Running Example: OSU All-reduce
CPU: AMD Ryzen 7 6800H Open MPI 4.1.3 Disable YHCCL export OMPI_MCA_coll_yhccl_priority=0 mpiexec -n 8 ./osu_allreduce -m 65536:268435456
Enable YHCCL export OMPI_MCA_coll_yhccl_priority=100 mpiexec -n 8 ./osu_allreduce -m 65536:268435456