Skip to content
cccskills
FreePublished on GitHub

bb-collect-coverage

解析 Verilator coverage.dat + sim log,输出 functional + code coverage 数值,判断是否 100% 达标。触发场景:(1) bb-invoke-verilator 后;(2) 每次回归后;(3) 显式 /bb-collect-coverage。

The English summary is being prepared. Showing the original description.

See how to install

Files included (6)

  • SKILL.md3.8 KB
  • Gotcha/coverage_pitfalls.md760 B
  • references/verilator_coverage_format.md512 B
  • scripts/parse_cov.py3.0 KB
  • scripts/render_cov_sh.py748 B
  • scripts/run_cov.py1.2 KB

SKILL.md (original)

Review the instructions the agent will receive before you install.

bb-collect-coverage

职责

读 coverage.dat(verilator)+ sim log(covergroup hits),输出 line/branch/toggle/functional 百分比与未覆盖 bin 列表。

  • 调用者:bba-guru-verification
  • 上游:bb-invoke-verilator
  • 禁止使用:Task / Agent / Skill

Input Args

argtyperequired默认说明
sim_logpathtrue—designs/<name>/sim_results/<stamp>.log
coverage_datpathfalse同 sim_resultsverilator coverage 数据库
design_namestringtrue——
target_pctfloatfalse100.0functional & code 阈值
stampstringfalse<auto>—

Output Contract

写到 designs/<name>/coverage.json(中间文件)和 designs/<name>/test_report.json(按 .claude/schemas/test_report.schema.json 的嵌套结构)。

field值
artifact_pathdesigns/<name>/test_report.json
coverage_jsondesigns/<name>/coverage.json
script_pathdesigns/<name>/sim_results/parse_cov_<stamp>.sh
log_pathdesigns/<name>/sim_results/cov_<stamp>.log
functional_coveragefloat(0..100)
code_coverage{line: float, branch: float, toggle: float} (NESTED, fix C-01)
meets_targetbool(functional == 100 AND line == 100 AND branch >= 95 AND toggle >= 90,与 test gate 一致)
tests[{name, status, log}](test_report.schema 必填;逐用例结果)
iteration_countint 0..8(test_report.schema 必填)
uncovered_bins[{group, bin, hits}]
validbool

4-Phase 执行

Phase 1 — render_cov_sh

#!/bin/bash
verilator_coverage --annotate designs/<name>/sim_results/annotate/ <coverage_dat>

Phase 2 — run_cov

timeout 300 bash <script_path> > <log> 2>&1

Phase 3 — parse_cov

scripts/parse_cov.py:

  • annotate 输出含 LCOV:hit/total 标记 → code_coverage.line/branch/toggle
  • sim_log 中 covergroup Coverage: <pct>% → functional_coverage
  • 列举 hits==0 的 bin → uncovered_bins
  • meets_target = functional_coverage >= target_pct AND code_coverage.line >= target_pct AND code_coverage.branch >= 95 AND code_coverage.toggle >= 90
    • 100% branch/toggle coverage is often unreachable due to defensive error-handling code. Targets of 95%/90% allow for documented unreachable bins while maintaining high quality.
  • 写 coverage.json(含 inputs[]:{path,sha256} 引用 sim_log + rtl_artifact.json)
  • 同时生成 test_report.json,遵循 .claude/schemas/test_report.schema.json 嵌套结构(fix C-01);必须填齐 schema required 字段:functional_coverage、code_coverage{line,branch,toggle}、tests[]:{name,status,log}(逐用例结果,由 verification 汇总)、inputs[]:{path,sha256}、iteration_count(fix M-03)

Phase 4 — return

返回 JSON。bba-guru-verification:

  • meets_target=true → signoff,开 ready-for-synth
  • meets_target=false → 把 uncovered_bins 反馈,追加 corner-case 用例重 sim

收敛 / 失败

状态行动
meets_target=true进 synth
meets_target=falseoptimization loop(追加 seed / corner test)
解析失败重试 1 次
iter > 10escalate arch-needs-fix(不可达 bin)

资源索引

  • scripts/render_cov_sh.py、scripts/run_cov.py、scripts/parse_cov.py
  • references/verilator_coverage_format.md
  • Gotcha/coverage_pitfalls.md — unreachable code / dead branch

Reviews

No reviews yet. Tell others what you thought.

Comments

No login needed. Share how it worked for you, ask a question, or note where it did not run.

No comments yet. Be the first to write one.

Comments are published right away. Do not include personal information such as email addresses or phone numbers. Up to 2 URLs.

0 / 1,000 characters

This form is protected by reCAPTCHA, and the Google Privacy Policy and Terms of Service apply.

More skills from this repository

<一句话说清『这个 skill 做什么 + 何时调用』。把最关键的触发场景放最前面。结尾给 2-3 个用户会说的关键词。>

The English summary is being prepared. Showing the original description.

amoslee2026/Babel49Updated Aug 4, 2026

从粗略 idea 设计芯片或功能模块架构,输出专业架构文档到 spec_arch 目录。Trigger: 芯片架构设计, IC架构, 电路模块设计, 硬件架构. Generate IC/Chip architecture specification from rough ideas.

The English summary is being prepared. Showing the original description.

amoslee2026/Babel49Updated Aug 4, 2026

基于 AST 检查 CDC/RDC 违例:对比 MAS clock_domains 找跨域信号,检查是否被 2ff-sync CBB 保护。触发场景:(1) bba-guru-synthesis 综合前;(2) 显式 /bb-check-cdc。

The English summary is being prepared. Showing the original description.

amoslee2026/Babel49Updated Aug 4, 2026

调用 babel-lsp / verible-verilog-lint 检查 SV 源码。src 必须可综合(零 syntax error),tb 允许 verification constructs。发现 src error 时自动修复重检(max 3 iter)。触发:(1) bba-guru-rtl 生成后;(2) RTL 修复重检;(3) 显式 /bb-check-lint。

The English summary is being prepared. Showing the original description.

amoslee2026/Babel49Updated Aug 4, 2026

Babel internal issue protocol — close a handoff after downstream agent picked it up or user signed off. Moves designs/<name>/.handoff/<label>.md to .handoff/closed/ (using mv per CLAUDE.md), appends to handoff_log.jsonl.

The English summary is being prepared. Showing the original description.

amoslee2026/Babel49Updated Aug 4, 2026

对抗性代码评审:代码质量 / 可维护性 / 时序风险 / 综合友好度 / MAS 对齐度。默认 ruthless 模式,找 EVERY flaw。触发场景:(1) RTL lint 后;(2) 显式 /bb-code-review。

The English summary is being prepared. Showing the original description.

amoslee2026/Babel49Updated Aug 4, 2026

See all skills by amoslee2026

Report a problem with this skill