Segmentation fault on extrude

Bug #886607 reported by sk
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
gmsh (Ubuntu)
Fix Released
Undecided
Unassigned

Bug Description

If I run "gmsh atm2.geo" (cf. attachment) it opens a window but crashes before any interaction or display action happens.

ProblemType: Bug
DistroRelease: Ubuntu 11.10
Package: gmsh 2.5.1~beta1~svn9724~dfsg-1
ProcVersionSignature: Ubuntu 3.0.0-12.20-generic 3.0.4
Uname: Linux 3.0.0-12-generic i686
ApportVersion: 1.23-0ubuntu3
Architecture: i386
CheckboxSubmission: 6e1841f6bc8ee04eadffe3696ff1f973
CheckboxSystem: 4189098363a6ad0365cb1899f85542bb
Date: Sat Nov 5 18:07:46 2011
EcryptfsInUse: Yes
ProcEnviron:
 PATH=(custom, no user)
 LANG=de_DE.UTF-8
 SHELL=/bin/bash
SourcePackage: gmsh
UpgradeStatus: Upgraded to oneiric on 2011-10-23 (13 days ago)

Revision history for this message
sk (s-karrmann) wrote :
Revision history for this message
Anton Gladky (gladky-anton) wrote :

The question was redirected to GMSH-developers. They gave the following answer [1]:

================================================================

You should extrude with an angle strictly smaller than Pi.
Otherwise, there is a problem with indetermination.
Try:

// foil parameters
r1 = 2.0;
d = 0.1;
r2 = r1 + d;

// size parameters
h1 = 4.0;
h2 = 10.0;
b = 2*r2*1.1;

// Mesh
lc = 2*d;
lv = 5;

p1 = newp; Point(p1) = {r1 , 0, h1+d, lc};

wall[] = Extrude {0, 0, h2-h1-2*d} { Point{ p1 }; Layers{ lv }; };

wall1[] = Extrude { {0, 0, 1}, {0, 0, 0},Pi/2 } { Line { wall[1] }; };
wall2[] = Extrude { {0, 0, 1}, {0, 0, 0}, Pi/2 } { Line { wall1[0] }; };

wall3[] = Extrude { {0, 0, 1}, {0, 0, 0},-Pi/2 } { Line { wall[1] }; };
wall4[] = Extrude { {0, 0, 1}, {0, 0, 0},-Pi/2 } { Line { wall3[0] }; };

Regards,
Ruth

--
Dr. Ir. Ruth V. Sabariego

================================================================

Please, close the bug, if the problem is solved.

[1] http://geuz.org/pipermail/gmsh/2011/006829.html

Revision history for this message
sk (s-karrmann) wrote :

IMHO the program should at least report an error instead of seg-faulting.

Because of the work around, I'll close this bug.

Revision history for this message
sk (s-karrmann) wrote :

Well, it's not a fix but a work around.

Changed in gmsh (Ubuntu):
status: New → Fix Committed
Changed in gmsh (Ubuntu):
status: Fix Committed → Invalid
status: Invalid → Fix Released
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.