Hardware/software partitioning is a critical phase in the codesign approach of embedded system design. Most of the partitioning algorithms proposed assume that an embedded system has distinct sub-modules. The issue of partitioning of resource constrained embedded systems consisting of multiple occurrences of a single functionality, needs to be addressed. In this paper, we propose a genetic algorithm (GA) based partitioning algorithm which addresses the utilization of the already existing hardware instances on the chip for the repeated functionalities of the system, based on the constraints and concurrency of the functionalities.