octave-3.0.3 crashes while executing 'expfit' ("free(): invalid pointer")

[Sergei Steshenko]

--- On Wed, 12/24/08, John W. Eaton <jwe at octave.org> wrote:

> From: John W. Eaton <jwe at octave.org>
> Subject: Re: octave-3.0.3 crashes while executing 'expfit' ("free(): invalid pointer")
> To: "Sergei Steshenko" <sergstesh at yahoo.com>
> Cc: bug-octave at octave.org
> Date: Wednesday, December 24, 2008, 5:06 PM
> On 16-Dec-2008, Sergei Steshenko wrote:
> 
> | > From: John W. Eaton <jwe at octave.org>
> | > Subject: Re: octave-3.0.3 crashes while executing
> 'expfit' ("free(): invalid pointer")
> | > To: sergstesh at yahoo.com
> | > Cc: "John W. Eaton"
> <jwe at octave.org>, "Ben Abbott"
> <bpabbott at mac.com>, bug-octave at octave.org
> | > Date: Tuesday, December 16, 2008, 7:59 AM
> | > On 16-Dec-2008, Sergei Steshenko wrote:
> | > 
> | > | > > What is expfit?
> | > | > > 
> | > | > > Can you please send a complete example?
> | > | > > 
> | > | > > jwe
> | > | > 
> | > | > Perhaps expfit is from OctaveForge?
> | > | > 
> | > | >
> 	http://octave.sourceforge.net/doc/f/expfit.html
> | > | > 
> | > | > Ben
> | > | 
> | > | Yes.
> | > 
> | > Then please send it and whatever it depends on so
> that I
> | > have a
> | > complete example that includes precisely the code
> you are
> | > using.  If
> | > you are reporting a bug in Octave itself, then
> please
> | > don't assume we
> | > have the same Octave Forge packages installed that
> you do.
> | > 
> | > jwe
> | 
> | Please find 'expfit.m' attached, it's from
> 'optim-1.0.4' package.
> | 
> | The latter is from 'octave-forge-bundle-20080831'
> which was downloaded from
> | SourceForge the regular way.
> | 
> | I haven't changed anything in 'octave' or
> 'octave-forge-bundle-20080831'.
> 
> I can't reproduce this problem on a Debian AMD64
> system.
> 
> jwe

IIRC and IIUIC, you are the second person who cannot reproduce the problem
on a 64 bits system, however, the problem occurred on a 32 bits system.

My wild guess there is 'new'/'malloc' somewhere which doesn't check pointer
for NULL; on a 64 bits system memory allocation error is less likely than
on a 32 bits system.

The wild guess is based on the fact that peak memory consumption is about
2G.

Thanks,
  Sergei.