Commit Graph
49 Commits
Author SHA1 Message Date
Stowell, Mark L cb8566ed91 Merge remote-tracking branch 'origin/master' into no-tet-reorder-dev
# Conflicts:
#	fem/fespace.cpp
#	fem/gridfunc.cpp
#	fem/pfespace.cpp
#	tests/unit/fem/test_domain_int.cpp
2021-05-21 10:02:19 -07:00
Julian Andrej a9ea9cba15 Merge branch 'master' into hypre-mixedint 2021-04-09 12:00:35 -07:00
Stowell, Mark L 1b16d8fbca Merge remote-tracking branch 'origin/master' into no-tet-reorder-dev
# Conflicts:
#	fem/fespace.cpp
#	fem/fespace.hpp
#	fem/pfespace.cpp
#	linalg/densemat.hpp
#	mesh/mesh.cpp
#	mesh/pmesh.cpp
#	mesh/pmesh.hpp
#	tests/unit/fem/test_3d_bilininteg.cpp
#	tests/unit/fem/test_get_value.cpp
#	tests/unit/fem/test_lin_interp.cpp
2021-04-08 19:44:11 -07:00
Andrew T. Barker 6c5831d15e MatrixFreeAMS: add optional smoother argument to constructor 2021-01-29 10:15:59 -08:00
Andrew T. Barker 6799097db9 Various changes based on code review, mostly in comments / documentation. 2021-01-29 09:41:16 -08:00
Andrew T. Barker 53be0a3c9f Make sure this compiles in serial. 2021-01-19 11:27:04 -08:00
Andrew T. Barker 9b847ddcff Merge remote-tracking branch 'origin/master' into barker29/matrix-free-ams 2021-01-19 08:01:28 -08:00
Stowell, Mark L 15f839e349 Merge remote-tracking branch 'origin/master' into no-tet-reorder-dev 2020-12-01 09:22:41 -08:00
camierjs 3d314e419a Remove duplicate ex3p tests 2020-11-17 08:29:03 -08:00
camierjs 72fe130e7e Put back np to 4 for no-pa runs 2020-11-16 12:20:59 -08:00
camierjs e0efc56aa2 Switch debug device for cuda in ex3p sample no-pa run 2020-11-16 12:12:33 -08:00
camierjs 64eb077e50 Add device w/o partial assembly runs 2020-11-16 11:26:27 -08:00
camierjs 020a97f0fc FullSimplify & revert ex3p 2020-11-13 16:37:55 -08:00
camierjs f82d078562 Simplify 2020-11-13 16:17:22 -08:00
camierjs dab294dabf Initial run 2020-11-13 15:25:36 -08:00
Dylan Copeland a3a99b0345 Added the option to use AmgX for the LOR solvers in MatrixFreeAMS. 2020-10-21 11:07:06 -07:00
Dylan Copeland 22f2591c9e Adding support for beta to be a MatrixCoefficient. 2020-10-15 15:39:40 -07:00
Andrew T. Barker 7e3d262c63 Clean up some debug information and std::cout stuff. 2020-09-24 09:38:28 -07:00
Andrew T. Barker 222d13eabf Boundary condition tweak.
This now matches results from old 2D templated code, used for the NLA paper.
2020-09-22 09:15:29 -07:00
Andrew T. Barker 0f6555e9ae Begin implementing matrix-free PA AMS cycle, does not work yet. 2020-09-15 14:18:39 -07:00
Veselin Dobrev 785486f12f Update examples, miniapps, tests to use HYPRE_BigInt instead
of HYPRE_Int where appropriate.
2020-07-25 20:47:06 -07:00
Stowell, Mark L c029b16d34 Merge remote-tracking branch 'origin/master' into no-tet-reorder-dev
# Conflicts:
#	fem/fespace.hpp
#	fem/pfespace.cpp
#	tests/unit/fem/test_3d_bilininteg.cpp
2020-03-18 11:06:04 -07:00
Tzanio 67093a6fd7 minor 2020-02-29 09:07:23 -08:00
Stowell, Mark L 9decf3a4aa Removing comment which may lead to confusion 2020-02-15 14:51:46 -08:00
Stowell, Mark L 8514657bf5 Reverting to standard refinement strategies 2020-02-15 14:49:12 -08:00
Dylan Copeland 603bec893a Simplified the Jacobi smoother definitions. Now using FormSystemMatrix in ex24p for the correct global solve. 2020-02-13 15:03:30 -08:00
camierjs 3bf5003fbe Remove unsuported device sample runs 2020-02-12 15:14:59 -08:00
Dylan Copeland 7ff48315ef Added PA to ex3, as well as cuda options and sample runs in comments. 2020-02-04 10:17:36 -08:00
Dylan Copeland a02409f8c9 Removed absolute value in diagonal assembly, by using a new MultTransposeUnsigned function in ElementRestriction. Made some changes for gpu support. 2020-02-03 09:58:28 -08:00
Dylan Copeland 17ec281338 Minor edits. 2020-01-16 15:49:02 -08:00
Dylan Copeland a13d0703f2 Fixed bilininteg_mass.cpp to not call class functions within MFEM_FORALL loops. Fixed Jacobi smoother usage, which was broken since merging from master. 2020-01-16 12:24:30 -08:00
Tzanio e51add7742 make style 2019-12-21 22:24:07 -08:00
Dylan Copeland b89cf4b497 Added Jacobi preconditioner to PA solver in ex3p. Removed BC in exNDH1. 2019-12-19 11:58:26 -08:00
Dylan Copeland c222f1deb1 Implemented 2D version of PA for ND_QuadrilateralElement. Cleaned up ex3p. 2019-12-16 09:48:54 -08:00
Dylan Copeland a381bc9647 Extended ex3p to work with PA. 2019-12-12 14:53:24 -08:00
Veselin Dobrev f5e20f33e3 Fix a few bugs and revert some debugging edits. 2019-11-12 19:40:20 -08:00
Jakub Červený 2617ebc59e Debugging shared triangle orientation (mpirun -n 3 ex3p -m ../data/inline-tet.mesh) 2019-11-05 18:19:20 +01:00
Stowell, Mark L f370b414d5 Moving call to ExchangeFaceNbrData into ParFiniteElementSpace 2019-05-21 11:29:13 -07:00
Stowell, Mark L 32f254b50f Replacing ReorientTetMesh with ExchangeFaceNbrData (also changing related comments) 2019-05-18 10:54:22 -07:00
Stowell, Mark L 7325eec877 Adding refinement level controls to ex3 and ex3p 2019-05-18 10:39:28 -07:00
Stowell, Mark L d5dfb6c52e Replacing ReorientTetMesh with ExchangeFaceNbrData 2019-05-18 10:39:01 -07:00
Stowell, Mark L 4d51451bfb Removing ReorientTetMesh methods (and breaking high order ND tets) 2018-11-13 21:48:09 -08:00
Veselin Dobrev da2fb88a17 Improved uniform refinement of tetrahedral meshes
In class Mesh/ParMesh:
  * Move the serial implementation of UniformRefinement3D to a new
    method: UniformRefinement3D_base. The implementations of the virtual
    method UniformRefinement3D (which now have no parameters) use the
    new UniformRefinement3D_base method.
  * In UniformRefinement3D_base, implemented two algorithms for choosing
    how to split the middle octahedron when refining a tetrahedron by
    cutting off its four corner tets. (These four tets have the same
    shape as the original tet and half the edge-length.) The choice of
    the algorithm is hard-coded in a const variable for now.
  * Add an optional parameter to UniformRefinement which is used to
    choose how to refine tet-only meshes: the default choice is to use
    the new algorithm defined by UniformRefinement3D; the second option
    is to use the old default - call LocalRefinement (marking all
    elements) to perform 3 levels of bisection. The new algorithm
    always produces elements with better shape (aspect ratio) than the
    old default (at least for the meshes in the data/ directory and a
    few other meshes).
  * Make the method Finalize virtual - its implementation in parallel
    requires updates in the ParMesh data.
  * Add a consistency check in ParMesh::ReorientTetMesh that verifies
    the assumption made in the method about the update of the shared
    triangles.
  * Simplify implementation of some methods in class ParMesh by
    separating common code in a new protected method: FinalizeParTopo.

Other updates:
  * In the examples and miniapps, when using a tet-only mesh which is
    first refined uniformly and then locally, it is now necessary to
    call the method Mesh::Finalize(true) (which is now virtual) in order
    to mark the elements for local refinement after the uniform
    refinement.
  * In example 12p, use better random seed values.
  * In examples 3/3p, add a sample run with order=2 on a tet mesh - this
    will test the methods {Mesh,ParMesh}::ReorientTetMesh. Previously,
    these were only tested by one sample run in example 4p.
  * In the mesh-explorer miniapp, add a refinement option to perform
    uniform refinement of tet-only meshes using bisection.
  * In the MFEM_LOCATION macro print the <file> and <line> location
    using a standard format: <file>:<line>, as used by most compilers
    when reporting warnings and errors.
  * Remove FIXME comments about mesh format v1.0.1.
2018-09-27 21:04:58 -07:00
Jakub Cerveny 125ff226e4 Version 3.2 2016-06-30 10:37:30 -07:00
Tzanio dbae60fe32 Version 3.1 2016-02-16 15:45:48 -08:00
Tzanio Kolev 9d7774a03e Version 3.0 2015-01-26 15:07:34 -08:00
Tzanio Kolev fa169e4b92 Version 2.0 2011-11-18 11:39:53 -08:00
Tzanio Kolev f74eaae422 Version 1.2.1 2011-04-11 16:09:41 -07:00
Tzanio Kolev 29ae8fb6ac Version 1.2 2011-04-08 15:27:24 -07:00