Hi all! Been using NME/OpenFL for the past couple years and loving it! That being said, I’ve run into an issue today that I’m not entirely sure how to continue debugging. In short, after migrating our code from OpenFL 2.2.7 to OpenFL 3.3.4 to the best of my ability, a very strange hanging issue is occurring.
The application starts and generally runs for 10-15 seconds (though I saw it hang almost immediately one time), but once it comes to a point in the game where there’s some animation and particle effects the game’s graphics hang (and music continues to play).
The backtrace generally looks like so at the time of the hang:
* thread #1: tid = 0x225109, 0x00007fff8df1b716 libsystem_kernel.dylib`__psynch_cvwait + 10, queue = 'com.apple.main-thread', stop reason = signal SIGSTOP * frame #0: 0x00007fff8df1b716 libsystem_kernel.dylib`__psynch_cvwait + 10 frame #1: 0x00007fff875d7c3b libsystem_pthread.dylib`_pthread_cond_wait + 727 frame #2: 0x00000001010145b7 CineMagic`MySemaphore::Wait(this=0x000000010270bf68) + 87 at Thread.h:335 frame #3: 0x000000010100a487 CineMagic`LocalAllocator::PauseForCollect(this=0x000000010270be00) + 151 at GCInternal.cpp:2781 frame #4: 0x000000010100aa03 CineMagic`LocalAllocator::Alloc(this=0x000000010270be00, inSize=16, inIsObject=false) + 83 at GCInternal.cpp:2851 frame #5: 0x000000010100855d CineMagic`hx::Object::operator new(inSize=16, inType=NewObjAlloc, inName=0x000000010122cfd3) + 253 at GCInternal.cpp:3331 frame #6: 0x000000010107b5b3 CineMagic`hx::DoubleData::operator new(inSize=16, inAlloc=NewObjAlloc, inName=0x000000010122cfd3) + 35 at Dynamic.cpp:101 frame #7: 0x00000001010761e1 CineMagic`Dynamic::Dynamic(this=0x00007fff5fbfa400, inVal=4340845056) + 337 at Dynamic.cpp:283 frame #8: 0x000000010107626f CineMagic`Dynamic::Dynamic(this=0x00007fff5fbfa400, inVal=4340845056) + 31 at Dynamic.cpp:284 frame #9: 0x000000010057e056 CineMagic`lime::graphics::cairo::Cairo_obj::destroy(this=0x0000000103610404) + 150 at Cairo.cpp:219 frame #10: 0x000000010057e0fa CineMagic`lime::graphics::cairo::__Cairo_objdestroy(inObj=0x0000000103610404) + 42 at Cairo.cpp:225
To me this looks like something squirrelly is happening with memory/GC/Threading, but I’m at a bit of a loss of where to even start debugging such an issue. Here’s a link for the full backtrace. I’m running the mac target on a machine running OSX 10.9.5. Any help would be appreciated!