector, Trilinos matrix -> Trilinos vector, matrix-free -> deal.II vector. deal.II vector can use multithreading unlike PETSc vector but if you are using MPI, I

Re: [deal.II] Re: Installation/Tutorial 1 Issue

2016-08-25 Thread Bruno Turcksin
Peter, 2016-08-25 11:03 GMT-04:00 Wolfgang Bangerth : > >> In the case of trying to compile the tbb.cc code using the command >> sudo gcc tbb.cc -std=c++11 >> >> I get a long string of errors that look something like this >> >> >>

Re: [deal.II] Re: Parallel Algebraic Operations

2016-08-24 Thread Bruno Turcksin
Bruce, 2016-08-24 12:41 GMT-04:00 : > If I try adding ghosts to the delta_rhs vector, I get the error > > An error occurred in line <909> of file > tsc_vector_base.cc> in function > void dealii::PETScWrappers::VectorBase::sadd(PetscScalar, const >

[deal.II] Re: This site can’t be reached, www.dealii.org took too long to respond.

2016-09-27 Thread Bruno Turcksin
Yes it is to be expected. See email below: Dear all, We are expecting more power outages on September 26 to 28. They are trying to fix the emergency supply such that the recent failures don't repeat. Let's keep our fingers crossed. The dealii web site will be down during these outages, which

[deal.II] Re: Function parser error

2016-10-03 Thread Bruno Turcksin
Jinhyun, On Monday, October 3, 2016 at 11:38:57 AM UTC-4, Jinhyun wrote: > Below I copy the exception message. It seems that the problem relates to > the size of input and stems from deal.II itself. Any help would be > appreciated! > Actually the problem is that muparser is being initialized

[deal.II] Re: what do SolverControl do for Direct Solvers?

2016-10-27 Thread Bruno Turcksin
Hamed, On Thursday, October 27, 2016 at 5:40:04 PM UTC-4, Hamed Babaei wrote: > First: I have parallelized a code in which I use > TrilinosWrappers::SolverDirect > . The problem is that with around 2 DOFs the code is run on my own > machine and on the cluster as well but when I increase

[deal.II] Re: Error with the content "dealii::numbers::is_finite(sum)"

2016-10-25 Thread Bruno Turcksin
Benhour, On Tuesday, October 25, 2016 at 12:59:29 PM UTC-4, benhour.amiria...@gmail.com wrote: > > I am solving a thermoelastic problem in cylindrical coordinate system. I > used CG solver for thermal problem and Newton-Raphson method for elasticity > section. For solving linear system, I used

Re: [deal.II] Re: How to use Trilinos instead of Petsc in step-40

2016-10-19 Thread Bruno Turcksin
2016-10-19 11:51 GMT-04:00 Hamed Babaei : > I need to compute the determinant of system_matrix to check if its positive > or negative so that I know my system_matrix is positive definite or not. That's what I thought. You don't need to know the determinant to do that, you

Re: [deal.II] Re: Help with time stepping methods

2016-11-21 Thread Bruno Turcksin
Vaibhav, 2016-11-10 5:21 GMT-05:00 Vaibhav Palkar : > Looking at the error message I get, the non convergence is probably not due > to the Newton iterations. I am using an iterative solver(Iterative Inverse > class of dealii) for computing the ((M-tau*J)^-1)*M*y term

Re: [deal.II] Choosing between adaptive BDF and IRK integrators: adaptive meshing effect?

2016-11-20 Thread Bruno Turcksin
Aslan, 2016-11-17 19:31 GMT-05:00 : > On Thursday, November 17, 2016 at 5:22:00 PM UTC-7, Wolfgang Bangerth wrote: >> > I can see that there is an implementation of IRK method in deal.ii. If >> > you do >> > not mind, may I ask what was the reason that that particular

Re: [deal.II] Re: Help with time stepping methods

2016-11-01 Thread Bruno Turcksin
Vaibhav, 2016-11-01 4:24 GMT-04:00 Vaibhav Palkar : > To quickly restate, I'm solving a reaction-diffusion problem using a > framework much like Step-52. My equations are stiff and hence I'm using > implicit methods for solving the resulting ODEs. Unfortunately, it

[deal.II] Re: How can I find the neighbouring element if I'm on a given face in a given cell?

2016-11-02 Thread Bruno Turcksin
Stephen, On Wednesday, November 2, 2016 at 1:20:58 PM UTC-4, smetcalfe...@gmail.com wrote: > I'm currently implementing a blow-up problem with space-time adaptivity > controlled by a derived pointwise a posteriori error bound and I'm having > trouble implementing the estimator (previously, I

[deal.II] Re: Visit error...?

2016-11-03 Thread Bruno Turcksin
Kyusik, On Thursday, November 3, 2016 at 5:51:20 AM UTC-4, hanks0...@gmail.com wrote: > > > And, I also calculated the {l2_norm of (exact_solution - numeric_solution) > / sqrt(#cells)} (I divided l2norm by sqrt(#cell) Because I want to know > kind of average value of difference between two

Re: [deal.II] Re: Convergence problem arising for large number of DoFs

2016-11-04 Thread Bruno Turcksin
Hamed, 2016-11-04 11:25 GMT-04:00 Hamed Babaei : > I have a quick question for you. I was wondering if the system_matrix of the > sequential code should be exactly the same as parallel on for the same > problem condition. It should be the same except for round-off error.

[deal.II] Re: hp convergence for non-linear solver.

2016-11-01 Thread Bruno Turcksin
Jaekwang, On Tuesday, November 1, 2016 at 5:43:36 PM UTC-4, Jaekwang Kim wrote: > > I think that I didn't make any mistake on my code lines because at least > my numerical solution is converging for the case p=1. > > but I couldn't get such convergence for the case p=2, and 3. I have tried >

Re: [deal.II] Re: Error with the content "dealii::numbers::is_finite(sum)"

2016-10-26 Thread Bruno Turcksin
Benhour, 2016-10-25 19:31 GMT-04:00 : > I am running in one process. What should i do for solving this error? Does > it mean that my vector has complex component? Like Daniel said, it probably means that your matrix is singular and therefore not invertible. Best,

Re: [deal.II] Re: what do SolverControl do for Direct Solvers?

2016-10-28 Thread Bruno Turcksin
Hamed, 2016-10-28 10:36 GMT-04:00 Hamed Babaei : > If I solve a problem with one million DOFs on a cluster of 64 processors, > every processor will be dealing with around 15000 DOFs. > So you mean even in that case direct solver would be slow? I thought direct > solver

[deal.II] Re: Convergence problem arising for large number of DoFs

2016-10-31 Thread Bruno Turcksin
Hamed, On Monday, October 31, 2016 at 3:01:20 PM UTC-4, Hamed Babaei wrote: > > This error message can indicate that you have simply not allowed a > sufficiently large number of iterations for your iterative solver to > converge. This often happens when you increase the size of your problem. In

Re: [deal.II] trouble running new step-55

2016-10-17 Thread Bruno Turcksin
Franck, On Monday, October 17, 2016 at 8:23:20 AM UTC-4, franck75 wrote: > Where is developer version 8.5.0-pre? > I installed dealii using candi, which I find very easy. > candi can install dev version 8.5.0-pre? > You can use candi to install the development version of deal.II by changing

Re: [deal.II] Re: MPI problem with petsc-3.7.4

2016-11-29 Thread Bruno Turcksin
2016-11-29 16:25 GMT-05:00 seyedali88 via deal.II User Group : > I have only Intel MPI installed at the moment. It's strange that it works > for 2 or 4 cores. Single core works flawlessly. > Maybe I should try the older compiler and see, if it works... I don't think it's a

Re: [deal.II] Re: Problem with installation

2016-12-08 Thread Bruno Turcksin
2016-12-08 8:57 GMT-05:00 sudarshan kumar : > I have complied netcdf in different directory and i pass the command > -DNETCDF_DIR=/path/to/netcdf to cmake when I configure dealii, but still > it search for the old directory and end up in error, Did you configure

[deal.II] Re: Problem with installation

2016-12-06 Thread Bruno Turcksin
Hi, On Tuesday, December 6, 2016 at 9:22:35 AM UTC-5, Sudarshan Kumar wrote: > /usr/bin/ld.gold: error: /usr/local/lib/libnetcdf_c++.a(netcdf.o): > requires unsupported dynamic reloc 11; recompile with -fPIC > The problem is that libnetcdf was compiled as a static library but you are building

[deal.II] Re: Literature sources of Krylov subspace solvers

2016-12-14 Thread Bruno Turcksin
Ingo, Krylov solvers have been around for a while and the implementation in deal.II is pretty standard. Here is a reference that a few people use https://web.stanford.edu/class/cme324/saad.pdf but you can just use google//google scholar to get more references. Best, Bruno On Wednesday,

Re: [deal.II] Re: Trouble building dealii with p4est - undeclared variables in scope

2016-12-02 Thread Bruno Turcksin
ange > anything - it still fails with variables undefined in the current scope. See > attached cmake and make logs for detailed output, but I don't noticed any > difference. > > Regards, > Chris > > > On Thursday, December 1, 2016 at 7:54:53 PM UTC+1, Bruno Turcksin wrote:

[deal.II] Re: Is deal.II for me? Or, I got burned elsewhere

2016-12-02 Thread Bruno Turcksin
Franco, I'll try to answer your questions but be aware that deal.II is a C++ library and that there are no python wrappers unlike Fenics: On Friday, December 2, 2016 at 6:59:16 AM UTC-5, Franco Milicchio wrote: > > 1 - Is it truly thread-safe? As in running a single analysis, or running >

[deal.II] Re: Trouble building dealii with p4est - undeclared variables in scope

2016-12-01 Thread Bruno Turcksin
Chris, On Thursday, December 1, 2016 at 10:56:13 AM UTC-5, Chris Coutinho wrote: > > With only p4est enabled, and P4EST_DIR pointing to the DEBUG folder of > version 0.3.4.1, I get the following errors associated with undeclared > variables in scope. The rest of the output can be found in the

[deal.II] Re: MPI problem with petsc-3.7.4

2016-11-29 Thread Bruno Turcksin
On Tuesday, November 29, 2016 at 11:13:07 AM UTC-5, seyedal...@googlemail.com wrote: > > I reported the problem to the PETSc team. Now I have setup my desktop > system with Intel compiler 2017 which compiled now without problems and the > step-40 works better now, but still gives an error. On

Re: [deal.II] Re: Is deal.II for me? Or, I got burned elsewhere

2016-12-02 Thread Bruno Turcksin
Franco, 2016-12-02 10:57 GMT-05:00 Franco Milicchio : > So can I use solvers as in Tutorial 3, for instance, and I'll be safe and > sound? Yes. > However, is someone planning some additions to deal.II to make it less > verbose, as in, for example, providing a DSL for

[deal.II] Re: Adding new finite element to the library

2017-01-06 Thread Bruno Turcksin
Hi, you also need to add your .cc and .inst.in files to the CMakeLists.txt https://github.com/dealii/dealii/blob/master/source/fe/CMakeLists.txt Then you need to reconfigure deal, so that the .inst files can be created, and finally you can recompile. Best, Bruno On Friday, January 6, 2017

[deal.II] Re: MUMPS overflow

2017-01-09 Thread Bruno Turcksin
Hi, On Monday, January 9, 2017 at 7:00:54 AM UTC-5, Pushparaj Bhosale wrote: > Solver: Linear solver failed because: MUMPS memory overflow - potentially > singular matrix > If the matrix of your system is singular, this means that your problem is ill-posed or that there is a bug in your code.

[deal.II] Re: Library file sizes

2017-01-10 Thread Bruno Turcksin
Praveen, yes it's normal. The debug library needs to keep a list of symbols that can be used by the debugger (if you use the release version in a debugger, you won't get much information). There may also partly due to the optimization of the code. Best, Bruno On Tuesday, January 10, 2017 at

[deal.II] Re: Postprocessing on Vector Valued Problem

2016-12-21 Thread Bruno Turcksin
Jaekwang, On Wednesday, December 21, 2016 at 4:05:54 PM UTC-5, Jaekwang Kim wrote: > > > but How can I relate my 'cellwise_shear_rate' with dof handler so that I > finally able to see the values in 'vtk' format? > for now, it just seems that no more than a list of arrays > You mean a list

Re: [deal.II] Re: Problem with installation

2016-12-06 Thread Bruno Turcksin
2016-12-06 15:46 GMT-05:00 Sudarshan Kumar : > I have done option 1 above. Still I get the following error when I run > make install, Did you reinstall deal in an empty directory? If not please do, cmake uses a cache and since it picked the static library the first time you

  1   2   3   4   5   >