[SciPy-user] Re: bug crash

Olivier Ravard olivier.ravard at novagrid.com
Wed Jun 16 03:04:19 CDT 2004


----- Original Message -----
From: "Berthold Höllmann" <bhoel at web.de>
To: <scipy-user at scipy.org>
Sent: Tuesday, June 15, 2004 7:22 PM
Subject: [SciPy-user] Re: bug crash


> Pearu Peterson <pearu at scipy.org> writes:
>
> > On Tue, 15 Jun 2004, Olivier Ravard wrote:
> >
> >> Hi,
> >>
> >> I used scipy under windows, and this simple code crashes python
> >> with no error message :
> >>
> >> from scipy import *
> >> b=arange(1000000).astype('f')
> >> b=b*1j
> >>
> >> but this one works :
> >>
> >> from scipy import *
> >> b=arange(100000).astype('f')
> >> b=b*1j
> >>
> >> Looks like a memory problem...
> >
> > Indeed..
> >
> >> Anyone have an idea ?
> >
> > Try installing more memory to your machine. On a machine with 1GB
physical
> > RAM I have:
> >

I have 512Mo RAM and I think that this is enough to reach an 1 million
elements vector...

> > In [1]: from scipy import *
> >
> > In [2]: b=arange(1000000).astype('f')
> >
> > In [3]: b=b*1j
> >
> > In [4]: b=arange(10000000).astype('f')
> >
> > In [5]: b=b*1j
> >
> > In [6]: b=arange(100000000).astype('f')
> >
> > In [7]: b=b*1j
> > Segmentation fault
>
> Hello,
>
> My machine has only 512MB RAM, but I get:
>
> >python
> Python 2.3.4 (#2, May 28 2004, 22:06:09)
> [GCC 3.3.1 (SuSE Linux)] on linux2
> Type "help", "copyright", "credits" or "license" for more information.
> >>> from scipy import *
> >>> b=arange(1000000).astype('f')
> >>> b=b*1j
> >>> b=arange(10000000).astype('f')
> >>> b=b*1j
> >>>
> >free -m
>              total       used       free     shared    buffers     cached
> Mem:           502        177        325          0         16         34
> -/+ buffers/cache:        126        375
> Swap:         2445        223       2222
>
> OK, it's Linux

Yes, I work a lot with linux and very big vectors do not make any crash !!!
I think this problem is really a scipy problem because if we use Numeric,
then it works...
even with ten millions element vector...

from Numeric import *
b=arange(10000000).astype('f')
b=b*1j

works.


>
> Kind Regards
> Berthold
> --
> bhoel at web.de / http://starship.python.net/crew/bhoel/
>
> _______________________________________________
> SciPy-user mailing list
> SciPy-user at scipy.net
> http://www.scipy.net/mailman/listinfo/scipy-user



More information about the SciPy-user mailing list