Application-aware spinlock control using a hardware scheduler in MPSoC platforms

Spinlocks are a common technique in Multi-Processor Systems-on-Chip (MPSoCs) to protect shared resources and prevent data corruption. Without a priori application knowledge, the control of spinlocks has high randomness which can degrade the system performance significantly. This paper presents a centralized control mechanism of spinlocks by using a hardware scheduler called OSIP, that increases system performance by utilizing application-specific information. A complete spinlock control flow, starting from integrating high-level user-defined information down to a low-level realization of the control, is introduced. Two case studies demonstrate the high efficiency of this mechanism.