[Insight-developers] Failing test on SunOS-CC-i386

[Bradley Lowekamp]

Hello all,

With ITKv4 are we going to re-intorduce compiler dependent ifdefs where needed or should we introduce CMAKE checks to detect the compiler "feature"?

Bellow is a discussion where this issue has already come up, with trying to get a Sun test to pass.

http://itk.org/gitweb?p=ITK.git;a=commitdiff;h=a162e545278fce7bba0fb698e7645038e0e495f5

Should we use the compiler ifdef or introduce a cmake/itkConfigure define?


On Aug 17, 2010, at 9:28 AM, Gaëtan Lehmann wrote:

> 
> Le 17 août 10 à 15:18, Bradley Lowekamp a écrit :
> 
>> Thanks,
>> 
>> This looks like a very odd bug with this compiler, and it might change in the next minor version.
> 
> yes, that's odd
> 
>> 
>> Do you think we should try to check for this with CMake?
> 
> At this stage, I guess we are supposed to clean things up, not to add more special cases, so I'm not so pleased with my fix.
> Do you think it would be cleaner with a check in CMake?
> 
> Gaëtan
> 
> 
>> 
>> Brad
>> 
>> On Aug 17, 2010, at 8:04 AM, Gaëtan Lehmann wrote:
>> 
>>> 
>>> Le 16 août 10 à 14:59, Bradley Lowekamp a écrit :
>>> 
>>>> Gaetan,
>>>> 
>>>> It looks like after this commit:
>>>> 
>>>> http://itk.org/gitweb?p=ITK.git;a=commitdiff;h=9e3c91c911371b99bc77d518e0b19ccd2732e511;hp=c6d83112bd10bff2a465244aa2999b2a8bb3586b
>>>> 
>>>> This test began failing on your system:
>>>> 
>>>> http://www.cdash.org/CDash/testDetails.php?test=62524993&build=693360
>>>> 
>>>> It looks like it is segment faulting when it is trying to print along double! Which really doesn't make since to me. Do you have any insight as to what is going on here?
>>>> 
>>> 
>>> Hi Brad,
>>> 
>>> It actually segfaults while trying to build a long double. I don't know why.
>>> As a workaround, I've set double as the PrintType of long double for that platform.
>>> 
>>> Gaëtan
>>> 
>>> -- 
>>> Gaëtan Lehmann
>>> Biologie du Développement et de la Reproduction
>>> INRA de Jouy-en-Josas (France)
>>> tel: +33 1 34 65 29 66    fax: 01 34 65 29 09
>>> http://voxel.jouy.inra.fr  http://www.itk.org
>>> http://www.mandriva.org  http://www.bepo.fr
>>> 



========================================================
Bradley Lowekamp  
Lockheed Martin 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://www.itk.org/mailman/private/insight-developers/attachments/20100817/00a7e294/attachment.htm>