Automation Optimization

This document describes the scaffolding tools, CI/CD automation, parameter sweeps, RTL optimization guidelines, and static-analysis reporting for RVSvKit, consistent with Overview and Module Inventory.

1. Code Scaffolding (in scripts/)

• scaffold_module.py
  Generates new module directory under modules/<category>/<name>/ with src/, tb/, and README stub.

• scaffold_pkg.py
  Creates package skeleton in common/pkg/ with standard header comment and CHANGELOG.md entry.

2. CI/CD Automation

• GitHub Actions

  • Functional CI (verilator.yml) on PR/push.

  • Synthesis CI (vivado_docker.yml) nightly.

• Scheduled Tasks

  • Weekly static timing analysis on examples/simple_soc.

  • Daily security scan of Docker images for known CVEs.

3. Parameter Sweeps & Auto‑Tuning

• sweep_params.py

  • Automates multi-dimensional sweeps over DATA_WIDTH, UNICORE flag, pipeline depth.

  • Collects area, timing, and power metrics into ci/benchmarks/ CSV files.

• Auto‑Tuner
  Future integration: use scripts to search for Pareto‑optimal configurations and update thresholds.json automatically.

4. RTL Optimization Guidelines

• Multi‑Cycle Operator Pipelining

  • Automate insertion of pipeline registers around long operators (e.g. multipliers) via scripts.

• Resource Utilization Hooks

  • Embed synthesis pragmas (/* synthesis syn_use_dsp=1 */) parameterized by USE_DSP flag.

• Consistency with Style

  • Generated code honors 03-coding-conventions.md (indentation, naming, non‑blocking assignments).

5. Static Analysis & Reporting

• Linting

  • svlint scans for anti-patterns, mixed blocking/non‑blocking usage.

• Assertion Coverage

  • Report assertion pass/fail counts per module.

• Security Checks

  • Enforce no unauthorized DPI imports; flag use of import "DPI-C".

• Dashboard Generation

  • Use a script to convert CSV benchmarks into HTML dashboard under docs/benchmarks/.