Jim Wilson wrote: > Andy Ross writes: > > 3. Reading information that the FDM hasn't set shouldn't cause a crash. > > Same with reverse?...seems like the problem with JSBsim and using > threads has to do with shared memory, one thread reading before or > while the other writes.
Eeep, no. Not what I meant. :) I was talking high level: If component XXX isn't prepared to exporting feature YYY, the simulator should still work, and not crash. This can be done in both C++ and properties, but with properties you get it for free; no chance of human error. What you describe is called a race condition, and that's a bug. Always was, always will be. Sometimes it's an unavoidable (or very hard to avoid) bug, in which case you say Just Don't Do That and mark your code as non-threadable. Which brings up a really good point: is the property library threadsafe? I'm guessing not, which means we need to write up a spec for how to do mutex access, or else punt and say you can only use it from the main() thread. Probably not a big deal. Andy -- Andrew J. Ross NextBus Information Systems Senior Software Engineer Emeryville, CA [EMAIL PROTECTED] http://www.nextbus.com "Men go crazy in conflagrations. They only get better one by one." - Sting (misquoted) _______________________________________________ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
