[Insight-developers] Cygwin tests
Lydia Ng
lng@insightful.com
Fri, 30 Nov 2001 11:29:48 -0800
I blasted by Insight-Cygwin directory and then ran cmake to create new
Makefiles.
I watched it build the itk libraries and itkCurvatureFlowTest
and D_PTHREADS are definitely set in c++ commands.
Here are two different backtraces I got:
======Trace #1=================================
(gdb) run
Starting program:
/cygdrive/e/users/lng/Insight/Insight-Cygwin/Testing/Code/Algorithms/itkCurv
atureFlowTest.exe
Program received signal SIGSEGV, Segmentation fault.
[Switching to thread 2144.0x508]
0x0046075d in itk::DenseFiniteDifferenceImageFilter<itk::Image<float, 2,
itk::DefaultImageTraits<float, 2, itk::ValarrayImageContainer<unsigned long,
float> > >, itk::Image<float, 2, itk::DefaultImageTraits<float, 2,
itk::ValarrayImageContainer<unsigned long, float> > >
>::ThreadedCalculateChange (this=0xa016f40, regionToProcess=@0x1a4af364,
threadId=1) at /usr/include/g++-3/stl_alloc.h:443
Current language: auto; currently c++
(gdb) bt
#0 0x0046075d in itk::DenseFiniteDifferenceImageFilter<itk::Image<float, 2,
itk::DefaultImageTraits<float, 2, itk::ValarrayImageContainer<unsigned long,
float> > >, itk::Image<float, 2, itk::DefaultImageTraits<float, 2,
itk::ValarrayImageContainer<unsigned long, float> > >
>::ThreadedCalculateChange (this=0xa016f40, regionToProcess=@0x1a4af364,
threadId=1) at /usr/include/g++-3/stl_alloc.h:443
#1 0x0043a2ad in itk::DenseFiniteDifferenceImageFilter<itk::Image<float, 2,
itk::DefaultImageTraits<float, 2, itk::ValarrayImageContainer<unsigned long,
float> > >, itk::Image<float, 2, itk::DefaultImageTraits<float, 2,
itk::ValarrayImageContainer<unsigned long, float> > >
>::CalculateChangeThreaderCallback (arg=0xa017010) at
/cygdrive/e/users/lng/Insight/Insight/Code/Common/itkDenseFiniteDifferenceIm
ageFilter.txx:170
#2 0x61063bcc in _libuser32_a_iname ()
#3 0x77e8758a in _libuser32_a_iname ()
===============Trace #2=======================================
(gdb) run
Starting program:
/cygdrive/e/users/lng/Insight/Insight-Cygwin/Testing/Code/Algorithms/itkCurv
atureFlowTest.exe
Program received signal SIGSEGV, Segmentation fault.
0x0046c976 in vector<itk::ImageRegion<2>, allocator<itk::ImageRegion<2> >
>::_M_insert_aux (this=0x22ef64, __position=0xa01fef0, __x=@0x22eef4) at
/usr/include/g++-3/stl_alloc.h:440
Current language: auto; currently c++
(gdb) bt
#0 0x0046c976 in vector<itk::ImageRegion<2>, allocator<itk::ImageRegion<2>
> >::_M_insert_aux (this=0x22ef64, __position=0xa01fef0, __x=@0x22eef4) at
/usr/include/g++-3/stl_alloc.h:440
#1 0x0045c70c in itk::DenseFiniteDifferenceImageFilter<itk::Image<float, 2,
itk::DefaultImageTraits<float, 2, itk::ValarrayImageContainer<unsigned long,
float> > >, itk::Image<float, 2, itk::DefaultImageTraits<float, 2,
itk::ValarrayImageContainer<unsigned long, float> > >
>::ThreadedCalculateChange (this=0xa016f40, regionToProcess=@0x22f004,
threadId=0) at /usr/include/g++-3/stl_vector.h:325
#2 0x0043a2ad in itk::DenseFiniteDifferenceImageFilter<itk::Image<float, 2,
itk::DefaultImageTraits<float, 2, itk::ValarrayImageContainer<unsigned long,
float> > >, itk::Image<float, 2, itk::DefaultImageTraits<float, 2,
itk::ValarrayImageContainer<unsigned long, float> > >
>::CalculateChangeThreaderCallback (arg=0xa016ffc) at
/cygdrive/e/users/lng/Insight/Insight/Code/Common/itkDenseFiniteDifferenceIm
ageFilter.txx:170
#3 0x0041539c in itk::MultiThreader::SingleMethodExecute (this=0xa016fe0)
at
/cygdrive/e/users/lng/Insight/Insight/Code/Common/itkMultiThreader.cxx:375
#4 0x0043a549 in itk::DenseFiniteDifferenceImageFilter<itk::Image<float, 2,
itk::DefaultImageTraits<float, 2, itk::ValarrayImageContainer<unsigned long,
float> > >, itk::Image<float, 2, itk::DefaultImageTraits<float, 2,
itk::ValarrayImageContainer<unsigned long, float> > > >::CalculateChange
(this=0xa016f40) at
/cygdrive/e/users/lng/Insight/Insight/Code/Common/itkDenseFiniteDifferenceIm
ageFilter.txx:136
#5 0x0043d1e8 in itk::FiniteDifferenceImageFilter<itk::Image<float, 2,
itk::DefaultImageTraits<float, 2, itk::ValarrayImageContainer<unsigned long,
float> > >, itk::Image<float, 2, itk::DefaultImageTraits<float, 2,
itk::ValarrayImageContainer<unsigned long, float> > > >::GenerateData
(this=0xa016f40) at
/cygdrive/e/users/lng/Insight/Insight/Code/Common/itkFiniteDifferenceImageFi
lter.txx:94
#6 0x00425850 in itk::ProcessObject::UpdateOutputData (this=0xa016f40) at
/cygdrive/e/users/lng/Insight/Insight/Code/Common/itkProcessObject.cxx:813
#7 0x004125d3 in itk::DataObject::UpdateOutputData (this=0xa0177b8) at
/cygdrive/e/users/lng/Insight/Insight/Code/Common/itkDataObject.cxx:423
#8 0x00411b59 in itk::DataObject::Update (this=0xa0177b8) at
/cygdrive/e/users/lng/Insight/Insight/Code/Common/itkDataObject.cxx:348
#9 0x00424921 in itk::ProcessObject::Update (this=0xa016f40) at
/cygdrive/e/users/lng/Insight/Insight/Code/Common/itkProcessObject.cxx:503
#10 0x00402cb5 in main () at
/cygdrive/e/users/lng/Insight/Insight/Testing/Code/Algorithms/itkCurvatureFl
owTest.cxx:118
#11 0x6100401e in _libuser32_a_iname ()
#12 0x6100421d in _libuser32_a_iname ()
#13 0x6100425c in _libuser32_a_iname ()
#14 0x00428f1f in cygwin_crt0 () at
/cygnus/netrel/src/cygwin-1.3.3-2/winsup/cygwin/lib/cygwin_crt0.c:33
> -----Original Message-----
> From: Lorensen, William E (CRD) [mailto:lorensen@crd.ge.com]
> Sent: Friday, November 30, 2001 11:02 AM
> To: 'lng@insightful.com'; Insight-developers (E-mail)
> Cc: Miller, James V (CRD)
> Subject: RE: [Insight-developers] Cygwin tests
>
>
> Lydia,
> Could you make sure -D_PTHREADS is defined on your command
> lines. Also, can you send a debug
> traceback? If there are line numbers displayed in stl_alloc.h
> that would be great.
>
> Bill
>
> -----Original Message-----
> From: Lydia Ng [mailto:lng@insightful.com]
> Sent: Thursday, July 26, 2001 3:44 PM
> To: Insight-developers (E-mail)
> Subject: [Insight-developers] Cygwin tests
>
>
> Hi,
>
> You guys might already know this ...
> I think the 20+ failure on Cygwin is do
> to multi-threading on a dual processor machine.
>
> Tests run okay on my single proc - but
> seg faults on the dual proc.
> I believe that the GE machine is dual proc.
>
> Does anybody know how to fix this?
> Volunteers?
>
> If not, is there some way for CMake to
> detect that its a dual proc and turn
> multi-threading off.
>
> -------
> Lydia
>
>
>
> _______________________________________________
> Insight-developers mailing list
> Insight-developers@public.kitware.com
> http://public.kitware.com/mailman/listinfo/insight-developers