<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=iso-8859-1">
<META NAME="Generator" CONTENT="MS Exchange Server version 6.5.7652.24">
<TITLE>RE[4]: [Gmsh] unstable behaviour with 2nd order elements</TITLE>
</HEAD>
<BODY>
<!-- Converted from text/plain format -->
<BR>
<P><FONT SIZE=2>Hi again Christophe,<BR>
<BR>
below is the result of your indication:<BR>
<BR>
<BR>
<BR>
GNU gdb 6.8<BR>
Copyright (C) 2008 Free Software Foundation, Inc.<BR>
License GPLv3+: GNU GPL version 3 or later <<A HREF="http://gnu.org/licenses/gpl.html">http://gnu.org/licenses/gpl.html</A>><BR>
This is free software: you are free to change and redistribute it.<BR>
There is NO WARRANTY, to the extent permitted by law. Type "show copying"<BR>
and "show warranty" for details.<BR>
This GDB was configured as "i586-suse-linux"...<BR>
(gdb) r fail_2nd_order.geo -3 -order 2<BR>
Starting program: /home/fgilabert/FEM/tmp/demo_gmsh/gmsh fail_2nd_order.geo -3 -order 2<BR>
Info : Reading 'fail_2nd_order.geo'<BR>
Info : Read 'fail_2nd_order.geo'<BR>
Info : Meshing 1D...<BR>
Info : Meshing curve 1 (Line)<BR>
Info : Meshing curve 2 (Line)<BR>
Info : Meshing curve 3 (Line)<BR>
Info : Meshing curve 4 (Line)<BR>
Info : Mesh 1D complete (0 s)<BR>
Info : Mesh<BR>
Info : Meshing 2D...<BR>
Info : Meshing surface 6 (Plane, MeshAdapt+Delaunay)<BR>
Info : Recombining triangles for surface 6<BR>
Info : Meshing surface 15 (extruded)<BR>
Info : Meshing surface 19 (extruded)<BR>
Info : Meshing surface 23 (extruded)<BR>
Info : Meshing surface 27 (extruded)<BR>
Info : Meshing surface 28 (extruded)<BR>
Info : Mesh 2D complete (0.036002 s)<BR>
Info : Mesh<BR>
Info : Meshing 3D...<BR>
Info : Meshing volume 1 (extruded)<BR>
Info : Mesh 3D complete (0.004 s)<BR>
Info : Mesh<BR>
Info : Generating High Order Nodes (q = 2) ...<BR>
<BR>
Program received signal SIGSEGV, Segmentation fault.<BR>
0x080c6f9c in getFaceVertices (gr=0x8477be8, ele=0x8496d68, vf=@0xbfb0fc94, blocked=@0xbfb0fc70, faceVertices=@0xbfb0fe18, edgeVertices=@0xbfb0fe30,<BR>
linear=false, nPts=1) at HighOrder.cpp:624<BR>
624 for (int i = 0; i < 3; i++) tmp[(i + orientation) % 3] = vtcs[i];<BR>
(gdb) backtrace<BR>
#0 0x080c6f9c in getFaceVertices (gr=0x8477be8, ele=0x8496d68, vf=@0xbfb0fc94, blocked=@0xbfb0fc70, faceVertices=@0xbfb0fe18, edgeVertices=@0xbfb0fe30,<BR>
linear=false, nPts=1) at HighOrder.cpp:624<BR>
#1 0x080cbef3 in setHighOrder (gr=0x8477be8, edgeVertices=@0xbfb0fe30, faceVertices=@0xbfb0fe18, linear=false, incomplete=false, nPts=1, displ2D=0x0,<BR>
displ3D=0x0) at HighOrder.cpp:933<BR>
#2 0x080ce101 in SetOrderN (m=0x8466148, order=2, linear=false, incomplete=false) at HighOrder.cpp:1151<BR>
#3 0x0806a7f4 in GenerateMesh (m=0x8466148, ask=3) at Generator.cpp:564<BR>
#4 0x080ef6d8 in GModel::mesh (this=0x8466148, dimension=3) at GModel.cpp:332<BR>
#5 0x0804fd09 in GmshBatch () at Gmsh.cpp:111<BR>
#6 0x0804d371 in main (argc=Cannot access memory at address 0x0<BR>
) at Main.cpp:30<BR>
<BR>
<BR>
<BR>
Thanks a lot<BR>
<BR>
Francisco<BR>
<BR>
<BR>
<BR>
<BR>
<BR>
<BR>
<BR>
<BR>
<BR>
<BR>
<BR>
<BR>
Dr. Francisco A. Gilabert Villegas<BR>
Mecánica de Medios Continuos y Modelización<BR>
<BR>
Instituto de Tecnología Cerámica<BR>
Campus Universitario Riu Sec | Avda. Vicent Sos Baynat, s/n<BR>
12006 Castellón (Spain)<BR>
Tel. 34 964 34 24 24<BR>
Fax 34 964 34 24 25<BR>
<BR>
CONFIDENCIALIDAD<BR>
La información contenida en este mensaje y/o archivo(s) adjunto(s) es confidencial/privilegiada y está destinada<BR>
a ser leída sólo por la(s) persona(s) a la(s) que va dirigida. Si usted lee este mensaje y no es el destinatario señalado,<BR>
el empleado o el agente responsable de entregar el mensaje al destinatario, o ha recibido esta comunicación por<BR>
error le informamos que está totalmente prohibida, y puede ser ilegal, cualquier divulgación, distribución o<BR>
reproducción de esta comunicación, y le rogamos que nos lo notifique inmediatamente y nos devuelva el mensaje<BR>
original al e-mail: itc@itc.uji.es. Gracias.<BR>
<BR>
<BR>
<BR>
-----Mensaje original-----<BR>
De: Geuzaine Christophe [<A HREF="mailto:cgeuzaine@ulg.ac.be">mailto:cgeuzaine@ulg.ac.be</A>]<BR>
Enviado el: jue 20/11/2008 17:18<BR>
Para: Francisco Gilabert Villegas<BR>
CC: Christophe Geuzaine; Jean-François Remacle<BR>
Asunto: Re: RE[3]: [Gmsh] unstable behaviour with 2nd order elements<BR>
<BR>
> Dear Christophe,<BR>
><BR>
> Thanks for your help.<BR>
><BR>
> I recompiled gmsh using -nogui flag. The compilation seemed to be ok.<BR>
><BR>
> Then, I run gmsh in the following way:<BR>
><BR>
> ./gmsh -3 -order 2 fail_2nd_order.geo<BR>
><BR>
> And the result was:<BR>
><BR>
> Info : Reading 'fail_2nd_order.geo'<BR>
> Info : Read 'fail_2nd_order.geo'<BR>
> Info : Meshing 1D...<BR>
> Info : Meshing curve 1 (Line)<BR>
> Info : Meshing curve 2 (Line)<BR>
> Info : Meshing curve 3 (Line)<BR>
> Info : Meshing curve 4 (Line)<BR>
> Info : Mesh 1D complete (0 s)<BR>
> Info : Mesh<BR>
> Info : Meshing 2D...<BR>
> Info : Meshing surface 6 (Plane, MeshAdapt+Delaunay)<BR>
> Info : Recombining triangles for surface 6<BR>
> Info : Meshing surface 15 (extruded)<BR>
> Info : Meshing surface 19 (extruded)<BR>
> Info : Meshing surface 23 (extruded)<BR>
> Info : Meshing surface 27 (extruded)<BR>
> Info : Meshing surface 28 (extruded)<BR>
> Info : Mesh 2D complete (0.052003 s)<BR>
> Info : Mesh<BR>
> Info : Meshing 3D...<BR>
> Info : Meshing volume 1 (extruded)<BR>
> Info : Mesh 3D complete (0.004 s)<BR>
> Info : Mesh<BR>
> Info : Generating High Order Nodes (q = 2) ...<BR>
> Violación de segmento<BR>
><BR>
><BR>
> The last sentence is saying "segmentation fault" (in spanish, sorry).<BR>
><BR>
> Maybe it can be useful for you the result of the compilation:<BR>
> The attached files "recompilation_part1.txt" and "recompilation_part2.txt"<BR>
> store the out of this process.<BR>
><BR>
<BR>
The compilation looks ok...<BR>
<BR>
Could you run the same example in the debugger with<BR>
<BR>
> gdb gmsh<BR>
(gdb) r fail_2nd_order.geo -3 -order 2<BR>
<BR>
and when it crashes type<BR>
<BR>
(gdb) backtrace<BR>
<BR>
<BR>
<BR>
<BR>
<BR>
<BR>
<BR>
> If you detect that I am making something in a strange way, please<BR>
> tell me and I will verify.<BR>
><BR>
><BR>
><BR>
> Thank you again<BR>
><BR>
> Francisco<BR>
><BR>
><BR>
><BR>
><BR>
><BR>
><BR>
><BR>
><BR>
><BR>
><BR>
> Dr. Francisco A. Gilabert Villegas<BR>
> Mecánica de Medios Continuos y Modelización<BR>
><BR>
> Instituto de Tecnología Cerámica<BR>
> Campus Universitario Riu Sec | Avda. Vicent Sos Baynat, s/n<BR>
> 12006 Castellón (Spain)<BR>
> Tel. 34 964 34 24 24<BR>
> Fax 34 964 34 24 25<BR>
><BR>
> CONFIDENCIALIDAD<BR>
> La información contenida en este mensaje y/o archivo(s) adjunto(s) es<BR>
> confidencial/privilegiada y está destinada<BR>
> a ser leída sólo por la(s) persona(s) a la(s) que va dirigida. Si usted<BR>
> lee este mensaje y no es el destinatario señalado,<BR>
> el empleado o el agente responsable de entregar el mensaje al<BR>
> destinatario, o ha recibido esta comunicación por<BR>
> error le informamos que está totalmente prohibida, y puede ser ilegal,<BR>
> cualquier divulgación, distribución o<BR>
> reproducción de esta comunicación, y le rogamos que nos lo notifique<BR>
> inmediatamente y nos devuelva el mensaje<BR>
> original al e-mail: itc@itc.uji.es. Gracias.<BR>
><BR>
><BR>
><BR>
> -----Mensaje original-----<BR>
> De: Christophe Geuzaine [<A HREF="mailto:cgeuzaine@ulg.ac.be">mailto:cgeuzaine@ulg.ac.be</A>]<BR>
> Enviado el: jue 20/11/2008 13:28<BR>
> Para: Francisco Gilabert Villegas<BR>
> CC: Jean-François Remacle<BR>
> Asunto: Re: [Gmsh] unstable behaviour with 2nd order elements<BR>
><BR>
> Francisco Gilabert Villegas wrote:<BR>
>> Dear Christophe,<BR>
>><BR>
>> Thanks for your swift answer.<BR>
>><BR>
>> I checked the nightly version (20-Nov-2008 01:11) on Linux.<BR>
>> In 2D, everything is correct. Additionally I tested some complex<BR>
>> geometries and all was ok.<BR>
>><BR>
>> But in 3D, the anomalous behaviour persist. At least, this time,<BR>
>> at the end of the execution you can read "segmentation fault" in<BR>
>> the command line.<BR>
>><BR>
><BR>
> Hi Francisco - Unfortunately I cannot reproduce this on our Linux box.<BR>
> I've run it with valgrind, which also returns no errors...<BR>
><BR>
> Could you try to recompile the code from source? To make it easy you<BR>
> could build the command-line version of the code (no GUI):<BR>
><BR>
> * download the latest source snapshot<BR>
><BR>
> * configure it with "./configure --disable-gui"<BR>
><BR>
> * then build with "make"<BR>
><BR>
> The only external dependency should be the GNU scientific library (GSL).<BR>
> Installing the GSL should be relatively simple (select the "gsl-devel"<BR>
> package or something similar in your Linux distro package manager).<BR>
><BR>
> Christophe<BR>
><BR>
><BR>
><BR>
><BR>
>> I used the same simple code a in the previous message, but this<BR>
>> time adding a simple extrusion of the square (see below):<BR>
>><BR>
>> //--------------------------------------------<BR>
>> mm = 0.001;<BR>
>> Lx = 1000*mm;<BR>
>> Ly = 1000*mm;<BR>
>> Lz = 1000*mm;<BR>
>> lc = 100*mm;<BR>
>> p1ewp; Point(p1)={-Lx/2,-Ly/2,-Lz/2,lc};<BR>
>> p2ewp; Point(p2)={-Lx/2,+Ly/2,-Lz/2,lc};<BR>
>> p3ewp; Point(p3)={+Lx/2,+Ly/2,-Lz/2,lc};<BR>
>> p4ewp; Point(p4)={+Lx/2,-Ly/2,-Lz/2,lc};<BR>
>> l1ewl; Line(l1) = {p1, p2};<BR>
>> l2ewl; Line(l2) = {p2, p3};<BR>
>> l3ewl; Line(l3) = {p3, p4};<BR>
>> l4ewl; Line(l4) = {p4, p1};<BR>
>> ll1ewll; Line Loop(ll1) = {l1, l2, l3, l4};<BR>
>> s1ews; Plane Surface(s1) = {ll1};<BR>
>> Recombine Surface{s1};<BR>
>><BR>
>> Ext1[]=Extrude {0, 0, Lz} { // NEW<BR>
>> Surface{s1}; // NEW<BR>
>> Layers{10,1}; // NEW<BR>
>> Recombine; // NEW<BR>
>> }; // NEW<BR>
>> //--------------------------------------------<BR>
>><BR>
>><BR>
>> Thank you very much for your help<BR>
>><BR>
>> Francisco<BR>
>><BR>
>><BR>
>><BR>
>><BR>
>><BR>
>><BR>
>><BR>
>><BR>
>><BR>
>> Dr. Francisco A. Gilabert Villegas<BR>
>> Mecánica de Medios Continuos y Modelización<BR>
>><BR>
>> Instituto de Tecnología Cerámica<BR>
>> Campus Universitario Riu Sec | Avda. Vicent Sos Baynat, s/n<BR>
>> 12006 Castellón (Spain)<BR>
>> Tel. 34 964 34 24 24<BR>
>> Fax 34 964 34 24 25<BR>
>><BR>
>> CONFIDENCIALIDAD<BR>
>> La información contenida en este mensaje y/o archivo(s) adjunto(s) es<BR>
>> confidencial/privilegiada y está destinada<BR>
>> a ser leída sólo por la(s) persona(s) a la(s) que va dirigida. Si usted<BR>
>> lee este mensaje y no es el destinatario señalado,<BR>
>> el empleado o el agente responsable de entregar el mensaje al<BR>
>> destinatario, o ha recibido esta comunicación por<BR>
>> error le informamos que está totalmente prohibida, y puede ser ilegal,<BR>
>> cualquier divulgación, distribución o<BR>
>> reproducción de esta comunicación, y le rogamos que nos lo notifique<BR>
>> inmediatamente y nos devuelva el mensaje<BR>
>> original al e-mail: itc@itc.uji.es. Gracias.<BR>
>><BR>
>><BR>
>><BR>
>><BR>
>><BR>
>> ------------------------------------------------------------------------<BR>
>><BR>
>> _______________________________________________<BR>
>> gmsh mailing list<BR>
>> gmsh@geuz.org<BR>
>> <A HREF="http://www.geuz.org/mailman/listinfo/gmsh">http://www.geuz.org/mailman/listinfo/gmsh</A><BR>
><BR>
><BR>
> --<BR>
> Prof. Christophe Geuzaine<BR>
> University of Liege, Electrical Engineering and Computer Science<BR>
> <A HREF="http://www.montefiore.ulg.ac.be/~geuzaine">http://www.montefiore.ulg.ac.be/~geuzaine</A><BR>
><BR>
><BR>
><BR>
><BR>
><BR>
<BR>
<BR>
<BR>
<BR>
</FONT>
</P>
</BODY>
</HTML>