getdp issueshttps://gitlab.onelab.info/getdp/getdp/-/issues2019-06-14T09:59:52Zhttps://gitlab.onelab.info/getdp/getdp/-/issues/56getdp 3.1.0 FTBS on Debina/Buster2019-06-14T09:59:52ZChristophe Trophimegetdp 3.1.0 FTBS on Debina/BusterHi,
trying to rebuild latest getdp with Sparskit support FTBS:
"-DENABLE_PETSC=0 -DENABLE_SPARSKIT=1"
```
[ 57%] Building CXX object CMakeFiles/getdp.dir/Kernel/Operation_Vector.cpp.o
/usr/bin/mpicxx -I/build/getdp-3.1.0+dfsg1/contri...Hi,
trying to rebuild latest getdp with Sparskit support FTBS:
"-DENABLE_PETSC=0 -DENABLE_SPARSKIT=1"
```
[ 57%] Building CXX object CMakeFiles/getdp.dir/Kernel/Operation_Vector.cpp.o
/usr/bin/mpicxx -I/build/getdp-3.1.0+dfsg1/contrib/Sparskit -I/build/getdp-3.1.0+dfsg1/Common -I/build/getdp-3.1.0+dfsg1/Interface -I/build/getdp-3.1.0+dfsg1/Kernel -I/build/getdp-3.1.0+dfsg1/Numeric -I/build/getdp-3.1.0+dfsg1/Functions -I/usr/lib/x86_64-linux-gnu/openmpi/include/openmpi -I/usr/lib/x86_64-linux-gnu/openmpi/include -I/usr/include/python2.7 -I/build/getdp-3.1.0+dfsg1/debian/build/sparskit/Common -g -O2 -fdebug-prefix-map=/build/getdp-3.1.0+dfsg1=. -fstack-protector-strong -Wformat -Werror=format-security -Wall -pedantic -Wdate-time -D_FORTIFY_SOURCE=2 -Wall -Wno-misleading-indentation -o CMakeFiles/getdp.dir/Kernel/Operation_Vector.cpp.o -c /build/getdp-3.1.0+dfsg1/Kernel/Operation_Vector.cpp
/build/getdp-3.1.0+dfsg1/Kernel/Operation_Vector.cpp:15:10: fatal error: petscvec.h: No such file or directory
#include "petscvec.h"
^~~~~~~~~~~~
compilation terminated.
```https://gitlab.onelab.info/getdp/getdp/-/issues/54entity missing in MSH file2019-04-27T07:02:43ZGiuseppeentity missing in MSH fileHi,
i'm trying to run the wrench2d model but, after meshing it, when performing getdp calculation, the process is aborted because something seems to miss in entity definition (see figure attached). I checked the MSH file but i can't unt...Hi,
i'm trying to run the wrench2d model but, after meshing it, when performing getdp calculation, the process is aborted because something seems to miss in entity definition (see figure attached). I checked the MSH file but i can't unterstand what's wrong.
I attached .geo .msh and .pro files.
Many thanks![Wrench2d](/uploads/18857481d9dcb403d070af42c5d56f96/Wrench2d.JPG)[wrench2D.geo](/uploads/1f7c4dbdfae4da17dbea279c79e831b9/wrench2D.geo)
[wrench2D.pro](/uploads/7889dcce2388c7e8864fcac7a33b9754/wrench2D.pro)
[wrench2D.msh](/uploads/21a8243d9d910040078c225306f2b68f/wrench2D.msh)
[wrench2D_common.pro](/uploads/f35a1dde515993f01ed8f56c6c7e0401/wrench2D_common.pro)https://gitlab.onelab.info/getdp/getdp/-/issues/53Allow build without PETSc2019-04-20T18:22:12ZMarcus Calhoun-LopezAllow build without PETScThe most recent release of getdp does not seem to build without PETSc.
If PETSc is not installed, `petscvec.h` does not exist and [Kernel/Operation_Vector.cpp](https://gitlab.onelab.info/getdp/getdp/blob/master/Kernel/Operation_Vector....The most recent release of getdp does not seem to build without PETSc.
If PETSc is not installed, `petscvec.h` does not exist and [Kernel/Operation_Vector.cpp](https://gitlab.onelab.info/getdp/getdp/blob/master/Kernel/Operation_Vector.cpp#L15) does not compile (see https://gitlab.onelab.info/getdp/getdp/commit/ff2ae098761367b7d030f94849ba85b931e2976e).
Testing for `HAVE_PETSC` fixed the problem for me.
````c++
#if defined(HAVE_PETSC)
#include "petscvec.h"
#endif
````https://gitlab.onelab.info/getdp/getdp/-/issues/52Double moving band machine model2019-05-20T14:50:37ZCássio Krugerkrugercassio@gmail.comDouble moving band machine modelHello guys! I'm working on a double moving band machine model for my final graduation project and I'm having some problems. The last problem I had was related to the "DomainM" group at the "machine_magstadyn_a_2rotors.pro" file. It seems...Hello guys! I'm working on a double moving band machine model for my final graduation project and I'm having some problems. The last problem I had was related to the "DomainM" group at the "machine_magstadyn_a_2rotors.pro" file. It seems to allow only one kind of region, I mean, only "Rotor_Magnets", as the examples files.
However, my model has a "Stator_Magnets" region and I can't make it work beside the "Rotor_Magnets" region. If I attach only one kind of region to "DomainM", it works, but not both at the same time.
Do you guys have any idea of the reason it doesn't work with two kind o regions (Stator_Magnets and Rotor_Magnets)? If so, please let me know as soon as possible.
Here is my github repo, so you can see the codes about that machine:
https://github.com/CassioKruger/PDD1
Thanks for the attention!https://gitlab.onelab.info/getdp/getdp/-/issues/51Started using Ubuntu and got an error2019-04-09T19:51:16ZCássio Krugerkrugercassio@gmail.comStarted using Ubuntu and got an errorHello, I recently started using Ubuntu and, when I click "Run" at gmsh to run a model that I'm working for a long time, I got this message: "Erro: GetDP - Could not read entities"
I don't know what it means, since it works perfectly on ...Hello, I recently started using Ubuntu and, when I click "Run" at gmsh to run a model that I'm working for a long time, I got this message: "Erro: GetDP - Could not read entities"
I don't know what it means, since it works perfectly on windows...
What can I do?
Thank you for the attention!https://gitlab.onelab.info/getdp/getdp/-/issues/50recent version of "minimal gmsh lib"2018-12-13T19:15:59ZGuillaume Demesyrecent version of "minimal gmsh lib"Hello Christophe,
I am facing trying to build recent getdp's with msh support (gmshread etc).
For the minimal Gmsh lib, I am following the relevant step in : https://gitlab.onelab.info/getdp/getdp/wikis/GetDP-compilation
When reading t...Hello Christophe,
I am facing trying to build recent getdp's with msh support (gmshread etc).
For the minimal Gmsh lib, I am following the relevant step in : https://gitlab.onelab.info/getdp/getdp/wikis/GetDP-compilation
When reading the mesh, getdp complains:
```
Info : Reading 'test_mesh_paramaille_5.msh'...
Info : 5170 vertices
Info : 29033 elements
Error : Matrix inversion requires LAPACK
```
How can we "minimally" add some lapack? -DENABLE_PETSC=1 does not do the trick...
Thanks! Guillaumehttps://gitlab.onelab.info/getdp/getdp/-/issues/47Break command not working2018-10-17T17:55:50ZGuillaume DilasserBreak command not workingIt looks like the Break command (used to exit from While or Time loops) is not working anymore iwith the latest GetDP 3.0.2 release: when it is called, the current loop continues and the message “Operation: ? ? ?” is printed on the conso...It looks like the Break command (used to exit from While or Time loops) is not working anymore iwith the latest GetDP 3.0.2 release: when it is called, the current loop continues and the message “Operation: ? ? ?” is printed on the console. From what I have tested, it was still working fine up to version 2.11.3. but I have not checked the other versions after 2.11.3.https://gitlab.onelab.info/getdp/getdp/-/issues/46Found a virus in GetDP documents2019-02-20T09:38:48ZCássio Krugerkrugercassio@gmail.comFound a virus in GetDP documentsHi, my name is Cássio and I'm an engineering student (Federal University of Pelotas, Brazil - Control and Automation Engineer). I'm was going to use GetDP (3.0.2, 10 september) to run a simulation for study, but, when I extracted the fi...Hi, my name is Cássio and I'm an engineering student (Federal University of Pelotas, Brazil - Control and Automation Engineer). I'm was going to use GetDP (3.0.2, 10 september) to run a simulation for study, but, when I extracted the files from the zip, my PC accused a virus (Trojan:Win32/Spursint.Q!cl). What should I do?
Thanks for the attention!https://gitlab.onelab.info/getdp/getdp/-/issues/45bug : libgfortran missing in getdp binaries for mac2019-02-20T09:38:55ZGuillaume Demesybug : libgfortran missing in getdp binaries for macSalut Christophe,
I tried the latest ONELAB bundle on for mac and, in short, libgfortran is missing.
In detail:
=> ONELAB side
`Info : Reading 'NonLinearEVP.pro'...`
`Info : Calling '"/Users/demesy/Desktop/Gmsh.app/Contents/Mac...Salut Christophe,
I tried the latest ONELAB bundle on for mac and, in short, libgfortran is missing.
In detail:
=> ONELAB side
`Info : Reading 'NonLinearEVP.pro'...`
`Info : Calling '"/Users/demesy/Desktop/Gmsh.app/Contents/MacOS/getdp" -onelab "GetDP" /Users/demesy/.gmshsock2 &'`
`Error : Abnormal server termination (Socket listening timeout on socket /Users/demesy/.gmshsock2)`
=> getdp side (from terminal):
`(py36) ➜ NonLinearEVP git:(master) ✗ /Users/demesy/Desktop/Gmsh.app/Contents/MacOS/gmsh NonLinearEVP.pro`
`dyld: Library not loaded: /opt/local/lib/libgcc/libgfortran.3.dylib`
`Referenced from: /Users/demesy/Desktop/Gmsh.app/Contents/MacOS/getdp`
`Reason: image not found`
I have a libgfortran.3 in, i.e. : `/usr/local/gfortran/lib`
Creating a symbolic link pointing there was enough to solve the issue:
`cd /opt/local/lib`
`sudo ln -s /usr/local/gfortran/lib libgcc`
Maybe this is related to this commit?
https://gitlab.onelab.info/getdp/getdp/commit/cefa47a60ba7a80dc9171aa01a76f24189637bdc
@+
GuillaumeChristophe GeuzaineChristophe Geuzainehttps://gitlab.onelab.info/getdp/getdp/-/issues/44bug with Periodic?2018-09-07T09:32:27ZGuillaume Demesybug with Periodic?Hello Christophe,
When using the latest git versions of gmsh/getdp with models/NonLinearEVP/NonLinearEVP.pro, I noticed that "Periodic Line {} = {} Translate {};" causes this weird Fatal error:
Info : Running '/Users/demesy/Desktop/...Hello Christophe,
When using the latest git versions of gmsh/getdp with models/NonLinearEVP/NonLinearEVP.pro, I noticed that "Periodic Line {} = {} Translate {};" causes this weird Fatal error:
Info : Running '/Users/demesy/Desktop/Gmsh.app/Contents/MacOS/getdp FEMfile3_dispersive_grating.pro -pre res_o3_ExEy_2dom_WFB_nleig_rat -msh FEMfile2_dispersive_grating_Tmesh.msh -cal -pos postop_o3_ExEy_2dom_WFB_nleig_rat -bin -slepc -nep_max_it 1000 -nep_type nleigs -nep_nleigs_rational -nep_target_magnitude -rg_interval_endpoints 0.00000000,18.83651567,0.00000000,0.62788386' [GetDP 3.0.0-git-376ec5d, 1 node, max. 1 thread]
Info : Started (Tue Aug 28 11:57:51 2018, Wall = 0.0198288s, CPU = 0.050849s, Mem = 5.00391Mb)
Info : Initializing Gmsh
Info : Loading problem definition 'FEMfile3_dispersive_grating.pro'
Info : Loading problem definition 'parameters_gmsh_getdp.dat'
Info : Selected Resolution 'res_o3_ExEy_2dom_WFB_nleig_rat'
Info : Loading Geometric data 'FEMfile2_dispersive_grating_Tmesh.msh'
Info : No current model available: creating one
Info : Reading 'FEMfile2_dispersive_grating_Tmesh.msh'...
Info : 4579 vertices
Info : 9325 elements
Fatal : Gmsh should be compiled using ANN
BTW, are periodic meshes on lines still required for "LinkCplx" by getdp? I noticed you don't use it in rhombus... and the NLeig model works without it!
Thanks, @+ Guillaumehttps://gitlab.onelab.info/getdp/getdp/-/issues/41Kernel/F_MultiHar.cpp:95: bad if ?2017-12-31T15:45:42ZDavid BindermanKernel/F_MultiHar.cpp:95: bad if ?getdp-2.11.3-source/Kernel/F_MultiHar.cpp:95:28: warning: logical 'and' of mutually exclusive tests is always false [-Wlogical-op]
Source code is
else if (Type1 == VECTOR && Type1 == SCALAR) {
Maybe better code
else if (Type1 == ...getdp-2.11.3-source/Kernel/F_MultiHar.cpp:95:28: warning: logical 'and' of mutually exclusive tests is always false [-Wlogical-op]
Source code is
else if (Type1 == VECTOR && Type1 == SCALAR) {
Maybe better code
else if (Type1 == VECTOR && Type2 == SCALAR) {https://gitlab.onelab.info/getdp/getdp/-/issues/40getdp does not build reproducibly2017-07-25T09:09:31ZBernhard M. Wiedemanngetdp does not build reproduciblyI'm trying to build all packages in the openSUSE Linux distribution twice and found that doing so for getdp gives a [diff in /usr/bin/getdp](http://rb.zq1.de/compare.factory-20170721/getdp-compare.out)
that can be fixed with the attache...I'm trying to build all packages in the openSUSE Linux distribution twice and found that doing so for getdp gives a [diff in /usr/bin/getdp](http://rb.zq1.de/compare.factory-20170721/getdp-compare.out)
that can be fixed with the attached patch [0001-use-cmake-TIMESTAMP.patch](/uploads/ffdf083564d159a354e32f1cb7b7789c/0001-use-cmake-TIMESTAMP.patch)
The other diffs come from texinfo and latex and will hopefully be solved by patches in those soon.https://gitlab.onelab.info/getdp/getdp/-/issues/39GenerateJac not working when GetDP compiled with SPARSKIT2017-08-05T11:52:58ZChristophe GeuzaineGenerateJac not working when GetDP compiled with SPARSKITWhen GetDP is compiled with SPARSKIT, the GenerateJac operation always triggers a GetDP error "Wrong arguments in LinAlg_AddMatrixMatrix()". I think this is due to a mistake in the function call at SolvingOperation.cpp:732 :
```
LinAlg_...When GetDP is compiled with SPARSKIT, the GenerateJac operation always triggers a GetDP error "Wrong arguments in LinAlg_AddMatrixMatrix()". I think this is due to a mistake in the function call at SolvingOperation.cpp:732 :
```
LinAlg_AddMatrixMatrix(&DofData_P->A, &DofData_P->Jac, &DofData_P->Jac) ;
```
Arguments 1 and 3 are not the same as they are expected to be, judging from the function definition in LinAlg_SPARSKIT.cpp :
```
void LinAlg_AddMatrixMatrix(gMatrix *M1, gMatrix *M2, gMatrix *M3)
{
if(M3 == M1)
add_matrix_matrix(&M1->M, &M2->M) ;
else
Message::Error("Wrong arguments in 'LinAlg_AddMatrixMatrix'");
}
```
I would say this is a simple inversion of the first and second argument.Christophe GeuzaineChristophe Geuzainehttps://gitlab.onelab.info/getdp/getdp/-/issues/36Higher order mesh and higher order basis functions not possible together2019-04-19T15:49:05ZChristophe GeuzaineHigher order mesh and higher order basis functions not possible togetherI attached an example where a second order mesh is used in a second order magnetodynamic problem (2D vector potential (0,0,Az), or more precisely the equivalent electric field formulation).
The Basisfunctions BF_PerpendicularEdge_2E see...I attached an example where a second order mesh is used in a second order magnetodynamic problem (2D vector potential (0,0,Az), or more precisely the equivalent electric field formulation).
The Basisfunctions BF_PerpendicularEdge_2E seems not to be supported:
```
P r o c e s s i n g . . .
Info : Generate[Sys_Mag]
Error : Unknown type of Element in BF_GradNode_2E
```
Is this a defect or not yet supported? I would really like to use higher order meshes with higher order basis functions.https://gitlab.onelab.info/getdp/getdp/-/issues/35Segmentation fault when running getdp from gmsh (gmsh-2.12.0, getdp 2.8)2017-08-05T11:59:00ZChristophe GeuzaineSegmentation fault when running getdp from gmsh (gmsh-2.12.0, getdp 2.8)Hi,
I receive the following error when trying to run OneLAB from gmsh:
[quaxo:12456] *** Process received signal ***
[quaxo:12456] Signal: Segmentation fault (11)
[quaxo:12456] Signal code: (128)
[quaxo:12456] Failing at address: (nil...Hi,
I receive the following error when trying to run OneLAB from gmsh:
[quaxo:12456] *** Process received signal ***
[quaxo:12456] Signal: Segmentation fault (11)
[quaxo:12456] Signal code: (128)
[quaxo:12456] Failing at address: (nil)
[quaxo:12456] [ 0] /lib64/libpthread.so.0(+0xf670)[0x2ada824b8670]
[quaxo:12456] [ 1] /opt/gmsh/lib/libGmsh.so.2.12(_ZNSt8_Rb_treeIP4CellSt4pairIKS1_6BdInfoESt10_Select1stIS5_E9Less_CellSaIS5_EE8_M_eraseEPSt13_Rb_tree_nodeIS5_E+0x24)[0x2ada7b7ae794]
[quaxo:12456] [ 2] /usr/lib64/octave/4.0.1/liboctinterp.so.3(_ZN10octave_mapD2Ev+0x28)[0x2ada807ae078]
[quaxo:12456] [ 3] /usr/lib64/octave/4.0.1/liboctinterp.so.3(_Z32initialize_default_warning_statev+0x124)[0x2ada80d087e4]
[quaxo:12456] [ 4] /usr/lib64/octave/4.0.1/liboctinterp.so.3(octave_initialize_interpreter+0x1b5)[0x2ada807a99e5]
[quaxo:12456] [ 5] /usr/lib64/octave/4.0.1/liboctinterp.so.3(octave_main+0x28)[0x2ada807ab028]
[quaxo:12456] [ 6] /opt/GetDP/bin/getdp(_ZN7Message10InitializeEiPPc+0x28f)[0x4303af]
[quaxo:12456] [ 7] /opt/GetDP/bin/getdp(_Z10MainLegacyiPPc+0x99)[0x48d479]
[quaxo:12456] [ 8] /opt/GetDP/bin/getdp(main+0x13)[0x42ad13]
[quaxo:12456] [ 9] /lib64/libc.so.6(__libc_start_main+0xf5)[0x2ada83218d05]
[quaxo:12456] [10] /opt/GetDP/bin/getdp[0x42b78d]
[quaxo:12456] *** End of error message ***
[quaxo:12460] *** Process received signal ***
[quaxo:12460] Signal: Segmentation fault (11)
[quaxo:12460] Signal code: (128)
[quaxo:12460] Failing at address: (nil)
[quaxo:12460] [ 0] /lib64/libpthread.so.0(+0xf670)[0x2b23970f3670]
[quaxo:12460] [ 1] /opt/gmsh/lib/libGmsh.so.2.12(_ZNSt8_Rb_treeIP4CellSt4pairIKS1_6BdInfoESt10_Select1stIS5_E9Less_CellSaIS5_EE8_M_eraseEPSt13_Rb_tree_nodeIS5_E+0x24)[0x2b23903e9794]
[quaxo:12460] [ 2] /usr/lib64/octave/4.0.1/liboctinterp.so.3(_ZN10octave_mapD2Ev+0x28)[0x2b23953e9078]
[quaxo:12460] [ 3] /usr/lib64/octave/4.0.1/liboctinterp.so.3(_Z32initialize_default_warning_statev+0x124)[0x2b23959437e4]
[quaxo:12460] [ 4] /usr/lib64/octave/4.0.1/liboctinterp.so.3(octave_initialize_interpreter+0x1b5)[0x2b23953e49e5]
[quaxo:12460] [ 5] /usr/lib64/octave/4.0.1/liboctinterp.so.3(octave_main+0x28)[0x2b23953e6028]
[quaxo:12460] [ 6] /opt/GetDP/bin/getdp(_ZN7Message10InitializeEiPPc+0x28f)[0x4303af]
[quaxo:12460] [ 7] /opt/GetDP/bin/getdp(_Z10MainLegacyiPPc+0x99)[0x48d479]
[quaxo:12460] [ 8] /opt/GetDP/bin/getdp(main+0x13)[0x42ad13]
[quaxo:12460] [ 9] /lib64/libc.so.6(__libc_start_main+0xf5)[0x2b2397e53d05]
[quaxo:12460] [10] /opt/GetDP/bin/getdp[0x42b78d]
[quaxo:12460] *** End of error message ***
I'm not sure what is going on? Maybe I am compiling getdp wrong. I have openmpi 1.10.2, PetSC 3.6.3 and octave 4.0.1 installed on Slackware64-14.1 Previous versions of gmsh and getdp worked very well.
If this has been covered previously, my apologies. Otherwise, thank you for gmsh, getdp and all of your work.
Cheers,
Fredhttps://gitlab.onelab.info/getdp/getdp/-/issues/34Build fails under fedora 232017-08-05T11:59:22ZChristophe GeuzaineBuild fails under fedora 23I compile getdp from sources on fedora 23 (gcc-5.3.1).
The compilation fails with the following error message:
[ 2%] Building CXX object CMakeFiles/getdp.dir/Common/Message.cpp.o
/home/collette/repositories/FEM/getdp/Common/Message.cpp...I compile getdp from sources on fedora 23 (gcc-5.3.1).
The compilation fails with the following error message:
[ 2%] Building CXX object CMakeFiles/getdp.dir/Common/Message.cpp.o
/home/collette/repositories/FEM/getdp/Common/Message.cpp: Dans la fonction membre statique ‘static void Message::Initialize(int, char**)’:
/home/collette/repositories/FEM/getdp/Common/Message.cpp:126:44: erreur: ‘Py_DecodeLocale’ was not declared in this scope
wargv[i] = Py_DecodeLocale(argv[i], NULL);
^
/home/collette/repositories/FEM/getdp/Common/Message.cpp:127:34: erreur: too many arguments to function ‘void Py_SetProgramName(wchar_t*)’
Py_SetProgramName(wargv[0], NULL);
^
In file included from /usr/include/python3.4m/Python.h:114:0,
from /home/collette/repositories/FEM/getdp/Common/Message.cpp:54:
/usr/include/python3.4m/pythonrun.h:25:6: note: declared here
PyAPI_FUNC(void) Py_SetProgramName(wchar_t *);
^
CMakeFiles/getdp.dir/build.make:182: recipe for target 'CMakeFiles/getdp.dir/Common/Message.cpp.o' failed
make[2]: *** [CMakeFiles/getdp.dir/Common/Message.cpp.o] Error 1
CMakeFiles/Makefile2:1059: recipe for target 'CMakeFiles/getdp.dir/all' failedhttps://gitlab.onelab.info/getdp/getdp/-/issues/33MovingBand2D doesn't support higher order Element2017-08-05T11:59:38ZChristophe GeuzaineMovingBand2D doesn't support higher order ElementDear sir,
once i set Mesh.ElementOrder =2, following Error Message appears:
MovingBand2D_contour must contain only line elements.Dear sir,
once i set Mesh.ElementOrder =2, following Error Message appears:
MovingBand2D_contour must contain only line elements.https://gitlab.onelab.info/getdp/getdp/-/issues/32python 3.5 compatibility2017-08-05T11:59:50ZChristophe Geuzainepython 3.5 compatibility```
[ 32%] Building CXX object CMakeFiles/shared.dir/Legacy/BF_Node_2.cpp.o
/usr/lib64/ccache/c++ -Dshared_EXPORTS -I/home/brain/rpmbuild/BUILD/getdp-2.7.0-source/Common -I/home/brain/rpmbuild/BUILD/getdp-2.7.0-source/Interface -I/home...```
[ 32%] Building CXX object CMakeFiles/shared.dir/Legacy/BF_Node_2.cpp.o
/usr/lib64/ccache/c++ -Dshared_EXPORTS -I/home/brain/rpmbuild/BUILD/getdp-2.7.0-source/Common -I/home/brain/rpmbuild/BUILD/getdp-2.7.0-source/Interface -I/home/brain/rpmbuild/BUILD/getdp-2.7.0-source/Legacy -I/home/brain/rpmbuild/BUILD/getdp-2.7.0-source/Numeric -I/usr/include/python3.5m -I/home/brain/rpmbuild/BUILD/getdp-2.7.0-source/build/Common -O2 -g -pipe -Wall -Werror=format-security -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector-strong --param=ssp-buffer-size=4 -grecord-gcc-switches -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1 -m64 -mtune=generic -fopenmp -O2 -g -DNDEBUG -fPIC -Wall -o CMakeFiles/shared.dir/Legacy/BF_Node_2.cpp.o -c /home/brain/rpmbuild/BUILD/getdp-2.7.0-source/Legacy/BF_Node_2.cpp
In file included from /usr/include/python3.5m/pyconfig.h:6:0,
from /usr/include/python3.5m/Python.h:8,
from /home/brain/rpmbuild/BUILD/getdp-2.7.0-source/Common/Message.cpp:65:
/usr/include/python3.5m/pyconfig-64.h:207:0: warning: "HAVE_DLOPEN" redefined
#define HAVE_DLOPEN 1
^
In file included from /usr/include/gmsh/GmshMessage.h:14:0,
from /usr/include/gmsh/Gmsh.h:10,
from /home/brain/rpmbuild/BUILD/getdp-2.7.0-source/Common/Message.cpp:50:
/usr/include/gmsh/GmshConfig.h:19:0: note: this is the location of the previous definition
#define HAVE_DLOPEN
^
/home/brain/rpmbuild/BUILD/getdp-2.7.0-source/Common/Message.cpp: In static member function 'static void Message::Initialize(int, char**)':
/home/brain/rpmbuild/BUILD/getdp-2.7.0-source/Common/Message.cpp:134:28: error: cannot convert 'char*' to 'wchar_t*' for argument '1' to 'void Py_SetProgramName(wchar_t*)'
Py_SetProgramName(argv[0]);
^
/home/brain/rpmbuild/BUILD/getdp-2.7.0-source/Common/Message.cpp:136:27: error: cannot convert 'char**' to 'wchar_t**' for argument '2' to 'void PySys_SetArgv(int, wchar_t**)'
PySys_SetArgv(argc, argv);
^
CMakeFiles/shared.dir/build.make:161: recipe for target 'CMakeFiles/shared.dir/Common/Message.cpp.o' failed
```https://gitlab.onelab.info/getdp/getdp/-/issues/31Doesn't compile with PETSc master2017-03-27T09:20:48ZChristophe GeuzaineDoesn't compile with PETSc masterGetDP 2.7.0 doesn't compile together with the newest PETSc master, because there has been a change with the PetscOptions API in PETSc:
https://bitbucket.org/petsc/petsc/commits/13cf9f3a154671946d9fc1df5182efb9d0780d74
On that page it i...GetDP 2.7.0 doesn't compile together with the newest PETSc master, because there has been a change with the PetscOptions API in PETSc:
https://bitbucket.org/petsc/petsc/commits/13cf9f3a154671946d9fc1df5182efb9d0780d74
On that page it is possible to see how e.g. all the PETSc examples and such were changed due to the change in PetscOptions. I think that for most of the errors it amounts to just inserting "PETSC_NULL" as an additional argument.
The compile-time error message is:
[ 60%] Building CXX object CMakeFiles/getdp.dir/Legacy/LinAlg.cpp.o
/home/tmac/src/upstream/getdp-2.7.0-source/Legacy/GeoEntity.cpp: In function ‘void Geo_CreateEdgesOfElement(Geo_Element*)’:
/home/tmac/src/upstream/getdp-2.7.0-source/Legacy/GeoEntity.cpp:55:39: warning: ‘Nbr_Entities2’ may be used uninitialized in this function [-Wmaybe-uninitialized]
CurrentGeoData->EdgesXNodes) ;
^
/home/tmac/src/upstream/getdp-2.7.0-source/Legacy/GeoEntity.cpp: In function ‘void Geo_CreateFacetsOfElement(Geo_Element*)’:
/home/tmac/src/upstream/getdp-2.7.0-source/Legacy/GeoEntity.cpp:96:40: warning: ‘Nbr_Entities2’ may be used uninitialized in this function [-Wmaybe-uninitialized]
CurrentGeoData->FacetsXEdges) ;
^
[ 61%] Building CXX object CMakeFiles/getdp.dir/Legacy/LinAlg_PETSC.cpp.o
[ 62%] Building CXX object CMakeFiles/getdp.dir/Legacy/LinAlg_SPARSKIT.cpp.o
[ 63%] Building CXX object CMakeFiles/getdp.dir/Legacy/SolvingAnalyse.cpp.o
[ 64%] Building CXX object CMakeFiles/getdp.dir/Legacy/SolvingOperations.cpp.o
/home/tmac/src/upstream/getdp-2.7.0-source/Legacy/LinAlg_PETSC.cpp: In function ‘void LinAlg_InitializeSolver(int*, char***)’:
/home/tmac/src/upstream/getdp-2.7.0-source/Legacy/LinAlg_PETSC.cpp:100:63: error: cannot convert ‘char*’ to ‘PetscOptions {aka _n_PetscOptions*}’ for argument ‘2’ to ‘PetscErrorCode PetscOptionsInsertFile(MPI_Comm, PetscOptions, const char*, PetscBool)’
PetscOptionsInsertFile(MyComm, (*argv)[i+1], PETSC_FALSE);
^
/home/tmac/src/upstream/getdp-2.7.0-source/Legacy/LinAlg_PETSC.cpp: In function ‘void LinAlg_CreateMatrix(gMatrix*, gSolver*, int, int)’:
/home/tmac/src/upstream/getdp-2.7.0-source/Legacy/LinAlg_PETSC.cpp:204:68: error: cannot convert ‘PetscInt* {aka int*}’ to ‘const char*’ for argument ‘3’ to ‘PetscErrorCode PetscOptionsGetInt(PetscOptions, const char*, const char*, PetscInt*, PetscBool*)’
PetscOptionsGetInt(PETSC_NULL, "-petsc_prealloc", &prealloc, &set);
^
/home/tmac/src/upstream/getdp-2.7.0-source/Legacy/LinAlg_PETSC.cpp:205:78: error: cannot convert ‘PetscInt* {aka int*}’ to ‘const char*’ for argument ‘3’ to ‘PetscErrorCode PetscOptionsGetInt(PetscOptions, const char*, const char*, PetscInt*, PetscBool*)’
PetscOptionsGetInt(PETSC_NULL, "-petsc_prealloc_full", &prealloc_full, &set);
^
/home/tmac/src/upstream/getdp-2.7.0-source/Legacy/LinAlg_PETSC.cpp: In function ‘void _solve(gMatrix*, gVector*, gSolver*, gVector*, int, int)’:
/home/tmac/src/upstream/getdp-2.7.0-source/Legacy/LinAlg_PETSC.cpp:1280:76: error: cannot convert ‘const char*’ to ‘PetscOptions {aka _n_PetscOptions*}’ for argument ‘1’ to ‘PetscErrorCode PetscOptionsSetValue(PetscOptions, const char*, const char*)’
_try(PetscOptionsSetValue("-pc_factor_nonzeros_along_diagonal", "1e-12"));
^
/home/tmac/src/upstream/getdp-2.7.0-source/Legacy/LinAlg_PETSC.cpp: In function ‘void LinAlg_SetGlobalSolverOptions(const string&)’:
/home/tmac/src/upstream/getdp-2.7.0-source/Legacy/LinAlg_PETSC.cpp:1360:44: error: cannot convert ‘const char*’ to ‘PetscOptions {aka _n_PetscOptions*}’ for argument ‘1’ to ‘PetscErrorCode PetscOptionsInsertString(PetscOptions, const char*)’
_try(PetscOptionsInsertString(opt.c_str()));
^
/home/tmac/src/upstream/getdp-2.7.0-source/Legacy/LinAlg_PETSC.cpp: In function ‘void _solveNL(gMatrix*, gVector*, gMatrix*, gVector*, gSolver*, gVector*, int, int)’:
/home/tmac/src/upstream/getdp-2.7.0-source/Legacy/LinAlg_PETSC.cpp:1466:69: error: cannot convert ‘PetscBool*’ to ‘const char*’ for argument ‘3’ to ‘PetscErrorCode PetscOptionsGetBool(PetscOptions, const char*, const char*, PetscBool*, PetscBool*)’
PetscOptionsGetTruth(PETSC_NULL, "-fd_jacobian", &fd_jacobian, 0);
^
/home/tmac/src/upstream/getdp-2.7.0-source/Legacy/LinAlg_PETSC.cpp:1467:61: error: cannot convert ‘PetscBool*’ to ‘const char*’ for argument ‘3’ to ‘PetscErrorCode PetscOptionsGetBool(PetscOptions, const char*, const char*, PetscBool*, PetscBool*)’
PetscOptionsGetTruth(PETSC_NULL, "-snes_fd", &snes_fd, 0);
^
CMakeFiles/getdp.dir/build.make:1598: recipe for target 'CMakeFiles/getdp.dir/Legacy/LinAlg_PETSC.cpp.o' failedhttps://gitlab.onelab.info/getdp/getdp/-/issues/29error magnetic field direction in H-formulation2022-07-07T15:42:22ZChristophe Geuzaineerror magnetic field direction in H-formulationDear Sir,
I found a strange behavior of GetDP. I ran the same case based on h-formulation with 2 different meshes, one from Salome platform and the other from Gmsh. Both mesh generates a different error in the computed magnetic field di...Dear Sir,
I found a strange behavior of GetDP. I ran the same case based on h-formulation with 2 different meshes, one from Salome platform and the other from Gmsh. Both mesh generates a different error in the computed magnetic field direction around the line where boundary conditions are applied. Here are the files to reproduce the issue.
Best,