trappy.compare_runs module

trappy.compare_runs.compare_runs(actor_order, map_label, runs, **kwords)

A side by side comparison of multiple runs

Plots include temperature, utilization, frequencies, PID controller and power.

Parameters:

• actor_order (list) –

  An array showing the order in which the actors were registered. The array values are the labels that will be used in the input and output power plots.

  For Example:

  ["GPU", "A15", "A7"]
  

• map_label (dict) –

  A dict that matches cpumasks (as found in the trace) with their proper name. This “proper name” will be used as a label for the load and allfreqs plots. It’s recommended that the names of the cpus matches those in actor_order.

  For Example:

  {"0000000f": "A7", "000000f0": "A15"}
  

• runs (list) –

  An array of tuples consisting of a name and the path to the directory where the trace.dat is.

  For example:

  [("experiment1", "wa_output/antutu_antutu_1"),
   ("known good", "good/antutu_antutu_1")]
  

• tz_id (int) – thermal zone id as it appears in the id field of the thermal_temperature trace event

trappy.compare_runs.summary_plots(actor_order, map_label, **kwords)

A summary of plots for a given run

Warning

This is a wrapper around compare_runs(). Use that instead.