I recompiled tcl8.4.10 and xotcl1.3.6 for -DTCL_MEM_DEBUG=1 -g and -00
and then run valgrind on  the attached scripts. 
Unfortunately I couldnt find anything in the debug outputs to pin point
the problem.  However I did find that the scripts memLeakMixin and
memLeakMultiMixin do increase memory over time.  memLeakClass does
not increase memory over time.&nbsp; <br>
<br>
It seems that the memory leak is related to the number of instmixins on
the class A.&nbsp;&nbsp; memLeakMulitMixin uses far more memory than
does memLeakMixin.<br>
<br>
Any suggestions where to go next?<br>
<br>
Ben<br>
<br>
<br><br><div><span class="gmail_quote">On 30/09/05, <b class="gmail_sendername">Zoran Vasiljevic</b> &lt;<a href="mailto:zv@archiware.com">zv@archiware.com</a>&gt; wrote:</span><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
<br>Am 30.09.2005 um 20:15 schrieb Jeff Hobbs:<br><br>&gt;&nbsp;&nbsp; We have<br>&gt; a copy that I have used many times.&nbsp;&nbsp;Valgrind comes close,<br>&gt; finds different things, but I still prefer Purify.&nbsp;&nbsp;Valgrind<br>&gt; is free though, so there's not excuse not to try that.
<br><br>Unfortunately (as you already know) ...<br>Those will not report regular memory usage (C-wise) when<br>you just expand an array, extend the list, flood the event<br>loop, forget to destoy objects, etc. pp.,&nbsp;&nbsp;indefinitetly.
<br>Hence, I found the Poor Man's Purify ([time] + glancing at<br>the top display) equaly useful.<br><br>What I believe, (under assumption that Ben is using only<br>Tcl/XOTcl code) is that he's hitting something like the<br>
above. I had those problems en mass until I persuaded<br>Gustaf to make the &quot;-volatile&quot; mechanism in XOTcl...<br><br>Zoran<br><br></blockquote></div><br><br clear="all"><br>-- <br>You
know you have reached perfection of design not when you have nothing
more to add, but when you have nothing more to take away. --Antoine de
Saint Exupéry