--- Quote Start ---
Ah, I put the wrong code. Suppose,
clEnqueueTask(command, task1, 0, NULL, &eventtask1);
clEnqueueTask(command, task2, 0, NULL, &eventtask2);
clEnqueueTask(command, task3, 0, NULL, &eventtask3);
Say, Task 1 load 1 Million floats from global memory and channel pass it to task 2.
Task 2 then perform summation and channel it to Task 3
Task 3 then write back to Global memory.
--- Quote End ---
If you want the kernels to execute in pipeline fashion like your picture, you should use different command queues and pass data via channels, no need for event.
Altera have excellent examples like FFT example and channelizer here (
https://www.altera.com/products/design-software/embedded-software-developers/opencl/developer-zone.html).
Honored Contributor
