]>
https://gitweb.dealii.org/ - dealii.git/log
Niklas Fehn [Wed, 16 Oct 2019 16:15:24 +0000 (18:15 +0200)]
extend GridGenerator::torus<3,3>() so that one can generate an open torus with angle 0 < phi <= 2*pi
Niklas Fehn [Wed, 16 Oct 2019 09:25:37 +0000 (11:25 +0200)]
fix bug in GridGenerator::torus<3,3>()
Daniel Arndt [Tue, 15 Oct 2019 18:19:01 +0000 (14:19 -0400)]
Merge pull request #8901 from dougshidong/master
Fix DoFHandler>::n_boundary_dofs() in issue #8897
David Wells [Mon, 14 Oct 2019 21:15:20 +0000 (17:15 -0400)]
Merge pull request #8908 from masterleinad/fix_typo_tensor
Fix another typo in tensor.h
Daniel Arndt [Mon, 14 Oct 2019 20:34:01 +0000 (16:34 -0400)]
Fix another typo in tensor.h
Daniel Arndt [Mon, 14 Oct 2019 20:30:19 +0000 (16:30 -0400)]
Merge pull request #8878 from dangars/complex_integrate_difference
Complex integrate_difference()
Martin Kronbichler [Mon, 14 Oct 2019 20:17:23 +0000 (22:17 +0200)]
Merge pull request #8907 from peterrum/tensor_docu_fix
Fix docu in Tensor
Peter Munch [Mon, 14 Oct 2019 20:06:01 +0000 (22:06 +0200)]
Fix docu in Tensor
David Wells [Mon, 14 Oct 2019 19:54:28 +0000 (15:54 -0400)]
Merge pull request #8903 from kronbichler/doc_vectorized_array_pod
VectorizedArray: Update documentation POD -> StandardLayoutType
Martin Kronbichler [Mon, 14 Oct 2019 13:25:36 +0000 (15:25 +0200)]
Update documentation about POD -> StandardLayoutType
Doug Shi-Dong [Fri, 11 Oct 2019 19:19:10 +0000 (15:19 -0400)]
Fix DoFHandler::n_boundary_dofs() in issue #8897
Only count locally owned dofs in the case of a parallel::distributed::Triangulation.
Use suggested syntax on cell iterator.
Co-Authored-By: Daniel Arndt <arndtd@ornl.gov>
Add test case for n_boundary_dofs in parallel
Fix n_boundary_dofs() for hp version.
Also implement suggested change
Start 3212: dofs/n_boundary_dofs_01.mpirun=1.debug
1/8 Test # 3212: dofs/n_boundary_dofs_01.mpirun=1.debug ..... Passed 0.50 sec
Start 3213: dofs/n_boundary_dofs_01.mpirun=1.release
2/8 Test # 3213: dofs/n_boundary_dofs_01.mpirun=1.release ... Passed 0.49 sec
Start 3214: dofs/n_boundary_dofs_01.mpirun=2.debug
3/8 Test # 3214: dofs/n_boundary_dofs_01.mpirun=2.debug ..... Passed 0.50 sec
Start 3215: dofs/n_boundary_dofs_01.mpirun=2.release
4/8 Test # 3215: dofs/n_boundary_dofs_01.mpirun=2.release ... Passed 0.51 sec
Start 3216: dofs/n_boundary_dofs_02.debug
5/8 Test # 3216: dofs/n_boundary_dofs_02.debug .............. Passed 0.51 sec
Start 3217: dofs/n_boundary_dofs_02.release
6/8 Test # 3217: dofs/n_boundary_dofs_02.release ............ Passed 0.52 sec
Start 3218: dofs/n_boundary_dofs_03.debug
7/8 Test # 3218: dofs/n_boundary_dofs_03.debug .............. Passed 0.52 sec
Start 3219: dofs/n_boundary_dofs_03.release
8/8 Test # 3219: dofs/n_boundary_dofs_03.release ............ Passed 0.54 sec
100% tests passed, 0 tests failed out of 8
Add hp n_boundary_dofs test.
Use unordered_set instead.
Fix a || to a && from switching the if-continue statement.
Start 3212: dofs/n_boundary_dofs_01.mpirun=1.debug
1/18 Test # 3212: dofs/n_boundary_dofs_01.mpirun=1.debug ..... Passed 51.27 sec
Start 3213: dofs/n_boundary_dofs_01.mpirun=1.release
2/18 Test # 3213: dofs/n_boundary_dofs_01.mpirun=1.release ... Passed 17.03 sec
Start 3214: dofs/n_boundary_dofs_01.mpirun=2.debug
3/18 Test # 3214: dofs/n_boundary_dofs_01.mpirun=2.debug ..... Passed 1.90 sec
Start 3215: dofs/n_boundary_dofs_01.mpirun=2.release
4/18 Test # 3215: dofs/n_boundary_dofs_01.mpirun=2.release ... Passed 1.45 sec
Start 3216: dofs/n_boundary_dofs_02.debug
5/18 Test # 3216: dofs/n_boundary_dofs_02.debug .............. Passed 50.17 sec
Start 3217: dofs/n_boundary_dofs_02.release
6/18 Test # 3217: dofs/n_boundary_dofs_02.release ............ Passed 16.30 sec
Start 3218: dofs/n_boundary_dofs_03.debug
7/18 Test # 3218: dofs/n_boundary_dofs_03.debug .............. Passed 50.67 sec
Start 3219: dofs/n_boundary_dofs_03.release
8/18 Test # 3219: dofs/n_boundary_dofs_03.release ............ Passed 16.71 sec
Start 5498: hp/n_boundary_dofs.debug
9/18 Test # 5498: hp/n_boundary_dofs.debug ................... Passed 44.68 sec
Start 5499: hp/n_boundary_dofs.release
10/18 Test # 5499: hp/n_boundary_dofs.release ................. Passed 11.83 sec
Start 5500: hp/n_boundary_dofs_01.mpirun=1.debug
11/18 Test # 5500: hp/n_boundary_dofs_01.mpirun=1.debug ....... Passed 34.48 sec
Start 5501: hp/n_boundary_dofs_01.mpirun=1.release
12/18 Test # 5501: hp/n_boundary_dofs_01.mpirun=1.release ..... Passed 12.46 sec
Start 5502: hp/n_boundary_dofs_01.mpirun=2.debug
13/18 Test # 5502: hp/n_boundary_dofs_01.mpirun=2.debug ....... Passed 1.75 sec
Start 5503: hp/n_boundary_dofs_01.mpirun=2.release
14/18 Test # 5503: hp/n_boundary_dofs_01.mpirun=2.release ..... Passed 1.36 sec
Start 5504: hp/n_boundary_dofs_02.debug
15/18 Test # 5504: hp/n_boundary_dofs_02.debug ................ Passed 34.33 sec
Start 5505: hp/n_boundary_dofs_02.release
16/18 Test # 5505: hp/n_boundary_dofs_02.release .............. Passed 12.25 sec
Start 5506: hp/n_boundary_dofs_03.debug
17/18 Test # 5506: hp/n_boundary_dofs_03.debug ................ Passed 41.88 sec
Start 5507: hp/n_boundary_dofs_03.release
18/18 Test # 5507: hp/n_boundary_dofs_03.release .............. Passed 15.45 sec
100% tests passed, 0 tests failed out of 18
Martin Kronbichler [Mon, 14 Oct 2019 07:08:16 +0000 (09:08 +0200)]
Merge pull request #8900 from peterrum/vectorizedarray-iterator
Introduce VectorizedArrayBase and VectorizedArrayIterator
Martin Kronbichler [Mon, 14 Oct 2019 07:07:42 +0000 (09:07 +0200)]
Merge pull request #8894 from nfehn/master
New functionality to compute maximum aspect ratio of a mesh
Niklas Fehn [Tue, 8 Oct 2019 11:10:56 +0000 (13:10 +0200)]
new functionality to compute mesh aspect ratio
Peter Munch [Fri, 11 Oct 2019 17:39:25 +0000 (19:39 +0200)]
Introduce VectorizedArrayBase and VectorizedArrayIterator
Martin Kronbichler [Sat, 12 Oct 2019 10:08:49 +0000 (12:08 +0200)]
Merge pull request #8869 from bangerth/nedelec
Update the description of the Nedelec element.
Wolfgang Bangerth [Thu, 26 Sep 2019 21:17:02 +0000 (15:17 -0600)]
Reference FE_Nedelec from the documentation of FE_NedelecSZ.
Daniel Arndt [Wed, 9 Oct 2019 23:01:39 +0000 (19:01 -0400)]
Merge pull request #8895 from starki0815/merge_triangulations_bug
Fix a bug in merge triangulations()
Daniel Garcia-Sanchez [Wed, 9 Oct 2019 17:20:16 +0000 (19:20 +0200)]
Add changelog entry
Daniel Arndt [Wed, 9 Oct 2019 12:15:47 +0000 (08:15 -0400)]
Merge pull request #8893 from kronbichler/vectorized_array_size
Add VectorizedArray::size() member function
Sebastian Stark [Wed, 9 Oct 2019 12:09:47 +0000 (14:09 +0200)]
Update source/grid/tria.cc
Co-Authored-By: Daniel Arndt <arndtd@ornl.gov>
Sebastian Stark [Wed, 9 Oct 2019 12:09:41 +0000 (14:09 +0200)]
Update source/grid/tria.cc
Co-Authored-By: Daniel Arndt <arndtd@ornl.gov>
Daniel Garcia-Sanchez [Wed, 9 Oct 2019 10:42:12 +0000 (12:42 +0200)]
Add test integrate_difference_01_complex_04
Daniel Garcia-Sanchez [Wed, 9 Oct 2019 08:57:15 +0000 (10:57 +0200)]
Use a std::complex function for the tests integrate_difference_01_complex_0{1,2,3}
Daniel Garcia-Sanchez [Wed, 9 Oct 2019 08:41:38 +0000 (10:41 +0200)]
Use a std::complex function for the test fe_nedelec_singularity_01
Daniel Garcia-Sanchez [Wed, 9 Oct 2019 07:43:41 +0000 (09:43 +0200)]
Add std::complex support to integrate_difference()
The type of exact_solution was double and fe_function could have any numerical
type (double, float, std::complex<double> or std::complex<float>). This would
lead in certain cases to unnecessary casts, for example from double to float.
In addition it was not possible to compare a std::complex exact_solution to a
std::complex fe_function.
Now the types of exact_solution and fe_function must be the same, as a result
integrate_difference() can be used to compare a std::complex exact_solution to
a std::complex fe_function.
This change is not backward-compatible. The old version of the function has
been deprecated.
Martin Kronbichler [Tue, 8 Oct 2019 07:17:57 +0000 (09:17 +0200)]
Changelog
Martin Kronbichler [Tue, 8 Oct 2019 07:16:10 +0000 (09:16 +0200)]
Test size() function
Martin Kronbichler [Tue, 8 Oct 2019 07:15:59 +0000 (09:15 +0200)]
Add VectorizedArray::size() function
Sebastian Stark [Tue, 8 Oct 2019 11:53:47 +0000 (13:53 +0200)]
Fix a bug in merge triangulations()
This patch makes sure that merge_triangulations() does not forget about where the boundary is (the problem has actually been in create_triangulation(), which assigned numbers::internal_face_boundary_id to boundary_faces)
Daniel Arndt [Mon, 7 Oct 2019 18:44:48 +0000 (14:44 -0400)]
Merge pull request #8889 from rezarastak/grid_tools_doc
Small rewording of doc for compute_point_locations()
Reza Rastak [Mon, 7 Oct 2019 17:16:58 +0000 (10:16 -0700)]
Apply suggestions from code review
Co-Authored-By: Daniel Arndt <arndtd@ornl.gov>
Martin Kronbichler [Mon, 7 Oct 2019 10:01:21 +0000 (12:01 +0200)]
Merge pull request #8888 from masterleinad/bind_thread_management
Remove some uses of std::bind from thread_management.h
Daniel Arndt [Sun, 6 Oct 2019 23:58:06 +0000 (19:58 -0400)]
Merge pull request #8887 from masterleinad/avoid_thread_managenent
Avoid including thread_management.h
Daniel Arndt [Sun, 6 Oct 2019 15:48:09 +0000 (11:48 -0400)]
Remove some uses of std::bind from thread_management.h
Timo Heister [Sun, 6 Oct 2019 22:16:49 +0000 (18:16 -0400)]
Merge pull request #8890 from rezarastak/doxygen_perl
PERL_PATH removed from options.dox
Reza Rastak [Sun, 6 Oct 2019 19:04:46 +0000 (12:04 -0700)]
PERL_PATH removed from options.dox
Reza Rastak [Sun, 6 Oct 2019 18:42:56 +0000 (11:42 -0700)]
Rewording of the doc in compute_point_locations
Daniel Arndt [Sun, 6 Oct 2019 15:43:39 +0000 (11:43 -0400)]
Avoid including thread_management.h
Martin Kronbichler [Sat, 5 Oct 2019 19:14:22 +0000 (21:14 +0200)]
Merge pull request #8886 from rezarastak/dist_solution_trans_doc_fix
Minor documentation fix in P::D::SolutionTransfer
Reza Rastak [Fri, 4 Oct 2019 23:17:30 +0000 (16:17 -0700)]
reorder part of the doc in P::D::SolutionTransfer
Reza Rastak [Thu, 3 Oct 2019 06:15:33 +0000 (23:15 -0700)]
Minor documentation fix in P::D::SolutionTransfer
Martin Kronbichler [Wed, 2 Oct 2019 14:36:20 +0000 (16:36 +0200)]
Merge pull request #8881 from masterleinad/fix_cuda_aware_mpi
Fix behavior for CUDA-aware MPI
Daniel Arndt [Wed, 2 Oct 2019 12:28:24 +0000 (08:28 -0400)]
Fixup style for preprocessor check
Daniel Arndt [Tue, 1 Oct 2019 21:20:01 +0000 (21:20 +0000)]
Avoid setting values when inserting ghost entries
Daniel Arndt [Tue, 1 Oct 2019 21:19:13 +0000 (21:19 +0000)]
Avoid using std::memset for device data
Bruno Turcksin [Tue, 1 Oct 2019 13:05:10 +0000 (09:05 -0400)]
Merge pull request #8880 from kronbichler/cuda_synchronize_mpi
Synchronize CUDA device in MPI ghost exchange
Martin Kronbichler [Tue, 1 Oct 2019 12:55:49 +0000 (14:55 +0200)]
Synchronize CUDA device in MPI ghost exchange
Daniel Arndt [Tue, 1 Oct 2019 05:03:21 +0000 (01:03 -0400)]
Merge pull request #8879 from masterleinad/Fix_ms_host_la_templates_campi
Fix using MemorySpace::Host with la_parallel_vector.templates.h and CUDA-aware MPI
Daniel Arndt [Mon, 30 Sep 2019 20:10:35 +0000 (16:10 -0400)]
Fix using MemorySpace::Host with la_parallel_vector.templates.h and CUDA-aware MPI
David Wells [Mon, 30 Sep 2019 18:41:16 +0000 (14:41 -0400)]
Merge pull request #8736 from fvanmaele/perl-canonical-doxygen
doxygen: rewrite set_canonical_doxygen.py in Perl
Daniel Arndt [Mon, 30 Sep 2019 18:40:21 +0000 (14:40 -0400)]
Merge pull request #8875 from kronbichler/cuda_improve_mf_loop
Bypass vector copy in MatrixFree::cell_loop with MPI
Martin Kronbichler [Fri, 27 Sep 2019 14:12:21 +0000 (16:12 +0200)]
Changelog
Daniel Arndt [Mon, 30 Sep 2019 02:23:19 +0000 (22:23 -0400)]
Merge pull request #8847 from tjhei/kinematics_always_inline
kinematics.h: fix always_inline
Bruno Turcksin [Sat, 28 Sep 2019 21:33:12 +0000 (17:33 -0400)]
Merge pull request #8813 from kronbichler/partitioner_use_ranges
Process indices in ComputeIndexOwner by intervals
Martin Kronbichler [Sat, 28 Sep 2019 12:24:57 +0000 (14:24 +0200)]
Merge pull request #8876 from masterleinad/fix_host_compiler
Fix setting host compiler
Daniel Arndt [Sat, 28 Sep 2019 04:26:08 +0000 (00:26 -0400)]
Merge pull request #8871 from peterrum/cuda-step64-update
Update step-64
Peter Munch [Fri, 27 Sep 2019 05:06:51 +0000 (07:06 +0200)]
Update step-64
Daniel Arndt [Fri, 27 Sep 2019 22:00:11 +0000 (18:00 -0400)]
Merge pull request #8811 from kronbichler/speedup_mg_transfer_copy_indices
Avoid sorting a large array in MG fill_copy_indices
Daniel Arndt [Fri, 27 Sep 2019 17:10:00 +0000 (17:10 +0000)]
Fix setting host compiler
Martin Kronbichler [Fri, 27 Sep 2019 16:06:33 +0000 (18:06 +0200)]
Merge pull request #8874 from masterleinad/clang-9-tidy
Fix some clang-tidy 9 findings
Martin Kronbichler [Fri, 27 Sep 2019 14:11:39 +0000 (16:11 +0200)]
Use proper partitioner in step-64
Martin Kronbichler [Fri, 27 Sep 2019 14:10:52 +0000 (16:10 +0200)]
Bypass vector copies if partitioners match
Martin Kronbichler [Fri, 20 Sep 2019 07:46:25 +0000 (09:46 +0200)]
Avoid sorting a large array in MG fill_copy_indices
Martin Kronbichler [Fri, 27 Sep 2019 07:03:29 +0000 (09:03 +0200)]
Merge pull request #8872 from peterrum/gpu-comment-fix2
Update the GPU matrix-free documentation (2) [WIP]
Peter Munch [Fri, 27 Sep 2019 05:37:40 +0000 (07:37 +0200)]
Update the GPU matrix-free documentation (2)
Wolfgang Bangerth [Thu, 26 Sep 2019 23:11:06 +0000 (17:11 -0600)]
Merge pull request #8870 from masterleinad/require_mpirun_p4est
Require mpirun when requesting p4est in tests
Daniel Arndt [Thu, 26 Sep 2019 22:32:55 +0000 (18:32 -0400)]
Require mpirun when requesting p4est in tests
Daniel Arndt [Thu, 26 Sep 2019 22:37:04 +0000 (18:37 -0400)]
Merge pull request #8858 from jwitte08/get-cell-index
add MatrixFree::get_cell_level_and_index()
Bruno Turcksin [Thu, 26 Sep 2019 21:56:41 +0000 (17:56 -0400)]
Merge pull request #8860 from masterleinad/fix_fe_nothing_coarsening
Fix SolutionTransfer when coarsening from FE_Nothing
Bruno Turcksin [Thu, 26 Sep 2019 21:50:53 +0000 (17:50 -0400)]
Merge pull request #8868 from peterrum/parallel-computeverticeswithghostneighbors-assert
Enable assert again
Wolfgang Bangerth [Thu, 26 Sep 2019 21:15:24 +0000 (15:15 -0600)]
Update the description of the Nedelec element.
Daniel Arndt [Thu, 26 Sep 2019 18:07:49 +0000 (18:07 +0000)]
Fix some clang-tidy 9 findings
Peter Munch [Thu, 26 Sep 2019 20:04:03 +0000 (22:04 +0200)]
Enable assert again
Wolfgang Bangerth [Thu, 26 Sep 2019 20:03:15 +0000 (14:03 -0600)]
Merge pull request #8867 from dougshidong/fix_manifold_get_new_point
Add changelog for #8861
Doug Shi-Dong [Thu, 26 Sep 2019 19:35:30 +0000 (15:35 -0400)]
Add changelog for #8861
Daniel Arndt [Thu, 26 Sep 2019 17:47:05 +0000 (13:47 -0400)]
Merge pull request #8864 from bangerth/direction
Change 'direction' of make_periodicity_constraint() to 'unsigned int'.
Bruno Turcksin [Thu, 26 Sep 2019 16:25:01 +0000 (12:25 -0400)]
Merge pull request #8865 from masterleinad/fix_cuda_tests
Fix failing CUDA tests
Daniel Arndt [Thu, 26 Sep 2019 16:18:09 +0000 (12:18 -0400)]
Merge pull request #8838 from peterrum/gpu-qpoint
Eliminate q_point from functor and other GPU functions
Daniel Arndt [Thu, 26 Sep 2019 16:11:41 +0000 (16:11 +0000)]
Fix failing CUDA tests
Martin Kronbichler [Thu, 26 Sep 2019 15:44:33 +0000 (17:44 +0200)]
Merge pull request #8854 from peterrum/shapeinfo-collocation
Compute shape_gradients/hessians_collocation in ShapeInfo
Daniel Arndt [Thu, 26 Sep 2019 15:23:44 +0000 (11:23 -0400)]
Merge pull request #8861 from dougshidong/fix_manifold_get_new_point
Fix Manifold get_new_point()
Wolfgang Bangerth [Thu, 26 Sep 2019 14:19:59 +0000 (08:19 -0600)]
Add a changelog entry.
Wolfgang Bangerth [Thu, 26 Sep 2019 14:19:24 +0000 (08:19 -0600)]
Change 'direction' of make_periodicity_constraint() to 'unsigned int'.
Julius Witte [Wed, 25 Sep 2019 16:38:09 +0000 (12:38 -0400)]
Store underlying Mapping
Julius Witte [Wed, 25 Sep 2019 16:45:12 +0000 (12:45 -0400)]
Store 1D quadrature
Julius Witte [Wed, 25 Sep 2019 18:17:50 +0000 (14:17 -0400)]
add and test MatrixFree::get_cell_level_and_index()
Wolfgang Bangerth [Thu, 26 Sep 2019 13:42:48 +0000 (07:42 -0600)]
Merge pull request #8859 from masterleinad/document_periodic_direction
Document direction parameter for periodic constraints
Daniel Arndt [Thu, 26 Sep 2019 13:31:14 +0000 (09:31 -0400)]
Merge pull request #8863 from peterrum/parallel-computeverticeswithghostneighbors-newtest
Add missing test output
Peter Munch [Wed, 25 Sep 2019 10:55:34 +0000 (12:55 +0200)]
Compute shape_gradients_collocation and shape_hessians_collocation in ShapeInfo
Daniel Arndt [Thu, 26 Sep 2019 13:04:17 +0000 (09:04 -0400)]
Remove unused variable
Martin Kronbichler [Thu, 26 Sep 2019 12:25:49 +0000 (14:25 +0200)]
Merge pull request #8862 from peterrum/parallel-computeverticeswithghostneighbors-fix
Fix a bug introduced in #8783
Peter Munch [Thu, 26 Sep 2019 12:08:39 +0000 (14:08 +0200)]
Add missing test
Martin Kronbichler [Thu, 26 Sep 2019 10:01:54 +0000 (12:01 +0200)]
Merge pull request #8857 from jwitte08/extend-tpmatrix
Extend TensorProductMatrixSymmetricSum
Peter Munch [Thu, 26 Sep 2019 08:09:43 +0000 (10:09 +0200)]
Fix a bug introduced in #8783
Doug Shi-Dong [Thu, 26 Sep 2019 05:47:41 +0000 (01:47 -0400)]
Fix get_new_point optimization.
A performance-if ignores next point when current point weight = 0.
However, since the current point weight is 0, we should be replacing the current point with the next point, not skipping the next point.
Doug Shi-Dong [Thu, 26 Sep 2019 04:31:07 +0000 (00:31 -0400)]
Fix Manifold get_new_point()
The weights can be negative, we simply want to avoid dividing by 0.
Daniel Arndt [Thu, 26 Sep 2019 04:12:54 +0000 (00:12 -0400)]
Use alias
Daniel Arndt [Thu, 26 Sep 2019 04:01:44 +0000 (00:01 -0400)]
Fix SolutionTransfer when corsening from FE_Nothing
Julius Witte [Thu, 26 Sep 2019 01:14:52 +0000 (03:14 +0200)]
Rename template parameter size to n_rows_1d
In the beginning the Universe was created. This has made a lot of
people very angry and has been widely regarded as a bad move.
Douglas Adams
Typeset in
Trocchi and Trocchi Bold Sans Serif.