Since AMDGPU v0.5.0 gridsize represents the number of "workgroups" (or blocks in CUDA) and no longer "workitems workgroups" (or threads blocks in CUDA) as HIP is used for kernel launches instead of HSA.
This means that previous AMDGPU kernel launch param gridsize need to be adapted from gridsize = nx to gridsize = cld(nx, groupsize).
Also, queues and signals are now abstracted into streams.
Since AMDGPU v0.5.0
gridsize
represents the number of "workgroups" (or blocks in CUDA) and no longer "workitems workgroups" (or threads blocks in CUDA) as HIP is used for kernel launches instead of HSA.This means that previous AMDGPU kernel launch param
gridsize
need to be adapted fromgridsize = nx
togridsize = cld(nx, groupsize)
.Also,
queues
andsignals
are now abstracted intostreams
.