[Insight-developers] Help to debug on LInux64 and Mac 10.5

Alexandre GOUAILLARD hanfei at caltech.edu
Fri Dec 21 12:08:07 EST 2007


Great to see that the code is now working.

There are a lot of type problems in the mesh part of ITK. Everytime I test
an existing filter for bacward compatibility I find some. I'm also sure I'm
missing some too, as I don't test on 64 bits yet (I have a 64 bits machine
coming in january). This part of the toolkit might have been undertested /
underused so far.

I'm going to test all the existing classes in the first 2 quarters 2008. If
any of your 64-to-32 bits warnings are on some Mesh classes, please send
them over.

For the rest, you'd better see with bill , he's better than me.

Thanks for the help.

Happy hollydays.

Alex.



On 12/21/07 8:11 AM, "Sean McBride" <sean at rogue-research.com> wrote:

> On 12/21/07 10:25 AM, Bill Lorensen said:
> 
>> I noticed that the "extra" stuff in 32 bits seems to be all zeroes. Not true
>> in 64 bits.
> 
> I am still curious about what this stuff is...
> 
>> Anyway, I've made changes that restrict the hashing to the Point part of the
>> object.
>> 
>> I just checked in the code. Can you update and tell me if it works? I made
>> changes in Code/Common, Code/Algorithms and Code/Review.
> 
> Tada!  Passes in both 32 and 64, key size now 12 bytes in both cases,
> and the output is identical.
> 
> Although not crucial, this investigation uncovered a couple of 64-to-32
> truncations, and I have attached a small patch for your review.  I think
> it's quite safe, if you agree I/you can commit it.
> 
> Now how about the other 42 failing tests on our dashboard?  :)
> 
> Happy holidays!
> 
> --
> ____________________________________________________________
> Sean McBride, B. Eng                 sean at rogue-research.com
> Rogue Research                        www.rogue-research.com
> Mac Software Developer              Montréal, Québec, Canada




More information about the Insight-developers mailing list