An implementation of the flexible spin-lock model in ERIKA Enterprise on a multi-core platform

Recently, the flexible spin-lock model (FSLM) has been introduced, unifying spin-based and suspension-based resource sharing protocols for real-time multiprocessor platforms by explicitly identifying the spin-lock priority as a parameter. Earlier work focused on the definition of a protocol for FSLM and its corresponding analysis under the assumption that various types of implementation overhead could be ignored. In this paper, we briefly describe an implementation of the FSLM for a selected range of spin-lock priorities in the ERIKA Enterprise RTOS as instantiated on an Altera Nios II platform using 4 soft-core processors. Moreover, we present measurement results for the protocol specific overhead of FSLM as well as the natively provided multiprocessor stack resource policy (MSRP). Given these results, we are now in a position to judge when it is advantageous to use either MSRP or FMLP for our system set-up for given global resource access times of tasks