GitLab

clean/refactor tests

git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/scalfmm/scalfmm/trunk@85 2616d619-271b-44dc-8df4-d4a8f33a7222

git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/scalfmm/scalfmm/trunk@84 2616d619-271b-44dc-8df4-d4a8f33a7222

+
Remove a bug in the mpi version.

git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/scalfmm/scalfmm/trunk@83 2616d619-271b-44dc-8df4-d4a8f33a7222

git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/scalfmm/scalfmm/trunk@82 2616d619-271b-44dc-8df4-d4a8f33a7222

Now it works well the results are equal when comparing mpi & no mpi vers.

git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/scalfmm/scalfmm/trunk@81 2616d619-271b-44dc-8df4-d4a8f33a7222

because the width was not correct.
Also change the testMortonIndex.cpp to make
the octree default height value to 6

git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/scalfmm/scalfmm/trunk@80 2616d619-271b-44dc-8df4-d4a8f33a7222

- to print box properties from files to user
- to because of a mistake in the file reading (width was * 2!
so the program used 4 when width was 2)

git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/scalfmm/scalfmm/trunk@79 2616d619-271b-44dc-8df4-d4a8f33a7222

Valgrind says there is not memory lack (excepted because of openmp)

git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/scalfmm/scalfmm/trunk@78 2616d619-271b-44dc-8df4-d4a8f33a7222

But here is a new tools :
testMortonIndex.cpp
and
FHLoader.hpp
and
Data/testH.txt

This is loading a file and printing all the morton index of the boxes containing particles.

git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/scalfmm/scalfmm/trunk@77 2616d619-271b-44dc-8df4-d4a8f33a7222

git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/scalfmm/scalfmm/trunk@76 2616d619-271b-44dc-8df4-d4a8f33a7222

But the program failled to dealloc a memory in hybrid mode.
Because of MPI an array is used, and when it is used with openmp a dealloc problem exists.
But this problem does not exist on platfrim...
Need to check again.

git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/scalfmm/scalfmm/trunk@75 2616d619-271b-44dc-8df4-d4a8f33a7222

Improve a little the M2M mpi-openmp version.

git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/scalfmm/scalfmm/trunk@74 2616d619-271b-44dc-8df4-d4a8f33a7222

git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/scalfmm/scalfmm/trunk@73 2616d619-271b-44dc-8df4-d4a8f33a7222

git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/scalfmm/scalfmm/trunk@72 2616d619-271b-44dc-8df4-d4a8f33a7222

Still a problem with fmb & mpi.

git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/scalfmm/scalfmm/trunk@71 2616d619-271b-44dc-8df4-d4a8f33a7222

Need to tcheck again the restult.

git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/scalfmm/scalfmm/trunk@70 2616d619-271b-44dc-8df4-d4a8f33a7222

To enable buffering there is an abstractClass Sendable.
So we are not using the mpi buffering system but our own model
with an abstraction to cell data class that need to be sent.

git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/scalfmm/scalfmm/trunk@69 2616d619-271b-44dc-8df4-d4a8f33a7222

git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/scalfmm/scalfmm/trunk@68 2616d619-271b-44dc-8df4-d4a8f33a7222

Remove the previous AbstractApplication >> MpiApplication.
Change it into a single file with define system.

git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/scalfmm/scalfmm/trunk@67 2616d619-271b-44dc-8df4-d4a8f33a7222

+
4 test files (without explaination to use them)

git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/scalfmm/scalfmm/trunk@66 2616d619-271b-44dc-8df4-d4a8f33a7222

(adding comments etc.)

git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/scalfmm/scalfmm/trunk@65 2616d619-271b-44dc-8df4-d4a8f33a7222

But need to be validate for a the FMB computation.

git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/scalfmm/scalfmm/trunk@64 2616d619-271b-44dc-8df4-d4a8f33a7222

git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/scalfmm/scalfmm/trunk@63 2616d619-271b-44dc-8df4-d4a8f33a7222

git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/scalfmm/scalfmm/trunk@62 2616d619-271b-44dc-8df4-d4a8f33a7222

Here it the new src dir with some minor changes in CMake files.

git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/scalfmm/scalfmm/trunk@61 2616d619-271b-44dc-8df4-d4a8f33a7222

Change Macro FUSE into SCALFMM_USE

git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/scalfmm/scalfmm/trunk@60 2616d619-271b-44dc-8df4-d4a8f33a7222

to get values like FMB.

Also change FMA to Fma to be consistent.

git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/scalfmm/scalfmm/trunk@59 2616d619-271b-44dc-8df4-d4a8f33a7222

This dir has to be used to build with cmake.

This dir will still empty with git (using .gitignore)
I am not sure about the behaviour under svn (I will test with an update)

git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/scalfmm/scalfmm/trunk@58 2616d619-271b-44dc-8df4-d4a8f33a7222

git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/scalfmm/scalfmm/trunk@57 2616d619-271b-44dc-8df4-d4a8f33a7222

Change the Cmake :
× add comments
× add mpi into cmake (not used yet)
× change variables names

git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/scalfmm/scalfmm/trunk@56 2616d619-271b-44dc-8df4-d4a8f33a7222

But we can change it in a next commit if needed.

git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/scalfmm/scalfmm/trunk@55 2616d619-271b-44dc-8df4-d4a8f33a7222

Also comment a define in globale include.

git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/scalfmm/scalfmm/trunk@54 2616d619-271b-44dc-8df4-d4a8f33a7222

git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/scalfmm/scalfmm/trunk@53 2616d619-271b-44dc-8df4-d4a8f33a7222

and have created a basic mpi version just for fun.

git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/scalfmm/scalfmm/trunk@52 2616d619-271b-44dc-8df4-d4a8f33a7222

git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/scalfmm/scalfmm/trunk@51 2616d619-271b-44dc-8df4-d4a8f33a7222

Now compile with Intel compiler and MKL library. Add new options in CMakeLists.ixt SCALFMM_USE_CBLAS and  SCALFMM_USE_MKL

git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/scalfmm/scalfmm/trunk@50 2616d619-271b-44dc-8df4-d4a8f33a7222

× remove the potentials or forces only
× remove the abstract fmb kernel
× put a attribute to know if we use the mutual P2P or not (a template or inherite model may be faster)

git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/scalfmm/scalfmm/trunk@49 2616d619-271b-44dc-8df4-d4a8f33a7222

(Doxygen doc + fmm old P2P version)

git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/scalfmm/scalfmm/trunk@48 2616d619-271b-44dc-8df4-d4a8f33a7222

After cmake command, need "make doc" to generate the entire lib doc.
+
Add an Fmm algorithm version with old P2P system (need to be renamed)

git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/scalfmm/scalfmm/trunk@47 2616d619-271b-44dc-8df4-d4a8f33a7222

The Fmm is in charge of dispatching the leafs compution.
Anyway, this is in the kernel that the optimization can be done.

git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/scalfmm/scalfmm/trunk@46 2616d619-271b-44dc-8df4-d4a8f33a7222