Re: Installing Octave 2.1.57 on Redhat 9

[David Bateman]

According to Robert French <address@hidden> (on 04/22/04):
> I recently started experimenting with Octave, as it appears to have some nice
> capabilities in the image processing area.  The version that came with the
> Redhat 9 distribution (2.1.40) seems to have a few quirks when displaying
> images, so I decided to try a newer version (2.1.57).   I built this from
> source, and this appeared to build ok without apparent errors, and did solve
> the problems that I had experienced with the 2.1.40 version when displaying
> images.  However, when attempting to run a fft (eg: Y=fft(x,512);), Octave 
> dies
> with a segmentation fault.
> 

Could you give a full example that gives the segfault. I can not generate a
seg-fault in the manner you suggest, with any of

x = randn(300,2); y = fft(x,512);
x = randn(600,2); y = fft(x,512);
x = randn(300,2,2); y = fft(x,512);
x = randn(600,2,2); y = fft(x,512);
x = randn(2,300,2); y = fft(x,512,2);
x = randn(2,600,2); y = fft(x,512,2);

So without a more complete bug-report I can do nothing to help you...

One possibility I checked is that due to the fact that there are two
different versions of the FFT code that are built depending on whether
you have FFTW 3.x installed or not, you might have had a version that
doesn't use FFTW. I've tried with a version of octave using FFTPACK as
well and also can't generate this seg-fault, so its not that. In any
case, when you send the example that causes the seg-fault can you also
send the output of octave_config_info so I know what you have
installed.
 
> 
> I m looking for advice on how to best get a recent version of Octave working 
> on
> my system.  Is there a way to troubleshoot the segmentation fault to determine
> its cause? Or, is there a fairly recent version of a pre-built Octave that is
> known to work with RH9?

gdb octave
run
<code that generates seg-fault>
backtrace

This will at least tell you where the seg-fault is happening, and would be
a valuable aid in debugging..

Regards
David

-- 
David Bateman                                address@hidden
Motorola CRM                                 +33 1 69 35 48 04 (Ph) 
Parc Les Algorithmes, Commune de St Aubin    +33 1 69 35 77 01 (Fax) 
91193 Gif-Sur-Yvette FRANCE

The information contained in this communication has been classified as: 

[x] General Business Information 
[ ] Motorola Internal Use Only 
[ ] Motorola Confidential Proprietary



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

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