openroad-debug

Diagnose and iterate on an OpenROAD/ORFS failure (crash, hang, race, nondeterministic result) and set up a fast edit/measure loop before you package a reproducer.

설치로 이동

저장소: The-OpenROAD-Project/bazel-orfs
최근 소스 활동: 2026년 8월 23일 17:38
감지된 SKILL.md 언어: 영어
스타: 26
포크: 23

설치 방법

기본적으로 소스를 먼저 확인하는 Prompt가 선택됩니다. 직접 명령으로 전환하거나 로컬 사본을 다운로드할 수도 있습니다.

Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.

직접 명령은 검토 Prompt를 거치지 않습니다. 실행하기 전에 소스를 확인하세요.

npx skills add https://github.com/The-OpenROAD-Project/bazel-orfs --skill openroad-debug

명령은 한 줄로 유지됩니다. 복사하기 전에 가로로 스크롤해 전체 내용을 확인하세요.

로컬 사본을 원하시나요? SkillsMP에서 현재 제공할 수 있는 파일을 다운로드하세요.

소스 파일 검토

설치 여부를 결정하기 전에 SKILL.md와 SkillsMP에 표시된 보조 파일을 읽어 보세요.

SKILL.md

소스 지침 · 읽기 전용 미리보기

name: openroad-debug
description: Diagnose and iterate on an OpenROAD/ORFS failure (crash, hang, race, nondeterministic result) and set up a fast edit/measure loop before you package a reproducer.

# openroad-debug The instructions for this skill are maintained in the shared Claude slash command. Please read and follow the single source of truth: [../../../.claude/commands/openroad-debug.md](../../../.claude/commands/openroad-debug.md)

GitHub에서 보기

검색으로 돌아가기 설치로 돌아가기

debug-rtl-sim

Debug an RTL or gate-level simulation that hangs, times out, produces no output, or computes the wrong answer, without dumping logs or waveforms into context. Covers the cheapest-first ladder (which gate failed, how the run stopped, a bounded fetch trace, bounded disassembly, reduce to the smallest failing case, and only then a windowed waveform), the harness properties that make each rung possible (distinct exit codes, a cycle budget, a probe port, the program as a runtime plusarg), and why a plausible-looking number from a broken configuration is the failure mode to fear most. Use when a simulation "just hangs", when a core produces no output, when a workload passes on one configuration and not another, or before reaching for a waveform.

2026-09-14

study-pr

Run a measurement study on the flow and publish it as a closed, reference-only bazel-orfs pull request. Covers the two kinds of patch a study carries into OpenROAD, ORFS or yosys (instrumentation that stays on the study branch, candidate fixes carried in patches/ one concern each), the edit/measure loop, the noise discipline that makes a runtime or QoR number trustworthy (byte-identical inputs, fixed and witnessed knobs, repeats with 2σ and a stated resolution, an idle machine asserted not hoped), the harness shape (manual flow targets, unit-tested parsers, a report that discovers results and says "not yet measured"), and the PR shape (study/<slug> branch, the finding as the title, graphs and tables in the body with images committed to the branch, raw data as a CSV comment, a closing comment). Use when asked to "study", "measure", "find out where the time goes", "A/B a knob or patch", or to document design decisions that go into an upstream PR.

2026-09-10

repair-timing-grt

Investigate OpenROAD global-route repair_timing correctly — the grt "grind"/timeout, hold- and setup-buffer insertion. Covers why a skip-repair (fast) grt measures nothing about repair, why repair_timing must run inside global_route.tcl on the pre-GR ODB, how to split the grind into setup vs hold before blaming either, why the win is measured as clk_period-WNS in picoseconds rather than as a percentage of WNS, and why a design that closes measures nothing about a repair knob. Use when a grt stage times out or runs for hours, when a change is meant to affect repair_timing, or when reasoning about hold vs setup buffer insertion at global route.

2026-09-08

이 저장소의 모든 Skills 보기

openroad-debug

소스 정보

설치 방법

소스 파일 검토

이 저장소의 다른 Skills

이 저장소의 다른 Skills