RE: 2nd post of the night!
"Travis Kriplean" <[email protected]> Thu, 3 Mar 2005 14:53:58 -0600
| Newsgroups | gmane.comp.breve |
|---|---|
| Message-ID | <[email protected]> |
Hi Jon, If its not too much trouble, is there anyway I can get a patch for the 2.1 release for the disable-freed-instance problem? I'm running a fairly hefty multi-leveled GP simulation...thousands of objects are indeed being created and freed and it gets bogged down fairly quickly (quick enough that actual large scale simulation is not possible, while testing is possible). I'm making a push right now to submit a paper for GECCO's undergrad workshop, whose due date is rapidly approaching ;) and it would be amazing if I could have some solid data by then. Thanks, Travis -----Original Message----- From: [email protected] [mailto:[email protected]] On Behalf Of jon klein Sent: Thursday, March 03, 2005 9:24 AM To: [email protected] Subject: Re: [breve] 2nd post of the night! On Mar 3, 2005, at 2:25 AM, Travis Kriplean wrote: > It crashes immediately whenever I call it on the controller (whether > within the control class or outside of it). I know that it crashes > immediately when it reaches the disable call, because I surrounded it > by print calls and it reaches the preceding one but not that the > latter. > I also added the disable-freed-instance-protection call to > swarmevolve-1.0.tz, and it also crashed upon reaching the statement. Thanks for the bug report -- I've confirmed and fixed the problem for the next release. In the meantime, I would just have to suggest not disabling freed instance protection. Unless you're creating and destroying many thousands of objects, it shouldn't make too much of a difference to memory usage. - jon klein _______________________________________________ breve mailing list [email protected] http://lists.spiderland.org/mailman/listinfo/breve _______________________________________________ breve mailing list [email protected] http://lists.spiderland.org/mailman/listinfo/breve