Android Performance Ecosystem

July 16, 2026 · View on GitHub

简体中文

An independent navigation Hub for Gracker's seven Android performance projects. The projects keep separate releases and runtime boundaries; this repository documents how they fit together and validates the shared public navigation.

Android performance ecosystem

The Android Performance Ecosystem brings its navigation Hub and seven core projects into an optional path from instrumentation and capture to analysis, system knowledge, and reproducible cases.

StageProjectPurposeAddress
NavigateAndroid Performance EcosystemMaintain the shared project map, handoff metadata, generated README navigation, and drift checks.GitHub
InstrumentTraceFixInject app-side android.os.Trace sections at build time so method work is visible at runtime.GitHub
Capture and measurePerfetto ToolsCapture repeatable Perfetto traces and collect FPS or Simpleperf measurements.GitHub
AnalyzeSmartPerfettoInvestigate traces with an AI-assisted Web UI, CLI, reports, sessions, comparisons, and evidence workflow.GitHub
Agent analysisPerfetto SkillsGive agents a portable Perfetto analysis Skill for Android, Linux, and Chromium, with selected assets synchronized through pinned workflows.GitHub
LearnAndroid Performance BlogTeach Perfetto and Systrace analysis through articles, system explanations, and case studies.AndroidPerformance.com · GitHub
System knowledgeAndroid Internal WikiAn alpha knowledge base for Android mechanisms from App to Framework, Native, and Kernel.Coming soon
ReproduceTrace for Blog (SystraceForBlog)Provide the Perfetto, Systrace, and related case files used by articles for hands-on reproduction.GitHub

Use the loop

The path is optional rather than a mandatory dependency chain:

  1. Add app-side sections with TraceFix when platform tracks are not enough.
  2. Capture a repeatable trace and measurements with Perfetto Tools.
  3. Choose SmartPerfetto for a complete product experience, or Perfetto Skills when an existing coding agent should analyze the trace locally.
  4. Use the Android Performance Blog and Android Internal Wiki to understand, explain, and review the result.
  5. Reproduce published cases from Trace for Blog, then feed proven lessons back into capture configs, Skills, SmartPerfetto workflows, and articles.

For example, a trace produced by Perfetto Tools can be uploaded to SmartPerfetto or passed to an agent with $perfetto-performance-analysis. Neither analysis surface requires TraceFix, and they do not require each other.

What this hub owns

ecosystem.json is authoritative only for cross-project metadata: project names, lifecycle roles, public entry points, languages, default branches, and README targets. Each project remains the source of truth for its own versions, capabilities, installation, releases, data, and license.

The first implementation closes three concrete gaps:

  • one discoverable map and handoff path across all seven projects;
  • generated README marker blocks plus a read-only remote drift check;
  • a provenance, privacy, license, and hash catalog for the article trace cases.

The remaining gaps and acceptance criteria are tracked in ROADMAP.md.

Validate

python3 -m unittest discover -s tests -v
python3 scripts/ecosystem.py validate
python3 scripts/ecosystem.py check-readmes --remote

The remote audit reads public repositories only. It has no cross-repository write token and never attempts to access non-public repositories.

License and affiliation

This hub's original documentation, catalog, and scripts are MIT-licensed. Every linked project and trace asset keeps its own license or data-use status. This is an independent community project and is not an official Google, Android, or Perfetto project.