Useful Skew Tradeoffs: Theory Deep Dive

Foundational theory

Useful Skew Tradeoffs is central to ECO Timing Closure. Intentional clock skew can borrow setup time but increases hold risk and clock tree complexity; useful skew is a negotiated margin. Senior STA engineers always tie slack movement to corner, mode, constraint set, and parasitic view — not just a single report line.

Core concepts explained

• Intentional clock skew can borrow setup time but increases hold risk and clock tree complexity; useful skew is a negotiated margin.

• Primary metric: skew budget used, hold impact, clock tree power

• Primary artifact: CTS skew report, useful skew map, hold regression

• Owners: CTS owner, STA lead, PD owner

• Setup: max delay path; Hold: min delay path

• WNS/TNS summarize tail risk across endpoints

Why this matters at signoff

At signoff, Useful Skew Tradeoffs failures block tapeout or force risky ECO. ECO is surgical timing surgery with regression discipline. Wrong analysis setup wastes weeks of PD effort.

Mental model

CLOCK AT CAPTURE FF delayed intentionally

launch edge  |
data path    |-------->
capture edge |    +skew borrow
             more setup margin, hold risk increases

Worked intuition

1. Name corner, mode, and view for the failing run.

2. Open skew budget used, hold impact, clock tree power and identify worst path group.

3. Read critical path: cell vs net, launch/capture clocks.

4. Check constraints on that path (exceptions, generated clocks).

5. Collect CTS skew report, useful skew map, hold regression and tag database versions.

6. Classify: constraint bug, view mismatch, or real path.

7. Propose minimal ECO and list MMMC+SI regression.

Common misconceptions

• One typical corner is enough for signoff.

• False paths can be applied broadly to green-wash violations.

• Implementation WNS equals signoff WNS without view mapping.

• Setup fix automatically preserves hold.

Visual reinforcement

Useful skew borrow

CLOCK AT CAPTURE FF delayed intentionally

launch edge  |
data path    |-------->
capture edge |    +skew borrow
             more setup margin, hold risk increases

Layer responsibilities

STA OWNERSHIP LAYERS — Useful Skew Tradeoffs

layer              owns                         failure mode
----------------   --------------------------   ---------------------
constraints        clocks, IO, exceptions       false violations
implementation     cells, nets, placement       real path delay
extraction         SPEF/RC corners              view mismatch
signoff policy     derate, MMMC matrix          margin disputes
closure           ECO order, regression        fix breaks other corner

STA deep dive

ECO without regression is a coin flip.

Concept diagram

ECO LOOP

failing path -> fix -> setup check -> hold check -> SI -> MMMC matrix

Metric graph

ECO PROGRESS

WNS
 0 |              o target
-50|        o
-100|   o baseline
    +-----------------> ECO #

Reports and artifacts

• pre/post WNS

• hold delta

• leakage delta

• ECO ticket log

Mini case study

48-hour tapeout crunch: VT swap fixed setup; hold fixed with route on short path — order mattered.

Debug branches

• Never ship setup-only proof

• Logical ECO needs equivalence

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.

Theory reinforcement

ECO is surgical timing surgery with regression discipline.