Timing Analysis of Fixed Priority Self-Suspending Sporadic Tasks

Abstract

Many real-time systems include tasks that need to suspend their execution in order to externalize some of their operations or to wait for data, events or shared resources. Although commonly encountered in real-world systems, study of their timing analysis is still limited due to the problem complexity. In this paper, we invalidate a claim made in one of the earlier works [1], that led to the common belief that the timing analysis of one self-suspending task interacting with non-self-suspending sporadic tasks is much easier than in the periodic case. This work highlights the complexity of the problem and presents a method to compute the exact worst-case response time (WCRT) of a self-suspending task with one suspension region. However, as the complexity of the analysis might rapidly grow with the number of tasks, we also define an optimization formulationto compute an upper-bound on the WCRT for tasks with multiple suspendion regions. In the experiments, our optimization framework outperforms all previous analysis techniques and oftenfinds the exact WCRT. Abstract-Many real-time systems include tasks that need to suspend their execution in order to externalize some of their operations or to wait for data, events or shared resources. Although commonly encountered in real-world systems, study of their timing analysis is still limited due to the problem complexity. In this paper, we invalidate a claim made in one of the earlier works [1], that led to the common belief that the timing analysis of one self-suspending task interacting with non-self-suspending sporadic tasks is much easier than in the periodic case. This work highlights the complexity of the problem and presents a method to compute the exact worst-case response time (WCRT) of a self-suspending task with one suspension region. However, as the complexity of the analysis might rapidly grow with the number of tasks, we also define an optimization formulation to compute an upper-bound on the WCRT for tasks with multiple suspendion regions. In the experiments, our optimization framework outperforms all previous analysis techniques and often finds the exact WCRT.

Many suspensions, many problems: a review of self-suspending tasks in real-time systems

“…-Incorrect segmented fixed-priority scheduling with period enforcement [KAdNR13,DTT09]. -Incorrect conversion of higher-priority self-suspending tasks into sporadic tasks with release jitter [NFRN15].…”

“…The problem of the schedulability analysis of segmented self-suspending tasks has been treated in [LR10,NFRN15], again assuming only one self-suspending task τ k . The proposed solutions are based on the notion of the critical instant.…”

“…For example, in Section VI of [NFRN15], the higher-priority segmented self-suspending tasks are converted into ordinary sporadic tasks with jitters but the suspension time of the task under analysis is not converted into computation. We will discuss this misconception in Section 5.6.…”

“…Thus, the total response time of τ 3 in a scenario where not all higher-priority tasks release a job synchronously with τ 3 is equal to R 3 = R The synchronous release of all tasks does not necessarily generate the maximum interference for the self-suspending task τ k and is thus not always a critical instant for τ k . It was however proven in [NFRN15] that in the critical instant of a self-suspending task τ k , every higher-priority task releases a job synchronously with the arrival of at least one computation segment of τ k , but not all higher-priority tasks must release a job synchronously with the same computation segment.…”

“…Unfortunately, the analysis in [LR10] has been proven to be flawed in [NFRN15]. Further details are provided in Section 5.3.…”