Steve Valliere 2010-04-27 07:15
I just installed an evaluation copy of BoundsChecker in hopes of finding a bug in one of the programs I'm working on. When my program exits, BoundsChecker find two memory leaks in my program, and 53 in RTSUltraMonHook.dll! 51 are from offset 0x000220A5 and 2 from offset 0x000183D9.
This thought you would like to know, just in case these can be corrected.
FWIW, I'm running 3.0.10 on XP SP3 with three monitors.
|
Christian Studer 2010-04-27 10:03
I'm not aware of any memory leaks in RTSUltraMonHook.dll, but will test with leak detection enabled for the next release. Thanks for the heads-up.
Christian Studer - www.realtimesoft.com
|
Steve Valliere 2010-04-28 01:07
Thanks! I suppose the leaks could caused by something in my code, but I don't understand how. The only connection I know of between my app and UltraMon is that UltraMon adds some items to the program's system menu (I don't have UltraMon's extra buttons enabled). My code doesn't mess with the system menu or non-client areas, so I'm not sure how we could interact.
But I'm willing to look, if you know of anything I should beware of...
|
Steve Valliere 2010-04-28 06:18
In case you're interested, I copied the info generated by BoundsChecker about the memory leaks and API call failures (though I think those are probably expected/anticipated failures) to a text file. I'm going to put it on our web site if you want to download it. The address will be http://www.electronicvisions.com/files/umleak.log.
I copied the RTSUltraMonHook.dll entries from the various BoundsChecker reports into a single text file and added the dividing lines that say which report the following items are from.
|
Christian Studer 2010-04-28 09:00
Thanks, the resource leaks are probably false alarms, but I'll need to take a closer look at this. I'll check for memory leaks myself with the CRT debug heap.
Christian Studer - www.realtimesoft.com
|
