[ff3d-users] 'Segmentation fault' always?!

[Arun Viswanathan]

Hi Stephane and others, 

After receiving this reply (below) form Stephane, I am
no longer using 100 or more objects in my geometry.
Right now, I only have 25 objects in my geometry. But
when I try to run this code, I consistently get
"Segmentation fault" everytime. I am sure that my mesh
is finer than my leakage source (the smallest object).
I have tried running the code for several mesh
granularities, but I only get seg faults everytime. I
know the code is right, because when I tried running
it with a huge leakage source and a suitable mesh,
then it produced the desired result. Could you please
tell me why this is happening, what is the reason for
these seg faults? I have attached all relevant files
(with comments on what works and what doesn't), incase
that is of any help.

Thanks so much,
Arun.

--- Stephane Del Pino <address@hidden> wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> Le lundi 19 Juillet 2004 15:52, Arun Viswanathan a
> écrit :
> > I was just thinking about this error I get....Is
> it
> > possible that I am getting this error because I
> have
> > several objects referenced by the same color?? My
> > geometry contains 24 pallets in 1 color and 96
> > cylinders in another. May be this error is because
> of
> > my referencing multiple objects in the same way??
> I
> > shall try changing the code and running it, but I
> just
> > thought I'd first ask to find out if I am right!!
> >
> > Would appreciate your thoughts/suggestions. Thanks
> and
> > regards, Arun.
> Well. It is quite complex.
> 
> First I have to recall that the surface mesh
> generator is still in 
> development. Cecile and Pascal are writting a more
> stable and accurate 
> algorithm, but they are waiting a contribution of
> mine to end their 
> implementation. The problem is that my contribution
> requires quite deep 
> modification of the code, much deeper than I first
> expected. So, I do not 
> know when ff3d will have a stable surface mesher...
> 
> The error messages you get are development messages.
> The problem here is that 
> you are using more than a hundred objects which may
> be quite rude for ff3d...
> 
> By the way, building such a complex mesh probably
> requires a very huge 
> structured mesh. If I remember your problem you are
> imposing neumann 
> condition on their border, this requires not used
> degrees of freedom *inside* 
> each object (which means outside the computational
> domain). This is due to 
> the use of the fictitious domain method. You may
> consider building an 
> unstructured 3D mesh to get an accurate solution...
> 
> Stephane.
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.2.4 (GNU/Linux)
> 
>
iD8DBQFA/YgNbZyhKhZwdc0RAohkAKCVUSa0HRAwxsl55ccvRWY6/+vfkwCgwHIm
> 06hyjsv+/ldfrRbYtcV1KkU=
> =7iHH
> -----END PGP SIGNATURE-----
> 
> 
> _______________________________________________
> ff3d-users mailing list
> address@hidden
> http://lists.nongnu.org/mailman/listinfo/ff3d-users
> 


                
__________________________________
Do you Yahoo!?
Yahoo! Mail - 50x more storage than other providers!
http://promotions.yahoo.com/new_mail

solve.ff
Description: solve.ff

geov0-3.pov
Description: geov0-3.pov