Skip to content

Latest commit

 

History

History
35 lines (28 loc) · 2.18 KB

CHANGELOG.md

File metadata and controls

35 lines (28 loc) · 2.18 KB

0.3.3 Change Log

Fixed

  • Init: Missing empty_page.html fix.

0.3.2 Change Log

Added

  • CLI: Added --headless flag Accepts true or false.
  • CLI: Added --keep-tab flag Accepts true or false. If true, headless is set to false and keeping the browser tab(s) open after test/benchmark. This is useful to examine failures.
  • Benchmark: Added task_config.load_libraries = true or false. This is true by default. If set to false, pFreak will not load libraries defined in the config.json. This is useful to benchmark library loading times & initiation.

Changes

  • Results: All KPIs are now selected by default.

Fixed

  • Results: Fixed base candidates list not populating correctly if some categories excluded in the overall results.
  • Results: Fixed NaN in overall slowness table when selected different base candidates.

0.3.1 Change Log

Added

  • Results: Added coefficient of variation (How much iterations deviate from its mean execution duration).
  • Results: Devtools Timeline & raw trace data JSON file is now easy to access via a popup by clicking on the number cells.
  • Benchmark Engine: Trying multiple times after a task iteration fail. This is useful to over come issues like dynamic file loading failures & DevtoolsTimelineModel errors (refer below).
  • Benchmark Engine: Added task_config.repeat = support. This is useful to benchmark small tasks by repeating them multiple times to increase the execution duration. This can be set to 1 during only-testing mode. (refer _task_template.js)

Changes

  • Benchmark Engine: Added all JavaScript trace types available at the github.com/ChromeDevTools/devtools-frontend
    • All traces types will be included in the results & categorization is now performed when displaying results.
  • Results: KPI types now populate based on available trace data instead of previous fixed set at index.html
  • Increased waiting time after the start of Tracing.
    • Sometimes the DevtoolsTimelineModel throws an error.
    • The task is set to as failed when that happen & excluded from the results.
    • Re-running the benchmark on that specific task works.
    • Exact reason behind this is not clear. Probably because the tracing started middle of an execution.