Skip to content
Snippets Groups Projects

Repository graph

You can move around the graph by using the arrow keys.
Select Git revision
  • feature/debian_package
  • feature/dirichlet_bc
  • feature/parallel_rosenbrock
  • issue/petsc_3-7
  • master default
  • rainer_production
6 results
Created with Raphaël 2.2.029Oct272625242322212019181716151110985428Sep27262524191776530Aug292827242322212017161514109876432131Jul3029262423221817121110543229Jun282726252221181413121174130May25242322211615141110987432130Apr2725232018171613121110543231Mar302827222016151413129765128Feb27242322201614131098763131Jan3027262524201918171614131210643222Dec16151211952130Nov2928252321151412118714Oct117515Sep1497125Aug16151312230Jul272522212015108727Jun2624232216151098763131May30Fixed compiling problem in release mode.Just a change of class name to make consistant with my diss.Navier stokes solver ready to be used.Merge to juropaWork on this totaly bull shit navier stokes what ever I have never heard about solver, i will go home and eat one or two cakes.Some work on navier stokes solver.Hopefully fixed problem with local-global matrix indices for the last.Fixed problem with new KSP solver interface.Fixed compilation problem.Fixed some issues for removing periodic boundary conditions.Added framework for a PETSc based navier stokes solver. Still must, or Sebastian should do it, implement the specific algorithm.Fixed a lot of small problems.Fixed problem for Dirichlet boundary conditions on non diagonal blocks in parallel.bo eh, part iiibo eh, part iibo ehUps, mal ein code refactoring dass nicht gleich in 100 bugfixes enden. So, nun sollte der ganze parallele spass auch fuer gekoppelte Probleme ohne Tricks funktionieren. Nur testen will es mal wieder keiner, daher geh ich mittag essen.Fixed FETI-DP for Stokes problem.Code refactoring, add comments, dirichlet rows in FETI-DP defined per component.And now really fixed, go ahead to write productive codes....No, does not work, so merge to juropa and go on with debugging.Small bug fix in parallel DOF mapping, it really works ....Fixed last issues in code change of parallel DOF mapping, now go on for debugging.Implemented iterator interface for ParallelDOFMapping.WUA, IT COMPILES....DOES NOT COMPILE, BUT I MUST MERGE TO WORK HARD ON WEEKEND....So onFixed for using petsc-dev, have to remove the petsc multigrid class which is far away from being usefull.correct for error due to "Name look up changes" in gcc 4.7 :Fixed problem for explicit computation of augmented schur primal system.Add support for BDDCML 2.0 and work on scalable FETI-DP in 3d using arithmetic constraints.Does not work, so do not checkout this revision, but still have fun to commit this shit of code.Fixed augmented coarse space, added first support for symmetric FETI-DP problems.Merge with local changes.Explicit Schur primal matrix creation for augmented FETI-DP finished.Added PETSc helper namespace, simplicies explicit schur primal computations in FETI-DP code.Work in Dirichlet preconditioner for FETI-DP.Fixed lagrange scaling for general 3D problems.Augmented FETI-DP is running now, test it for 3D problems.Changed augmented coarse grid problem to its transpose.