site map | contact | login | Protection des données personnelles | Powered by FluxBB | réalisation artaban
You are not logged in.
This is not planned for the moment.
Building code_aster on arm64 is not possible for the moment
see https: //gitlab.com/codeaster-opensource-documentation/opensource-installation-development
In the salome_meca container, you have to compile yourself the MPI version of code_aster then link it to aster_study
This an internal name. Never use the name of the finite element but only the modelization. Here, 2D_FLUIDE
Fixed in 16.3
This solution is ok for code_aster < 15.3. After that, your fonction should work directly (and the syntax of MACRO has complety changed)
I suppose that works for quadratique cells.
Try and see.
The most probable is that ChatGPT gives non-sense results.
MECA_STATIQUE has 3 different version (fortran : v14, c++ : v15, python: v16)
In v15, I advise you to use RESULTAT in place of reuse in the command file
In v16.3< : You can fix as you do it
In >v16.3: It is fixed and works as in v14
There was no test to verify this configuration before
Ok this is a bug.
I fix it in the current version and I have updated the doc.
The doc is not up to date.
Have you a complete output to be sure that it is work as I am expected ?
from 15.3.17 until 15.8 : salome_meca-lgpl-2020.0.1-7-20210601-scibian-9
They do not exist for version 15. The first one for which the new procedure is given is 16.2.4 (20220817).
From 15.3.17, this is possible with the appropriate singularity container
This is planned in the HPC-roadmap (in the two next years probably)
This is not possible in THER_NON_LINE. You have to define manually your time discretization
Which version do tou use ?
There is a known bug for ELGE_ELNO (memory leak) not yet solved
C'est un problème de doc. Une modif en v16 a été mise en v15. Il faut utilser GROT_GDEP en v15
run_testcase from src directory and run_aster (in install/bin) from where you want (before source the file profile.sh in install/share_aster)
the stable version is 15.6 now on gitlab
You can build with the container but with appropriate version
https: //gitlab.com/codeaster-opensource-documentation/opensource-installation-development/-/blob/main/devel/changelog.md
Have you verified that your installation is ok with lauching all verification test cases with run_testcases ?
You can use run_aster instead to have more informations
The only difference between containter salome_meca_2021.1.0-***** is the prerequisities for the unstable version of code_aster and nothing change for salome_meca.
They are given to build unstable version and not stable version 15.*. Moreover, they work a specific interval of unstavle version
If you don't plan to use unstable version, this is not necessecary to change salome_meca container.
Use git repository instead source forge
Hello,
Is there an outpout ? .mess ? .o ? or something like that ?
The mpi installation is presented because it is always harder
For sequential version, discard installation of MPI librairies and at the end to configure prerequisities, set
ARCH=gcc-openblas-seq for exemple. seq is for sequential version (it is mentionned at the end)
For aster, use waf_std instead of waf (waf=waf_mpi now)
Maybe the macro ASSEMBLAGE is a good choice
Read the following doc to:
https /codeaster.readthedocs.io/en/latest/devguide/recommendations.html?highlight=macro#macro-commands
There is no backward compatibility since too many things have changes.
Your links is a good starting point. I advise you to see macro inside code_aster that have converted