help-octave
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Segfaults with octave (any version) on Sempron machine running Cygwi


From: Robert A. Macy
Subject: Re: Segfaults with octave (any version) on Sempron machine running Cygwin
Date: Mon, 09 May 2005 07:36:04 -0700

This may not relate to your problem, but the phrase
"segment fault" caught my eye.  

I run binary 
octave 2.1.50a-inst.exe
on both an XP and a Win98

I used to have problems with segment fault problem crashes
whenever I tried to analyze large data sets.  

In the XP I had to manually increase memory allocation
using octave commands.  That solved the problem for
crashing, but running the program took over 20 minutes. You
could hear the hard drive continually swapping.  After
installing more RAM, 512MB; the swapping sounds stopped and
the program only takes 2 minutes to run.  And now have had
no segment fault crashes PLUS the oct.hist is always
correctly updated AND octave completely disappears upon
exit.  Historically, octave sometimes used to "hang around"
leaving multiple copies running, or erase the oct.hist
file.  But, again, all those nuisances went away after
expanding the octave memory allocation and adding RAM.

There was no easy way to get the Win98 to expand octave's
memory allocation - due to Win98 DOS not supporting weirdly
long file names, so I never expanded octave's memory
allocation.  Thus, I don't analyze large data sets on the
Win98.  However, I recently increased RAM memory to 256MB
and I rarely have any segment fault crashes now.

              - Robert -


On Mon, 9 May 2005 02:44:12 -0500
 <address@hidden> wrote:
> Hello all,
> 
> I have the following problem:
> I compiled octave (2.9xx tarball) on a Windows XP laptop
> with Centrino
> (Pentium-M? CPU) on a Cygwin installation with gcc-3.2.x.
> (I downloaded
> this from the Ptolomy-II website following a hint from
> this website,
> since the compilation using gcc-3.3x produced a very slow
> octave due to
> sjlj exception handling).
> The result works fine and loops are executed at least 10
> times faster
> than the version compiled with gcc-3.3x.
> Now I also need to run this on a AMD-Sempron machine
> running Win98. I
> cannot compile on this machine because I keep getting an
> error
> (according to Cygwin mailing list running out of PIDs)
> making it
> impossible to configure and make. Supposedly "rebaseall"
> should help
> with that problem, but is not included in the Cygwin that
> I'm using.
> Running the binary produced on the Win-XP machine works
> for a while, but
> it crashes frequently (seg-fault).
> Is this perhaps due to different CPU? Is there a
> configure-flag I can
> set when configuring on the Centrino PC such that the
> result will also
> work fine with the Sempron CPU?
> 
> Thanks for any help
> Reiner Suikat
> 
> 
> 
>
-------------------------------------------------------------
> 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
>
-------------------------------------------------------------
> 



-------------------------------------------------------------
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
-------------------------------------------------------------



reply via email to

[Prev in Thread] Current Thread [Next in Thread]