Corner / Mode Matrix Design: Expanded Case Study

Extended case study

Signoff review: corner coverage completeness, mode cross-product size, signoff WNS per scenario fails after an apparently minor change affecting Corner / Mode Matrix Design.

Background

Team had closure last week on full MMMC matrix. New run shows regression clustered in one path group.

Symptoms observed

• corner coverage completeness, mode cross-product size, signoff WNS per scenario regression

• Disagreement between synth and STA WNS

• Hold violations after setup ECO

Investigation timeline

1. Hour 0: freeze netlist, SDC, SPEF, lib tags

2. Hour 1: diff constraints and run manifest

3. Hour 2: isolate corner/mode that fails

4. Hour 3: read worst path — cell vs net, clocks

5. Hour 4: classify root cause

6. Hour 5: minimal ECO or SDC fix

7. Hour 6: full regression + signoff memo

Root cause

Root cause tied to Corner / Mode Matrix Design: MMMC runs STA across PVT corners and functional modes; the matrix must cover real silicon without exploding runtime.

Fix and validation

• Bounded ECO or audited SDC change

• Re-run MMMC scenario table, signoff matrix spreadsheet, scenario WNS dashboard

• Full MMMC + SI regression

Lessons learned

• Tag every run

• Diff constraints before blaming PD

• Setup and hold together

CASE STUDY — Corner / Mode Matrix Design
baseline WNS / regressed WNS / post-fix WNS

Sequence under stress

TIMING PATH — Corner / Mode Matrix Design

launch FF --logic stages--> capture FF
     |                           |
  launch clock                capture clock
     |                           |
  required: arrival < capture_edge - setup
  metric: corner coverage completeness, mode cross-product size, signoff WNS per scenario

STA deep dive

Signoff WNS is the minimum across the full scenario matrix.

Concept diagram

MMMC

corners (ss/tt/ff) x modes (func/test) x views (signoff RC)
= scenario set -> min(WNS) is signoff

Metric graph

WNS HEATMAP BY CORNER

        ss    tt    ff
setup  -80   +10   +5
hold   -15   -2    -8

Reports and artifacts

• per-corner WNS

• mode summary

• lib map audit

• view delta

Mini case study

Passed TT typical but failed SS hold — signoff matrix was incomplete; added low-voltage mode.

Debug branches

• Isolate failing scenario first

• Never compare unlike SPEF tags

Senior review question

Ask: what corner/mode/view proves this topic is closed or failing?

Key takeaways

• State corner, mode, view, and database tag with every slack claim.

• Run setup and hold plus MMMC regression after every ECO.

Common pitfalls

• Comparing STA runs with different SPEF or SDC tags.

• Broad false_path to green-wash violations.

• Setup-only ECO without hold check.

Principal STA review addendum

MMMC runs STA across PVT corners and functional modes; the matrix must cover real silicon without exploding runtime.

Metric: corner coverage completeness, mode cross-product size, signoff WNS per scenario