Library prosa.implementation.priority.tiebreaking_edf

Require Export prosa.implementation.priority.edf.
Require Export prosa.implementation.priority.tiebreaks_by_id.

EDF Priority Policy with Tie-Breaking

This EDF implementation resolves equal absolute deadlines by favoring lower task IDs, and then lower job IDs.

Section TiebreakingEDF.

Consider tasks with numeric identifiers ...

Context {Task : TaskType} `{TaskId Task}.

... and their jobs, with absolute deadlines and numeric identifiers.

Context {Job : JobType} `{JobTask Job Task} `{JobDeadline Job} `{JobId Job}.

We apply identifier-based tie-breaking to EDF's deadline order.

#[local] Instance tiebreaking_edf : JLFP_policy Job :=
tiebreaks_by_id (EDF Job).

The following facts connect the implementation to the generic priority interface used by schedulers and analyses.

We expose the comparison rule for use in proofs about this policy.

  Fact tiebreaking_edf_hep_job :
    ∀ j1 j2,
      @hep_job Job tiebreaking_edf j1 j2
      = ((job_deadline j1 < job_deadline j2 )
         || ((job_deadline j1 == job_deadline j2 )
             && ((task_id (job_task j1) < task_id (job_task j2))
                 || ((task_id (job_task j1) == task_id (job_task j2))
                     && (job_id j1 ≤ job_id j2 ))))).

Every job has at least its own priority.

Fact tiebreaking_edf_is_reflexive :
reflexive_job_priorities tiebreaking_edf.

Priority comparisons remain consistent across chains of jobs.

Fact tiebreaking_edf_is_transitive :
transitive_job_priorities tiebreaking_edf.

The scheduler can compare every pair of jobs.

Fact tiebreaking_edf_is_total :
total_job_priorities tiebreaking_edf.

Valid job identifiers make the final tie-breaking step unambiguous for arriving jobs.

  Fact tiebreaking_edf_is_antisymmetric :
    ∀ arr_seq,
      valid_job_ids arr_seq →
      antisymmetric_job_priorities tiebreaking_edf arr_seq.

The tie-breaking rule preserves EDF's deadline order and satisfies the priority properties required by EDF analyses.

Fact tiebreaking_edf_is_edf_policy :
policy_is_EDF tiebreaking_edf.

End TiebreakingEDF.

The identifier tie-breaks repeat together with EDF's deadline order.

Section HyperperiodPriorities.

Consider periodic tasks with relative deadlines and numeric identifiers ...

Context {Task : TaskType} `{PeriodicModel Task} `{TaskDeadline Task} `{TaskId Task}.

... and jobs with task-derived absolute deadlines and numeric identifiers.

Context {Job : JobType} `{JobTask Job Task} `{JobArrival Job} `{JobId Job}.

We name the task-derived absolute deadline for clarity.

Let absolute_deadline (j : Job) := job_arrival j + task_deadline (job_task j).

Consider a task set with valid periods and unique task identifiers ...

  Variable ts : TaskSet Task.
  Hypothesis H_valid_periods : valid_periods ts.
  Hypothesis H_valid_task_ids : valid_task_ids ts.

... generating a valid arrival sequence of these tasks ...

  Variable arr_seq : arrival_sequence Job.
  Hypothesis H_valid_arrival_sequence : valid_arrival_sequence arr_seq.
  Hypothesis H_all_jobs_from_taskset : all_jobs_from_taskset arr_seq ts.

... with periodic releases continuing indefinitely.

Hypothesis H_periodic_arrivals : taskset_respects_periodic_task_model arr_seq ts.
Hypothesis H_infinite_jobs : tasks_have_infinite_arrivals arr_seq ts.

Unique task IDs preserve ties between tasks. Within a task, equal deadlines identify the same release, whose job-ID comparison is reflexive.

Fact tiebreaking_edf_priorities_consistent_across_hyperperiods :
priorities_consistent_across_hyperperiods ts arr_seq tiebreaking_edf.

End HyperperiodPriorities.

We register the policy facts so Rocq can apply them automatically.

Global Hint Resolve
  tiebreaking_edf_is_edf_policy
  tiebreaking_edf_is_reflexive
  tiebreaking_edf_is_transitive
  tiebreaking_edf_is_total
  tiebreaking_edf_is_antisymmetric
  tiebreaking_edf_priorities_consistent_across_hyperperiods
  : basic_rt_facts.