Skip to content

Gap: console.time/timeLog timer resolution — prints near-zero instead of real elapsed ms #155

Description

@proggeramlug

Summary

test_gap_console_methods.ts exercises console.time / console.timeLog / console.timeEnd. The elapsed value prints as 0.005ms / 0.010ms where Node reports 1.271ms / 2.517ms. The actual work between calls is a real setTimeout(resolve, N) so the correct elapsed should be at least the sleep duration.

Actual vs Expected (excerpt)

< timer1: 0.005ms            (perry)
> timer1: 1.271ms            (node)
< timer1: 0.010ms            (perry)
> timer1: 2.517ms            (node)

Scope

  • console.time(label) stores the start time at performance.now() precision
  • console.timeLog(label) / console.timeEnd(label) compute now - start and format to ms with 3 decimals
  • Perry's current impl appears to use a clock that either doesn't advance during the timer wait or uses the wrong monotonic source

Related: see #74 (Date.now not advancing during tight CPU loops — different root cause but similar symptom class).

Impact

One of six currently-failing gap tests (baseline 22/28). Likely 3-4 lines of runtime change in console.rs or builtins.rs.

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions