Commentary to: An exact schedulability test for fixed-priority preemptive mixed-criticality real-time systems

Abstract

In this paper we point to some errors in recent paper by Asyaban et al. in which they devise an exact schedulability test. These errors are critical for the correct operation of the algorithm and they should be corrected.

Analysis-Runtime Co-design for Adaptive Mixed Criticality Scheduling

“…AMC is considered the most effective fixed priority scheme [13], and has been extended to account for many additional aspects including: preemption thresholds [14], [15], multiple criticality levels [16], criticality-specific periods [17], changes in priority [18], communications [19], deferred preemption [20], weakly-hard timing constraints [21], probabilistic task models [22], context switch costs [23], and robust [24] and semiclairvoyant [25] timing behavior. An exact analysis has also been developed for periodic task sets with offsets [26], [27].…”

Compensating Adaptive Mixed Criticality Scheduling

“…AMC is considered the most effective fixed-priority scheme [38], and has been extended to account for many additional aspects including: preemption thresholds [59,60], multiple criticality levels [31], criticality-specific periods [11], changes in priority [7], communications [19], deferred preemption [20], weakly-hard timing constraints [33], probabilistic task models [48], design optimization [62], context switch costs [25], robustness and resilience [24], implementation overheads [44], and semi-clairvoyant timing behavior [23,61]. An exact analysis has also been developed for periodic task sets [4,50].…”

Mixed Criticality on Multi-cores Accounting for Resource Stress and Resource Sensitivity

“…AMC is considered the most effective fixed priority scheme [43] for single cores, and has been extended to account for many additional aspects including: preemption thresholds [68,69], multiple criticality levels [37], criticality-specific task periods [13], changes in priority [10], communications [18], deferred preemption [19], a fast return to LO-criticality behavior [15,16], weakly-hard timing constraints [38], probabilistic task models [54], design optimization [71], context switch costs [28], robustness and resilience [23], implementation overheads [51], and semi-clairvoyant timing behavior [22,70]. An exact analysis for AMC has also been developed for periodic task sets with offsets [6,58]. Finally, a modified AMCR runtime protocol [17] has been developed that delays the onset of degraded behaviour where LO-criticality jobs are dropped.…”