]>
https://gitweb.dealii.org/ - dealii.git/log
Wolfgang Bangerth [Tue, 22 Aug 2017 15:31:47 +0000 (09:31 -0600)]
Merge pull request #4926 from masterleinad/fix_integrators_laplacian_01
Fix integrators/laplacian_01
Daniel Arndt [Tue, 22 Aug 2017 12:33:05 +0000 (14:33 +0200)]
Fix integrators/laplacian_01
Martin Kronbichler [Tue, 22 Aug 2017 06:29:27 +0000 (08:29 +0200)]
Merge pull request #4881 from bangerth/branch-8
Enable hp::DoFHandler on p::d::Triangulation and add tests
Martin Kronbichler [Tue, 22 Aug 2017 06:28:02 +0000 (08:28 +0200)]
Merge pull request #4924 from bangerth/simplify-code
Simplify code that uses variadic templates.
Martin Kronbichler [Tue, 22 Aug 2017 06:25:43 +0000 (08:25 +0200)]
Merge pull request #4925 from bangerth/simplify-pointer-class
Simplify VectorMemory::Pointer.
Wolfgang Bangerth [Mon, 17 Jul 2017 23:17:01 +0000 (17:17 -0600)]
Simplify the VectorMemory::Pointer class.
Specifically, it looks a lot like std::unique_ptr except that it
does something different during destruction. This can easily be
modeled using a custom deleter with std::unique_ptr and avoids
a couple of dozen lines of code.
David Wells [Mon, 21 Aug 2017 23:33:53 +0000 (19:33 -0400)]
Merge pull request #4923 from bangerth/fix-warning
Avoid a warning with gcc 4.8.
Wolfgang Bangerth [Mon, 21 Aug 2017 16:28:12 +0000 (10:28 -0600)]
Update the style of tests.
While there, also update a couple of outdated comments.
Wolfgang Bangerth [Wed, 16 Aug 2017 16:30:49 +0000 (10:30 -0600)]
Add a whole bunch of tests for hp::DoFHandler on p::d::Triangulation.
Wolfgang Bangerth [Sat, 22 Jul 2017 00:56:07 +0000 (18:56 -0600)]
Enable hp::DoFHandler on parallel meshes.
Wolfgang Bangerth [Mon, 21 Aug 2017 23:07:20 +0000 (17:07 -0600)]
Use standard language to call things 'argument' instead of 'parameter'.
Wolfgang Bangerth [Mon, 21 Aug 2017 23:05:11 +0000 (17:05 -0600)]
Simplify code that uses variadic templates.
It turns out that we can do '&args...' to get the addresses of the objects, and then
'{ &args... }' to get a std::initializer_list that we can iterate over. This makes the
code significantly easier to read.
Wolfgang Bangerth [Mon, 21 Aug 2017 22:29:07 +0000 (16:29 -0600)]
Avoid a warning with gcc 4.8.
GCC 4.8 warns about an 'unused argument' for constructors marked '=default'.
This is annoying, but easily worked around.
Wolfgang Bangerth [Mon, 21 Aug 2017 18:55:33 +0000 (12:55 -0600)]
Merge pull request #4893 from jppelteret/tensor_symm_tensor_function_generalisation
Generalisation of tensor and symmetric tensor functions
Jean-Paul Pelteret [Thu, 17 Aug 2017 22:02:35 +0000 (00:02 +0200)]
Added changelog entries for Tensor and SymmetricTensor class changes.
Jean-Paul Pelteret [Thu, 17 Aug 2017 16:54:08 +0000 (18:54 +0200)]
Added test for symmetric tensor operations with generic number types
Jean-Paul Pelteret [Thu, 17 Aug 2017 16:53:36 +0000 (18:53 +0200)]
Added single contraction operator* for symm tensor and tensor args
Jean-Paul Pelteret [Thu, 17 Aug 2017 13:12:17 +0000 (15:12 +0200)]
Generalised symmetric tensor operator /
Jean-Paul Pelteret [Wed, 16 Aug 2017 19:31:56 +0000 (13:31 -0600)]
Fixed symmetric tensor operator * (scalar * symm_ten)
Jean-Paul Pelteret [Wed, 16 Aug 2017 19:35:24 +0000 (13:35 -0600)]
Generalised symmetric tensor operator* (tensor contrac.) on number type
Jean-Paul Pelteret [Wed, 16 Aug 2017 19:31:29 +0000 (13:31 -0600)]
Generalised symmetric tensor operator* (double contract) on number type
Jean-Paul Pelteret [Wed, 16 Aug 2017 19:06:08 +0000 (13:06 -0600)]
Generalised symmetric tensor constructor (from tensor) on number type
Jean-Paul Pelteret [Wed, 16 Aug 2017 09:06:46 +0000 (03:06 -0600)]
Generalised dim 3 rank 4 symmetric tensor invert() function
Jean-Paul Pelteret [Wed, 16 Aug 2017 08:58:47 +0000 (02:58 -0600)]
Generalised symmetric tensor double_contract() on number type
Jean-Paul Pelteret [Wed, 16 Aug 2017 08:57:57 +0000 (02:57 -0600)]
Generalised symmetric tensor scalar product on number type
Jean-Paul Pelteret [Wed, 16 Aug 2017 08:57:03 +0000 (02:57 -0600)]
Generalised tensor linfty_norm() on number type
Jean-Paul Pelteret [Wed, 16 Aug 2017 08:56:36 +0000 (02:56 -0600)]
Generalised tensor l1_norm() on number type
Wolfgang Bangerth [Mon, 21 Aug 2017 16:17:23 +0000 (10:17 -0600)]
Merge pull request #4922 from masterleinad/improve_error_fe_collection
Improve error message in hp::FECollection constructor
Daniel Arndt [Mon, 21 Aug 2017 14:28:50 +0000 (16:28 +0200)]
Improve error message in hp::FECollection constructor
Bruno Turcksin [Mon, 21 Aug 2017 00:33:01 +0000 (20:33 -0400)]
Merge pull request #4902 from davydden/fix_zeor_const_functions
move ConstantFunction and ZeroFunction to Functions:: namespace
Daniel Arndt [Sun, 20 Aug 2017 20:54:56 +0000 (22:54 +0200)]
Merge pull request #4915 from drwells/fix-gsl-compilation
Fix compilation with GSL.
Denis Davydov [Sun, 20 Aug 2017 20:28:25 +0000 (22:28 +0200)]
remove redundant destructors
Wolfgang Bangerth [Sun, 20 Aug 2017 20:24:48 +0000 (14:24 -0600)]
Merge pull request #4911 from drwells/remove-have-sys-types
Remove the check for sys/types.h.
David Wells [Sun, 20 Aug 2017 19:40:26 +0000 (15:40 -0400)]
Fix compilation with GSL.
As of
2d311f0f48 the conversion constructor from Tensor to SymmetricTensor is
explicit.
David Wells [Sun, 20 Aug 2017 19:35:43 +0000 (15:35 -0400)]
Merge pull request #4905 from drwells/timer-mpi-fixes
Timer mpi fixes
Wolfgang Bangerth [Sun, 20 Aug 2017 19:17:40 +0000 (13:17 -0600)]
Merge pull request #4913 from masterleinad/explicit_tensor
Mark conversion constructors of SymmetricTensor and Tensor as 'explicit'
David Wells [Thu, 17 Aug 2017 23:31:09 +0000 (19:31 -0400)]
Give the MPI::MinMaxAvg members of Timer more descriptive names.
Daniel Arndt [Sun, 20 Aug 2017 15:29:12 +0000 (17:29 +0200)]
Changelog entry
Daniel Arndt [Sun, 20 Aug 2017 15:28:30 +0000 (17:28 +0200)]
Improve error message for odd rank SymmetricTensor
Daniel Arndt [Sat, 19 Aug 2017 00:37:30 +0000 (02:37 +0200)]
Mark conversion constructors of SymmetricTensor and Tensor as 'explicit'
Jean-Paul Pelteret [Sun, 20 Aug 2017 15:22:14 +0000 (17:22 +0200)]
Merge pull request #4912 from masterleinad/fix_include_adolc
Fix include directories for ADOLC
Martin Kronbichler [Sun, 20 Aug 2017 13:07:48 +0000 (15:07 +0200)]
Merge pull request #4676 from davydden/mg_block_transfer_simple
MGTransferBlockMatrixFree for LA::d::BlockVector (version 2)
Martin Kronbichler [Sun, 20 Aug 2017 12:54:24 +0000 (14:54 +0200)]
Merge pull request #4909 from masterleinad/improve_fe_collection
Clean up hp::FECollection
Martin Kronbichler [Sun, 20 Aug 2017 12:52:46 +0000 (14:52 +0200)]
Merge pull request #4910 from Rombur/fix_cuda
Fix a bug in CUDA matix-free when there is no constraint
Daniel Arndt [Sun, 20 Aug 2017 12:05:13 +0000 (14:05 +0200)]
Fix include directories for ADOLC
David Wells [Sun, 20 Aug 2017 02:18:41 +0000 (22:18 -0400)]
David Wells [Sat, 19 Aug 2017 23:30:46 +0000 (19:30 -0400)]
Remove the check for sys/types.h.
This was added in
7b34010c81d but apparently never used.
David Wells [Sat, 19 Aug 2017 23:04:08 +0000 (19:04 -0400)]
Merge pull request #4862 from masterleinad/nothrow_exception
Make noexcept for AssertNoThrow more explicit
Bruno Turcksin [Sat, 19 Aug 2017 22:45:12 +0000 (18:45 -0400)]
Fix a bug in CUDA matix-free when there is no constraint
Daniel Arndt [Sat, 19 Aug 2017 22:18:08 +0000 (00:18 +0200)]
Add changelog entry
Daniel Arndt [Sat, 19 Aug 2017 21:41:01 +0000 (23:41 +0200)]
Improve hp::FECollection
Daniel Arndt [Sat, 19 Aug 2017 17:31:22 +0000 (19:31 +0200)]
Merge pull request #4906 from drwells/actually-remove-explicit-constructor-bug
Actually remove DEAL_II_EXPLICIT_CONSTRUCTOR_BUG.
Denis Davydov [Sat, 19 Aug 2017 14:22:32 +0000 (16:22 +0200)]
change inheritance structure of ZeroFunction and ConstantFunction
Denis Davydov [Sat, 19 Aug 2017 13:55:59 +0000 (15:55 +0200)]
replace occurances of ZeroFunction and ConstantFunction in all .cc
Timo Heister [Sat, 19 Aug 2017 12:23:29 +0000 (06:23 -0600)]
Merge pull request #4907 from drwells/remove-wreturn-type-const-qualifier-bug
Remove DEAL_II_WRETURN_TYPE_CONST_QUALIFIER_BUG.
Denis Davydov [Fri, 18 Aug 2017 13:56:07 +0000 (15:56 +0200)]
fix typos
Denis Davydov [Fri, 18 Aug 2017 13:44:35 +0000 (15:44 +0200)]
move ConstantFunction and ZeroFunction to Functions:: namespace
David Wells [Sat, 19 Aug 2017 03:43:24 +0000 (23:43 -0400)]
Remove DEAL_II_WRETURN_TYPE_CONST_QUALIFIER_BUG.
This was a workaround for a bug in gcc 4.3 that has long since been fixed.
David Wells [Sat, 19 Aug 2017 02:51:03 +0000 (22:51 -0400)]
Actually remove DEAL_II_EXPLICIT_CONSTRUCTOR_BUG.
Follow-up to
ac33ee15e72 .
Matthias Maier [Sat, 19 Aug 2017 02:00:44 +0000 (21:00 -0500)]
Merge pull request #4904 from masterleinad/fix_mpi_tests
Fix failing MPI tests
Daniel Arndt [Fri, 18 Aug 2017 17:13:30 +0000 (19:13 +0200)]
Fix failing MPI tests
David Wells [Thu, 17 Aug 2017 23:14:13 +0000 (19:14 -0400)]
Write an '#error' message with a string.
This looks nicer in my editor.
David Wells [Thu, 17 Aug 2017 23:12:58 +0000 (19:12 -0400)]
Avoid mentioning a deprecated function in the docs.
David Wells [Thu, 17 Aug 2017 23:12:37 +0000 (19:12 -0400)]
Clean things up with a delegating constructor call.
David Wells [Thu, 17 Aug 2017 23:11:53 +0000 (19:11 -0400)]
Enable MPI timer features when MPI is not present.
All of the relevant functions have redefinitions in the case that MPI is not
available.
Wolfgang Bangerth [Fri, 18 Aug 2017 20:12:11 +0000 (14:12 -0600)]
Merge pull request #4903 from masterleinad/fix_number_type
Fix Assert with the correct number type
Daniel Arndt [Fri, 18 Aug 2017 18:07:42 +0000 (20:07 +0200)]
More vertex_dof_offset fixes
Daniel Arndt [Fri, 18 Aug 2017 16:34:06 +0000 (18:34 +0200)]
Fix Assert with the correct number type
Martin Kronbichler [Fri, 18 Aug 2017 14:46:43 +0000 (16:46 +0200)]
Merge pull request #4900 from davydden/mf_operator_fix
fix a minor bug in MatrixFreeOperators::Base::vmult_interface_down()
Martin Kronbichler [Fri, 18 Aug 2017 14:43:38 +0000 (16:43 +0200)]
Merge pull request #4901 from davydden/la_parallel_vector/ghost_assertion2
make an Assert in LA::distributed::Vector::compress_finish() less rigid
Denis Davydov [Fri, 4 Aug 2017 08:27:39 +0000 (10:27 +0200)]
add MGTransferBlockMatrixFree
Denis Davydov [Fri, 18 Aug 2017 11:42:11 +0000 (13:42 +0200)]
make an Assert in LA::distributed::Vector::compress_finish() less rigid
Denis Davydov [Tue, 15 Aug 2017 15:05:39 +0000 (17:05 +0200)]
fix a minor bug in MatrixFreeOperators::Base::vmult_interface_down()
Martin Kronbichler [Fri, 18 Aug 2017 10:04:16 +0000 (12:04 +0200)]
Merge pull request #4892 from drwells/rename-timer-variables
Rename timer variables
Denis Davydov [Fri, 18 Aug 2017 05:46:40 +0000 (07:46 +0200)]
Merge pull request #4897 from Rombur/tensor
Fix a bug when using Tensor with CUDA.
Wolfgang Bangerth [Fri, 18 Aug 2017 02:57:55 +0000 (20:57 -0600)]
Merge pull request #4896 from Rombur/dim3
Do not use dim3 for non-CUDA code
Bruno Turcksin [Fri, 18 Aug 2017 02:44:49 +0000 (22:44 -0400)]
Fix a bug when using Tensor with CUDA.
Timo Heister [Fri, 18 Aug 2017 02:37:22 +0000 (20:37 -0600)]
Merge pull request #4895 from Rombur/underscores
Remove double underscores from include guards
Daniel Arndt [Sun, 13 Aug 2017 22:15:48 +0000 (00:15 +0200)]
Treat the nothrow exception handling special
Bruno Turcksin [Thu, 17 Aug 2017 21:23:52 +0000 (17:23 -0400)]
Do not use dim3 for non-CUDA code
dim3 is an array of size 3 defined by CUDA. Remove non CUDA-related dim3
because it creates an error due to conflicting declaration.
Bruno Turcksin [Thu, 17 Aug 2017 21:13:48 +0000 (17:13 -0400)]
Remove double underscores from include guards
Wolfgang Bangerth [Thu, 17 Aug 2017 20:17:18 +0000 (14:17 -0600)]
Merge pull request #4870 from davydden/consecutive_control
minor fix of ConsecutiveControl
Matthias Maier [Thu, 17 Aug 2017 17:25:41 +0000 (12:25 -0500)]
Merge pull request #4891 from tjhei/fix_distr_grid_tools_serialize
fix distributed GridTools::exchange_cell_data_to_ghost serialization …
Timo Heister [Thu, 17 Aug 2017 17:22:25 +0000 (13:22 -0400)]
fix distributed GridTools::exchange_cell_data_to_ghost serialization call
Wolfgang Bangerth [Thu, 17 Aug 2017 16:47:20 +0000 (10:47 -0600)]
Merge pull request #4889 from masterleinad/fix_integrator_tests_fix
Follow up to #4888 fix regex substitution error
David Wells [Thu, 17 Aug 2017 16:46:11 +0000 (12:46 -0400)]
Rename some private Timer variables.
The new names are more descriptive, e.g., the CPU time variables did not mention
that they measured CPU time.
David Wells [Thu, 17 Aug 2017 16:44:17 +0000 (12:44 -0400)]
Improve some of the Timer documentation.
This commit fixes a few small things:
1. The example at the top of the Timer class used a deprecated function; this
example now uses the replacement function.
2. The MPI communicator documentation was written for a different variable.
3. A few sentences have been rewritten to improve clarity.
Daniel Arndt [Thu, 17 Aug 2017 16:43:31 +0000 (18:43 +0200)]
Follow up to #4888 fix regex substitution error
Wolfgang Bangerth [Thu, 17 Aug 2017 16:37:18 +0000 (10:37 -0600)]
Merge pull request #4888 from masterleinad/fix_integrator_tests
Fix test output for integrators tests
Daniel Arndt [Thu, 17 Aug 2017 16:25:47 +0000 (18:25 +0200)]
Fix test output for integrators tests
Wolfgang Bangerth [Thu, 17 Aug 2017 15:30:15 +0000 (09:30 -0600)]
Merge pull request #4887 from kronbichler/fix_tensor_product_matrix_test_02
Fix tensor product matrix test case in float mode.
Martin Kronbichler [Thu, 17 Aug 2017 15:20:26 +0000 (17:20 +0200)]
Fix tensor product matrix test case in float mode.
Wolfgang Bangerth [Thu, 17 Aug 2017 13:22:50 +0000 (07:22 -0600)]
Merge pull request #4884 from drwells/remove-timer-child-measurement
Remove unused child (process) resource usage checks.
Wolfgang Bangerth [Thu, 17 Aug 2017 13:11:51 +0000 (07:11 -0600)]
Merge pull request #4885 from tamiko/lalala
CMake: Bugfix: Use correct cmake variable in external projects
Matthias Maier [Thu, 17 Aug 2017 05:25:37 +0000 (00:25 -0500)]
Add a test
Matthias Maier [Thu, 17 Aug 2017 05:15:00 +0000 (00:15 -0500)]
CMake: Bugfix: Use correct cmake variable in external projects
In macros for external projects we have to use
DEAL_II_USER_DEFINITIONS
instead of
DEAL_II_DEFINITIONS
David Wells [Thu, 17 Aug 2017 04:56:49 +0000 (00:56 -0400)]
Remove unused child (process) resource usage checks.
The option RUSAGE_CHILDREN (see man 2 getrusage) returns resource usage
statistics for terminated children of the current process. Since we don't use
this type of multiprocessing these values are always zero and we never needed to
measure them.
Reverts
5592e89cfa8 .
Bruno Turcksin [Thu, 17 Aug 2017 02:36:42 +0000 (22:36 -0400)]
Merge pull request #4874 from tamiko/fix_cuda_configuration
CMake: Second round of CUDA fixes
David Wells [Thu, 17 Aug 2017 02:20:48 +0000 (22:20 -0400)]
Merge pull request #4882 from masterleinad/fix_boost_1_64_serialize_array
Include correct headers for serialization of std::array
Daniel Arndt [Wed, 16 Aug 2017 22:38:54 +0000 (00:38 +0200)]
Include correct headers for serialization of std::array
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.