The Message Passing Interface (MPI) is designed as an architecture independent interface for parallel programming in the shared-nothing, message passing paradigm. We briefly summarize basic requirements to a high-quality implementation of MPI fo efficient progamming of SMP clusters and related architectures, and discuss possible, mild extensions of the topology functionality of MPI, which, while retaining a high degree of architecture independence, can make MPI more useful and efficient for message-passing programming of SMP clusters. We show that the discussed extensions can all be implemented on top of MPI with very little environmental support.