[Insight-users] Writing RGBPixel images problem, Bug?
Bradley Lowekamp
blowekamp at mail.nih.gov
Thu Jul 20 10:06:24 EDT 2006
Yes! This is correct. The problem I am encountering is only on
release builds. My dashboard submission yesterday was not a release
build even though it said it was. Xcode build has a configuration
argument that is case sensitive, but will not complain if the
configuration name in unknown. I don't think the CMAKE_BUILD_TYPE
does anything.
CMAKE_BUILD_TYPE:STRING=Release
MAKECOMMAND:STRING=xcodebuild -target ALL_BUILD -configuration RELEASE
Should have been:
MAKECOMMAND:STRING=xcodebuild -target ALL_BUILD -configuration Release
A build was fired off yesterday and it produced the same errors I
posted, but some how it didn't show on the board. A correctly
configured nightly release build is just started running, and should
submit at some point this afternoon.
On Jul 18, 2006, at 6:34 PM, Karthik Krishnan wrote:
> Is it something that happens in release and not debug ?
>
> CMAKE_BUILD_TYPE:STRING=Debug
>
> http://www.itk.org/Testing/Sites/salmon/Darwin-g5-g++4.0.1-static-
> Xcode/20060718-0100-Nightly/Notes.html
>
>
>
>
> Bradley Lowekamp wrote:
>
>> So I built ITK with xcode and no shared libraries. Then ran the
>> tests with the following results:
>>
>> 95% tests passed, 47 tests failed out of 885
>>
>> The following tests FAILED:
>> 202 - itkCenteredRigid2DTransformTest (Failed)
>> 215 - itkEuler2DTransformTest (Failed)
>> 318 - itkVectorImageTest (Failed)
>> 398 - itkInverseDeformationFieldImageFilterTest (Failed)
>> 399 - itkIterativeInverseDeformationFieldImageFilterTest
>> (Failed)
>> 400 - itkDeformationFieldSourceTest (Failed)
>> 541 - itkConnectedComponentImageFilterTest (Failed)
>> 542 - itkConnectedComponentImageFilterTest2 (Failed)
>> 543 - itkConnectedComponentImageFilterTest3 (Failed)
>> 544 - itkMaskConnectedComponentImageFilterTest (Failed)
>> 545 - itkScalarConnectedComponentImageFilterTest (Failed)
>> 568 - itkTileImageFilterTest (Failed)
>> 569 - itkPushPopTileImageFilterTest (Failed)
>> 581 - itkVectorResampleImageFilterTest (Failed)
>> 631 - itkFEMRegistrationFilterTest (Failed)
>> 732 - itkVoronoiPartitioningImageFilterTest1 (Failed)
>> 733 - itkVoronoiPartitioningImageFilterTest2 (Failed)
>> ccmake 754 - itkTransformIOTest (Failed)
>> 760 - itkVectorImageReadWriteTest (Failed)
>> 761 - itkVectorImageReadWriteTest2 (Failed)
>> 764 - itkPolygonGroupSpatialObjectXMLFileTest (SEGFAULT)
>> 769 - itkLSMImageIOTest (Failed)
>> 771 - itkBMPImageIOTest (Failed)
>> 772 - itkBMPImageIOTest2 (Failed)
>> 778 - itkTIFFImageIOTest (Failed)
>> 779 - itkTIFFImageIOTest1 (Failed)
>> 780 - itkTIFFImageIOTest2 (Failed)
>> 781 - itkTIFFImageIOTest3 (Failed)
>> 782 - itkTIFFImageIOTest4 (Failed)
>> 783 - itkTIFFImageIOTest5 (Failed)
>> 784 - itkTIFFImageIOTest6 (Failed)
>> 785 - itkTIFFImageIOTest7 (Failed)
>> 786 - itkTIFFImageIOTest8 (Failed)
>> 787 - itkTIFFImageIOTest9 (Failed)
>> 788 - itkTIFFImageIOTest10 (Failed)
>> 792 - itkImageSeriesWriterTest (Failed)
>> 795 - itkPNGImageIOTest1 (Failed)
>> 796 - itkPNGImageIOTest2 (Failed)
>> 849 - itkNrrdRGBImageReadWriteTest0 (Failed)
>> 850 - itkNrrdRGBImageReadWriteTest1 (Failed)
>> 851 - itkNrrdRGBImageReadWriteTest2 (Failed)
>> 852 - itkNrrdRGBAImageReadWriteTest (Failed)
>> 856 - itkNrrdDiffusionTensor3DImageReadWriteTest (Failed)
>> 858 - itkNrrdComplexImageReadWriteTest (Failed)
>> 860 - itkNrrdVectorImageReadWriteTest (Failed)
>> 862 - itkNrrdCovariantVectorImageReadWriteTest (Failed)
>> 883 - itkSpatialObjectDuplicatorTest (SEGFAULT)
>>
>> I think that this rules out that it is a linking problem or gcc
>> versioning. I am going to work to get a nightly build submitted to
>> the dashboard, so this problem is known.
>>
>> On Jul 17, 2006, at 12:08 PM, Luis Ibanez wrote:
>>
>>>
>>> Hi Bradley,
>>>
>>> Do you have any other versions of the GCC compiler installed
>>> in your machine ?
>>>
>>> It is very likely that you are building the application with a
>>> different compiler that you use for building ITK, and therefore
>>> the typeid signatures of the RGBPixel may be different, despite
>>> that the typeid.name() is the same.
>>>
>>>
>>> Also, are you building ITK as shared libraries ?
>>> if yes, are you sure that you don't have old .so libraries in
>>> your LD_LIBRARY_PATH that may be linking against your current
>>> application ?
>>>
>>>
>>>
>>> Since the error is showing up at run time, it seems to indicate
>>> a mistmatch between the configuration used for builgind the
>>> ImageIOBase.cxx file, and the one used for building your current
>>> application.
>>>
>>>
>>> BTW, Yes, we are missing an XCode nightly submission to the ITK
>>> Dashboard. It will be great if you can enable testing in your
>>> Xcode build of ITK and contribute a submission to the Dashboard.
>>> That will help a lot to identify the source of the problem. The
>>> ITK test suite already includes about 10 different IO tests for
>>> RGBPixel images.
>>>
>>>
>>> Please let us know what you find.
>>>
>>>
>>> Thanks
>>>
>>>
>>> Luis
>>>
>>>
>>>
>>> --------------------------
>>> Bradley Lowekamp wrote:
>>>
>>>> That is what it produced for both compilations when I tried it.
>>>> On Jul 14, 2006, at 5:08 PM, Karthik Krishnan wrote:
>>>>
>>>>> What does the following piece of code compiled on the ppc
>>>>> darwin with xcode and with unix makefiles tell you ?
>>>>>
>>>>> #include "itkRGBPixel.h"
>>>>> #include <iostream>
>>>>> int main() { std::cout << typeid(itk::RGBPixel<unsigned
>>>>> char>).name() << std::endl; return 1; }
>>>>>
>>>>> Is it not "N3itk8RGBPixelIhEE" ?
>>>>>
>>>> ========================================================
>>>> Bradley Lowekamp Management Systems Designers Contractor for
>>>> Office of High Performance Computing and Communications
>>>> National Library of Medicine blowekamp at mail.nih.gov
>>>> <mailto:blowekamp at mail.nih.gov>
>>>> -------------------------------------------------------------------
>>>> -----
>>>> _______________________________________________
>>>> Insight-users mailing list
>>>> Insight-users at itk.org <mailto:Insight-users at itk.org>
>>>> http://www.itk.org/mailman/listinfo/insight-users
>>>
>>>
>>
>> ========================================================
>>
>> Bradley Lowekamp
>> Management Systems Designers Contractor for
>>
>> Office of High Performance Computing and Communications
>>
>> National Library of Medicine
>> blowekamp at mail.nih.gov <mailto:blowekamp at mail.nih.gov>
>>
>>
>> ---------------------------------------------------------------------
>> ---
>>
>> _______________________________________________
>> Insight-users mailing list
>> Insight-users at itk.org
>> http://www.itk.org/mailman/listinfo/insight-users
>>
========================================================
Bradley Lowekamp
Management Systems Designers Contractor for
Office of High Performance Computing and Communications
National Library of Medicine
blowekamp at mail.nih.gov
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://public.kitware.com/pipermail/insight-users/attachments/20060720/5a68e52f/attachment-0001.htm
More information about the Insight-users
mailing list