From bug-octave-request at bevo dot che dot wisc dot edu  Tue Jan 25 03:44:52 2000
Subject: Lack of memory -> segmentation fault using svd()
From: "John W. Eaton" <jwe at bevo dot che dot wisc dot edu>
To: Markus Svensen <svensen at cns dot mpg dot de>
Cc: bug-octave at bevo dot che dot wisc dot edu
Date: Tue, 25 Jan 2000 03:44:52 -0600 (CST)

On 10-Aug-1999, Markus Svensen <svensen at cns dot mpg dot de> wrote:

|  Bug report for Octave 2.0.14 configured for mips-sgi-irix6.4
| 
| Description:
| -----------
| 
| This is not a very serious bug, more of a beautyspot, but nevertheless:
| 
| Doing singular value decomposition of a matrix with a large number of
| rows or large number of columns, Octave falls over with a segmentation
| fault and core dumps. The matrix in the demo below has 256^2=65536
| columns, which is large, but fairly common e.g. in image processing.
| 
| What I think happens is that a memory shortage arises somewhere in the
| computations. Economy-sized svd still works fine! However, its not a
| very elegant behaviour in this situation, to simply throw up a core
| and then die.
| 
| Repeat-By:
| ---------
| 
| data = rand(5,256^2);
| [pcscrs, eVls, eVts] = svd(data);

I've implemented a kluge for the development sources that attempts to
avoid this problem.

Thanks,

jwe



-----------------------------------------------------------------------
Octave is freely available under the terms of the GNU GPL.

Octave's home on the web:  http://www.che.wisc.edu/octave/octave.html
How to fund new projects:  http://www.che.wisc.edu/octave/funding.html
Subscription information:  http://www.che.wisc.edu/octave/archive.html
-----------------------------------------------------------------------