phynexis-deps/cgal
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
98,097 Commits 3 Branches 0 Tags
b9de94de9300a1d0a30de704af1bca7b183acdda
Commit Graph
286 Commits
Author SHA1 Message Date
Clément Jamin b268598665 Replace LINKED_WITH_TBB by CGAL_LINKED_WITH_TBB. 2012-06-13 09:33:08 +00:00
Clément Jamin b948543cc6 Use of a Tag (template parameter) to activate/deactivate concurrency in Mesh_3, instead of #ifdef macros. 2012-06-09 08:52:37 +00:00
Clément Jamin dfbddff185 Bugfixes + some clean-up 2012-05-30 14:49:02 +00:00
Laurent Rineau fd77f801dc Bug fix: g++-4.7 (and C++03/11) requires to qualify the call. 2012-05-21 06:53:58 +00:00
Clément Jamin d2792dff8d - Added the ability to add points on a sphere outside the domain in the sequential case => better performance for the fandisk model (x2). I'm still wondering why... 
- Code refactoring/clean-up
 2012-05-04 15:20:28 +00:00
Clément Jamin 9da17676e9 Clean-up / formatting 2012-05-03 16:37:48 +00:00
Clément Jamin 5a628edd4e - New worksharing technique: we use a coarse version of the mesh to split the space. 
- Some clean-up (removed g_global_mutex)
 2012-05-03 16:05:36 +00:00
Clément Jamin 107a428fb5 - Added the ability to sort work batches according to quality, in order to reduce the number of inserted vertices 
- Modified config parameters manager (Concurrent_mesher_config class)
 2012-05-02 15:21:31 +00:00
Clément Jamin 0eee3dc040 Fixed a deadlock introduced recently. 
Added some assertions in the way...
 2012-04-25 13:00:12 +00:00
Clément Jamin fb7d171271 Oops. Forgot to remove a line. 2012-04-24 16:01:54 +00:00
Clément Jamin 107b268756 - Now, we first start by a little bit of sequential refinement, in order to get a better approximation of the bounding box 
- New worksharing strategy, which groups tasks by batches and let TBB manage the load balancing => the results are slightly improved, compared to parallel_do
- New params in config file
 2012-04-24 15:06:10 +00:00
Clément Jamin fc121a451d - Some clean-up 
- Function get_number_of_bad_elements in the mesher levels (for debugging)
- option CGAL_MESH_3_ADD_OUTSIDE_POINTS_ON_A_FAR_SPHERE to reduce contention on the infinite vertex
 2012-04-24 07:59:29 +00:00
Clément Jamin 9492b2eb84 Added the option to sort the refinement queue after the initial scans (facets and cells), when the queue is an unsorted container. 2012-04-20 09:45:58 +00:00
Clément Jamin 8002220aed Clean-up and bugfixes 2012-04-13 13:57:54 +00:00
Clément Jamin ce132eb34a The concurrency parameters are now stored in a config file which is handled via boost::program_options. 2012-04-12 13:37:15 +00:00
Clément Jamin 9933dd83a4 - Some clean-up: moved global variables (locking/worksharing data structures) to member variables of Mesher_3. 
- Some tests on the worksharing strategies
 2012-04-12 08:33:43 +00:00
Clément Jamin 69272e4d9a Test with a worksharing strategy (based on TBB task scheduler) + test with parallel_do. 2012-04-10 13:23:51 +00:00
Clément Jamin a0a92cdf64 The problem "A facet is not in conflict with its refinement point" (can occur in sequential version too) is now solved => we switch to exact computation (for the dual) when it happens. 
Note: don't try to #define CGAL_MESH_3_DO_NOT_LOCK_INFINITE_VERTEX, it doesn't work yet.
 2012-04-04 07:05:46 +00:00
Clément Jamin a520f3cbf0 Corrected a rare race condition. 
The few bad facets that are created during the "cell refinement" step are treated immediately inside the thread which created them.
 2012-03-29 15:05:28 +00:00
Clément Jamin a3f27eff3c Finally, a fully parallel version of the refinement. Not very efficient, though, but the idea was to identify all data races and to protect it using locks, atomics, TLS... Needs some tests now, to check if we didn't miss any rare data race. 2012-03-26 07:39:58 +00:00
Clément Jamin da923ba811 Forget about yesterday. There was a studid mutex making the whole thing nearly sequential. 
Anyway, now the refinement of the facets is parallel. The speedup is around 2-3 on a 12-core CPU. Can do better, but at least it shows it's not sequential.
The refinement of the cells is not parallel, yet.
One last thing: the program needs to be restarted between two refinements (should be fixed soon).
 2012-03-23 18:05:24 +00:00
Clément Jamin 4ccf8e0447 First working parallel version. It's not faster than the sequential one, but not slower neither (depends if you're optimistic or not). The good news is that we found all (?) the sections to protect or to make thread-local. 2012-03-22 17:07:03 +00:00
Clément Jamin 7137347f95 Added::process_a_batch_of_elements(...) which use parallel_for to insert batch of facets/cells. For now, the body is protected by a scoped_lock to avoid races => no speedup! 2012-03-07 13:57:16 +00:00
Laurent Rineau 8f59fd9592 Turn QPL into LGPLv3+ 2012-01-13 16:33:35 +00:00
Laurent Rineau 9024d5bc04 Fix the text of the assertion msg. Thanks to Sebastien L. 2011-11-16 18:25:18 +00:00
Laurent Rineau 3d1dc999a0 Add an assertion that checks that a point constructed as the midpoint of an 
edge is located either on the edge or inside one of the two incident faces.
 2011-11-16 17:58:22 +00:00
Laurent Rineau 0a363c52b5 Fix a bug introduced in trunk: criteria for Mesh_2 needs a geom traits 2011-06-04 17:00:31 +00:00
Laurent Rineau 5e1de228a1 Do not use systematically a default constructed geom traits 
If one wants to use Mesh_2 with Triangulation_2_projection_traits_3 (a
projection traits constructed with a normal vector), one cannot use a
default constructed geom traits, but only one passed on argument. Then the
criteria class needs to be constructed with tr.geom_traits() as argument.
 2011-03-23 17:45:59 +00:00
Sébastien Loriot ae62b2ae4d remove warnings 2011-02-24 16:36:04 +00:00
Andreas Fabri 43aea3bd18 Remove unused variable 2010-11-12 08:06:17 +00:00
Sébastien Loriot 5c8f3946f4 call to is_conforming_[Gabriel,Delaunay] was no longer possible: fix proposal. 
Add a test into the testsuite
 2010-10-19 14:23:04 +00:00
Stéphane Tayeb 61f6dcbc10 previous data member is now a reference in Null_mesher_visitor_level: visitor used by Mesh_3 as Previous_level has a state, so doing a copy is not an option. 2010-06-09 15:39:04 +00:00
Andreas Fabri 8e47168183 Remove extra ';' 2010-05-17 14:05:54 +00:00
Andreas Fabri 397bc16d9f Remove ';' 2010-05-14 09:13:45 +00:00
Laurent Rineau 49b095589d Strict-ansi compliance 2010-05-08 11:49:52 +00:00
Sébastien Loriot 72ef527fc6 remove extra ; 2010-04-26 09:00:17 +00:00
Laurent Rineau d8e2cfaecd Fix yet another bug of Mesh_2: if a vertex of the CDT has all its incident 
edges in a cluster of constrained edges, then the construction of the
cluster was buggy.
 2010-03-23 14:59:11 +00:00
Laurent Rineau 73c9d4fb3d Determinize Mesh_2 (comparison of Face_handles in Double_map). 2010-03-10 18:05:57 +00:00
Laurent Rineau 1c84cc189b Fix a bug of Mesh_2 that exists for years (since r23141, 2004-10-19)! 2010-03-10 16:46:10 +00:00
Laurent Rineau 274143fa4d - Fix the Mesh_2 test suite: previously it was no longer testing anything! 
- Fix the handling of the cdt with the triangulation plus_2.
- Add tests with CDT_plus_2.
 2010-03-09 15:25:12 +00:00
Stéphane Tayeb a9813aa1ef Fix double map container: pop_next_element_impl now pops the first element of the internal double map. 2010-02-26 10:45:22 +00:00
Stéphane Tayeb 88b3b85630 Fix a bug concerning Boost.Bimap (boost 1.41.0) and _GLIBCXX_PARALLEL g++ flag. 2010-01-27 17:09:55 +00:00
Sylvain Pion 5fda817f95 Remove a few "warning: extra ';'" with g++ -pedantic. 2009-11-14 11:49:52 +00:00
Laurent Rineau c5e779d227 Add a edges mesher level that prevents edges to be refined, and let some bad triangles in the mesh (when their circumcenter encroaches a constrained edge, or when their circumcenter lies outside the meshing domain. 2009-06-24 14:12:12 +00:00
Sylvain Pion 55b427d531 lenght -> length 2009-04-07 19:30:59 +00:00
Andreas Fabri 0879f84971 Don't use deprecated functions internally 2008-11-02 20:43:30 +00:00
Laurent Rineau 324db7bcee Rename variables: add "sq" in their names, so that one know that the 
distances are squared.
 2008-09-26 13:46:54 +00:00
Laurent Rineau a8cdc7ca4a Remove Boost.Bimap FEATURE: it is now hard-coded that bimap is supported iff boost version is >= 1.35 2008-09-24 09:13:57 +00:00
Andreas Fabri 64d8e77840 Overload meshing function for domains selevted by the user 2008-09-24 08:58:02 +00:00
Andreas Fabri 9e667d5865 Put parenthesis around block even if it is only one statement 2008-09-24 08:57:10 +00:00