[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Plans for ITK version 4



> On i386, seven tests segfault which I can reproduce on my machine in a
> 32-bit chroot.
and those segfaults seems to go back to FFT facilities:

> As with 4.0.0, it builds and tests OK on my amd64, but has not yet
> been successful on any other architecture [1] although some builds
> have not yet been attempted at this time.

wouldn't it be useful to may be build it in Debug mode by default for now
instead of default Release so failures could be readily troubleshooted?  then
it might provide an ITK knowledgeable person (i.e. not me) clues, such as:

  PlanRigor: FFTW_EXHAUSTIVE (8)
ITKFFTTestDriver: /tmp/buildd/insighttoolkit4-4.1.0/Modules/Filtering/FFT/include/itkFFTWCommon.h:247: static fftwf_plan_s* itk::fftw::Proxy<float>::Plan_dft_r2c(int, const int*, itk::fftw::Proxy<float>::PixelType*, float (*)[2], unsigned int, int, bool): Assertion `plan != __null' failed.
Aborted

Happens on the 2nd invocation of 

 Breakpoint 1, itk::fftw::Proxy<float>::Plan_dft_r2c (rank=1, n=0x86f75a0, in=0x86f9758, out=0x86f8028, flags=24, threads=8, canDestroyInput=false) at /tmp/buildd/insighttoolkit4-4.1.0/Modules/Filtering/FFT/include/itkFFTWCommon.h:242
242             plan = fftwf_plan_dft_r2c(rank,n,in,out,roflags);


ie whenever rank=2 and it is ok with rank=1 

-- 
=------------------------------------------------------------------=
Keep in touch                                     www.onerussian.com
Yaroslav Halchenko                 www.ohloh.net/accounts/yarikoptic


Reply to: