Problem whit too many libraries
Hi! I'm using SDL, SDL image, Paragui, Boost, OpenTNL and STL.
All of these compile just fine together but after I added TNL there has been memory leaks. Even I'd simbly have int main(){ return 0; } the MS visual c++ debug library shows assertion failed message box saying: "_CrtIsValidHeapPointer(pUserPointer)"
Now I've had that before whit other projects but now I have no clue what's causing it since I dont put any code in main().
So any idea what could be wrong?
Thanks for any feedback. I hope this wasn't too hopeless question. :)
Take each library away one at a time, or start with one and keep adding until you find the culprit causing the leak and then go see if theirs an update for it :)
Quote:Original post by GCoda
Hi! I'm using SDL, SDL image, Paragui, Boost, OpenTNL and STL.
So any idea what could be wrong?
I think the problem is either with OpenTNL or Paragui. I've never had leaks with the SDL and SDL_Image and STL can't leak memory when it's not used [wink]. Boost from what I've heard is fine as well, since you are not using any code yet. So along with what others have said, just take it one library at a time. However, I would also suggust using MMGR to find your problem. It should be able to tell you where the leak is - but it may not in release mode. However it will confirm any leaks you will have.
- Drew
As OpenTNL was essentially used in Tribes 2, I'd be more inclined to look elsewhere for the leak. However, post for help on the GarageGames forums for OpenTNL or use the mailing lists on opentnl.org.
Thanks for your replies!
I believe the problem is solved now. It was because ParaGUI was compiled whit Multithreaded DLL run-time library while all the other LIBs we're compiled whit Debug Multithreaded DLL.
I believe the problem is solved now. It was because ParaGUI was compiled whit Multithreaded DLL run-time library while all the other LIBs we're compiled whit Debug Multithreaded DLL.
Quote:Original post by Rob Loach
What do you think of OpenTNL?
Very nice. I wrote master server for torque using it. I haven't used any other networking libraries though. But that just proves it's easy to work whit and I especially like the platform support and all the classes you get whit openTNL.
This topic is closed to new replies.
Advertisement
Popular Topics
Advertisement