Xfiles issueshttp://git.gem.ec-nantes.fr/eXlibris/Xfiles/-/issues2018-05-18T07:23:38Zhttp://git.gem.ec-nantes.fr/eXlibris/Xfiles/-/issues/21Removing xFitToVertexKeyManager in xFitToVertices class2018-05-18T07:23:38ZAlexis SALZMANRemoving xFitToVertexKeyManager in xFitToVertices classFollowing commit 5d34c604 the key manager xFitToVertexKeyManager should be remove and usage replace by new keyManagerSendOrReceiveFollowing commit 5d34c604 the key manager xFitToVertexKeyManager should be remove and usage replace by new keyManagerSendOrReceivehttp://git.gem.ec-nantes.fr/eXlibris/Xfiles/-/issues/22Major memory leak in Lshaped testcase2018-05-18T07:23:37ZOlivier PIERARDMajor memory leak in Lshaped testcaseI ran a vasic valgrind analysis on the Lshaped testcase (with max_number_iterations=2). Numerous issues are reported (see [memcheck.log](/uploads/8c535eda899414c53c3887183f1ac17f/memcheck.log))
Most critical ones seem to be the 2 'defin...I ran a vasic valgrind analysis on the Lshaped testcase (with max_number_iterations=2). Numerous issues are reported (see [memcheck.log](/uploads/8c535eda899414c53c3887183f1ac17f/memcheck.log))
Most critical ones seem to be the 2 'definitely lost' (around line 1624), e.g.:
`==30008== 5,787,792 bytes in 1 blocks are definitely lost in loss record 83 of 83
==30008== at 0x4A079EE: malloc (vg_replace_malloc.c:270)
==30008== by 0x8C5A027: taucs_dccs_create (in /home/cenaero/pierard/TLS/lib/x86_64_Linux_Debug/libTaucs.so)
==30008== by 0x8C5BED7: taucs_dccs_permute_symmetrically (in /home/cenaero/pierard/TLS/lib/x86_64_Linux_Debug/libTaucs.so)
==30008== by 0x8C447FD: lalg::xLinearSystemSolverTaucsBase::permuteSymmetrically(void*, int*, int*) (xLinearSystemSolverTaucs.cc:269)
==30008== by 0x547DB29: lalg::xLinearSystemSolverTaucs<double>::reorder(int*, int*, int**) (xLinearSystemSolverTaucs_imp.h:171)
==30008== by 0x547C883: lalg::xLinearSystemSolverTaucs<double>::symb() (xLinearSystemSolverTaucs_imp.h:277)
==30008== by 0x547E139: void lalg::xLinearSystemSolverTaucs<double>::connectMatrix<lalg::xGenericSparseMatrix<double, xTraitMatrixDefinitePositive, xTraitMatrixLowerSym, xTraitMatrixSparceCSC, xTraitMatrixCindex> >(lalg::xGenericSparseMatrix<double, xTraitMatrixDefinitePositive, xTraitMatrixLowerSym, xTraitMatrixSparceCSC, xTraitMatrixCindex>&) (xLinearSystemSolverTaucs_imp.h:229)
==30008== by 0x54D9BC5: void DamageGrowthMechanics::resolveMechanicalProblem_V1<lalg::xLinearSystemSolverTaucs<double>, lalg::xGenericSparseMatrix<double, xTraitMatrixDefinitePositive, xTraitMatrixLowerSym, xTraitMatrixSparceCSC, xTraitMatrixCindex>, xEvalIsotropDamagedHooke2DPlaneStrain, xEvalIsotropDamagedStress2DPlaneStrain, xEvalIsotropElasticHook2DPlaneStrain, xtls::xEvalTLSDamageFunction>(lalg::xLinearSystemSolverTaucs<double>&, xEvalIsotropDamagedHooke2DPlaneStrain&, xEvalIsotropDamagedStress2DPlaneStrain&, xEvalIsotropElasticHook2DPlaneStrain&, xfem::xEvalGradField<xfem::xSymmetrize, xfem::xField>&, xtls::xEvalTLSDamageFunction&, int) (DamageGrowthMechanics.ipp:277)
==30008== by 0x54D7426: void DamageGrowthMechanics::compute_elastic<lalg::xLinearSystemSolverTaucs<double>, lalg::xGenericSparseMatrix<double, xTraitMatrixDefinitePositive, xTraitMatrixLowerSym, xTraitMatrixSparceCSC, xTraitMatrixCindex> >(lalg::xLinearSystemSolverTaucs<double>&) (DamageGrowthMechanics.ipp:84)
==30008== by 0x54CC6B3: DamageTestMesh::run() (DamageTest.cc:608)
==30008== by 0x42F04F: DamageTestContext<DamageTestMesh>::run_test() (DamageTest.h:117)
==30008== by 0x42D8F8: main (main.cc:100)`
A very first analysis shows that:
* Matrices are allocated within TAUCS, but should be freed outside taucs
* xLinearSystemSolverTaucsBase::PAPT stores the result of the call to taucs_dccs_permute_symmetrically
* This should be freed in destructor of xLinearSystemSolverTaucsBase, as it is implemented (with free instead of taucs_free - but that's another debate...)
* BUT call to to destructor of xLinearSystemSolverTaucsBase or xLinearSystemSolverTaucs is never done !!!
* I guess this should be done in DamageTest, but I'm a little bit lost with all these templates and typedefs...Olivier PIERARDOlivier PIERARDhttp://git.gem.ec-nantes.fr/eXlibris/Xfiles/-/issues/23xChrono/xDeltatime migration2019-01-29T12:40:55ZAlexis SALZMANxChrono/xDeltatime migrationxDeltaTime (in xChrono) is use here and there to obtain somme monitoring assesment on code performance.
Thus it should not be anylonger in Xfem. It's natural place is in eXlibriss_tools.
For example in solverinterface test some test ...xDeltaTime (in xChrono) is use here and there to obtain somme monitoring assesment on code performance.
Thus it should not be anylonger in Xfem. It's natural place is in eXlibriss_tools.
For example in solverinterface test some test do not need any xfem suff but are linking to this library just because xDeltaTime is locate there.
xChrono.h/cc to migrate in eXlibriss_tools with namespace change.http://git.gem.ec-nantes.fr/eXlibris/Xfiles/-/issues/24treillis with _DEBUG_ macro enable2018-05-18T07:23:37ZAlexis SALZMANtreillis with _DEBUG_ macro enableIn Treillis, in mVector and mPoint (at least) if _DEBUG_ macro is enable code does not compile.
Bug correction needed.
For now don't turn on USE_DEBUG_FLAG option in cmake (which is not only a printing stuff as message suggest)In Treillis, in mVector and mPoint (at least) if _DEBUG_ macro is enable code does not compile.
Bug correction needed.
For now don't turn on USE_DEBUG_FLAG option in cmake (which is not only a printing stuff as message suggest)http://git.gem.ec-nantes.fr/eXlibris/Xfiles/-/issues/25Keeping boost iostream isolated2018-05-18T07:23:37ZAlexis SALZMANKeeping boost iostream isolatedDiscussion around cmake point out that with some effort (little ?) we may hide boost iostram (used for gzip output) from Xfem interface.
That may also be treated as an extra interface in xInterface.
In one hand treating it in Xfem corr...Discussion around cmake point out that with some effort (little ?) we may hide boost iostram (used for gzip output) from Xfem interface.
That may also be treated as an extra interface in xInterface.
In one hand treating it in Xfem correspond to where it is used for now, in the other adding it as an Xinterface gives a way to use it everywhere.http://git.gem.ec-nantes.fr/eXlibris/Xfiles/-/issues/26CGAL interface not compatible with CGAL 4.112018-05-18T07:23:37ZGrégory LEGRAINCGAL interface not compatible with CGAL 4.11AABBPrimitiveExtension.h seems incompatible with new CGAL implementation
`pas de concordance pour l'appel à « (CGAL::Cartesian_base_no_ref_count<double, CGAL::Simple_cartesian<double> >::Construct_projected_point_3 {aka CGAL::CartesianK...AABBPrimitiveExtension.h seems incompatible with new CGAL implementation
`pas de concordance pour l'appel à « (CGAL::Cartesian_base_no_ref_count<double, CGAL::Simple_cartesian<double> >::Construct_projected_point_3 {aka CGAL::CartesianKernelFunctors::Construct_projected_point_3<CGAL::Simple_cartesian<double> >}) (CGAL::internal::Primitive_helper<CGAL::AABB_traits<CGAL::Simple_cartesian<double>, CGAL::AABB_point_primitive<CGAL::Simple_cartesian<double>, __gnu_cxx::__normal_iterator<CGAL::Point_3<CGAL::Simple_cartesian<double> >*, std::vector<CGAL::Point_3<CGAL::Simple_cartesian<double> > > > > >, false>::Datum_type, const Point&) »
Point closest_point = geom_traits.construct_projected_point_3_object()(
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^
internal::Primitive_helper<AT>::get_datum(pr,m_traits), p);
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
`
It complains that the point to point case (whici is de fined in AABBPrimitiveExtension. if I am correct) is missing.
Good to know when we will update the version which is available on the platformGrégory LEGRAINGrégory LEGRAINhttp://git.gem.ec-nantes.fr/eXlibris/Xfiles/-/issues/27test_ndiff.sh2020-04-01T13:07:31ZGilles MARCKMANNtest_ndiff.shXfiles/Util/cmakeUtil/test_ndiff.sh must be rewritten to be portable. The way to find the tool 'ndiff' is durty and hard coded for titan. It should not work on other platform as it is. The 'find 'command must be improved and a warning me...Xfiles/Util/cmakeUtil/test_ndiff.sh must be rewritten to be portable. The way to find the tool 'ndiff' is durty and hard coded for titan. It should not work on other platform as it is. The 'find 'command must be improved and a warning message should be written if 'ndiff' is not found. This bug may indicate that some ndiff_test passed while the test is not done in fact.Gilles MARCKMANNGilles MARCKMANNhttp://git.gem.ec-nantes.fr/eXlibris/Xfiles/-/issues/28Find Blas22018-05-18T07:23:37ZGilles MARCKMANNFind Blas2With the use of function FindNoHeaderLibraries(), the find process give a FATAL ERROR (Blas REQUIED) if no BLAS_VENDOR is given and when default vendors give no results. So, default libblas.so can not be found.
May be the corrective is...With the use of function FindNoHeaderLibraries(), the find process give a FATAL ERROR (Blas REQUIED) if no BLAS_VENDOR is given and when default vendors give no results. So, default libblas.so can not be found.
May be the corrective is useless if we try to use the default findBlas of cmake.Gilles MARCKMANNGilles MARCKMANNhttp://git.gem.ec-nantes.fr/eXlibris/Xfiles/-/issues/29FindMETIS2018-05-18T07:23:37ZGilles MARCKMANNFindMETISFindMETIS still have a reference to FindMPI4 which has been removed. MPI should be defined in the main CMakeLists.txt.FindMETIS still have a reference to FindMPI4 which has been removed. MPI should be defined in the main CMakeLists.txt.Gilles MARCKMANNGilles MARCKMANNhttp://git.gem.ec-nantes.fr/eXlibris/Xfiles/-/issues/30CMakeLists.txt of TTK2018-05-18T07:23:37ZGilles MARCKMANNCMakeLists.txt of TTKThe CMakeLists.txt of TTK must be updated to take into account modifications introduced by new_cmake.The CMakeLists.txt of TTK must be updated to take into account modifications introduced by new_cmake.Gilles MARCKMANNGilles MARCKMANNhttp://git.gem.ec-nantes.fr/eXlibris/Xfiles/-/issues/31xDoubleManager/xValManger distributed concept2018-05-18T07:23:37ZAlexis SALZMANxDoubleManager/xValManger distributed conceptHas mentioned in commit d1f3ed5561e76ed4c2b318220128f5af68a19cc0 the partition manager introduced in double manager may be generalized to xValManager.
Make a template version of xValKeyDataManager would be certainly a must. But how ...Has mentioned in commit d1f3ed5561e76ed4c2b318220128f5af68a19cc0 the partition manager introduced in double manager may be generalized to xValManager.
Make a template version of xValKeyDataManager would be certainly a must. But how genPartitionManager would be done in xValManager. In a rather specific way as it is in xDoubleManager where xMesh::partman_t is expected. Or will it be generic wit a template type for mesh partition manager. In this case maybe it will be a bit touchy to deal with a generic partman to create xValManager partion manager (for example at no stage we want to introduce AOMD struff) ...
First check that this concept is good and then pass some time on this generalization ...Parallel integrationhttp://git.gem.ec-nantes.fr/eXlibris/Xfiles/-/issues/32kill xPhysSurf ?2018-09-07T11:44:16ZGrégory LEGRAINkill xPhysSurf ?the (old) xPhysSurf is (in practice) deprecated in favor of xPhysSurfByTagging.
We should move it to attic.
Though, it necessitates some fixes in some testcases and classes...the (old) xPhysSurf is (in practice) deprecated in favor of xPhysSurfByTagging.
We should move it to attic.
Though, it necessitates some fixes in some testcases and classes...Grégory LEGRAINGrégory LEGRAINhttp://git.gem.ec-nantes.fr/eXlibris/Xfiles/-/issues/33xDistVector Test2019-12-05T13:16:16ZAlexis SALZMANxDistVector TestSome new method are not tested in the atomic test: to be done
Some new friend functions may be optimizedSome new method are not tested in the atomic test: to be done
Some new friend functions may be optimizedParallel integrationhttp://git.gem.ec-nantes.fr/eXlibris/Xfiles/-/issues/34geom: distance computation in distributed context with ann or brute force2018-04-25T07:45:54ZAlexis SALZMANgeom: distance computation in distributed context with ann or brute forceCommit 3b05590 introduce a way to compute distance to a surface in distributed context with CGAL. The same should be done with ANN at least and maybe for brute force. Test case functional_xDistanceNearestPointDist should be modified in c...Commit 3b05590 introduce a way to compute distance to a surface in distributed context with CGAL. The same should be done with ANN at least and maybe for brute force. Test case functional_xDistanceNearestPointDist should be modified in consequence.Parallel integrationhttp://git.gem.ec-nantes.fr/eXlibris/Xfiles/-/issues/35xDomainStringManager::getDomainId("zone")2020-05-15T13:19:38ZGilles MARCKMANNxDomainStringManager::getDomainId("zone")Potential Bug:
in xGeomElem, xDomainStringManager::getDomainId("zone") is used to manage and to get a tag (int) corresponding to the string "zone". This tag is used, after, in Trellis through the function getData(tag) which gives a mAtta...Potential Bug:
in xGeomElem, xDomainStringManager::getDomainId("zone") is used to manage and to get a tag (int) corresponding to the string "zone". This tag is used, after, in Trellis through the function getData(tag) which gives a mAttachableData (xDomainAttachableData).
The use of xDomainStringManager to manage the link between tags and strings may be in conflict with attachableDataIds of AOMD_Util::newMeshDataId which does the same thing: two different maps are used to manage tags of mAttachableData.
In fact, the bug never appears because the tag in xDomainStringManager:getDomainId("zone") is currently 0. This tag corresponds to a not-used tag of attachableDataIds. Note that the first tags declared in attachableDataIds of Trellis through AOMD_Util::newMeshDataId are :
* 0 <-> "_parametric"
* 1 <-> "_parent"
* 2 <-> "_fmod"
* 3 <-> "_emod"
* 4 <-> "_id"
* 5 <-> "_dn"
* 6 <-> "_size"
* 7 <-> "_ATT1"
* 8 <-> "_ATT2"
* 9 <-> "_ATT3"
* 10 <-> "_weight"
* 11 <-> "_mirror"
These tags are not used in Xfem, then 12 different domain names might be managed with xDomainStringManager before the bug appears.
This bug should nore affect applications which don't use MaterialManager (typically DamageBand which use xEval to manage materials).http://git.gem.ec-nantes.fr/eXlibris/Xfiles/-/issues/36Remove seq branch2019-12-11T15:54:57ZGrégory LEGRAINRemove seq branchHi,
the sequential branch was removed from DamageBand, if there is no opposition for this (**thumb-up** / **thumb-down**), I plan to remove the branch in Xfiles.Hi,
the sequential branch was removed from DamageBand, if there is no opposition for this (**thumb-up** / **thumb-down**), I plan to remove the branch in Xfiles.Grégory LEGRAINGrégory LEGRAINhttp://git.gem.ec-nantes.fr/eXlibris/Xfiles/-/issues/37Cleanup xPhysSurf, Export ensight2018-09-07T11:45:02ZGrégory LEGRAINCleanup xPhysSurf, Export ensightxPhysSurf is now deprecated for a while, as in practice, everybody (?) use xPhysSurfByTagging. I would make sense to move it to attic.
Maybe we could also clean some tags from xMesh that were only used by xPhysSurf
xExportEnsight has n...xPhysSurf is now deprecated for a while, as in practice, everybody (?) use xPhysSurfByTagging. I would make sense to move it to attic.
Maybe we could also clean some tags from xMesh that were only used by xPhysSurf
xExportEnsight has not been used for ages... we should move it to atticGrégory LEGRAINGrégory LEGRAINhttp://git.gem.ec-nantes.fr/eXlibris/Xfiles/-/issues/38Drop xTable ?2018-07-09T12:59:15ZGrégory LEGRAINDrop xTable ?xTable class (xTable.h) does not seem to be used in practical applications although it is used in xTensors.h.
Maybe we should drop it ?xTable class (xTable.h) does not seem to be used in practical applications although it is used in xTensors.h.
Maybe we should drop it ?http://git.gem.ec-nantes.fr/eXlibris/Xfiles/-/issues/39Potential error in setUVWForVertex method of xGeomElem2020-02-07T12:53:31ZAlexis SALZMANPotential error in setUVWForVertex method of xGeomElemIn setUVWForVertex method of xGeomElem we have:
```cpp
blabla
switch(e->getType()) {
case mEntity::TET:
case mEntity::TRI:
switch(inod){
case 0:
Upos = 0.0;Vpos = 0.0;Wpos = 0.0;break;
case 1:
Upos = 1.0;Vpos = 0.0;...In setUVWForVertex method of xGeomElem we have:
```cpp
blabla
switch(e->getType()) {
case mEntity::TET:
case mEntity::TRI:
switch(inod){
case 0:
Upos = 0.0;Vpos = 0.0;Wpos = 0.0;break;
case 1:
Upos = 1.0;Vpos = 0.0;Wpos = 0.0;break;
case 2:
Upos = 0.0;Vpos = 1.0;Wpos = 0.0;break;
case 3:
Upos = 0.0;Vpos = 0.0;Wpos = 1.0;break;
default: assert(0); throw; break;
}
break;
blabla
```
If by error some one use with a TRI a inod=3 it won't stop but continue. It will be just as if inod was 0 as in 2D "w" is not used. I do not know if there is some trick here but a more secure way to do this could be:
```cpp
blabla
switch(e->getType()) {
case mEntity::TET:
if (inod==3)
{
Upos = 0.0;Vpos = 0.0;Wpos = 1.0;
break;
}
case mEntity::TRI:
switch(inod){
case 0:
Upos = 0.0;Vpos = 0.0;Wpos = 0.0;break;
case 1:
Upos = 1.0;Vpos = 0.0;Wpos = 0.0;break;
case 2:
Upos = 0.0;Vpos = 1.0;Wpos = 0.0;break;
default: assert(0); throw; break;
}
break;
blabla
```
to be done if neededhttp://git.gem.ec-nantes.fr/eXlibris/Xfiles/-/issues/40Compiling with clang++2018-07-27T13:38:51ZAlexis SALZMANCompiling with clang++Now that clang++ 3.9.0 is installed on titan (/opt/llvm-3.9.0/bin/clang++) it is possible to do some compilation test with this compiler.
Modify your LocalConfig.cmake to use clang++ and rebuild everything.
A full Ctest on Xtest have ...Now that clang++ 3.9.0 is installed on titan (/opt/llvm-3.9.0/bin/clang++) it is possible to do some compilation test with this compiler.
Modify your LocalConfig.cmake to use clang++ and rebuild everything.
A full Ctest on Xtest have been done and give almost same results as g++.
First remark regarding CMAKEFILE, clang give the following message:
```
clang-3.9: warning: optimization flag '-frounding-math' is not supported
```
This mean that we have put this flag every-where or almost but it is compiler dependent !!! If we pass to modern cmake we will be able to use `$<COMPILER_ID:GNU>,$<COMPILER_ID:Clang>` stuff to appropriately use -frounding-math or not
When using MUMPS or PASTIX clang need to have an explicit -lpthread. Those libraries where compiled with g++. Maybe with this compiler the -lpthread is automatically set. Like above if we pass to modern cmake we may encapsulate MUMPS or PASTIX target dependencies and fix cleanly those dependencies with `$<COMPILER_ID:GNU>,$<COMPILER_ID:Clang>` stuff.
What is more interesting is that clang have note exactly the same behavior as g++ regarding errors and warnings. For example with -Wconversion clang check implicit conversion miss formed (this is not possible with g++):
```cpp
#include <iostream>
#include <vector>
using namespace std;
int main ()
{
size_t h=1234567891123456Lu;
int hi;
short hs;
hi=h;
hs=h;
cout<<"h "<<h<<endl;
cout<<"hi "<<hi<<endl;
cout<<"hs "<<hs<<endl;
hi=-1;
h=hi;
cout<<"h "<<h<<endl;
cout<<"hi "<<hi<<endl;
vector<int> v(3,1);
int l=v.size();
v[l-1]=4;
for (auto val : v)
cout<<val<<endl;
return 0;
}
```
With this code clang produce following messages:
```
clang++ -std=c++11 -Wconversion main.cc
main.cc:13:8: warning: implicit conversion loses integer precision: 'size_t' (aka 'unsigned long') to 'short' [-Wconversion]
hs=h;
~^
main.cc:20:7: warning: implicit conversion changes signedness: 'int' to 'size_t' (aka 'unsigned long') [-Wsign-conversion]
h=hi;
~^~
main.cc:27:8: warning: implicit conversion changes signedness: 'int' to 'size_type' (aka 'unsigned long') [-Wsign-conversion]
v[l-1]=4;
~ ~^~
main.cc:12:8: warning: implicit conversion loses integer precision: 'size_t' (aka 'unsigned long') to 'int' [-Wshorten-64-to-32]
hi=h;
~^
main.cc:26:13: warning: implicit conversion loses integer precision: 'size_type' (aka 'unsigned long') to 'int' [-Wshorten-64-to-32]
int l=v.size();
~ ~~^~~~~~
5 warnings generated.
```
With eXlibris many messages of those types are issued ... For many it is only a matter of using `int` instead of `size_t` when dealing with STL container. But unfortunately this is masking more interesting ones that may be mistakes or error !