> Hmmm, if I understand you correctly, you say the program reports a > leak of 192 bytes for every run of the program, no matter how long it > runs for, or how many times the [add|repeat]_timeout is called? >
That's right. > In that case, I suspect that what you are seeing is a false positive > from the detector (indeed, I'm surprised the fltk code doesn't > trigger more examples than that.) It really wouldn't surprise me either way. I haven't had a chance to try my example under my Linux virtual machine, so I have no idea whether or not valgrind will complain about it. I'm not too worried if I can't find anything to make the warning go away. I have tested the detector with Allegro programs and I've always managed to get zero reports of leaked objects, so as far as I can tell it's reliable. My research project involves code from another open source project. When I first ran it through the leak detector, I was getting leaked objects starting at about half a megabyte and growing as the program was open longer (due to problems in the borrowed code). I've worked it down to a stable 192 bytes, so if I need to be happy with that, I have no problems with it. I mainly posted because I was able to find a specific step that caused it, and I was wondering if I'm using add_timeout and repeat_timeout correctly or if I needed to do something else to avoid the leak. _______________________________________________ fltk mailing list [email protected] http://lists.easysw.com/mailman/listinfo/fltk
