Task dependency aware IP core for dynamic scheduling in MPSoC environment

This paper deals with Intellectual Property (IP) core design for dynamic task scheduling to support Out-of-Order (OoO) execution in Multiprocessor System-on-Chip (MPSoC) environment. MPSoC is one of the most promising future processor architecture. But such systems have to face challenges in the context of OoO execution during dynamic scheduling due to data dependencies like Read-after-Write (RAW), Write-after-Write (WAW) and Write-after-Read (WAR). Due to these dependencies stalling problem occur during OoO execution. In order to solve this stalling problem and achieve task level parallelism (TLP), Scoreboarding algorithm with register renaming technique is designed. As these dependencies impose challenging constraints on the direct use of techniques like OoO execution, register renaming and dynamic scheduling, a synthesizable IP core is designed. The simulation results show that the design can analyze all the task dependencies during run-time and resolves them at TLP. This algorithm is able to resolve 100% RAW, WAW and WAR hazards which are not solvable at instruction level parallelism (ILP).