]>
https://gitweb.dealii.org/ - dealii.git/log
Wolfgang Bangerth [Fri, 27 Feb 2015 03:00:56 +0000 (21:00 -0600)]
Merge pull request #609 from tjhei/nocxx11_test_fixes
no cxx11 test fixes
Timo Heister [Fri, 27 Feb 2015 02:28:24 +0000 (21:28 -0500)]
no cxx1 test fixes
Fix two tests that won't compile without c++ 11
Wolfgang Bangerth [Thu, 26 Feb 2015 00:02:47 +0000 (18:02 -0600)]
Merge pull request #607 from tjhei/intel_bug
add failing test on intel 15
Timo Heister [Wed, 25 Feb 2015 22:33:07 +0000 (17:33 -0500)]
add failing test on intel 15
This adds a simplified version of vector-vector.cc that fails in release mode on Intel.
Toby D. Young [Wed, 25 Feb 2015 14:35:25 +0000 (15:35 +0100)]
Merge pull request #604 from bangerth/more-tensor-fixes-2
More tensor fixes 2
Wolfgang Bangerth [Wed, 25 Feb 2015 14:22:38 +0000 (08:22 -0600)]
Provide even more specializations of ProductType in non-C++11 mode.
Wolfgang Bangerth [Wed, 25 Feb 2015 14:08:53 +0000 (08:08 -0600)]
Merge pull request #603 from tjhei/sparse_mic_optimization
SparseMIC optimization
Wolfgang Bangerth [Wed, 25 Feb 2015 13:57:18 +0000 (07:57 -0600)]
Explicitly provide ProductType for Tensor<rank,dim> in non-C++11 mode.
Wolfgang Bangerth [Wed, 25 Feb 2015 13:56:30 +0000 (07:56 -0600)]
Explicitly provide ProductType for Tensor<1,dim> in non-C++11 mode.
Wolfgang Bangerth [Wed, 25 Feb 2015 13:53:53 +0000 (07:53 -0600)]
Provide more specializations of ProductType for the non-C++11 case.
Wolfgang Bangerth [Wed, 25 Feb 2015 13:50:14 +0000 (07:50 -0600)]
Avoid C++11'isms.
Wolfgang Bangerth [Wed, 25 Feb 2015 13:49:02 +0000 (07:49 -0600)]
Add another template specialization for ProductType.
Timo Heister [Wed, 25 Feb 2015 14:03:04 +0000 (09:03 -0500)]
SparseMIC optimization
It seems like we are computing a mat-vec just to overwrite it a couple
of lines later. Remove it.
Wolfgang Bangerth [Wed, 25 Feb 2015 14:00:25 +0000 (08:00 -0600)]
Merge pull request #602 from tjhei/sparse_mic_fixes
fix SparseMIC
Wolfgang Bangerth [Wed, 25 Feb 2015 14:00:02 +0000 (08:00 -0600)]
Merge pull request #601 from tjhei/open_cascade_error
fix error in opencascade assert
Timo Heister [Wed, 25 Feb 2015 13:56:22 +0000 (08:56 -0500)]
fix SparseMIC
We need to initialize the decomposition before we can look at this->m()
and this->n(). Bug got introduced in
27b50d8 .
Timo Heister [Wed, 25 Feb 2015 13:52:47 +0000 (08:52 -0500)]
Merge pull request #599 from tjhei/fix_warnings
fix compiler warnings
that got introduced in
61a893a (#197)
Timo Heister [Wed, 25 Feb 2015 13:40:53 +0000 (08:40 -0500)]
fix error in opencascade assert
Timo Heister [Wed, 25 Feb 2015 13:18:32 +0000 (08:18 -0500)]
fix compiler warnings
that got introduced in
61a893a (#197)
Wolfgang Bangerth [Wed, 25 Feb 2015 13:02:45 +0000 (07:02 -0600)]
Merge pull request #597 from bangerth/add-test-4
Update Utilities::split_string_list
Wolfgang Bangerth [Tue, 24 Feb 2015 23:46:55 +0000 (17:46 -0600)]
Add another test. Update another.
Wolfgang Bangerth [Tue, 24 Feb 2015 23:45:39 +0000 (17:45 -0600)]
Clarify some of the corner cases of the split_string_list() function.
Wolfgang Bangerth [Tue, 24 Feb 2015 23:41:36 +0000 (17:41 -0600)]
Merge pull request #197 from rosskynch/fix_project_bv_curl_conf
Fix project bv curl conf
Wolfgang Bangerth [Tue, 24 Feb 2015 23:30:50 +0000 (17:30 -0600)]
Merge pull request #595 from mfherbst/master
Use AlignedVector::size_type in all of table.h
Wolfgang Bangerth [Tue, 24 Feb 2015 23:28:23 +0000 (17:28 -0600)]
Merge pull request #596 from tjhei/petsc_mat_compress
PETSc mat compress
Timo Heister [Tue, 24 Feb 2015 17:51:50 +0000 (12:51 -0500)]
fix a few broken tests
Looks like somebody messed up the VectorOperation in compress() in many
places.
Wolfgang Bangerth [Tue, 24 Feb 2015 22:33:33 +0000 (16:33 -0600)]
Add a test for a patch by Menno Fraters this morning.
Timo Heister [Tue, 24 Feb 2015 15:10:30 +0000 (10:10 -0500)]
add compress() error checks for PETSc vectors
Timo Heister [Tue, 24 Feb 2015 14:45:01 +0000 (09:45 -0500)]
remove compress without args from all PETSc sparse matrix types
- compress() without arguments was deprecated and now got remove
- replace redundant LastAction enum in by VectorOperation
- add additional compress() check.
Michael F. Herbst [Tue, 24 Feb 2015 15:14:04 +0000 (16:14 +0100)]
Merge branch 'master' of https://github.com/dealii/dealii
Michael F. Herbst [Tue, 24 Feb 2015 15:07:59 +0000 (16:07 +0100)]
Use AlignedVector::size_type in all of table.h
In all classes of table.h the type Use AlignedVector::size_type is now
used for the calculation of indices for the values vector and the
indexing of elements in the values vector.
Note: This commit changes the interface of the n_elements function
in TableBase<N,T>
Wolfgang Bangerth [Tue, 24 Feb 2015 14:21:51 +0000 (08:21 -0600)]
Merge pull request #594 from MFraters/patch-1
empty string in split_string_list fix
Ross Kynch [Tue, 14 Oct 2014 13:54:57 +0000 (14:54 +0100)]
Added project_boundary_values_curl_conforming_l2() to VectorTools for
use with FE_Nedelec elements in 2D & 3D.
This includes a test code included in tests/deal.II to check it works correctly.
This new function uses an L2-projection on edges and
faces (only edges in 2D, both for 3D) to compute
the constaints and factors in the residual from the edge
projection when computing the face constraints. The original
function project_boundary_values_curl_conforming() has
problems with non-rectangular faces, this one does not. The
function assumes that the face orientations are standard (e.g. a
sphere may cause problems due to mismatched face orientations).
The test shows that project_boundary_curl_conforming_l2() works on a
distorted cube (non-rectangular faces) for a simple test case
(polynomial function).
Martin Kronbichler [Tue, 24 Feb 2015 13:42:17 +0000 (14:42 +0100)]
Merge pull request #593 from bangerth/remove-deprecated-functions-13
Remove deprecated function SparseLUDecomposition::decompose.
MFraters [Tue, 24 Feb 2015 13:38:40 +0000 (14:38 +0100)]
Changed evaluation order
Wolfgang Bangerth [Tue, 24 Feb 2015 13:37:42 +0000 (07:37 -0600)]
Update documentation.
Wolfgang Bangerth [Tue, 24 Feb 2015 13:27:40 +0000 (07:27 -0600)]
Remove the SparseLUDecomposition::decomposed variable.
It is only ever written to, never read.
Michael F. Herbst [Tue, 24 Feb 2015 11:57:18 +0000 (12:57 +0100)]
Use AlignedVector::size_type in Table::operator()
Use AlignedVector<T>::size_type as the integer type to carry out the
calculation of the index in the values AlignedVector object for the
various operator() and el() functions of Table<N,T> and its explicit
specialisations.
MFraters [Tue, 24 Feb 2015 10:32:07 +0000 (11:32 +0100)]
empty string in split_string_list fix
When an empty string or a string with only spaces is given to this function, the split_string_list function will ask in line 222 for name[-1], which will result in unpredictable behaviour. I propose to prevent this by adding a condition that the length of the string may not be 0.
Martin Kronbichler [Tue, 24 Feb 2015 08:11:05 +0000 (09:11 +0100)]
Merge pull request #576 from dakshinai/master
modified filtered_matrix tests to work with square matrices only
Wolfgang Bangerth [Tue, 24 Feb 2015 03:23:27 +0000 (21:23 -0600)]
Remove deprecated function SparseLUDecomposition::decompose.
Wolfgang Bangerth [Mon, 23 Feb 2015 21:31:27 +0000 (15:31 -0600)]
Merge pull request #591 from tjhei/wrapcomments
wrapcomments
Wolfgang Bangerth [Mon, 23 Feb 2015 21:31:18 +0000 (15:31 -0600)]
Merge pull request #592 from tjhei/ez_const_fix
add missing const
Timo Heister [Mon, 23 Feb 2015 21:27:45 +0000 (16:27 -0500)]
add missing const
Doxygen was rightfully complaining
Timo Heister [Mon, 23 Feb 2015 21:25:20 +0000 (16:25 -0500)]
sneak in documentation improvements
Timo Heister [Mon, 23 Feb 2015 21:20:40 +0000 (16:20 -0500)]
run wrapcomments
Timo Heister [Mon, 23 Feb 2015 21:06:01 +0000 (16:06 -0500)]
manual edits to doxygen comments
Timo Heister [Mon, 23 Feb 2015 20:58:12 +0000 (15:58 -0500)]
update wrapcomments script
handle many additional doxygen commands correctly
Timo Heister [Mon, 23 Feb 2015 14:58:15 +0000 (09:58 -0500)]
Merge pull request #588 from bangerth/unique_ptr
Add a workaround class for std::unique_ptr that can also be used in the ...
Wolfgang Bangerth [Mon, 23 Feb 2015 14:34:16 +0000 (08:34 -0600)]
Merge pull request #590 from kronbichler/master
Comment on the behavior of FE_TraceQ as compared to FE_Q.
Wolfgang Bangerth [Mon, 23 Feb 2015 14:32:37 +0000 (08:32 -0600)]
Merge pull request #589 from mfherbst/master
Use AlignedVector::size_type in TableBase::position
Martin Kronbichler [Mon, 23 Feb 2015 14:29:36 +0000 (15:29 +0100)]
Comment on the behavior of FE_TraceQ as compared to FE_Q.
Michael F. Herbst [Mon, 23 Feb 2015 14:28:07 +0000 (15:28 +0100)]
Merge remote-tracking branch 'upstream/master'
Michael F. Herbst [Mon, 23 Feb 2015 13:59:05 +0000 (14:59 +0100)]
Use AlignedVector::size_type in TableBase::position
Use AlignedVector<T>::size_type as the integer type to carry out the
calculation of the position of the element indicated by the
TableIndices<N> object.
This has the advantage that a potential integer overflow for Tables
with sizes less than max long int in each dimension, but with more than
max long int entries is prevented.
Martin Kronbichler [Mon, 23 Feb 2015 12:09:42 +0000 (13:09 +0100)]
Merge pull request #585 from kronbichler/master
Enable make_hanging_node_constraints with FE_TraceQ.
Martin Kronbichler [Mon, 23 Feb 2015 12:04:19 +0000 (13:04 +0100)]
Add test case for static condensation with FE_TraceQ.
Martin Kronbichler [Sat, 21 Feb 2015 17:31:45 +0000 (18:31 +0100)]
Enable make_hanging_node_constraints with FE_TraceQ.
Patch by Anton Evgrafov.
Wolfgang Bangerth [Mon, 23 Feb 2015 04:58:52 +0000 (22:58 -0600)]
Add a workaround class for std::unique_ptr that can also be used in the non-C++11 context.
Timo Heister [Mon, 23 Feb 2015 02:33:59 +0000 (21:33 -0500)]
Merge pull request #581 from bangerth/doc-update-8
Better document FESystem.
Timo Heister [Mon, 23 Feb 2015 02:29:52 +0000 (21:29 -0500)]
Merge pull request #587 from bangerth/add-value_type-for-IndexSet
Add a local typedef IndexSet::value_type.
Wolfgang Bangerth [Mon, 23 Feb 2015 00:14:40 +0000 (18:14 -0600)]
Fix typo.
Wolfgang Bangerth [Sun, 22 Feb 2015 22:10:21 +0000 (16:10 -0600)]
Add a local typedef IndexSet::value_type.
Wolfgang Bangerth [Sun, 22 Feb 2015 19:30:41 +0000 (13:30 -0600)]
Merge pull request #584 from drwells/add-data-filter-to-hdf5-test
Add data filter to hdf5 test
Wolfgang Bangerth [Sun, 22 Feb 2015 04:42:14 +0000 (22:42 -0600)]
Merge pull request #586 from tjhei/doc_updates
documentation updates
Timo Heister [Sun, 22 Feb 2015 02:01:33 +0000 (21:01 -0500)]
improve TrilinosWrappers::PreconditionILU documentation
Timo Heister [Sun, 22 Feb 2015 02:00:20 +0000 (21:00 -0500)]
document SparsityPattern::bandwidth
David Wells [Sat, 21 Feb 2015 17:06:44 +0000 (12:06 -0500)]
Improved the style.
I ran astyle (deleted trailing whitespace) and removed some
commented-out code.
David Wells [Sat, 21 Feb 2015 17:04:21 +0000 (12:04 -0500)]
Removed deprecated call to `write_hdf5_parallel`.
This function has been superceded by one with the same name that
requires an additional parameter.
Wolfgang Bangerth [Sat, 21 Feb 2015 02:52:15 +0000 (20:52 -0600)]
Use correct doxygen tag for video lectures.
Wolfgang Bangerth [Fri, 20 Feb 2015 23:38:13 +0000 (17:38 -0600)]
Add a testcase.
Wolfgang Bangerth [Fri, 20 Feb 2015 23:37:41 +0000 (17:37 -0600)]
Figure out and document a way to call the FESystem constructor with multiple base elements.
This is non-trivial in C++ before C++11 and in particular if one wants to
avoid memory leaks.
Wolfgang Bangerth [Fri, 20 Feb 2015 22:30:32 +0000 (16:30 -0600)]
Reference video lectures.
Wolfgang Bangerth [Fri, 20 Feb 2015 22:29:50 +0000 (16:29 -0600)]
Update a few comments.
Timo Heister [Fri, 20 Feb 2015 19:29:57 +0000 (14:29 -0500)]
Merge pull request #520 from Rombur/threads_init
Add set_thread_limit in step-9.
Bruno Turcksin [Fri, 20 Feb 2015 17:14:29 +0000 (11:14 -0600)]
Add set_thread_limit in step-9.
Timo Heister [Fri, 20 Feb 2015 16:32:57 +0000 (11:32 -0500)]
Merge pull request #580 from bangerth/step-7-update
Update the documentation of step-7.
In particular, name the Method of Manufactured Solutions in the documentation.
Wolfgang Bangerth [Fri, 20 Feb 2015 16:21:19 +0000 (10:21 -0600)]
Update the documentation of step-7.
In particular, name the Method of Manufactured Solutions in the documentation.
Wolfgang Bangerth [Thu, 19 Feb 2015 22:01:07 +0000 (16:01 -0600)]
Merge pull request #575 from kronbichler/master
Improve documentation of VectorMemory in Solver
Wolfgang Bangerth [Thu, 19 Feb 2015 21:58:49 +0000 (15:58 -0600)]
Merge pull request #578 from tamiko/fix_bundled_umfpack
CMake: Bugfix: Also install bundled Umfpack headers
Matthias Maier [Thu, 19 Feb 2015 19:17:05 +0000 (20:17 +0100)]
CMake: Bugfix: Also install bundled Umfpack headers
With "sparse_direct.h" directly including "umfpack.h" we also have to make
sure to install the bundled header files...
Bruno Turcksin [Thu, 19 Feb 2015 19:11:44 +0000 (13:11 -0600)]
Merge pull request #577 from tjhei/mpi_file_fixes
fix status parameter in MPI_File_write
Timo Heister [Thu, 19 Feb 2015 18:27:09 +0000 (13:27 -0500)]
fix status parameter in MPI_File_write
Reported by Thomas Geenen. Setting the MPI_Status* parameter to NULL is
not correct and fails with intel MPI. The correct argument to pass is
MPI_STATUS_IGNORE.
Martin Kronbichler [Thu, 19 Feb 2015 13:58:11 +0000 (14:58 +0100)]
Improve documentation
Martin Kronbichler [Thu, 19 Feb 2015 13:40:14 +0000 (14:40 +0100)]
Merge pull request #574 from bangerth/doc-update-7
Improve error messages.
Wolfgang Bangerth [Thu, 19 Feb 2015 13:34:07 +0000 (07:34 -0600)]
Improve error messages.
Using bare exceptions without any explanatory text in so many places was a bad idea right from the beginning, though nobody seems to have noticed for a long time. This replaces a bare exception with an ExcMessage in a couple of places.
Wolfgang Bangerth [Thu, 19 Feb 2015 11:57:12 +0000 (05:57 -0600)]
Merge pull request #572 from kronbichler/cleanup_mapping
Remove a few 1D specializations for mapping.
Martin Kronbichler [Thu, 19 Feb 2015 09:50:55 +0000 (10:50 +0100)]
Remove a few 1D specializations for mapping.
The old implementation of MappingQEulerian only moved the vertices
in 1D, not interior points for higher order mappings. While it might
be debatable how useful non-linear mappings are in 1D, this was not
documented and lead to surprising behavior. By making the class more
general we can actually make our code base simpler because we can remove
a few template specialization and use the same code as in higher
dimensions.
Matthias Maier [Thu, 19 Feb 2015 10:04:37 +0000 (11:04 +0100)]
Merge pull request #566 from kronbichler/fix_vector_memory
Put implementation (Growing)VectorMemory into .templates.h file
Matthias Maier [Thu, 19 Feb 2015 10:03:31 +0000 (11:03 +0100)]
Merge pull request #571 from kronbichler/tensor_vs_point_fix
More Point vs Tensor fixes
Martin Kronbichler [Thu, 19 Feb 2015 07:50:40 +0000 (08:50 +0100)]
More Point vs Tensor fixes
Dakshina Ilangovan [Thu, 19 Feb 2015 02:57:26 +0000 (20:57 -0600)]
modified filtered_matrix tests to work with square matrices only
modified filtered_matrix tests to work with square matrices only
Martin Kronbichler [Wed, 18 Feb 2015 16:35:18 +0000 (17:35 +0100)]
Put implementation (Growing)VectorMemory into .templates.h file
This enables users to provide instantiations of the class for non-deal.II
vector types and their use in Solver classes.
Reported by Karl Ljungkvist.
Martin Kronbichler [Wed, 18 Feb 2015 21:01:43 +0000 (22:01 +0100)]
Merge pull request #568 from bangerth/further-tensor-fixes
More Point vs Tensor fixes.
Guido Kanschat [Wed, 18 Feb 2015 19:50:58 +0000 (20:50 +0100)]
Merge pull request #567 from tamiko/export_umfpack_include_dir
CMake: Bugfix: Also export suitesparse include directories in project config
Wolfgang Bangerth [Wed, 18 Feb 2015 19:42:32 +0000 (13:42 -0600)]
More Point vs Tensor fixes.
Matthias Maier [Wed, 18 Feb 2015 19:31:36 +0000 (20:31 +0100)]
CMake: Bugfix: Also export suitesparse include directories in project config
With
053bb82 a direct include of "umfpack.h" was introduced in
"sparse_direct.h" which makes it necessary for the user to have the umfpack
include directory location available
Closes #564
Wolfgang Bangerth [Wed, 18 Feb 2015 17:00:40 +0000 (11:00 -0600)]
Merge pull request #565 from bangerth/more-tensor-fixes
More changes hoping to make the compiler happy with OpenCascade interfaces again
Wolfgang Bangerth [Wed, 18 Feb 2015 16:29:38 +0000 (10:29 -0600)]
More changes hoping to make the compiler happy with OpenCascade interfaces again.
Wolfgang Bangerth [Wed, 18 Feb 2015 12:47:00 +0000 (06:47 -0600)]
Merge pull request #562 from bangerth/tensor-fixes
Tensor fixes
Wolfgang Bangerth [Wed, 18 Feb 2015 12:46:37 +0000 (06:46 -0600)]
Further fixes to accommodate new Point vs Tensor distinction.
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.