CDC / RDC · All levels
Multi-bit Crossing Strategies: Comparison Matrix
Comparison Matrix for Multi-bit Crossing Strategies.
Comparison matrix
Encoding and protocol patterns trade area, latency, and robustness.
+------------------+----------------+----------------+----------------+
| Approach | Strength | Weakness | Best when |
+------------------+----------------+----------------+----------------+
| Conservative | safe | extra latency | high-risk crossings |
| Balanced | practical | more setup | most production paths |
| Aggressive | fast schedule | escape risk | late-stage crunch |
| Refactor | structural gain | long cycle | recurring failures |
+------------------+----------------+----------------+----------------+When to choose each approach
Select approach by silicon risk, not only schedule urgency
Interview traps
Global waiver spray
Unproven protocol assumptions
Evidence comparison
CDC/RDC EVIDENCE MATRIX — Multi-bit Crossing Strategies
+-----------------------+----------------------------+----------------------------+---------------------------+
| Evidence | Tells you | Does not prove | Next action |
+-----------------------+----------------------------+----------------------------+---------------------------+
| structural report | crossing topology class | protocol liveness | add assertions/formal |
| reset map | release dependencies | dynamic startup behavior | run reset stress tests |
| formal proof | bounded safety/liveness | real firmware sequence | correlate with simulation |
| simulation stress | observed mode behavior | full state-space closure | target unhit corners |
| waiver log | accepted residual risk | technical correctness | periodic revalidation |
+-----------------------+----------------------------+----------------------------+---------------------------+CDC/RDC deep dive
Reconvergence failures are timing + logic coupling hazards.
Concept diagram
RECONVERGENCE
branch A sync -- -> merge logic -> illegal combination window
branch B sync --/Metric graph
HAZARD REPRODUCTION RATE
before fix: 1/2000 runs
after fix: 0/100000 runsReports and artifacts
reconvergence warnings
gray transition checks
hazard replay traces
multi-bit strategy map
Mini case study
Gray pointer was legal, but downstream merge decoded mixed-era values during burst stress.
Debug branches
Trace fanin cones
align event windows
verify decode assumptions
Senior review question
Ask: what evidence proves this risk is closed for silicon, not just tool-clean?
Key takeaways
State crossing class, assumptions, and owner with every issue.
Run structural and dynamic regressions after each fix.
Common pitfalls
Treating all warnings as equivalent risk.
Waiving issues without containment evidence.
Skipping reset and reconvergence stress after CDC fixes.
Principal CDC/RDC review addendum
Bundled-data handshakes, FIFOs, source-synchronous strobes, and encoded protocols each trade latency, area, and proof complexity.
Metric: coherency guarantee coverage, data corruption incidents, protocol overhead