CGAL users discussion list

[mastair <>]

Sorry, forgot the new Visual leak detector message:

---------- Block 2240060 at 0x059C0528: 28 bytes ----------

  Call Stack:

    c:\program files (x86)\microsoft visual studio 10.0\vc\include\xmemory (36): LoD2plus.exe!std::_Allocate<CGAL::Handle_for<CGAL::Gmpq_rep,std::allocator<CGAL::Gmpq_rep> >::RefCounted> + 0x15 bytes

    c:\program files (x86)\microsoft visual studio 10.0\vc\include\xmemory (187): LoD2plus.exe!std::allocator<CGAL::Handle_for<CGAL::Gmpq_rep,std::allocator<CGAL::Gmpq_rep> >::RefCounted>::allocate + 0xB bytes

    c:\dev\cgal-4.1\include\cgal\handle_for.h (62): LoD2plus.exe!CGAL::Handle_for<CGAL::Gmpq_rep,std::allocator<CGAL::Gmpq_rep> >::Handle_for<CGAL::Gmpq_rep,std::allocator<CGAL::Gmpq_rep> > + 0x4E bytes

    c:\dev\cgal-4.1\include\cgal\gmp\gmpq_type.h (86): LoD2plus.exe!CGAL::Gmpq::Gmpq + 0x4A bytes

    c:\dev\cgal-4.1\include\cgal\point_3.h (81): LoD2plus.exe!CGAL::Point_3<CGAL::Cartesian<CGAL::Gmpq> >::Point_3<CGAL::Cartesian<CGAL::Gmpq> ><int,int,int> + 0x80 bytes

    c:\dev\cgal-4.1\include\cgal\minkowski_sum_3\pointmark.h (35): LoD2plus.exe!CGAL::PointMark<CGAL::Cartesian<CGAL::Gmpq> >::PointMark<CGAL::Cartesian<CGAL::Gmpq> > + 0x5E bytes

    c:\dev\cgal-4.1\include\cgal\nef_s2\sm_decorator.h (275): LoD2plus.exe!CGAL::SM_decorator<CGAL::Sphere_map<CGAL::Sphere_geometry<CGAL::Cartesian<CGAL::Gmpq> >,CGAL::SM_items,CGAL::PointMark<CGAL::Cartesian<CGAL::Gmpq> > > >::new_svertex + 0x52 bytes

    c:\dev\cgal-4.1\include\cgal\minkowski_sum_3\gaussian_map.h (949): LoD2plus.exe!CGAL::Gaussian_map<CGAL::Cartesian<CGAL::Gmpq>,CGAL::Nef_polyhedron_3<CGAL::Cartesian<CGAL::Gmpq>,CGAL::SNC_indexed_items,bool>,CGAL::PointMark<CGAL::Cartesian<CGAL::Gmpq> > >::Gaussian_map<CGAL::Cartesian<CGAL::Gmpq>,CGAL::Nef_polyhedron_3<CGAL::Cartesia + 0x64 bytes

    c:\dev\cgal-4.1\include\cgal\minkowski_sum_3\bipartite_nary_union_sorted_combined.h (182): LoD2plus.exe!CGAL::bipartite_nary_union_sorted_combined<CGAL::Nef_polyhedron_3<CGAL::Cartesian<CGAL::Gmpq>,CGAL::SNC_indexed_items,bool> > + 0x4F bytes

    c:\dev\cgal-4.1\include\cgal\minkowski_sum_3.h (56): LoD2plus.exe!CGAL::minkowski_sum_3<CGAL::Nef_polyhedron_3<CGAL::Cartesian<CGAL::Gmpq>,CGAL::SNC_indexed_items,bool> > + 0x11 bytes

    d:\blabla\building.h (229): LoD2plus.exe!Building::removeExtrudedFacets + 0x14 bytes

    d:\blabla\building.h (251): LoD2plus.exe!Building::splitBuilding

    d:\blabla\lod2plus.cpp (51): LoD2plus.exe!main

    f:\dd\vctools\crt_bld\self_x86\crt\src\crtexe.c (555): LoD2plus.exe!__tmainCRTStartup + 0x19 bytes

    f:\dd\vctools\crt_bld\self_x86\crt\src\crtexe.c (371): LoD2plus.exe!mainCRTStartup

    0x75E88543 (File and line number not available): KERNEL32.DLL!BaseThreadInitThunk + 0xE bytes

    0x77C9AC69 (File and line number not available): ntdll.dll!RtlInitializeExceptionChain + 0x85 bytes

    0x77C9AC3C (File and line number not available): ntdll.dll!RtlInitializeExceptionChain + 0x58 bytes

  Data:

    01 00 00 00    00 00 00 00    78 CD D7 00    01 00 00 00     ........ x.......

    01 00 00 00    98 CD D7 00    01 00 00 00                    ........ ........

2013/2/7 Roeland Boeters <[hidden email]>

The same problem is present using CGAL::Cartesian<CGAL::Gmpq>.

2013/2/7 Philipp Moeller-2 [via cgal-discuss] <[hidden email]>

mastair <[hidden email]> writes:

> So apparently the problem is with BOOST_PP_REPEAT_FROM_TO(1, 9,
> CGAL_CONSTRUCTION_OPERATOR, _) on line 1381 of lazy.h.

That simply expands to a bunch of overloaded operator(). The new'd
pointer is handed off to a ref-counted class.

>
>
> Any idea?

It's not clear to me if Lazy leaks or the constructor wrapped by
Lazy. It would be a good try to run the program with a non-lazy exact
kernel and see if the checker comes up with the same problem.

--
You are currently subscribed to cgal-discuss.
To unsubscribe or access the archives, go to
https://sympa.inria.fr/sympa/info/cgal-discuss

---

If you reply to this email, your message will be added to the discussion below:

http://cgal-discuss.949826.n4.nabble.com/Minkowski-sum-3-memory-leak-tp4656672p4656677.html

To unsubscribe from Minkowski_sum_3 memory leak?, click here.
NAML

---

View this message in context: Re: Minkowski_sum_3 memory leak?
Sent from the cgal-discuss mailing list archive at Nabble.com.