If I do the following
(dim = 2)
GridGenerator::hyper_shell (triangulation,
Point(0,0),
1.0,
2.0);
static const SphericalManifold boundary;
triangulation.set_all_manifold_ids(0);
triangulation.
Thanks a lot, Bruno.
I will keep trying.
On Thu, Jun 1, 2017 at 5:50 PM, Bruno Turcksin
wrote:
> 2017-06-01 11:37 GMT-04:00 Tuanny Cajuhi :
> > You are right, thank you for the explanation. My code uses PetSc with
> MUMPS.
> > Just the output is done by Trilinos (the idea of joint_fe). In summar
2017-06-01 11:37 GMT-04:00 Tuanny Cajuhi :
> You are right, thank you for the explanation. My code uses PetSc with MUMPS.
> Just the output is done by Trilinos (the idea of joint_fe). In summary, I
> need to add p4est.
Download p4est, untar it, and try something like this:
./configure CC=cc CXX=CC
You are right, thank you for the explanation. My code uses PetSc with
MUMPS. Just the output is done by Trilinos (the idea of joint_fe). In
summary, I need to add p4est.
On Thu, Jun 1, 2017 at 5:29 PM, Bruno Turcksin
wrote:
> 2017-06-01 11:17 GMT-04:00 Tuanny Cajuhi :
> > I think numdiff and b
2017-06-01 11:17 GMT-04:00 Tuanny Cajuhi :
> I think numdiff and bzip2 are not available. From the list of available
> modules in https://www.hlrn.de/home/view/System3/CraySoftware . Petsc,
> Mumps, SuperLu, Trilinos are available. P4est I need to install.
But do you need all of these any way? Does
It seems that the Cray has PetSc, I am not sure about MUMPS and I would
need it.
>
> Well you don't need numdiff and you can load bzip2 from a module
> right? My question is more do you just need to install p4est and deal
> and you can load modules for everything else or do you need to install
>
2017-06-01 11:03 GMT-04:00 Tuanny Cajuhi :
> I tried to install using spack as you suggested in the previous message, but
> also got stuck. Basically I changed to module load cray-mpich/7.5.1, but
> then did not come further in (3) because I did not know how to change/what
> to change in packages.
>
>
> 1) ask your admin to install deal for you
>
I asked but it seems to not an option at the moment...
2) use spack that has support for cray but I don't think anyone has
> ever tried to use spack to install deal on a cray.
>
I tried to install using spack as you suggested in the previous mes
2017-06-01 10:40 GMT-04:00 Tuanny Cajuhi :
> I am on a cray (loaded module cray-mpich/7.5.1) and I don't know how to
> explicitly call one thing or the other
Oh that explains a lot. Not sure that candi is going to work then.
Cray don't work like other clusters :-/ One big difference is that
they do
I am on a cray (loaded module cray-mpich/7.5.1) and I don't know how to
explicitly call one thing or the other
On Thu, Jun 1, 2017 at 4:37 PM, Bruno Turcksin
wrote:
> 2017-06-01 10:34 GMT-04:00 Tuanny Cajuhi :
> > I did not find any of them... :/
> That's very strange how do you compile a mpi p
2017-06-01 10:34 GMT-04:00 Tuanny Cajuhi :
> I did not find any of them... :/
That's very strange how do you compile a mpi program by hand? Do you
have mpic++ ? Are you on a cray?
--
The deal.II project is located at http://www.dealii.org/
For mailing list/forum options, see
https://groups.googl
I did not find any of them... :/
On Thu, Jun 1, 2017 at 4:29 PM, Bruno Turcksin
wrote:
> This is the problem "mpif77: command not found" Does mpif77 exist? If
> not, does mpif90 exist? If yes try export F77=mpif90
>
> 2017-06-01 10:21 GMT-04:00 Tuanny Cajuhi :
> > Hi Bruno,
> >
> > thank you.
>
This is the problem "mpif77: command not found" Does mpif77 exist? If
not, does mpif90 exist? If yes try export F77=mpif90
2017-06-01 10:21 GMT-04:00 Tuanny Cajuhi :
> Hi Bruno,
>
> thank you.
>
>> Are you using the same version of gfortran and gcc? Can you send the
>> whole config.log. I need to
Hi Bruno,
thank you.
Are you using the same version of gfortran and gcc? Can you send the
> whole config.log. I need to see what's the environment that p4est is
> using.
>
/opt/gcc/6.2.0/bin/gfortran
/opt/gcc/6.2.0/bin/gcc
The log is attached. Thank you for the help!
--
The deal.II project i
Tuanny,
2017-06-01 9:56 GMT-04:00 Tuanny Cajuhi :
> My next issue is p4est...
:-(
> Build FAST version in /home/h/niitcaju/deal.ii-candi/tmp/build/p4est-2.0/FAST
> configure: error: in
> `/home/h/niitcaju/deal.ii-candi/tmp/build/p4est-2.0/FAST':
> configure: error: Fortran 77 compiler cannot cre
...
and the 2 lines with 'error' in the config.log:
configure:4404: error: in
`/home/h/niitcaju/deal.ii-candi/tmp/build/p4est-2.0/FAST':
configure:4406: error: Fortran 77 compiler cannot create executables
On Thu, Jun 1, 2017 at 3:56 PM, Tuanny Cajuhi wrote:
> Hi Bruno,
>
> I was able to build
Hi Bruno,
I was able to build boost after loading the module to update Python. Thanks
a lot for the hint!
My next issue is p4est...
Unpacking p4est-2.0.tar.gz
Building p4est 2.0
CFLAGS_FAST: -O2
CFLAGS_DEBUG: -O0 -g
This script tries configure and build the p4est library.
Build FAST: /home/h/ni
Tuanny,
2017-06-01 3:53 GMT-04:00 Tuanny Cajuhi :
> Failure with exit status: 1
> Exit message: There was a problem building boost 1_63_0.
> (attached you can find all the messages from the terminal)
The problem is this "libs/python/src/numpy/dtype.cpp:61: error:
‘NPY_FLOAT16’ was not declared in
Chuyi,
> i was trying to use Dealii to solve a 3-dimensional elastic-plastic
> question on a cubic by applying periodic boundary condition like in step-45
> (v. 8.5.0).
> std::vector::
> cell_iterator> > periodicity_vector;
> GridTools::collect_periodic_faces(dof_handler, 1, 2, 2, periodicity_
Dear Alberto,
Thanks for your feedback. We'll take a look at the documentation and try to
see how/where it can be enhanced.
Best,
Jean-Paul
On Thursday, June 1, 2017 at 12:51:42 PM UTC+2, Alberto Salvadori wrote:
>
> Thank you Jean Paul,
> very clear. Perhaps your notes could be added on the de
Dear all,
i was trying to use Dealii to solve a 3-dimensional elastic-plastic
question on a cubic by applying periodic boundary condition like in step-45
(v. 8.5.0).
std::vector::
cell_iterator> > periodicity_vector;
GridTools::collect_periodic_faces(dof_handler, 1, 2, 2,
periodicity_vector);/
Thank you Jean Paul,
very clear. Perhaps your notes could be added on the deal.II documentation
system, I believe this comprehensive description is not simple to been
acquired directly from the tutorials examples and class description.
Alberto
*Alberto Salvadori* Dipartimento di Ingegneria Civile
Thanks, that helps a lot.
So the QGaussLobatto or QIterated (from the compute_cod_array() function)
stay unchanged? In how far does changing them affect the computation?
Best,
Seyed Ali
--
The deal.II project is located at http://www.dealii.org/
For mailing list/forum options, see
https://gro
Hi Wolfram,
I was trying to write the Makefile for step-8 from scratch. I didn't
realize that cmake can be used to create the Makefiles for the step
programs.
The solution is quite simple:
cmake -DDEAL_II_DIR=/deal_folder
Thanks,
Daniel
--
The deal.II project is located at http://www.dealii.o
Dear Tuanny,
if you have zlib already installed on your system, *remove* the line with
zlib from the linux_cluster.platform,
it is located in
./candi/deal.II-toolchain/platforms/supported/linux_cluster.platform.
If you not have zlib installed, you should try to modify the package file
of zlib.
25 matches
Mail list logo
