[deal.II] Exec format error when using mpirun -n 2 ./parameters_sneddon_2d.prm

2017-02-12 Thread Qing Yin
Hi, I am a new deal.ii user. I am trying to follow the "readme.md" to run the example by command line /$ mpirun -n 2 ./parameters_sneddon_2d.prm An error is below: Open MPI tried to fork a new process via the "execve" system call but failed. Open MPI checks many things before attempting to lau

[deal.II] Problem with MUMPS Solver

2017-02-12 Thread RAJAT ARORA
Hello all, I am using deal.ii to solve a 3D solid mechanics problem along with Petsc and P4est. The problem has 9 Dofs per node. I run my code on stampede. This error may or may not be directly related to an issue with deal.ii (may be sparsity pattern ?) or its usage. But, given a diverse comm

Re: [deal.II] second derivatives of solution

2017-02-12 Thread hanks0227
OK Thank you very much! Kyusik. 2017년 2월 13일 월요일 오후 1시 3분 27초 UTC+9, Wolfgang Bangerth 님의 말: > > On 02/12/2017 06:15 PM, hank...@gmail.com wrote: > > > > Do you mean that it is right way to get second derivatives but it could > not be > > continuous? > > Yes, you are using the correct way

Re: [deal.II] second derivatives of solution

2017-02-12 Thread Wolfgang Bangerth
On 02/12/2017 06:15 PM, hanks0...@gmail.com wrote: Do you mean that it is right way to get second derivatives but it could not be continuous? Yes, you are using the correct way of computing something that is of questionable usefulness. As I mentioned, this is because for finite elements, the

Re: [deal.II] Run into a cluster

2017-02-12 Thread Timo Heister
Joaquin, this looks like your compilers are different now from when you compiled deal.II. On a cluster this can happen if you have different modules loaded or you are on a different machine than before. What does your detailed.log in you deal.II build directory say about what compiler you are usin

[deal.II] Run into a cluster

2017-02-12 Thread 'Joaquin' via deal.II User Group
Dear all: I am trying to run the tutorial programs into a cluster. For example in the case of step-3, after typing make, it sent the following error: [fjust@boqueron step-3]$ make Scanning dependencies of target step-3 [100%] Building CXX object CMakeFiles/step-3.dir/step-3.cc.o cc1plus: error:

Re: [deal.II] second derivatives of solution

2017-02-12 Thread Wolfgang Bangerth
On 02/12/2017 04:50 AM, hanks0...@gmail.com wrote: It seems that my question is very easy one, but I can't find the similar thing... I know how I can use solution and first derivative solution that comes from fe_values.get_function_values and fe_values.get_function_gradients So I thought that

Re: [deal.II] Parallelization of step-2: DynamicMatrix to SparsityMatrix

2017-02-12 Thread Timo Heister
Yes, it looks like you can not copy a DSP with an IndexSet into a SparsityPattern. We could make that work but it also is not a very useful operation. Can you try replacing DynamicSparsityPattern dsp (locally_relevant_dofs); with DynamicSparsityPattern dsp; ? This is not efficient for large com

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

2017-02-12 Thread Timo Heister
>> >> > fe_values.get_function_values() on the velocity vector. >> > ExcIndexNotPresent (n) >> >> get_function_values() needs to see the ghost values of your vector >> (all DoFs on a cell). > > > yes, that is what is being passed to it. I use the post flags _lo and _lr > for my vector names t

[deal.II] second derivatives of solution

2017-02-12 Thread hanks0227
Hi all, It seems that my question is very easy one, but I can't find the similar thing... I know how I can use solution and first derivative solution that comes from fe_values.get_function_values and fe_values.get_function_gradients So I thought that second derivative can be obtained by