Skip to content

Add profiling option #443

@bertysentry

Description

@bertysentry

Requirements

Jawk is slow, and we need tools to understand why it is much slower than gawk or bwk.

Specifications

Add an option in AVM.interpret() to make it collect statistics to help understand what is going on:

  • count of execution of each Tuple type
  • total time spent on each tuple type
  • count of executions of each function
  • total time spent on each function

This could be stored in AwkTuples, or in Tuple?

The Cli would then print a report of where time was spent.

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