ECO Regression Discipline: Theory Deep Dive

Foundational theory

ECO Regression Discipline is a core part of Incremental & ECO Synthesis. Regression discipline prevents late-stage surprises by enforcing setup/hold, power, and equivalence checks after every ECO wave. Senior synthesis engineers connect every QoR claim to constraint context, compile setup, and reproducible artifacts.

Core concepts explained

• Regression discipline prevents late-stage surprises by enforcing setup/hold, power, and equivalence checks after every ECO wave.

• Primary metric: post-ECO regressions caught, corner coverage, signoff approval latency

• Primary artifact: regression matrix, signoff checklist, release memo

• Owners: timing closure lead, verification owner, program owner

• Compare timing, area, and power together

• Preserve run manifest for every regression jump

Why this matters at closure

At tapeout pace, ECO Regression Discipline decisions can either shorten closure loops or create hidden debt. Incremental compile and ECO flow protect schedule by limiting unnecessary churn.

Mental model

[ ] setup corners
[ ] hold corners
[ ] power checks
[ ] equivalence
[ ] handoff report update

Worked intuition

1. Freeze RTL tag, constraint tag, and compile switches.

2. Open post-ECO regressions caught, corner coverage, signoff approval latency and isolate the first meaningful regression.

3. Classify whether issue is constraints, mapping transform, or physical estimate.

4. Collect regression matrix, signoff checklist, release memo and owner signoff evidence.

5. Pick minimal reversible fix and define rollback criteria.

6. Run timing + power + area regression matrix before merge.

Common misconceptions

• Better WNS always means better overall QoR.

• dont_touch is harmless if timing still passes.

• Retiming gain is free and always safe for formal.

• Topographical estimates are equivalent to signoff route outcomes.

Visual reinforcement

Regression matrix

[ ] setup corners
[ ] hold corners
[ ] power checks
[ ] equivalence
[ ] handoff report update

Layer responsibilities

SYNTHESIS OWNERSHIP LAYERS — ECO Regression Discipline

layer               owns                          failure mode
----------------    ---------------------------   -------------------------
constraints         clocks/exceptions/policy      fake QoR optimism
mapping             cell choices/structure        depth/fanout regressions
optimization        timing/power tradeoffs        one-metric overfitting
physical-aware      topo/congestion estimates     handoff delta surprises
closure             ECO order/regression          fixes break other corners

Synthesis deep dive

Incremental and ECO synthesis protect schedule only with strict regression discipline.

Concept diagram

INCREMENTAL ECO LOOP

identify delta -> localized compile -> patch -> regression -> release

Metric graph

ECO CYCLE TIME

full compile      ███████████
incremental       █████

Reports and artifacts

• incremental reuse

• ECO patch list

• LEC status

• regression matrix

Mini case study

Localized compile cut turnaround 40%, but skipped LEC caused rollback.

Debug branches

• Reuse ratio is not enough

• Patch locality check

• Always include equivalence hooks

Senior review question

Ask: what evidence proves this QoR move is real and stable?

Key takeaways

• State exact run context (RTL, SDC, libs, switches) with every QoR claim.

• Re-run timing, area, and power regressions after each synthesis ECO.

Common pitfalls

• Comparing runs with mismatched constraints or library views.

• Timing-only fixes that violate power or area budgets.

• Skipping equivalence checks after structural changes.

Theory reinforcement

Incremental compile and ECO flow protect schedule by limiting unnecessary churn.