[Insight-users] Re: Insight-users Digest, Vol 34, Issue 42

[Simon Warfield]

insight-users-request at itk.org wrote:
 >
 > Message: 2
 > Date: Wed, 14 Feb 2007 15:21:03 +0000
 > From: Florian Pierron <F.Pierron at exeter.ac.uk>
 > Subject: Re: [Insight-users] Resampling 64-bit crash
 > To: Thomas Lambertz <thomas at hexerei-software.de>,
 >     insight-users at itk.org
 > Message-ID: <45D328DF.9010502 at exeter.ac.uk>
 > Content-Type: text/plain; charset=ISO-8859-1; format=flowed
 >
 > Hi Thomas,
 >
 > Thank you for your answer! Yes, I agree with you the allocated memory 
seems to be modulo 4GB as you said and that's why I was wondering if 
there's a problem in ITK to go over this 4GB limit of 32-bit 
applications. Thank you for your input, maybe my question should be 
reformulated to: "Does anyone know if ITK support 64-bit addressing?"
 >
 > Thanks again,
 > Florian
 >  

In order to allocate more than 2GB of memory per alloc, you can use a 64 
bit machine, and ITK code compiled to support large allocation.  It is 
not enough that the pointers are 64 bits, the memory model needs to be 
appropriate too.

For example, with GCC, on x86_64 you would compile with -mcmodel=medium
http://gcc.gnu.org/onlinedocs/gcc-4.1.2/gcc/i386-and-x86_002d64-Options.html#index-mcmodel_003dmedium-1052



-- 
Simon K. Warfield, Ph.D.   simon.warfield at childrens.harvard.edu
Associate Professor of Radiology,     Harvard Medical School
Director, Computational Radiology Laboratory,
Department of Radiology, Wolbach 309,  Children's Hospital Boston
300 Longwood Avenue Boston MA 02115
Phone: 617-355-4566