<!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 6.5.7654.12">
<TITLE>GAList and GAGenome Destructors - Segmentation Fault</TITLE>
</HEAD>
<BODY>
<!-- Converted from text/plain format -->
<BR>
<P><FONT SIZE=2>To whom it concerns,<BR>
I have derived a class GEListGenome using the galib class GAListGenome. The version of galib I'm using is 2.4.7. Up until recently my experiments were dealing with small population sizes (usually < 90) and they were running to completion with no errors. When I increased the population size to 110 I got a segmentation fault. Using GDB I was able to isolate the line of code which caused the fault. It is in GAList.C in the destructor function. The following is the line of code<BR>
<BR>
while(hd) delete GAListBASE:: remove(DYN_CAST(GANode<T>*,hd));<BR>
<BR>
<BR>
Now another strange thing is happening also.<BR>
When I increased the popsize to 150 again my experiment crashed with a segmentation fault but this time the line of code is in GAGenome.C in the GAGenome destructor function.<BR>
<BR>
delete evd;<BR>
<BR>
<BR>
I have gone through my own code with a fine toothcomb and I can't see what could be causing the segmentation faults. If my code is the issue why doesn't the problem occur at smaller population sizes.<BR>
<BR>
Has anybody ever seen the two segmentation faults that I've described and if so how did they overcome them?<BR>
<BR>
I would greatly appreciate any help or advice that would help me solve the problems.<BR>
<BR>
Thanks,<BR>
Miriam<BR>
</FONT>
</P>
</BODY>
</HTML>