Sufficient temporal independence and improved interrupt latencies in a real-time hypervisor

Abstract

Virtualization techniques for hard real-time systems typically employ TDMA scheduling to achieve temporal isolation among partitions. The processing of user-level interrupt handlers is only performed within appropriate time slots, thus significantly increasing interrupt latencies. We propose a novel approach permitting execution of user-level interrupt handlers during time slots of other partitions hence reducing interrupt latencies. Sufficient temporal independence among partitions, as required by safety standards, is maintained through a monitoring mechanism, which bounds the interference of user-level interrupt handlers in other partitions. We show correctness of the approach and evaluate its performance in a hypervisor implementation.