Posted by rhatcher on Aug 13, 2012; 8:28pm
URL: https://forum.jogamp.org/AWT-JOGL-Deadlock-Revisited-tp4025772p4025798.html

Hi Julien,

I wish I could whittle it down to something small, and if I manage to do that I will be happy to provide a test case.  Unfortunately the situation is complex and I suspect the complexity is a big part of the problem.

Since my original post the problem has changed a little bit.  I built GlueGen and JOGL from repositories I cloned last Friday 08/10/2012 (I guess this would be something slightly after RC10 by JogAmp's reckoning) to see if the deadlock issue went away.  So far I haven't seen any more deadlocks, or at least nothing that's being reported as a deadlock.

However...

I am still frequently seeing situations where no drawing takes place and the AWT event queue seems to be locked up across the board (for example menus don't work either) even though no deadlock is reported by jstack.

Every time this happens I can get in the debugger and find a timer thread that appears to be stuck waiting on a java.awt.EventQueue$1InvocationLock with this backtrace:

"Timer-6" prio=10 tid=0xa5319c00 nid=0x5fce in Object.wait() [0xa5d69000]
   java.lang.Thread.State: WAITING (on object monitor)
    at java.lang.Object.wait(Native Method)
    - waiting on <0xad29c688> (a java.awt.EventQueue$1AWTInvocationLock)
    at java.lang.Object.wait(Object.java:503)
    at java.awt.EventQueue.invokeAndWait(EventQueue.java:1232)
    - locked <0xad29c688> (a java.awt.EventQueue$1AWTInvocationLock)
    at jogamp.common.awt.AWTEDTExecutor.invoke(AWTEDTExecutor.java:53)
    at jogamp.opengl.awt.AWTThreadingPlugin.invokeOnOpenGLThread(AWTThreadingPlugin.java:103)
    at jogamp.opengl.ThreadingImpl.invokeOnOpenGLThread(ThreadingImpl.java:206)
    at javax.media.opengl.Threading.invokeOnOpenGLThread(Threading.java:172)
    at javax.media.opengl.Threading.invoke(Threading.java:191)
    at javax.media.opengl.awt.GLCanvas.display(GLCanvas.java:428)
    at com.jogamp.opengl.util.AWTAnimatorImpl.display(AWTAnimatorImpl.java:74)
    at com.jogamp.opengl.util.AnimatorBase.display(AnimatorBase.java:142)
    at com.zedasoft.cba.view.DisplaySetAnimator.display(DisplaySetAnimator.java:105)
    at com.jogamp.opengl.util.FPSAnimator$1.run(FPSAnimator.java:128)
    at java.util.TimerThread.mainLoop(Timer.java:555)
    at java.util.TimerThread.run(Timer.java:505)

Here's the full jstack output from one failed run: jstack2.out

My observation is that once it's in this condition it never recovers.  Just for good measure I set a breakpoint higher up in the backtrace to verify that it's never returning from the display call.

I've been poking around in the lower levels of all this stuff for most of the day trying to make sense of where it's getting hung up.  The problem happens relatively frequently, so if you have any suggestions about ways I might trap the root cause I will be glad to try them.  Maybe that will help me learn enough to produce a smaller example.