Forum Discussion
Avinash1
New Contributor
New ContributorHow to run multiple work groups in parallel using ND Range
As far as I know , each work group is passed sequentially to the kernel and all the work-items are run in parallel. Example : a 3D array of 8*8*8 would have 512 points, each work group size is 4*4*4 ...
HRZ
Frequent Contributor
Frequent ContributorWork-items in the same work-group won't run in parallel, they will be pipelined. You will need to use the SIMD attribute to achieve work-item-level parallelism:
Multiple work-groups are also automatically pipelined one after the other inside the same compute unit, and the compiler will replicate local memory buffers inside your kernel to accommodate for this. If you want to have work-group-level parallelism, then you need to use the num_compute_units() attribute:
