Output Format.
Last updated on

Output Format

generates separate output files for each task it analyzes. A single thread can be mapped into multiple tasks when its properties change during execution (e.g., when the scheduling policy is modified).

TASK_ID Format

Each task is identified by a unique TASK_ID, formatted as $(pid)-$(identifier):

  • pid: The process ID of the thread
  • identifier: A unique identifier (0-255) that increments when task properties change

Output Files

For each task, generates the following files:

  • $(TASK_ID).models.json: Successfully extracted timing models
  • $(TASK_ID).jobs.json: Extracted jobs information (only generated when executing the extract-jobs command)
  • $(TASK_ID).events.json: All raw events captured for the task (when using offline mode)
  • $(TASK_ID).infos.json: Task information such as scheduling policy, PPID, comm and other task-specific details