[Gmsh] I have discovered a bug in Gmsh! :o)

[Christophe Geuzaine]

Laurent Klinger wrote:
> Hello,
> 
> running gmsh on the attached .geo file, with command line "gmsh -3
> mesh.geo", yields:
> 
> 
> [snip a lot of lines]
> Warning : Unable to add point 4317 (will try it later)
> Warning : Coplanar points in circum sphere computation
> Warning : (1.15701,0,0.180305) (1.1483,0,0.109934) (1.15013,0,0.132817)
> (1.15182,0,0.0961568)
> Fatal   : Segmentation violation (invalid memory reference)
>         : ------------------------------------------------------
>         : You have discovered a bug in Gmsh! You may report it
>         : by e-mail (together with any helpful data permitting to
>         : reproduce it) to <gmsh at geuz.org>
> debug!
> Abandon


Hi Laurent - give the new version 2.0 a try (it seems to work like a
charm on your geometry over here).


> 
> 
> gmsh version is 1.65.0, built with gcc (GCC) 4.1.1 20060724 (prerelease)
> (4.1.1-3mdk).
> 
> Just to give a quick intro to the geometry, the attached screenshot
> shows an overview of the thing. I'm varying the parameter freeze,
> which controls the distance between the two points in the red
> circle. Here is what happens :
> - with freeze = 0.1, 0.2, and cl = 0.02 (characteristic length), no
>   problems
> - setting freeze to 0.01, two domains out of six are not meshed (but 
>   no segfault).
>   BTW, the funny thing is, these two domains are the top ones (which
>   should be only remotely influenced by the freeze value)
> - trying a different value of cl (0.022), just to see if it helps;
>   quite the contrary, it makes gmsh segfault.
> 
> Hope this helps,
> 
>   laurent
> 
> 
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> gmsh mailing list
> gmsh at geuz.org
> http://www.geuz.org/mailman/listinfo/gmsh


-- 
Prof. Christophe Geuzaine
University of Liege, Electrical Engineering and Computer Science
http://www.montefiore.ulg.ac.be/~geuzaine