Library prosa.analysis.facts.busy_interval.ideal.inequalities

Require Export analysis.abstract.ideal_jlfp_rta.

In this file, we prove some inequalities that always hold inside the busy interval of a job. Throughout this file we assume the ideal uniprocessor model.
Section BusyIntervalInequalities.

Consider any kind of tasks and their jobs, each characterized by an arrival time, a cost, and an arbitrary notion of readiness.
  Context {Task : TaskType} {Job : JobType} `{JobTask Job Task}.
  Context `{JobArrival Job} `{JobCost Job} {JR :JobReady Job (ideal.processor_state Job)}.

Consider a JLFP policy that is reflexive and respects sequential tasks.
  Context `{JLFP_policy Job}.
  Hypothesis H_policy_is_reflexive : reflexive_priorities.
  Hypothesis H_policy_respecsts_sequential_tasks : policy_respects_sequential_tasks.

Consider a consistent arrival sequence that does not contain duplicates.
  Variable arr_seq : arrival_sequence Job.
  Hypothesis H_consistent_arrival_times : consistent_arrival_times arr_seq.
  Hypothesis H_arrival_sequence_is_a_set : arrival_sequence_uniq arr_seq.

Consider any valid ideal uniprocessor schedule defined over this arrival sequence.
  Variable sched : schedule (ideal.processor_state Job).
  Hypothesis H_valid_schedule : valid_schedule sched arr_seq.

Consider a set of tasks ts that contains all the jobs in the arrival sequence.
  Variable ts : list Task.
  Hypothesis H_all_jobs_from_taskset : all_jobs_from_taskset arr_seq ts.

Consider a task tsk.
  Variable tsk : Task.

Consider a job j of the task tsk that has a positive job cost.
  Variable j : Job.
  Hypothesis H_j_arrives : arrives_in arr_seq j.
  Hypothesis H_job_of_tsk : job_of_task tsk j.
  Hypothesis H_job_cost_positive: job_cost_positive j.

Consider the ideal JLFP definitions of interference and interfering workload.
  Let interference (j : Job) (t : instant) :=
        ideal_jlfp_rta.interference arr_seq sched j t.

  Let interfering_workload (j : Job) (t : instant) :=
        ideal_jlfp_rta.interfering_workload arr_seq sched j t.

Consider the busy interval for j is given by [t1,t2).
  Variable t1 t2 : duration.
  Hypothesis H_busy_interval :
    definitions.busy_interval sched interference interfering_workload j t1 t2.

Let us denote the relative arrival time by A.
  Let A := job_arrival j - t1.

Consider any arbitrary time Δ inside the busy interval.
  Variable Δ: duration.
  Hypothesis H_Δ_in_busy : t1 + Δ < t2.

First, we prove that if the priority inversion is bounded then, the cumulative priority inversion is also bounded.
  Section PIBound.

Consider the priority inversion in any given interval is bounded by a constant.
    Variable priority_inversion_bound: duration → duration.
    Hypothesis H_priority_inversion_is_bounded:
      priority_inversion_is_bounded_by
        arr_seq sched tsk priority_inversion_bound.

Then, the cumulative priority inversion in any interval is also bounded.
    Lemma cumulative_priority_inversion_is_bounded:
      cumulative_priority_inversion arr_seq sched j t1 (t1 + Δ) ≤
        priority_inversion_bound (job_arrival j - t1).
    Proof.
      apply leq_trans with (cumulative_priority_inversion arr_seq sched j t1 t2).
      - rewrite [X in _ ≤ X](@big_cat_nat _ _ _ (t1 + Δ)) //=; try by lia.
        by rewrite /priority_inversion leq_addr.
      - apply H_priority_inversion_is_bounded; try done.
        eapply instantiated_busy_interval_equivalent_busy_interval in H_busy_interval; rt_eauto.
        by move: H_busy_interval ⇒ [PREF _].
    Qed.

  End PIBound.

Let jobs denote the arrivals in the interval Δ.
  Let jobs := arrivals_between arr_seq t1 (t1 + Δ).

Next, we prove that the cumulative interference from higher priority jobs of other tasks in an interval is bounded by the total service received by the higher priority jobs of those tasks.
  Lemma cumulative_interference_is_bounded_by_total_service:
    cumulative_interference_from_hep_jobs_from_other_tasks sched j t1 (t1 + Δ)
    ≤ service_of_jobs sched (fun jo ⇒ another_task_hep_job jo j) jobs t1 (t1 + Δ).
  Proof.
    move: (H_busy_interval) ⇒ [[/andP [JINBI JINBI2] [QT _]] _].
    erewrite cumulative_i_thep_eq_service_of_othep;
      rt_eauto.
    - move: (H_job_of_tsk) ⇒ /eqP TSK.
      by rewrite /another_task_hep_job /jobs /service_of_jobs.
    - by rewrite instantiated_quiet_time_equivalent_quiet_time; rt_eauto.
  Qed.

  Section WorkloadRBF.

Consider a valid arrival curve that is followed by the task set ts.
    Context `{TaskCost Task} `{MaxArrivals Task}.
    Hypothesis H_valid_arrival_curve : valid_taskset_arrival_curve ts max_arrivals.
    Hypothesis H_is_arrival_curve : taskset_respects_max_arrivals arr_seq ts.

Suppose all arrivals have WCET-compliant job costs.
    Hypothesis H_valid_job_cost : arrivals_have_valid_job_costs arr_seq.

Consider a task tsk_o that is not equal to tsk.
    Variable tsk_o : Task.
    Hypothesis H_tsko_in_ts: tsk_o \in ts.
    Hypothesis H_neq: tsk_o != tsk.

We prove that the workload of higher priority jobs of a task tsk in any interval is bound by the request bound function of the task in that interval.
    Lemma workload_le_rbf:
      workload_of_jobs (fun jo ⇒ hep_job jo j && (job_task jo == tsk_o)) jobs ≤
        task_request_bound_function tsk_o Δ.
    Proof.
      apply leq_trans with (task_cost tsk_o × number_of_task_arrivals arr_seq tsk_o t1 (t1 + Δ)).
      { apply leq_trans with (\sum_(j0 <- arrivals_between arr_seq t1 (t1 + Δ) | job_task j0 == tsk_o)
                                job_cost j0).
        { rewrite /workload_of_jobs big_mkcond [X in _ ≤ X]big_mkcond //= leq_sum // ⇒ s _.
          by case (job_task s == tsk); case (hep_job s j).
        }
        { rewrite /number_of_task_arrivals /task.arrivals.number_of_task_arrivals
                  -sum1_size big_distrr /= big_filter muln1.
          apply leq_sum_seq; move ⇒ jo IN0 /eqP EQ.
          by rewrite -EQ; apply H_valid_job_cost; apply in_arrivals_implies_arrived in IN0. } }
      { rewrite leq_mul2l; apply/orP; right.
        rewrite -{2}[Δ](addKn t1).
        apply H_is_arrival_curve; auto using leq_addr. }
    Qed.

  End WorkloadRBF.

End BusyIntervalInequalities.