Nusrat Islam
aa8a5f3b2a
With Simple protocol, unroll factor of 4 offers better
performance for most of the collectives (on MI200. MI250, and
MI300) except large message allreduce with Ring algorithm
on MI250 and MI200). This PR changes the default unroll factor
to 4 while adding fine tuning for reduction operations.
[ROCm/rccl commit: 58e53dfd37]