Re: Unstructured locking bug

[Bryce McKinlay]

Etienne Gagnon wrote:

> Here's what the JNI spec says about it:
>
>  MonitorExit
>
>  Prototype jint MonitorExit(JNIEnv *env, jobject obj);
>
>  ...
>
>  Native code must not use MonitorExit to exit a monitor entered through
>  a synchronized method or a monitorenter Java virtual machine
>  instruction.
>
> So, the current AWT code clearly does something that the JNI spec does
> not allow for.


OK. It might be possible to fix this by moving the synchronization to 
the Java level. eg in EventQueue.getNextEvent():

synchronized (queueLock)
 {
   ...
   callGtkThreadsEnter();
 }
tk.iterateNativeQueue();
synchronized (queueLock)
 {
   callGtkThreadsLeave();
   ...
 }

However, implementing this approach is not quite as simple because 
iterateNativeQueue() is called more than once and also called within 
nested flow control. A redesign of getNextEvent() is probably needed - 
converting some of the if and while constructs into local state 
variables which determine whether or not iterateNativeQueue() should be 
called.

Bryce