Posted by Wibowit on Apr 20, 2011; 8:59pm
URL: https://forum.jogamp.org/Concurrent-Kernels-Synchronization-JOCL-tp2843784p2843796.html

See my post a little below in threads list: http://forum.jogamp.org/Concurrent-kernel-execution-in-OpenCL-implementations-tp2827852p2827852.html

Basically, registers and LDS (Local Data Storage) have lifetime the same as kernel that uses them and they formally cannot be shared by different work-groups (registers are even private for a work-item). The only way you can exchange data is by Global Memory on video card. But as there's no universal way to synchronize different work-groups you don't have any guarantee of order of computations, so you're left with a scheme where one kernel invocation waits for completion of another.

You can use out-of-order queues and/ or events. Events serve as a way to make a partial order on kernel invocations or they're used for retrieving profiling data.