AnsweredAssumed Answered

OpenCL 2 work group functions

Question asked by boxerab on Dec 15, 2016
Latest reply on Dec 31, 2016 by boxerab

From an Intel page on OpenCL:

Work-group functions, as the name implies, always operate in parallel over entire work-group. An implicit consequence from this fact is that any work-group function call acts as a barrier.

 

So, on AMD hardware, if I have a

work_group_any()

call,

 

do I also need a

 

barrier(CL_LOCAL_MEMORY_FENCE)

 

call as well?

 

In my kernels, it looks like I do need this barrier.

 

Thanks,

Aaron

Outcomes