To achieve a balance between high performance and energy efficiency embedded systems often use heterogeneous multiprocessor platforms tuned for a well defined application domain. However, due to extremely high design cost and NRE for deep submicron IC, not many applications can afford that. An alternative solution is FPGA-based multiprocessor system. In this paper, both high programmability and low risk can be obtained. Recent research shows such a system which loses only 2.6X in performance (normalized) compared to a dedicated ASIP for IP packet forwarding application.