Diagnose PPA
Read the selected rows in <project>/reports/flow_engine.json and open their
native reports. flow/steps.json names each current tool and
flow/native_limits.json retains its original limits. Review tool_ran,
native_steps, native_metrics, and each metric's value, operator and limit.
A missing native measurement is NOT_MEASURED. AI reviews the engineering
meaning of every result and fixes problems in the design, tool or plugin.
python3 programs/flow_engine.py report --project <project>
Start with the measured failing metric and the original requirement. Read the
native path, cell, net, power or congestion breakdown that accounts for it.
- For timing, separate cell delay, interconnect, logic depth, fanout and skew.
Check constraints and I/O loading before proposing an RTL change.
- For area, identify the dominant cells, memories, macros and unused reserve.
Keep intentional ECO spare area separate from removable logic.
- For power, identify activity, capacitance, voltage, leakage and the dominant
switching structures. State any missing activity or model coverage.
- For congestion or IR, connect the hotspot to the actual geometry, grid and
measured demand.
- Propose a bounded change and a falsifiable expected effect. Preserve the
previous design, run the owning tools, and compare the same metric and scope.
Report observations separately from hypotheses. Cite source paths and native
numbers. An untested explanation remains a hypothesis. Keep original limits
and disclose tradeoffs in every other affected metric. No context receipt or
separate diagnostic authorization is required.
Large repeated physical-verification counts need a setup diagnosis before a
claim about the design. Confirm the actual GDS/DEF/Magic view selected by the
step, merged cell geometry, cell GDS/CDL pin interfaces, metal/via/track maps,
taps/endcaps/fillers and supply/substrate connections. Group native markers by
rule and inspect their coordinates. Report the first LVS mismatch class and
separate missing library annotations from device/net/property differences.
A measured GF180 7-track case illustrates both causes. A vendor two-supply
GDS/LEF set lacked the declared library's explicit well/substrate pin texts;
DEF-based extraction and a powered netlist from that foreign database produced
thousands of spurious LVS mismatches. Native declared-SCL power connections
and missing library texts restored unique LVS matching without changing any
source polygon. The same layouts had real PDN errors: 0.26 µm cuts repeated
at 0.128 µm pitch overlapped, and a 100 µm tap-placement distance violated the
15 µm tap-distance rules. Name the measured shipped platform revision and
actual rule families; do not generalize its result to every ORFS version.
1---2name: ppa-diagnose3description: Explain why a PPA number is where it is — which paths, cells, nets or flow decisions account for a timing, power or area result — starting from what the deterministic diagnosis already found and adding only falsifiable hypotheses on top. Use when the user says "why is timing failing", "where is the area going", "what is burning the power", "diagnose this PPA result", or when a closure loop needs a reason before it needs a candidate.4metadata:5 source: https://skillmd.com/skills/vibeic/ppa-diagnose6---78# Diagnose PPA910Read the selected rows in `<project>/reports/flow_engine.json` and open their11native `reports`. `flow/steps.json` names each current tool and12`flow/native_limits.json` retains its original limits. Review `tool_ran`,13`native_steps`, `native_metrics`, and each metric's value, operator and limit.14A missing native measurement is `NOT_MEASURED`. AI reviews the engineering15meaning of every result and fixes problems in the design, tool or plugin.1617```bash18python3 programs/flow_engine.py report --project <project>19```2021Start with the measured failing metric and the original requirement. Read the22native path, cell, net, power or congestion breakdown that accounts for it.23241. For timing, separate cell delay, interconnect, logic depth, fanout and skew.25 Check constraints and I/O loading before proposing an RTL change.262. For area, identify the dominant cells, memories, macros and unused reserve.27 Keep intentional ECO spare area separate from removable logic.283. For power, identify activity, capacitance, voltage, leakage and the dominant29 switching structures. State any missing activity or model coverage.304. For congestion or IR, connect the hotspot to the actual geometry, grid and31 measured demand.325. Propose a bounded change and a falsifiable expected effect. Preserve the33 previous design, run the owning tools, and compare the same metric and scope.3435Report observations separately from hypotheses. Cite source paths and native36numbers. An untested explanation remains a hypothesis. Keep original limits37and disclose tradeoffs in every other affected metric. No context receipt or38separate diagnostic authorization is required.3940Large repeated physical-verification counts need a setup diagnosis before a41claim about the design. Confirm the actual GDS/DEF/Magic view selected by the42step, merged cell geometry, cell GDS/CDL pin interfaces, metal/via/track maps,43taps/endcaps/fillers and supply/substrate connections. Group native markers by44rule and inspect their coordinates. Report the first LVS mismatch class and45separate missing library annotations from device/net/property differences.4647A measured GF180 7-track case illustrates both causes. A vendor two-supply48GDS/LEF set lacked the declared library's explicit well/substrate pin texts;49DEF-based extraction and a powered netlist from that foreign database produced50thousands of spurious LVS mismatches. Native declared-SCL power connections51and missing library texts restored unique LVS matching without changing any52source polygon. The same layouts had real PDN errors: 0.26 µm cuts repeated53at 0.128 µm pitch overlapped, and a 100 µm tap-placement distance violated the5415 µm tap-distance rules. Name the measured shipped platform revision and55actual rule families; do not generalize its result to every ORFS version.