메뉴
Run and interpret the WoofWare.PawPrint BenchmarkDotNet performance harness. Use when measuring PawPrint interpreter performance, establishing allocation/time baselines, comparing before/after optimization work, or explaining how to run the repo's `WoofWare.PawPrint.Performance` benchmarks.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
| name | pawprint-performance-benchmark |
| description | Run and interpret the WoofWare.PawPrint BenchmarkDotNet performance harness. Use when measuring PawPrint interpreter performance, establishing allocation/time baselines, comparing before/after optimization work, or explaining how to run the repo's `WoofWare.PawPrint.Performance` benchmarks. |
Use the repo-local BenchmarkDotNet project for PawPrint performance baselines:
nix develop -c dotnet run --project WoofWare.PawPrint.Performance/WoofWare.PawPrint.Performance.fsproj -c Release -- --filter "*StackHeavy*"
For a quick smoke check of the benchmark wiring, use BDN's dry job:
nix develop -c dotnet run --project WoofWare.PawPrint.Performance/WoofWare.PawPrint.Performance.fsproj -c Release -- --filter "*StackHeavy*" --job Dry
Treat --job Dry output only as a path check. Use the non-dry command for before/after numbers.
The benchmark compiles a stack-heavy C# guest program in GlobalSetup, verifies PawPrint's exit code against the real .NET runtime, then measures Program.run with MemoryDiagnoser.
Compare the BenchmarkDotNet summary columns most relevant to interpreter-state work:
Mean: elapsed time per PawPrint run.Allocated: managed allocation per PawPrint run.Gen0/Gen1/Gen2: collection pressure.BenchmarkDotNet writes reports under BenchmarkDotNet.Artifacts/; the repo ignores that directory.