ExpertQuestion 4 of 10Source page 16

SPEF loaded without a fatal error, but one critical net reports almost zero net delay. Does that prove parasitic annotation is correct?

From PDVerse MMMC Interview Masterclass, part of the pdVerse Mentor Series

Direct answer

No. A clean read_spef command only confirms syntax validity. If net or instance names in the SPEF do not match the gate netlist (e.g. hierarchy delimiter mismatches or renamed ECO nets), the tool silently ignores the parasitics and defaults to zero wire delay.

Technical Reference DiagramSPEF loaded without a fatal error, but one critical net reports almost zero net delay. Does that prove parasitic annotation is correct?

Mentor explanation

Parasitic annotation must be verified with coverage reports, not just command exit codes.

Key terms

  • report_annotated_parasitics โ€” PrimeTime command reporting the percentage of nets with valid annotated R and C.
  • Hierarchy delimiter mismatch โ€” using / vs . in net names can cause 100% of SPEF nets to fail matching.
  • Silent default โ€” unannotated nets silently default to zero or estimated wire delay, masking real timing violations.
  • Unannotated net report โ€” report_annotated_parasitics -check lists every net that missed annotation.

Signoff requires 100% parasitic annotation on all internal nets before timing reports can be trusted.

Practical example

PrimeTime Parasitic Annotation Audit:

# 1. Check annotation percentage across all nets
report_annotated_parasitics -check
# Output:
# Total number of nets             = 1,452,300
# Number of fully annotated nets   = 1,450,100 (99.85%)
# Number of unannotated nets       =     2,200 ( 0.15%) <-- SIGNOFF ESCAPE RISK!

# 2. Inspect specific unannotated critical nets
report_annotated_parasitics -unannotated -max_nets 50

Interview trap

Relying on read_spef returning exit code 0 without running report_annotated_parasitics -check. Silent annotation mismatches cause wire delays to evaluate as zero, hiding real failures.

Key takeaways

  • Clean SPEF file reading does NOT guarantee that parasitics were correctly mapped to netlist pins.
  • Name mismatches cause tools to silently fall back to zero wire delay.
  • Signoff leads must audit report_annotated_parasitics to guarantee 100.00% annotation coverage.
Self-check: can you answer this aloud?

Try a 45-second answer using this structure:

  1. State the direct answer.
  2. Explain the timing or physical reason.
  3. Name one caveat.
  4. Say how you would verify it in a real flow.

MMMC Signoff Guide

Explore the full 10-chapter MMMC guide on modes, PVT corners, RC parasitics, analysis views, and correlation between implementation and signoff tools.

Offline PDF Bundle

Want all 1109 questions offline?

Get the complete 4-book PDF bundle (PnR, STA, MMMC, Low Power) with a clickable table of contents - no ads, no internet needed.

See what's inside the bundle
PnR Flow Physical Design Mentor Guide โ€” eight chaptersPnR Flow Mentor GuideEight chapters, library setup through to stream-out.