Closed Irvise closed 1 month ago
Hi @Irvise,
The correct way to launch the test is to use make check
that will first build all the libraries required for the tests and then run make
test`: see here for details.
I have installed OpenSUSE Tumbleweed in a virtualbox, I hope to confirm soon that everything is ok. If so, I will send you the packages that I had to install.
Best,
Thomas
Hi @thelfer,
thank you very much for your quick answer. You are 100% right and I should have read the documentation with a bit more care, make check
did indeed build the test dependencies and ran the tests. I just defaulted mentally to the standard UNIX make test
... Here are the wonderful results!
[...]
7593/7593 Test #7593: ComputePK1DerivativeTest ................................................................................................................... Passed 0.00 sec
100% tests passed, 0 tests failed out of 7593
Total Test time (real) = 144.33 sec
[100%] Built target check
Also, thank you for allowing the @Data
entry for @MaterialProperty
in 5.0. Tabulated data is quite common in engineering codes!
Best regards, Fer
P.S.: I am back again playing with MFront as I saw that ASME BPVC III Div. 5 (High temperature nuclear reactors code) 2023 edition added an accepted behaviour model for 316 Stainless Steel (HBB-Z appendix) and I wanted to see if I could code it. After playing around with @MaterialProperty
I think it will not be practical to create several files/@MaterialProperty
entries and I will see if the Chaboche model described in ASME (it includes viscosity and damage) can be used directly with the ElastoViscoPlastic
Brick + some damage calculation/post-processing.
Hi @Irvise,
Great !
I am back again playing with MFront as I saw that ASME BPVC III Div. 5 (High temperature nuclear reactors code) 2023 edition added an accepted behaviour model for 316 Stainless Steel (HBB-Z appendix) and I wanted to see if I could code it.
Seems interesting.
I will see if the Chaboche model described in ASME (it includes viscosity and damage) can be used directly with the ElastoViscoPlastic Brick + some damage calculation/post-processing.
The damage part could be tricky. Do not hesitate to ask for help !
Hi Thomas,
I was building the latest MFront (both master and the 4.2.1 release) and when I run
make test
I am getting thousands of error messages. They are all related to thegeneric
section of tests. The error message is always an exception, here is the manual run of CTest with the output of the test:All tests fail with basically the same error
libMFrontStandardElastoViscoPlasticityBrickBehaviours.so
or more commonly, thelibMFrontGenericBehaviours.so
library cannot be found:Basically, all tests fail in this manner, they cannot find one or another library, mainly
libMFrontGenericBehaviours.so
. Igrep
andfind
my way to try and find what is going on. I saw thatlibMFrontGenericBehaviours.so
had a final "s" at the end, which the.cxx
file does not. Ised
allGenericBehaviours
toGenericBehaviour
and tried recompiling MFront, but that did not fix the issue. I also tried to see if eitherlibMFrontGenericBehaviours.so
orlibMFrontGenericBehaviour.so
where somewhere to be found after compiling MFront, but that is not the case, the shared libraries are nowhere to be found... There are also libraries namedlibMFrontGenericBehaviours3.so
and similar...All this leads me to believe there is some kind of issue with CMake/CMakeFiles, which are not generating or compiling the required libraries and correctly setting up the
LIBRARY_PATH
for them to be used... Maybe it is related to my setup, to the options I am using, to a typo/mistake in the naming of libraries within MFront...What do you think?
Here is the information regarding my compilation setup:
ccmake
During CMake config, I see a line/config that says
Maybe it is related?
Thank you for your help.
Edit 1: added reference to the
ccmake
command.Edit 2: added
boost_python
info