]> https://gitweb.dealii.org/ - dealii.git/log
dealii.git
10 years agoTestsuite: Fix globbing command for all-headers tests, cleanup code 133/head
Matthias Maier [Wed, 3 Sep 2014 22:07:11 +0000 (00:07 +0200)]
Testsuite: Fix globbing command for all-headers tests, cleanup code

Turns out that globbing together a full path and manually extracting
relative paths is not the best idea. Clean up this mess by using
FILE(GLOB_RECURSE [base path] [...]) that produces nice relative paths.

10 years agoMerge pull request #132 from tamiko/fix_step53
Wolfgang Bangerth [Wed, 3 Sep 2014 22:12:29 +0000 (17:12 -0500)]
Merge pull request #132 from tamiko/fix_step53

Bugfix: step-53 requires enabled DEAL_II_WITH_ZLIB

10 years agoBugfix: step-53 requires enabled DEAL_II_WITH_ZLIB 132/head
Matthias Maier [Wed, 3 Sep 2014 21:57:24 +0000 (23:57 +0200)]
Bugfix: step-53 requires enabled DEAL_II_WITH_ZLIB

step-53 uses boost's <iostreams/filter/gzip.hpp> in order to unpack the
file topography.txt.gz.

If the internal boost variant is used this requires external linkage to
zlib.

10 years agoMerge pull request #131 from bangerth/sort-changelog
Luca Heltai [Wed, 3 Sep 2014 19:51:36 +0000 (21:51 +0200)]
Merge pull request #131 from bangerth/sort-changelog

A couple of entries had ended up in the wrong place (general changes instead of specific changes). While it's debatable where exactly the line is, these items should probably be in the second category. Move them there.

10 years agoMove a couple of entries in the changelog to their correct place. 131/head
Wolfgang Bangerth [Wed, 3 Sep 2014 19:46:23 +0000 (14:46 -0500)]
Move a couple of entries in the changelog to their correct place.

10 years agoMerge pull request #130 from bangerth/fix-up-step-53-2
Timo Heister [Wed, 3 Sep 2014 11:56:58 +0000 (07:56 -0400)]
Merge pull request #130 from bangerth/fix-up-step-53-2

Fix up step 53. This was supposed to be addressed by #127 but the merge of that issue apparently encountered some problems. Here's a respin.

10 years agoFix language. 130/head
Wolfgang Bangerth [Tue, 2 Sep 2014 15:13:44 +0000 (10:13 -0500)]
Fix language.

10 years agoReference the correct picture.
Wolfgang Bangerth [Tue, 2 Sep 2014 15:12:03 +0000 (10:12 -0500)]
Reference the correct picture.

10 years agoFix up a few places that accidentally didn't get changed when downsampling data.
Wolfgang Bangerth [Tue, 2 Sep 2014 14:43:26 +0000 (09:43 -0500)]
Fix up a few places that accidentally didn't get changed when downsampling data.

10 years agoMerge pull request #126 from tamiko/refactor_openmp_simd_setup
Bruno Turcksin [Tue, 2 Sep 2014 19:12:19 +0000 (14:12 -0500)]
Merge pull request #126 from tamiko/refactor_openmp_simd_setup

Refactor openmp simd setup

10 years agoRefactor openmp-simd setup 126/head
Matthias Maier [Tue, 2 Sep 2014 13:41:31 +0000 (15:41 +0200)]
Refactor openmp-simd setup

 * The check for support of -(f)openmp-simd is now done in
   checks/check_01_cpu_features.cmake and guarded by
   DEAL_II_ALLOW_PLATFORM_INTROSPECTION

   The guard is important in order to allow for platform independent
   compilations (as required by Linux distributions)

   Furthermore, this allows for manual override via
   DEAL_II_HAVE_OPENMP_SIMD.

 * The pragma statement is refactored into a macro defined in
   config.h.in. With this, no -Wno-unknown-pragmas workarounds are
   necessary and picky compilers that choke on "#pragma" are happy.

10 years agoSmall bugfixes wrt DEAL_II_HAVE_AVX512
Matthias Maier [Tue, 2 Sep 2014 12:50:03 +0000 (14:50 +0200)]
Small bugfixes wrt DEAL_II_HAVE_AVX512

 - Use the same logic as for DEAL_II_HAVE_(SSE|AVX)
 - Update documentation

10 years agoMerge pull request #129 from bangerth/install-step-53-topography.txt.gz
Timo Heister [Tue, 2 Sep 2014 15:34:37 +0000 (11:34 -0400)]
Merge pull request #129 from bangerth/install-step-53-topography.txt.gz

Install the topography.txt.gz file of step-53.

10 years agoInstall the topography.txt.gz file of step-53. 129/head
Wolfgang Bangerth [Tue, 2 Sep 2014 15:30:35 +0000 (10:30 -0500)]
Install the topography.txt.gz file of step-53.

10 years agoMerge pull request #125 from kronbichler/master
Bruno Turcksin [Tue, 2 Sep 2014 13:44:25 +0000 (08:44 -0500)]
Merge pull request #125 from kronbichler/master

Assert valid matrix row in TrilinosWrappers::SparseMatrix::add/set

10 years agoMerge branch 'master' of https://github.com/dealii/dealii 125/head
Martin Kronbichler [Tue, 2 Sep 2014 13:33:50 +0000 (15:33 +0200)]
Merge branch 'master' of https://github.com/dealii/dealii

10 years agoAdd assertion for row index in TrilinosWrappers::SparseMatrix::set/add
Martin Kronbichler [Mon, 1 Sep 2014 20:13:03 +0000 (22:13 +0200)]
Add assertion for row index in TrilinosWrappers::SparseMatrix::set/add

10 years agoMerge pull request #123 from kronbichler/master
Timo Heister [Tue, 2 Sep 2014 13:28:49 +0000 (09:28 -0400)]
Merge pull request #123 from kronbichler/master

Avoid setting the templated 'Number' to '0' (double) in trace(Tensor<2,dim,Number>)

10 years agoMerge pull request #120 from bangerth/step-53
Matthias Maier [Tue, 2 Sep 2014 08:17:39 +0000 (10:17 +0200)]
Merge pull request #120 from bangerth/step-53

Replace the original attempt at providing step-53 by something real.

In particular, provide a step-53 that shows how to describe geometries
using charts and use a section of Earth under Eastern Africa as an
example. This has about the right level of complexity, and produces
awesome pictures.

Closes #120

10 years agoCleanup stray merge conflict
Matthias Maier [Tue, 2 Sep 2014 08:12:40 +0000 (10:12 +0200)]
Cleanup stray merge conflict

10 years agoClarify where the data comes from.
Wolfgang Bangerth [Fri, 29 Aug 2014 12:32:32 +0000 (07:32 -0500)]
Clarify where the data comes from.

10 years agoAdd a changelog entry.
Wolfgang Bangerth [Fri, 29 Aug 2014 12:00:35 +0000 (07:00 -0500)]
Add a changelog entry.

10 years agoShow some more results.
Wolfgang Bangerth [Thu, 28 Aug 2014 22:45:16 +0000 (17:45 -0500)]
Show some more results.

10 years agoProvide some cross-references to the new step-53.
Wolfgang Bangerth [Thu, 28 Aug 2014 15:59:45 +0000 (10:59 -0500)]
Provide some cross-references to the new step-53.

10 years agoReplace the original attempt at providing step-53 by something real.
Wolfgang Bangerth [Mon, 25 Aug 2014 22:26:40 +0000 (17:26 -0500)]
Replace the original attempt at providing step-53 by something real.

In particular, provide a step-53 that shows how to describe geometries
using charts and use a section of Earth under Eastern Africa as an
example. This has about the right level of complexity, and produces
awesome pictures.

This commit was generated from a number of smaller one:

More text.

Implement the Africa model for the WGS 84 geometry (actual topography still missing).

Provide the actual data set. Provide a way of reading it when not compressed.

Results and finish the program. Much still to be done.

More text.

Finish documenting everything.

Remove a picture we now no longer need.

10 years agoMerge pull request #124 from bangerth/fix-c++11-namespace
Matthias Maier [Tue, 2 Sep 2014 07:25:13 +0000 (09:25 +0200)]
Merge pull request #124 from bangerth/fix-c++11-namespace

Fix an oversight in the previous patch in which we did not put stuff into
the correct namespace.

10 years agoAvoid setting the templated 'Number' to '0' (double) in trace(Tensor<2,dim,Number>). 123/head
Martin Kronbichler [Mon, 1 Sep 2014 20:10:58 +0000 (22:10 +0200)]
Avoid setting the templated 'Number' to '0' (double) in trace(Tensor<2,dim,Number>).

Enables using trace also with e.g. Tensor<2,dim,VectorizedArray<double> >.

10 years agoFix an oversight in the previous patch in which we did not put stuff into the correct... 124/head
Wolfgang Bangerth [Tue, 2 Sep 2014 00:21:20 +0000 (19:21 -0500)]
Fix an oversight in the previous patch in which we did not put stuff into the correct namespace.

10 years agoMerge pull request #111 from bangerth/replace-std_cxx1x-by-std_cxx11
Matthias Maier [Mon, 1 Sep 2014 20:01:55 +0000 (22:01 +0200)]
Merge pull request #111 from bangerth/replace-std_cxx1x-by-std_cxx11

Replace all occurrences of std_cxx1x by std_cxx11.

10 years agoReplace all occurrences of std_cxx1x by std_cxx11 also in tests/. 111/head
Wolfgang Bangerth [Sun, 24 Aug 2014 16:48:45 +0000 (11:48 -0500)]
Replace all occurrences of std_cxx1x by std_cxx11 also in tests/.

10 years agoReplace all occurrences of std_cxx1x by std_cxx11.
Wolfgang Bangerth [Sun, 24 Aug 2014 16:46:25 +0000 (11:46 -0500)]
Replace all occurrences of std_cxx1x by std_cxx11.

Keep the old files and namespace for backward compatibility.

10 years agoMerge pull request #121 from tjhei/clang_boost_warnings
Matthias Maier [Fri, 29 Aug 2014 21:02:15 +0000 (23:02 +0200)]
Merge pull request #121 from tjhei/clang_boost_warnings

Suppress clang warnings when compiling bundled boost 1.56

10 years agoSuppress clang warnings when compiling bundled boost 1.56 121/head
Timo Heister [Fri, 29 Aug 2014 20:25:39 +0000 (16:25 -0400)]
Suppress clang warnings when compiling bundled boost 1.56

Clang generates a bunch of warnings of the type:
bundled/boost-1.56.0/include/boost/preprocessor/tuple/eat.hpp:23:26: warning: variadic macros are a C99 feature [-Wvariadic-macros]
bundled/boost-1.56.0/include/boost/mpl/if.hpp:131:1: warning: empty macro arguments are a C99 feature [-Wc99-extensions]

This patch just disables those warnings. Also see https://code.google.com/p/dealii/issues/detail?id=252

10 years agoMerge pull request #119 from tjhei/fix_petsc_compile
Bruno Turcksin [Wed, 27 Aug 2014 15:17:20 +0000 (10:17 -0500)]
Merge pull request #119 from tjhei/fix_petsc_compile

Fix compilation of Vector when configured with PETSc

10 years agoFix compilation of Vector when configured with PETSc 119/head
Timo Heister [Wed, 27 Aug 2014 15:14:57 +0000 (11:14 -0400)]
Fix compilation of Vector when configured with PETSc

this was introduced in e5f7ade

10 years agoMerge pull request #109 from Rombur/branch_vectorization
Martin Kronbichler [Wed, 27 Aug 2014 11:58:39 +0000 (13:58 +0200)]
Merge pull request #109 from Rombur/branch_vectorization

Vectorization (SIMD) of Vector

10 years agoReplace transform by vectorized_transform. 109/head
Bruno Turcksin [Thu, 21 Aug 2014 23:48:57 +0000 (18:48 -0500)]
Replace transform by vectorized_transform.

Replace initialize_val by allocate and clear_val by deallocate.

Add vectorization to changes.h

10 years agoAlign the underlying array of Vector and add runtime switched between add and vectori...
Bruno Turcksin [Thu, 21 Aug 2014 15:12:54 +0000 (10:12 -0500)]
Align the underlying array of Vector and add runtime switched between add and vectorized add.

10 years agoAdd the first vectorized add and add flags to use OpenMP SIMD pragmas.
Bruno Turcksin [Wed, 20 Aug 2014 22:01:26 +0000 (17:01 -0500)]
Add the first vectorized add and add flags to use OpenMP SIMD pragmas.

10 years agoMerge pull request #118 from luca-heltai/fix_travis_ci
Timo Heister [Mon, 25 Aug 2014 15:24:23 +0000 (11:24 -0400)]
Merge pull request #118 from luca-heltai/fix_travis_ci

Removed all builds. Only indent test.

I implemented here Matthias' suggestion. Removing all unnecessary comments and lines from the Travis CI configuration scripts. This should also make it much faster, since we don't need to install anything anymore (except astyle). This is the one that closes #89.

10 years agoMoved download and install of astyle into before_install section. Renamed files. 118/head
Luca Heltai [Mon, 25 Aug 2014 15:01:48 +0000 (17:01 +0200)]
Moved download and install of astyle into before_install section. Renamed files.

10 years agoRemoved all builds. Only indent test.
Luca Heltai [Mon, 25 Aug 2014 14:28:01 +0000 (16:28 +0200)]
Removed all builds. Only indent test.

10 years agoMerge pull request #117 from luca-heltai/travis_ci_indent_check
Timo Heister [Mon, 25 Aug 2014 14:08:35 +0000 (10:08 -0400)]
Merge pull request #117 from luca-heltai/travis_ci_indent_check

Only test indentation on pull requests. This closes issue #89, where we discussed about possibly testing with Travis CI the indentation of the pull request.

Since most of the pull requests are failing anyway on Travis (time limit for a pull request is 25 minutes, and we hit it pretty regularly), this pull request configures Travis CI to only check indentation in pull requests, and do the full build only when the pull request is merged on master.

10 years agoOnly test indentation on pull requests. 117/head
Luca Heltai [Mon, 25 Aug 2014 14:06:01 +0000 (16:06 +0200)]
Only test indentation on pull requests.

10 years agoMerge pull request #116 from bangerth/add-missing-spirit-lib
Luca Heltai [Mon, 25 Aug 2014 13:35:37 +0000 (15:35 +0200)]
Merge pull request #116 from bangerth/add-missing-spirit-lib

Add back the spirit sublibrary of boost.

10 years agoAdd forgotten header file. 116/head
Wolfgang Bangerth [Mon, 25 Aug 2014 13:15:50 +0000 (08:15 -0500)]
Add forgotten header file.

10 years agoAdd back the spirit sublibrary of boost.
Wolfgang Bangerth [Mon, 25 Aug 2014 13:06:01 +0000 (08:06 -0500)]
Add back the spirit sublibrary of boost.

We apparently need it from the serialization sublibrary.

10 years agoMerge pull request #91 from tjhei/pr_astyle
Wolfgang Bangerth [Mon, 25 Aug 2014 12:55:14 +0000 (07:55 -0500)]
Merge pull request #91 from tjhei/pr_astyle

indent with astyle

10 years agoindent with astyle 91/head
Timo Heister [Mon, 25 Aug 2014 12:48:03 +0000 (08:48 -0400)]
indent with astyle

10 years agoMerge pull request #115 from bangerth/more-manifold-documentation
Wolfgang Bangerth [Mon, 25 Aug 2014 11:48:38 +0000 (06:48 -0500)]
Merge pull request #115 from bangerth/more-manifold-documentation

More manifold documentation

10 years agoAdd another cross reference. 115/head
Wolfgang Bangerth [Mon, 25 Aug 2014 00:22:04 +0000 (19:22 -0500)]
Add another cross reference.

10 years agoProvide a link to the manifold documentation module.
Wolfgang Bangerth [Mon, 25 Aug 2014 00:11:30 +0000 (19:11 -0500)]
Provide a link to the manifold documentation module.

Also provide more discussion and fix a number of oversights in the
current version of the documentation.

10 years agoAdd documentation to the manifold module.
Wolfgang Bangerth [Sun, 24 Aug 2014 23:47:52 +0000 (18:47 -0500)]
Add documentation to the manifold module.

In particular, give an example of why it matters and a few code snippets and pictures
that show how it works.

10 years agoMerge pull request #107 from bangerth/update-boost
Timo Heister [Mon, 25 Aug 2014 11:40:17 +0000 (07:40 -0400)]
Merge pull request #107 from bangerth/update-boost

This is a patch to update boost to the latest version, 1.56.0.

10 years agoMerge pull request #114 from bangerth/fix-default-template-argument
Luca Heltai [Mon, 25 Aug 2014 07:56:07 +0000 (09:56 +0200)]
Merge pull request #114 from bangerth/fix-default-template-argument

Add a default value for spacedim.

This matches all other existing places where we have both
'dim' and 'spacedim' as template arguments.

10 years agoAdd a default value for spacedim. 114/head
Wolfgang Bangerth [Sun, 24 Aug 2014 22:36:04 +0000 (17:36 -0500)]
Add a default value for spacedim.

This matches all other existing places where we have both
'dim' and 'spacedim' as template arguments.

10 years agoMerge pull request #113 from bangerth/fix-test-output
Luca Heltai [Sun, 24 Aug 2014 22:23:17 +0000 (00:23 +0200)]
Merge pull request #113 from bangerth/fix-test-output

Fix test output.

10 years agoFix test output. 113/head
Wolfgang Bangerth [Sun, 24 Aug 2014 21:46:08 +0000 (16:46 -0500)]
Fix test output.

The recent changes to what get_name() returns for finite element objects did not update this output file.
Fix this.

10 years agoMerge pull request #112 from bangerth/doc-fixes-7
Luca Heltai [Sun, 24 Aug 2014 18:27:26 +0000 (20:27 +0200)]
Merge pull request #112 from bangerth/doc-fixes-7

Remove an obscure comment that does not enlighten anything at all.

10 years agoRemove an obscure comment that does not enlighten anything at all. 112/head
Wolfgang Bangerth [Sun, 24 Aug 2014 16:55:20 +0000 (11:55 -0500)]
Remove an obscure comment that does not enlighten anything at all.

10 years agoMerge pull request #108 from tamiko/make_icc_happy
Matthias Maier [Fri, 22 Aug 2014 23:13:33 +0000 (01:13 +0200)]
Merge pull request #108 from tamiko/make_icc_happy

Do not output CG messages in memory_consumption_01

10 years agoDo not output CG messages in memory_consumption_01 108/head
Matthias Maier [Fri, 22 Aug 2014 09:10:44 +0000 (11:10 +0200)]
Do not output CG messages in memory_consumption_01

Do not output the number of cycle and final residual in the test
memory_consumption_01

10 years agoMerge pull request #110 from Rombur/fix_gcc_4_4
Wolfgang Bangerth [Fri, 22 Aug 2014 22:40:52 +0000 (17:40 -0500)]
Merge pull request #110 from Rombur/fix_gcc_4_4

Fix a compilation error with gcc 4.4.

10 years agoFix a compilation error with gcc 4.4. 110/head
Bruno Turcksin [Fri, 22 Aug 2014 22:29:01 +0000 (17:29 -0500)]
Fix a compilation error with gcc 4.4.

10 years agoAdapt a patch from around the internet to fix an issue with gcc 4.6. 107/head
Wolfgang Bangerth [Fri, 22 Aug 2014 02:12:28 +0000 (21:12 -0500)]
Adapt a patch from around the internet to fix an issue with gcc 4.6.

10 years agoUpdate test output.
Wolfgang Bangerth [Fri, 22 Aug 2014 00:06:04 +0000 (19:06 -0500)]
Update test output.

10 years agoImport those parts of BOOST 1.56.0 that we need for deal.II.
Wolfgang Bangerth [Thu, 21 Aug 2014 18:31:29 +0000 (13:31 -0500)]
Import those parts of BOOST 1.56.0 that we need for deal.II.

This includes renaming the existing directory and overwriting every
file with the corresponding one from BOOST 1.56.0. Then removing thos
parts of the boost sublibraries that we don't need but that for
some reasons were included previously. Specifically, this applies to
documentation and build directories in libs/ and to the fusion and
spirit libraries that consist exclusively of header files.

10 years agoMerge pull request #106 from agrayver/optimize_fe_nedelec
Wolfgang Bangerth [Fri, 22 Aug 2014 12:30:57 +0000 (07:30 -0500)]
Merge pull request #106 from agrayver/optimize_fe_nedelec

Optimize construction of high-order FE_Nedelec by moving out some non-essential computations

10 years agoMerge pull request #105 from bangerth/make-cast-explicit
Matthias Maier [Fri, 22 Aug 2014 10:53:34 +0000 (12:53 +0200)]
Merge pull request #105 from bangerth/make-cast-explicit

Use an explicit cast when need a bool from optional<...>.

10 years agoConstruct restriction and prolongation matrices on first request. This reduces constr... 106/head
Alexander Grayver [Thu, 21 Aug 2014 18:52:10 +0000 (20:52 +0200)]
Construct restriction and prolongation matrices on first request. This reduces construction time of high-order FE_Nedelec substantially.

Log changes.

10 years agoUse an explicit cast when need a bool from optional<...>. source/base/parameter_handl... 105/head
Wolfgang Bangerth [Thu, 21 Aug 2014 18:26:15 +0000 (13:26 -0500)]
Use an explicit cast when need a bool from optional<...>. source/base/parameter_handler.cc

The implicit cast was previously allowed, but newer versions of BOOST together with C++11 mark
the cast as 'explicit', and it fails in the current context.

10 years agoMerge pull request #104 from tamiko/cmake_fixes
Wolfgang Bangerth [Thu, 21 Aug 2014 18:04:53 +0000 (13:04 -0500)]
Merge pull request #104 from tamiko/cmake_fixes

Cmake fixes

10 years agoCMake: Remove all "help" targets 104/head
Matthias Maier [Thu, 21 Aug 2014 17:10:48 +0000 (19:10 +0200)]
CMake: Remove all "help" targets

It turns out that CMAKE_PROPERTY is too buggy to be useful. We cannot
set property CMP0037 to the old behavior in order to define a custom "help"
target in a macro without a warning. o_O

10 years agoCMake: Remove non-functional code
Matthias Maier [Thu, 21 Aug 2014 17:14:20 +0000 (19:14 +0200)]
CMake: Remove non-functional code

This actually never worked...

10 years agoMerge pull request #103 from bangerth/move-functions
Luca Heltai [Thu, 21 Aug 2014 13:28:37 +0000 (15:28 +0200)]
Merge pull request #103 from bangerth/move-functions

These functions were previously in GridTools, but conceptually they create
meshes and so should be in GridGenerator. Move them there but leave the
old functions in GridTools with a note that they are now deprecated.

Also structure the documentation of the functions in GridGenerator better.

10 years agoMove create_union_triangulation() and extract_boundary_mesh(). 103/head
Wolfgang Bangerth [Thu, 21 Aug 2014 11:24:06 +0000 (06:24 -0500)]
Move create_union_triangulation() and extract_boundary_mesh().

These functions were previously in GridTools, but conceptually they create
meshes and so should be in GridGenerator. Move them there but leave the
old functions in GridTools with a note that they are now deprecated.

Also structure the documentation of the functions in GridGenerator better.

10 years agoMerge pull request #102 from tamiko/make_icc_happy
Wolfgang Bangerth [Thu, 21 Aug 2014 11:18:07 +0000 (06:18 -0500)]
Merge pull request #102 from tamiko/make_icc_happy

Fix test memory_consumption_01 for icc

10 years agoFix memory_consumption_01 for icc 102/head
Matthias Maier [Thu, 21 Aug 2014 09:04:09 +0000 (11:04 +0200)]
Fix memory_consumption_01 for icc

The comparison file for compiler=Intel=yes was just invalid. Current icc
does not miscompile the test in release mode anymore. So remove all special
comparison files.

10 years agoMerge pull request #88 from guidokanschat/fe_from_name_codim_ready
Matthias Maier [Thu, 21 Aug 2014 08:24:52 +0000 (10:24 +0200)]
Merge pull request #88 from guidokanschat/fe_from_name_codim_ready

FETools::get_fe_from_name() implemented for nonzero codimension

- Existing testbits/get_fe_from_name passes
- Modified all get_name functions to include the spacedim parameter. …
- Added one test for codimension one finite element generation from names.
- Reintroduce get_fe_from_name<dim>() with one template parameter as deprecated
- New function get_fe_by_name<dim,spacedim>()

10 years agoModified all get_name functions to include the spacedim parameter. 88/head
Luca Heltai [Sat, 16 Aug 2014 16:47:09 +0000 (18:47 +0200)]
Modified all get_name functions to include the spacedim parameter.
Added one test for codimension one finite element generation from names.
Reintroduce get_fe_from_name<dim>() with one template parameter as deprecated
New function get_fe_by_name<dim,spacedim>()

10 years agoStart with instantiation of FETools::get_fe_from_name()
Guido Kanschat [Sat, 16 Aug 2014 12:31:47 +0000 (14:31 +0200)]
Start with instantiation of FETools::get_fe_from_name()
Existing testbits/get_fe_from_name passes

10 years agoMerge pull request #101 from bangerth/fix-uninitialized-variable
Timo Heister [Wed, 20 Aug 2014 19:05:05 +0000 (15:05 -0400)]
Merge pull request #101 from bangerth/fix-uninitialized-variable

Fix a warning about undefined behavior.

10 years agoMerge pull request #100 from bangerth/silence-warning-2
Timo Heister [Wed, 20 Aug 2014 19:02:09 +0000 (15:02 -0400)]
Merge pull request #100 from bangerth/silence-warning-2

Silence a warning from gcc 4.9.1. The warning pertained to the fact that we define a function in an anonymous
namespace that is only called from a function inside an #ifdef ... #endif
block. The solution is to move the function into #ifdef ... #endif as well.

This addresses this test result: http://cdash.kyomu.43-1.org/viewBuildError.php?type=1&buildid=11683

10 years agoFix a warning about undefined behavior. 101/head
Wolfgang Bangerth [Wed, 20 Aug 2014 18:54:05 +0000 (13:54 -0500)]
Fix a warning about undefined behavior.

I get an error of the form
  /u/bangerth/p/deal.II/1/dealii/include/deal.II/meshworker/dof_info.h:69:9: runtime error: load of value 216, which is not a valid value for type 'bool'
when running the testsuite with the ASAN framework. This is caused by the DoFInfo
class having a member variable that is left uninitialized in the constructor.
Fix this.

10 years agoSilence a warning from gcc 4.9.1. 100/head
Wolfgang Bangerth [Wed, 20 Aug 2014 18:49:26 +0000 (13:49 -0500)]
Silence a warning from gcc 4.9.1.

The warning pertained to the fact that we define a function in an anonymous
namespace that is only called from a function inside an #ifdef ... #endif
block. The solution is to move the function into #ifdef ... #endif as well.

This addresses this test result: http://cdash.kyomu.43-1.org/viewBuildError.php?type=1&buildid=11683

10 years agoMerge pull request #99 from bangerth/fix-test-signature
Timo Heister [Wed, 20 Aug 2014 18:24:03 +0000 (14:24 -0400)]
Merge pull request #99 from bangerth/fix-test-signature

Rename output of tests that need PETSc, not Trilinos. The tests actually use PETSc, but their output names suggested that they need Trilinos. This was wrong.

10 years agoRename output of tests that need PETSc, not Trilinos. 99/head
Wolfgang Bangerth [Wed, 20 Aug 2014 17:53:09 +0000 (12:53 -0500)]
Rename output of tests that need PETSc, not Trilinos.

10 years agoMerge pull request #95 from luca-heltai/flatten_triangulation
Luca Heltai [Wed, 20 Aug 2014 14:01:10 +0000 (16:01 +0200)]
Merge pull request #95 from luca-heltai/flatten_triangulation

GridGenerator::flatten_triangulation: Gives the ability to copy a flattened version of a Triangulation, with possibly different spacedims.

10 years agoGridGenerator::flatten_triangulation. 95/head
Luca Heltai [Tue, 19 Aug 2014 12:37:08 +0000 (14:37 +0200)]
GridGenerator::flatten_triangulation.

Added GridTools::flatten_triangulation + one test.

Fixed WB comments.

Moved flatten triangulation to GridGenerator.

Fixed TH and WB comments.

Fixed indentation issue.

10 years agoMerge pull request #98 from bangerth/require-newer-external-boost
Timo Heister [Tue, 19 Aug 2014 17:47:47 +0000 (13:47 -0400)]
Merge pull request #98 from bangerth/require-newer-external-boost

Require BOOST 1.48 or later when looking for an external installation. This addresses issue #246.

10 years agoRequire BOOST 1.48 or later when looking for an external installation. 98/head
Wolfgang Bangerth [Tue, 19 Aug 2014 16:18:16 +0000 (11:18 -0500)]
Require BOOST 1.48 or later when looking for an external installation.

10 years agoMerge pull request #96 from bangerth/implement-face-measure
Luca Heltai [Tue, 19 Aug 2014 16:03:54 +0000 (18:03 +0200)]
Merge pull request #96 from bangerth/implement-face-measure

Patch by k3daevin: Implement at least compute the area of planar faces of 3d cells.

10 years agoFurther improve test to avoid square roots. 96/head
Wolfgang Bangerth [Tue, 19 Aug 2014 16:00:58 +0000 (11:00 -0500)]
Further improve test to avoid square roots.

10 years agoPatch by k3daevin: Implement at least compute the area of planar faces of 3d cells.
Wolfgang Bangerth [Mon, 18 Aug 2014 21:47:44 +0000 (16:47 -0500)]
Patch by k3daevin: Implement at least compute the area of planar faces of 3d cells.

10 years agoMerge pull request #97 from tamiko/fix_a_test
Luca Heltai [Tue, 19 Aug 2014 13:47:44 +0000 (15:47 +0200)]
Merge pull request #97 from tamiko/fix_a_test

Remove JobId line so that after normalization the test succeeds.

10 years agoFix test output project_bv_curl_conf_02.output 97/head
Matthias Maier [Tue, 19 Aug 2014 13:38:36 +0000 (15:38 +0200)]
Fix test output project_bv_curl_conf_02.output

Remove JobId line so that after normalization the test succeeds.

10 years agoMerge pull request #94 from bangerth/silence-warning
Luca Heltai [Tue, 19 Aug 2014 13:35:24 +0000 (15:35 +0200)]
Merge pull request #94 from bangerth/silence-warning

Some newer compilers kept warning about an out-of-bounds array access
in Point::operator() but didn't indicate where exactly it came
from. This turned out to be in DataOutBase::write_eps(), a function
that is implemented only for dim==2 but was compiled also for dim==1
and dim==3 and contained this statement: 'pointsvertex'. The patch
reorganizes the code so that the function is only compiled for dim==2
and a more general specialization that does nothing but abort is used
for all other dimensions.

10 years agoSilence a warning. 94/head
Wolfgang Bangerth [Mon, 18 Aug 2014 22:44:55 +0000 (17:44 -0500)]
Silence a warning.

Some newer compilers kept warning about an out-of-bounds array access
in Point<1>::operator() but didn't indicate where exactly it came
from. This turned out to be in DataOutBase::write_eps(), a function
that is implemented only for dim==2 but was compiled also for dim==1
and dim==3 and contained this statement: 'points[vertex](1)'. The patch
reorganizes the code so that the function is only compiled for dim==2
and a more general specialization that does nothing but abort is used
for all other dimensions.

10 years agoMerge pull request #90 from bangerth/fix-more-c++11-auto-variables
Timo Heister [Mon, 18 Aug 2014 11:48:51 +0000 (07:48 -0400)]
Merge pull request #90 from bangerth/fix-more-c++11-auto-variables

Fix a couple more files that accidentally use C++11 auto variables.

10 years agoFix a couple more files that accidentally use C++11 auto variables. 90/head
Wolfgang Bangerth [Mon, 18 Aug 2014 10:27:03 +0000 (05:27 -0500)]
Fix a couple more files that accidentally use C++11 auto variables.


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.