phynexis-deps/cgal
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
87,411 Commits 3 Branches 0 Tags
b5412e7f42daa55bd83da170f981ccbe32e630c9
Commit Graph
815 Commits
Author SHA1 Message Date
Clément Jamin 4adf775814 Bugfix for sequential Mesh_3 using lazy refinement queue. 2012-04-13 15:18:22 +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
Laurent Rineau 95bec3999c Use Unique_has_map instead of std::map (patch from Andreas Fabri) 
The gain on copies of T2 or T3 is the following:


afabri@klimt /cygdrive/c/cgal/next/Triangulation_3/benchmark/Triangulation_3
$ ./copy 1000000 2
36.806 sec

afabri@klimt /cygdrive/c/cgal/next/Triangulation_3/benchmark/Triangulation_3
$ ./copy 1000000 2
4.865 sec

The benchmarck copy.cpp is the following (not commited):

#define  CGAL_T3_HASH
#include <CGAL/Exact_predicates_inexact_constructions_kernel.h>
#include <CGAL/Delaunay_triangulation_3.h>
#include <CGAL/Timer.h>
#include <CGAL/point_generators_3.h>


#include <iostream>
#include <fstream>
#include <string>
#include <vector>

typedef CGAL::Exact_predicates_inexact_constructions_kernel  K;
typedef CGAL::Delaunay_triangulation_3<K> Delaunay;
typedef K::Point_3                                     Point;
typedef CGAL::Creator_uniform_3<double,Point>  Creator;



int main(int argc, char **argv)
{
  int n=1000000;
  int rep=100;
  if (argc>=2)
    n=atoi(argv[1]);
  if (argc>=3)
    rep=atoi(argv[2]);
  std::vector<Point> points;
  points.reserve(n);  
  CGAL::Random_points_in_sphere_3<Point,Creator> g(1);
  CGAL::copy_n( g, n, std::back_inserter(points));
  Delaunay original;
  original.insert(points.begin(),points.end());
  
  double res = 0;
  for (int r=0;r<rep;++r){
    CGAL::Timer t;    t.start();
    Delaunay delaunay=original;
    t.stop();
    res+=t.time();
  }

  std::cout << res << std::endl;
            
  return 0;
}
 2012-03-30 12:37:19 +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 4fdf647428 "Re-tabulation" 2012-03-28 08:30:32 +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 9ae59efaaf Merge from next 2012-02-15 15:58:17 +00:00
Clément Jamin d284cf0e22 Lazy removal of cells from the refinement queue 2012-02-13 11:47:24 +00:00
Sébastien Loriot 498c0a6e1e Make the insert-by-range function more permissive 
The condition is now to be convertible to the point type, pair<point,info>,...
This was the behavior before the insert-by-range-with-info methods were
introduced.
 2012-01-24 17:25:20 +00:00
Clément Jamin 677190c9ec merge from next 2012-01-17 15:26:27 +00:00
Clément Jamin 5c19245264 Merge from next 2012-01-16 19:38:35 +00:00
Laurent Rineau 8f59fd9592 Turn QPL into LGPLv3+ 2012-01-13 16:33:35 +00:00
Laurent Rineau 7df0200a6b Reintegrate /branches/features/Power_test_3-VC_64bit-GF into next. 2011-11-17 17:11:42 +00:00
Laurent Rineau 3089c5d022 Revert that patch. Andreas agreed that he should not have commited that to 
next before the branch ShewchuckKernel-GF is tested in candidates

Reverted commit:
  | ------------------------------------------------------------------------
  | r66076 | afabri | 2011-10-26 21:29:58 +0200 (Wed, 26 Oct 2011) | 1 line
  | Changed paths:
  |    M /branches/next/Triangulation_3/include/CGAL/internal/Static_filters/Power_test_3.h
  | 
  | Split the operator in two functions to avoid an ICE for VC++ 64bit release compilations
  | ------------------------------------------------------------------------
 2011-10-27 13:57:12 +00:00
Andreas Fabri 1c055d6951 Split the operator in two functions to avoid an ICE for VC++ 64bit release compilations 2011-10-26 19:29:58 +00:00
Olivier Devillers bb02600abb better predicates in triangulation traits 2011-10-19 12:43:10 +00:00
Andreas Fabri 706e98cd3b Workaround for ICE on VC++ generating 64bit code 2011-10-14 13:04:06 +00:00
Laurent Rineau b7c33d7b2e Add a comment to mark the end of move_if_no_collision 
The function body of Triangulation_3<Gt,Tds>::move_if_no_collision is so
long that comments should be added to mark blocks ends. For now, mark the
end of the function body.
 2011-08-24 13:39:51 +00:00
Laurent Rineau 1fb7632ec3 Fix a bug in finite_adjacent_vertices when dimension()==1. 
When tr.dimension()==1, tr.finite_adjacent_vertices(...) and
tr.finite_incident_edges(...) returned infinite things.
 2011-07-28 16:53:11 +00:00
Andreas Fabri e8fb926dae Add missing return statements 2011-06-29 14:19:27 +00:00
Sébastien Loriot cc9929b7de qualify boost::bind to avoid ambiguity on windows. 2011-06-08 06:36:55 +00:00
Sébastien Loriot 98ddbf9c7e merge packages Spatial_sorting, Triangulation_2 and Triangulation_3 from branch spatial_sorting-add_info-sloriot 2011-05-24 15:48:31 +00:00
Sébastien Loriot 2c2ce21536 remove commented lines from example. 
int -> size_type
 2011-03-04 07:12:17 +00:00
Sébastien Loriot c1ff39de9a fix documentation from branch (code merged in r60697) 2011-02-28 17:22:38 +00:00
Andreas Fabri be782ec8e3 Use CGAL::abs instead of fabs 2011-02-16 15:53:26 +00:00
Laurent Rineau a2daef39ab Add a comment, that explain a tricky line of code. 
(Actually, seeing the code, I first thought there was a bug. But then I
understood. That is why I think the comment may be is helpful for others).
 2011-02-11 10:43:45 +00:00
Laurent Rineau 7e1f88e0cb Abort operator>> as soon as one of the error flags failbit or badbit is set 
on the stream. That avoids a possible very long loop if the stream format
is not the excepted one (imagine that is >> n sets n to INTMAX...)
 2011-02-08 16:13:38 +00:00
Sébastien Loriot 68c3dfbfa9 add missing default constructor 2011-02-07 11:11:10 +00:00
Sébastien Loriot af6f99ed1a protect code using insert points with info by range using macro 
CGAL_TRIANGULATION_3_DONT_INSERT_RANGE_OF_POINTS_WITH_INFO
 2011-01-10 17:54:57 +00:00
Sébastien Loriot 3181ed9401 merge only the code to insert points with info by range from the branch 
/branches/experimental-packages/Triangulation_insert_with_info/Triangulation_3
 2011-01-10 16:00:51 +00:00
Laurent Rineau eea1755591 Merge the candidate branch of 
https://cgal.geometryfactory.com/CGAL/Members/wiki/Features/Small_Features/Rt::vertices_inside_conflict_zone
in trunk.
 2010-12-21 16:36:04 +00:00
Laurent Rineau 96162627aa Merge the following small features in trunk: 
https://cgal.geometryfactory.com/CGAL/Members/wiki/Features/Small_Features/Remove_cluster
  https://cgal.geometryfactory.com/CGAL/Members/wiki/Features/Small_Features/Structural_filtering

Merge also a partial support of:
  https://cgal.geometryfactory.com/CGAL/Members/wiki/Features/Small_Features/Rt::vertices_inside_conflict_zone
    (vertices_inside_conflict_zone" is added in Regular_triangulation_3,  without documenting it)
 2010-12-10 17:02:10 +00:00
Sébastien Loriot dfa20899ef derecursion of create_star_3: 
I remade a benchmark and finally when using boost::thread_specific_ptr,
using a std::vector is faster than the dedicated vector-class I previously
use.
I also retry using a tuple instead of a dedicated class to store variables
useful for the emulated recursion but tuple is (on my machine) slower.

benchmark results are the following (points are randomly taken in a sphere,
time is in seconds and is the mean of 10 idendical runs)

nb points              |   1000        100000    100000    1000000   10000000
-----------------------------------------------------------------------------
trunk version          | 0.0088       0.0928     0.9865    10.216     102.99 
with vector            | 0.0096       0.0900     0.9817    10.075     102.24 
with vector and tuple  | 0.0100       0.0948     0.9981    10.219     104.58 
recursive version      | 0.0080       0.0936     0.9732    10.141     102.33
 2010-11-24 07:27:04 +00:00
Laurent Rineau 77a2bead80 Add missing svn properties (keywords and eol-style). 2010-09-20 12:12:46 +00:00
Stéphane Tayeb ca4f9f11e8 Add an overload of insert_in_hole() in class Regular_triangulation_3 to ensure that hidden points are properly treated. 2010-08-04 15:36:38 +00:00
Sylvain Pion 8bd594a8be Simplification : the case dim==1 should work for dim==0 as well. 2010-07-27 12:55:09 +00:00
Sylvain Pion fdd4243639 nearest_vertex_in_cell() : handle dimension == 0. 2010-07-27 12:48:53 +00:00
Sébastien Loriot 599b23129e missing using 2010-07-08 09:04:57 +00:00
Stéphane Tayeb 771b6f1b79 Fix bug on g++-4.5 -std=c++0x: use int[3] instead of unsigned char[3] to store vertex indices. 2010-07-07 11:26:48 +00:00
Sébastien Loriot 9567810ef8 add macro CGAL_TDS_USE_OLD_CREATE_STAR_3 in TDS_3.h so that user can 
still use the old recursive version of create_star_3.
 2010-06-28 14:06:14 +00:00
Sébastien Loriot 238f5a8479 change a private into protecter to have access to hidden_tester in derived class 
add a using find_conflicts so that can call T3 version from class inheriting from DT or RT
 2010-06-25 21:13:06 +00:00
Sébastien Loriot bb39c7d194 bug in Facet_extractor: missing * and ++ 2010-06-25 12:58:37 +00:00
Laurent Rineau 0625b6c6bd int -> std::size_t|size_type 2010-06-24 13:27:44 +00:00
Andreas Fabri 1e12152c5f int size_t fixes 2010-06-24 12:50:37 +00:00
Andreas Fabri d185e4152f Finally use ptrdiff_t as return type for insert(beg, end) 2010-06-23 13:35:35 +00:00
Andreas Fabri b456cdb994 int -> size_type/std::size_t 2010-06-21 20:31:19 +00:00