]> https://gitweb.dealii.org/ - dealii.git/log
dealii.git
7 years agoDocument FETools::compute_node_matrix. 3809/head
Wolfgang Bangerth [Mon, 16 Jan 2017 19:36:35 +0000 (12:36 -0700)]
Document FETools::compute_node_matrix.

7 years agoMerge pull request #3797 from bangerth/make-compile
Matthias Maier [Sat, 14 Jan 2017 16:13:58 +0000 (10:13 -0600)]
Merge pull request #3797 from bangerth/make-compile

Make things compile again.

7 years agoMake things compile again. 3797/head
Wolfgang Bangerth [Sat, 14 Jan 2017 16:05:02 +0000 (09:05 -0700)]
Make things compile again.

7 years agoMerge pull request #3791 from bangerth/fix-abf
Martin Kronbichler [Sat, 14 Jan 2017 11:27:23 +0000 (12:27 +0100)]
Merge pull request #3791 from bangerth/fix-abf

Fix the polynomial degree the ABF element reports about itself.

7 years agoMerge pull request #3796 from bangerth/make-function-safer
Martin Kronbichler [Sat, 14 Jan 2017 11:26:20 +0000 (12:26 +0100)]
Merge pull request #3796 from bangerth/make-function-safer

Add a few assertions to FETools::compute_node_matrix().

7 years agoMerge pull request #3795 from bangerth/indent-arguments
Martin Kronbichler [Sat, 14 Jan 2017 11:25:25 +0000 (12:25 +0100)]
Merge pull request #3795 from bangerth/indent-arguments

Indent function arguments to align them.

7 years agoMerge pull request #3794 from bangerth/fix-FE_PolyTensor
Martin Kronbichler [Sat, 14 Jan 2017 11:24:59 +0000 (12:24 +0100)]
Merge pull request #3794 from bangerth/fix-FE_PolyTensor

Guard access to mutable variables with a mutex.

7 years agoMerge pull request #3790 from kronbichler/cleanup_step-37
Martin Kronbichler [Sat, 14 Jan 2017 11:21:39 +0000 (12:21 +0100)]
Merge pull request #3790 from kronbichler/cleanup_step-37

Cleanup step-37

7 years agoMerge pull request #3766 from kronbichler/improve_vectorized_array
Martin Kronbichler [Sat, 14 Jan 2017 11:20:42 +0000 (12:20 +0100)]
Merge pull request #3766 from kronbichler/improve_vectorized_array

Gather and scatter operations in vectorized array

7 years agoAdd a few assertions to FETools::compute_node_matrix(). 3796/head
Wolfgang Bangerth [Sat, 14 Jan 2017 05:17:52 +0000 (22:17 -0700)]
Add a few assertions to FETools::compute_node_matrix().

7 years agoIndent function arguments to align them.include/deal.II/fe/fe.h 3795/head
Wolfgang Bangerth [Sat, 14 Jan 2017 05:14:57 +0000 (22:14 -0700)]
Indent function arguments to align them.include/deal.II/fe/fe.h

7 years agoAdd changelog entry. 3794/head
Wolfgang Bangerth [Sat, 14 Jan 2017 05:13:01 +0000 (22:13 -0700)]
Add changelog entry.

7 years agoGuard access to mutable variables with a mutex.
Wolfgang Bangerth [Sat, 14 Jan 2017 05:09:48 +0000 (22:09 -0700)]
Guard access to mutable variables with a mutex.

7 years agoAdd test. 3791/head
Wolfgang Bangerth [Fri, 13 Jan 2017 22:39:44 +0000 (15:39 -0700)]
Add test.

7 years agoAdd changelog entry.
Wolfgang Bangerth [Fri, 13 Jan 2017 22:32:29 +0000 (15:32 -0700)]
Add changelog entry.

7 years agoFix the polynomial degree the ABF element reports about itself.
Wolfgang Bangerth [Fri, 13 Jan 2017 22:29:50 +0000 (15:29 -0700)]
Fix the polynomial degree the ABF element reports about itself.

As stated in the original paper (http://www-users.math.umn.edu/~arnold/papers/vecquad.pdf),
section 5, and as also stated in our discussion of the PolynomialsABF element, the
ABF space of order 'r' actually contains polynomials of degree 'r+2'. Report this
accurately.

Without this, the computation of embedding matrices fails in 3d because we integrate
the least squares matrix terms with a quadrature formula of too low order.

7 years agoMerge pull request #3788 from masterleinad/check_add_indices_sizes
Martin Kronbichler [Fri, 13 Jan 2017 20:15:52 +0000 (21:15 +0100)]
Merge pull request #3788 from masterleinad/check_add_indices_sizes

Check that add_indices respects index_space_size

7 years agoCleanup step-37 regarding shared_ptr usage. 3790/head
Martin Kronbichler [Fri, 13 Jan 2017 17:47:45 +0000 (18:47 +0100)]
Cleanup step-37 regarding shared_ptr usage.

7 years agoDisable warning about deprecated functionality.
Martin Kronbichler [Fri, 13 Jan 2017 17:47:18 +0000 (18:47 +0100)]
Disable warning about deprecated functionality.

7 years agoAdd additional access method to the underlying shared ptr
Martin Kronbichler [Fri, 13 Jan 2017 17:47:02 +0000 (18:47 +0100)]
Add additional access method to the underlying shared ptr

7 years agoMerge pull request #3786 from Rombur/documentation
Bruno Turcksin [Fri, 13 Jan 2017 18:29:43 +0000 (13:29 -0500)]
Merge pull request #3786 from Rombur/documentation

Augment documentation for Triangulation.

7 years agoCheck that add_indices respects index_space_size 3788/head
Daniel Arndt [Fri, 13 Jan 2017 16:54:27 +0000 (17:54 +0100)]
Check that add_indices respects index_space_size

7 years agoAugment documentation for Triangulation. 3786/head
Bruno Turcksin [Fri, 13 Jan 2017 14:35:28 +0000 (09:35 -0500)]
Augment documentation for Triangulation.

7 years agoMerge pull request #3785 from tamiko/cuda_fixes_3
Bruno Turcksin [Fri, 13 Jan 2017 13:44:47 +0000 (08:44 -0500)]
Merge pull request #3785 from tamiko/cuda_fixes_3

CMake: Various fixes for the cuda configuration

7 years agoCMake: Remove obsolete file 3785/head
Matthias Maier [Fri, 13 Jan 2017 11:13:47 +0000 (05:13 -0600)]
CMake: Remove obsolete file

7 years agoMerge pull request #3763 from elddle/Fix-get_ria_vector
Daniel Arndt [Fri, 13 Jan 2017 10:14:41 +0000 (11:14 +0100)]
Merge pull request #3763 from elddle/Fix-get_ria_vector

Fix a get_ria_vector for FE_BDM

7 years agoMerge pull request #3761 from jppelteret/symm_tensor_contract_3
Matthias Maier [Fri, 13 Jan 2017 09:35:49 +0000 (03:35 -0600)]
Merge pull request #3761 from jppelteret/symm_tensor_contract_3

Add contract3() function with SymmetricTensor arguments.

7 years agoremove forgotten dot
Karl Ljungkvist [Fri, 13 Jan 2017 08:14:02 +0000 (09:14 +0100)]
remove forgotten dot

7 years agoadd a changes file
Matthias Maier [Thu, 12 Jan 2017 17:03:13 +0000 (11:03 -0600)]
add a changes file

7 years agoCMake: Make sure the cuda wrapper only uses -std=c++11
Matthias Maier [Thu, 12 Jan 2017 17:00:30 +0000 (11:00 -0600)]
CMake: Make sure the cuda wrapper only uses -std=c++11

7 years agoCMake: Properly propagate all compiler flags to the cuda wrapper
Matthias Maier [Thu, 12 Jan 2017 16:54:39 +0000 (10:54 -0600)]
CMake: Properly propagate all compiler flags to the cuda wrapper

7 years agoCMake: Rename all object targets to obj_<foo>_(debug|release)
Matthias Maier [Thu, 12 Jan 2017 16:34:42 +0000 (10:34 -0600)]
CMake: Rename all object targets to obj_<foo>_(debug|release)

This change is necessary to please the cuda wrapper nvcc that insists on
exporting the target name as preprocessor definition. Unfortunately, a
period "." is an invalid character and we end up with warnings of the
form

  "missing whitespace after the macro name"

Fix this by renaming all object targets to only contain underscores.

7 years agoCMake: Avoid bogus preprocessor definitions
Matthias Maier [Thu, 12 Jan 2017 16:24:07 +0000 (10:24 -0600)]
CMake: Avoid bogus preprocessor definitions

The macro CUDA_WARP_SRCS expects the target to have a set of
preprocessor definitions that does not contain an empty element.

7 years agoCMake: Fix a typo
Matthias Maier [Thu, 12 Jan 2017 15:46:50 +0000 (09:46 -0600)]
CMake: Fix a typo

7 years agoMerge pull request #3764 from Rombur/shared_ptr
Martin Kronbichler [Fri, 13 Jan 2017 08:01:34 +0000 (09:01 +0100)]
Merge pull request #3764 from Rombur/shared_ptr

Use shared_ptr to store MatrixFree data in MatrixFreeOperators::Base

7 years agoMerge pull request #3778 from tamiko/add_a_boost_check
Daniel Arndt [Thu, 12 Jan 2017 23:12:16 +0000 (00:12 +0100)]
Merge pull request #3778 from tamiko/add_a_boost_check

CMake: Define BOOST_NO_CXX11_HDR_UNORDERED_MAP for bundled boost if <gcc-4.8

7 years agoMerge pull request #3777 from jppelteret/step-18-update_results
Jean-Paul Pelteret [Thu, 12 Jan 2017 22:50:05 +0000 (23:50 +0100)]
Merge pull request #3777 from jppelteret/step-18-update_results

Update results for step-18

7 years agoUpdate results for step-18 3777/head
Jean-Paul Pelteret [Thu, 12 Jan 2017 13:06:26 +0000 (14:06 +0100)]
Update results for step-18

fixes #3757

7 years agoCMake: Refactor macros and augment comment 3778/head
Matthias Maier [Thu, 12 Jan 2017 22:20:22 +0000 (16:20 -0600)]
CMake: Refactor macros and augment comment

7 years agoCMake: Define BOOST_NO_CXX11_HDR_UNORDERED_MAP for bundled boost if <gcc-4.8
Matthias Maier [Thu, 12 Jan 2017 13:26:27 +0000 (07:26 -0600)]
CMake: Define BOOST_NO_CXX11_HDR_UNORDERED_MAP for bundled boost if <gcc-4.8

Closes #3624.

7 years agoMerge pull request #3779 from tamiko/cuda_fixes_1
Wolfgang Bangerth [Thu, 12 Jan 2017 21:47:55 +0000 (14:47 -0700)]
Merge pull request #3779 from tamiko/cuda_fixes_1

CMake: let *_cuda targets depend on *.inst targets

7 years agoFix get_ria_vector for BDM 3763/head
elddle [Thu, 12 Jan 2017 16:46:08 +0000 (11:46 -0500)]
Fix get_ria_vector for BDM

7 years agoMerge pull request #3783 from tjhei/fix_compressed_fe_index_is_active
David Wells [Thu, 12 Jan 2017 16:45:30 +0000 (11:45 -0500)]
Merge pull request #3783 from tjhei/fix_compressed_fe_index_is_active

fix hp fe_index_is_active()

7 years agoMerge pull request #3780 from drwells/doxygen-link-fixes
Wolfgang Bangerth [Thu, 12 Jan 2017 16:40:25 +0000 (09:40 -0700)]
Merge pull request #3780 from drwells/doxygen-link-fixes

Doxygen link fixes

7 years agoMerge pull request #3781 from drwells/analogous-grammar-fix
Wolfgang Bangerth [Thu, 12 Jan 2017 16:10:16 +0000 (09:10 -0700)]
Merge pull request #3781 from drwells/analogous-grammar-fix

Write 'is analogous to', not 'is analogy to'.

7 years agoFix some dead UNV links. 3780/head
David Wells [Thu, 12 Jan 2017 15:08:20 +0000 (10:08 -0500)]
Fix some dead UNV links.

7 years agofix hp fe_index_is_active() 3783/head
Timo Heister [Thu, 12 Jan 2017 15:50:43 +0000 (10:50 -0500)]
fix hp fe_index_is_active()

7 years agoWrite 'is analogous to', not 'is analogy to'. 3781/head
David Wells [Thu, 12 Jan 2017 15:23:53 +0000 (10:23 -0500)]
Write 'is analogous to', not 'is analogy to'.

7 years agoDocumentation update 3761/head
Jean-Paul Pelteret [Tue, 10 Jan 2017 16:21:00 +0000 (17:21 +0100)]
Documentation update

Change ".." to "\ldots" in tensor and symmetric tensor documentation

7 years agoAdd contract3() function with SymmetricTensor arguments.
Jean-Paul Pelteret [Tue, 10 Jan 2017 15:10:26 +0000 (16:10 +0100)]
Add contract3() function with SymmetricTensor arguments.

fixes #2691

7 years agoMerge pull request #3776 from tamiko/fix_spaces_testsuite
Timo Heister [Thu, 12 Jan 2017 15:12:13 +0000 (10:12 -0500)]
Merge pull request #3776 from tamiko/fix_spaces_testsuite

CMake: Simplify run_test.sh and avoid array handling magic

7 years agoFix doxygen formatting of URLs with hyphens.
David Wells [Thu, 12 Jan 2017 15:01:58 +0000 (10:01 -0500)]
Fix doxygen formatting of URLs with hyphens.

The script wrapcomments.py was previously allowed to split lines after hyphens
which results in broken links.

7 years agoCMake: let *_cuda targets depend on *.inst targets 3779/head
Matthias Maier [Thu, 12 Jan 2017 15:06:48 +0000 (09:06 -0600)]
CMake: let *_cuda targets depend on *.inst targets

7 years agoUse shared_ptr to store MatrixFree data in MatrixFreeOperators::Base 3764/head
Bruno Turcksin [Tue, 10 Jan 2017 23:08:04 +0000 (18:08 -0500)]
Use shared_ptr to store MatrixFree data in MatrixFreeOperators::Base

7 years agoCMake: Update DEAL_II_ADD_TEST macro 3776/head
Matthias Maier [Thu, 12 Jan 2017 11:47:55 +0000 (05:47 -0600)]
CMake: Update DEAL_II_ADD_TEST macro

7 years agoCMake: Simplify run_test.sh and avoid array handling magic
Matthias Maier [Thu, 12 Jan 2017 11:41:43 +0000 (05:41 -0600)]
CMake: Simplify run_test.sh and avoid array handling magic

7 years agoMerge pull request #3775 from kronbichler/fix_test_1
Matthias Maier [Thu, 12 Jan 2017 11:43:12 +0000 (05:43 -0600)]
Merge pull request #3775 from kronbichler/fix_test_1

Fix test output

7 years agoFix test output 3775/head
Martin Kronbichler [Thu, 12 Jan 2017 07:56:29 +0000 (08:56 +0100)]
Fix test output

7 years agoMerge pull request #3773 from kronbichler/mg_transfer_mf_legendre
Martin Kronbichler [Thu, 12 Jan 2017 05:58:36 +0000 (06:58 +0100)]
Merge pull request #3773 from kronbichler/mg_transfer_mf_legendre

Allow for FE_DGQLegendre in MGTransferMatrixFree

7 years agoMerge pull request #3771 from bangerth/fix-face-domination
Martin Kronbichler [Thu, 12 Jan 2017 05:48:50 +0000 (06:48 +0100)]
Merge pull request #3771 from bangerth/fix-face-domination

Fix face domination between FE_Q and FE_DGQ.

7 years agoAdd changelog entry. 3771/head
Wolfgang Bangerth [Wed, 11 Jan 2017 20:19:50 +0000 (13:19 -0700)]
Add changelog entry.

7 years agoAdd test.
Wolfgang Bangerth [Wed, 11 Jan 2017 20:19:47 +0000 (13:19 -0700)]
Add test.

7 years agoFix face domination between FE_Q and FE_DGQ.
Wolfgang Bangerth [Wed, 11 Jan 2017 20:18:32 +0000 (13:18 -0700)]
Fix face domination between FE_Q and FE_DGQ.

Right now, one got an answer if one asked the FE_DGQ whether or not
it dominates the FE_Q, but not the other way around. Make this symmetric.

While there, also include FE_DGP in the test.

7 years agoMerge pull request #3770 from kronbichler/introduce_fe_dgq_legendre
Daniel Arndt [Thu, 12 Jan 2017 00:40:43 +0000 (01:40 +0100)]
Merge pull request #3770 from kronbichler/introduce_fe_dgq_legendre

Introduce FE_DGQLegendre and FE_DGQHermite

7 years agoMerge pull request #3772 from bangerth/fix-type-mismatch
David Wells [Thu, 12 Jan 2017 00:09:45 +0000 (19:09 -0500)]
Merge pull request #3772 from bangerth/fix-type-mismatch

Fix missing 'spacedim' template argument.

7 years agoMerge pull request #3768 from masterleinad/test_fe_mf
Daniel Arndt [Wed, 11 Jan 2017 22:15:06 +0000 (23:15 +0100)]
Merge pull request #3768 from masterleinad/test_fe_mf

Implement function to check for MatrixFree support

7 years agoMerge pull request #3769 from kronbichler/fix_parallel_vector_comm
Martin Kronbichler [Wed, 11 Jan 2017 22:04:15 +0000 (23:04 +0100)]
Merge pull request #3769 from kronbichler/fix_parallel_vector_comm

Change communication pattern in parallel vector.

7 years agoAllow for FE_DGQLegendre in MGTransferMatrixFree 3773/head
Martin Kronbichler [Wed, 11 Jan 2017 21:58:42 +0000 (22:58 +0100)]
Allow for FE_DGQLegendre in MGTransferMatrixFree

7 years agoList incompatible change 3770/head
Martin Kronbichler [Wed, 11 Jan 2017 18:14:45 +0000 (19:14 +0100)]
List incompatible change

7 years agoAdd another test
Martin Kronbichler [Wed, 11 Jan 2017 18:14:34 +0000 (19:14 +0100)]
Add another test

7 years agoIntroduce FE_DGQLegendre and FE_DGQHermite
Martin Kronbichler [Wed, 11 Jan 2017 17:49:02 +0000 (18:49 +0100)]
Introduce FE_DGQLegendre and FE_DGQHermite

7 years agoAdd changelog entry. 3772/head
Wolfgang Bangerth [Wed, 11 Jan 2017 20:27:08 +0000 (13:27 -0700)]
Add changelog entry.

7 years agoFix missing 'spacedim' template argument.
Wolfgang Bangerth [Wed, 11 Jan 2017 20:26:56 +0000 (13:26 -0700)]
Fix missing 'spacedim' template argument.

7 years agoMerge pull request #3755 from jppelteret/fix_step-18_rotation_matrix
Bruno Turcksin [Wed, 11 Jan 2017 19:55:17 +0000 (14:55 -0500)]
Merge pull request #3755 from jppelteret/fix_step-18_rotation_matrix

Fix rotation matrix definition in step-18

7 years agoMerge pull request #3767 from kronbichler/chebyshev_common_vector_interface
Daniel Arndt [Wed, 11 Jan 2017 17:40:21 +0000 (18:40 +0100)]
Merge pull request #3767 from kronbichler/chebyshev_common_vector_interface

Use common vector interface for Chebyshev smoother

7 years agoImplement function to check for MatrixFree support 3768/head
Daniel Arndt [Mon, 9 Jan 2017 13:44:33 +0000 (14:44 +0100)]
Implement function to check for MatrixFree support

7 years agoChange communication pattern in parallel vector. 3769/head
Martin Kronbichler [Wed, 11 Jan 2017 17:29:33 +0000 (18:29 +0100)]
Change communication pattern in parallel vector.

7 years agoGather and scatter operations in vectorized array 3766/head
Martin Kronbichler [Wed, 11 Jan 2017 10:46:30 +0000 (11:46 +0100)]
Gather and scatter operations in vectorized array

7 years agoUpdate step-18 to use rotation matrix defined in physics module. 3755/head
Jean-Paul Pelteret [Mon, 9 Jan 2017 18:15:08 +0000 (19:15 +0100)]
Update step-18 to use rotation matrix defined in physics module.

7 years agoAdded functions to compute rotation matrices in 2d and 3d.
Jean-Paul Pelteret [Mon, 9 Jan 2017 17:06:18 +0000 (18:06 +0100)]
Added functions to compute rotation matrices in 2d and 3d.

7 years agoUse common vector interface for Chebyshev smoother 3767/head
Martin Kronbichler [Wed, 11 Jan 2017 14:58:28 +0000 (15:58 +0100)]
Use common vector interface for Chebyshev smoother

7 years agoMerge pull request #3754 from bangerth/fix-abf-polynomials-pointer
David Wells [Wed, 11 Jan 2017 13:48:54 +0000 (08:48 -0500)]
Merge pull request #3754 from bangerth/fix-abf-polynomials-pointer

Do not use a plain pointer in PolynomialsABF.

7 years agoAdd changelog entry. 3754/head
Wolfgang Bangerth [Tue, 10 Jan 2017 02:47:48 +0000 (19:47 -0700)]
Add changelog entry.

7 years agoDo not use a plain pointer in PolynomialsABF.
Wolfgang Bangerth [Mon, 9 Jan 2017 15:52:15 +0000 (08:52 -0700)]
Do not use a plain pointer in PolynomialsABF.

These objects are being copied by the FE_Poly* classes, so plain pointers without
dedicated copy constructors and operators are likely going to lead to memory
corruption. It's not clear to me how this ever worked, but it's easy to fix.

7 years agoMerge pull request #3762 from drwells/make-member-const
Bruno Turcksin [Tue, 10 Jan 2017 18:52:42 +0000 (13:52 -0500)]
Merge pull request #3762 from drwells/make-member-const

Make a member variable const.

7 years agoMake a member variable const. 3762/head
David Wells [Tue, 10 Jan 2017 16:10:38 +0000 (11:10 -0500)]
Make a member variable const.

This is copied from an input parameter and never changed so it may as
well be marked as const.

7 years agoMerge pull request #3758 from Rombur/reinit
Wolfgang Bangerth [Tue, 10 Jan 2017 13:36:23 +0000 (06:36 -0700)]
Merge pull request #3758 from Rombur/reinit

Do not use deprecated reinit function.

7 years agoMerge pull request #3760 from masterleinad/fix_filename
Martin Kronbichler [Tue, 10 Jan 2017 09:45:45 +0000 (10:45 +0100)]
Merge pull request #3760 from masterleinad/fix_filename

Rename misnamed file

7 years agoRename misnamed file 3760/head
Daniel Arndt [Tue, 10 Jan 2017 09:43:17 +0000 (10:43 +0100)]
Rename misnamed file

7 years agoDo not use deprecated reinit function. 3758/head
Bruno Turcksin [Tue, 10 Jan 2017 02:21:15 +0000 (21:21 -0500)]
Do not use deprecated reinit function.

7 years agoMerge pull request #3752 from bangerth/fix-abf-in-3d
Bruno Turcksin [Tue, 10 Jan 2017 02:20:20 +0000 (21:20 -0500)]
Merge pull request #3752 from bangerth/fix-abf-in-3d

Fix PolynomialsABF::compute_n_pols() in 3d.

7 years agoMerge pull request #3749 from jppelteret/fix_tests_01
Daniel Arndt [Mon, 9 Jan 2017 20:42:31 +0000 (21:42 +0100)]
Merge pull request #3749 from jppelteret/fix_tests_01

Redo output for physics/step-44* tests

7 years agoMerge pull request #3735 from guidokanschat/tangential_laplace
Daniel Arndt [Mon, 9 Jan 2017 20:41:36 +0000 (21:41 +0100)]
Merge pull request #3735 from guidokanschat/tangential_laplace

tangential jumps for Laplacian

7 years agoMerge pull request #3753 from bangerth/make-var-const
Daniel Arndt [Mon, 9 Jan 2017 20:36:36 +0000 (21:36 +0100)]
Merge pull request #3753 from bangerth/make-var-const

Make a member variable 'const'.

7 years agobe a bit more explicit 3735/head
Guido Kanschat [Mon, 9 Jan 2017 06:15:35 +0000 (07:15 +0100)]
be a bit more explicit

7 years agoMerge pull request #3691 from drwells/constraint-petsc-vector-fast-path
Martin Kronbichler [Mon, 9 Jan 2017 16:53:12 +0000 (17:53 +0100)]
Merge pull request #3691 from drwells/constraint-petsc-vector-fast-path

Add a faster code path for PETSc vectors.

7 years agoMake a member variable 'const'. 3753/head
Wolfgang Bangerth [Mon, 9 Jan 2017 15:16:54 +0000 (08:16 -0700)]
Make a member variable 'const'.

We can do so because it is initialized in the constructor.

7 years agoAdd changelog entry. 3752/head
Wolfgang Bangerth [Mon, 9 Jan 2017 15:10:28 +0000 (08:10 -0700)]
Add changelog entry.

7 years agoBulk write vector values in distribute_local_to_global. 3691/head
David Wells [Wed, 21 Dec 2016 17:28:48 +0000 (12:28 -0500)]
Bulk write vector values in distribute_local_to_global.

This patch lowers the total number of calls to VecSetValues in
ConstraintMatrix::distribute_local_to_global by writing new vector values to a
temporary array before sending them off to PETSc. There are no notable
performance improvements for the other vector classes (though this certainly
should not hurt).

I used step-40 to check that the performance improved. Here is the (filtered)
output with this patch:

[drwells@archway step-40]$ make run | grep assembly
| assembly                        |         1 |   0.00966s |       8.7% |
| assembly                        |         1 |    0.0122s |       6.4% |
| assembly                        |         1 |    0.0217s |       5.9% |
| assembly                        |         1 |    0.0431s |       5.6% |
| assembly                        |         1 |    0.0848s |       5.6% |
| assembly                        |         1 |     0.173s |       5.6% |
| assembly                        |         1 |     0.351s |       5.5% |
| assembly                        |         1 |     0.729s |       5.5% |

and on master:

[drwells@archway step-40]$ make run | grep assembly
| assembly                        |         1 |   0.00672s |       8.9% |
| assembly                        |         1 |    0.0133s |       7.2% |
| assembly                        |         1 |    0.0258s |       6.9% |
| assembly                        |         1 |    0.0513s |       6.8% |
| assembly                        |         1 |     0.101s |       6.6% |
| assembly                        |         1 |     0.203s |       6.5% |
| assembly                        |         1 |     0.414s |       6.4% |
| assembly                        |         1 |      0.93s |       6.9% |

so it appears that the whole assembly process takes about 10% less time.

7 years agoAdd an 'add' method to the native parallel vectors.
David Wells [Wed, 28 Dec 2016 17:39:22 +0000 (12:39 -0500)]
Add an 'add' method to the native parallel vectors.

This is necessary for an upcoming patch that uses the method inside
ConstraintMatrix::distribute_local_to_global.


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.