Skip to Coverage interview questions

Coverage interview preparation

Start with real interview questions, then learn how to define meaningful bins, sample the right events, diagnose coverage holes, and defend closure with evidence—not a headline percentage.

Question indexCoverage questions to work through first.

Filter the compact directory and open the exact selected question in the practice bank. Premium prompts remain access-gated.

Topic
Access

20 of 20 practice questions shown.

  1. Q0858Explain what functional coverage measuresStrategy & evidence · EasySeparate planned behavior from code exercise and correctness checking.Premium
  2. Q1104Build Traceable Verification Items from a SpecificationStrategy & evidence · MediumTrace each requirement to stimulus, checking, assertions, and coverage.Free
  3. Q0257Coverpoint binsCoverpoints & bins · EasyCompare per-value bins with one aggregate range bin.Free
  4. Q0188Ignored and illegal binsCoverpoints & bins · EasyDistinguish irrelevant values from forbidden observations.Free
  5. Q0031Arrayed coverage binsCoverpoints & bins · MediumPredict fixed-count bin partitioning and boundary allocation.Free
  6. Q0465Cross coverageCrosses & scale · EasyBuild the Cartesian product from the source coverpoint bins.Free
  7. Q0872Cross-bin filteringCrosses & scale · MediumRemove selected combinations with binsof and intersect.Free
  8. Q0522Cover opcode, error, and latency combinationsCrosses & scale · MediumBalance exact bins, bounded crosses, and illegal latency values.Premium
  9. Q0204Detect missing type-priority coverageCrosses & scale · MediumFind the first missing pair in a declared expected domain.Premium
  10. Q0205Report current cross-coverage percentageCrosses & scale · EasyKeep covered and expected counts aligned in percentage reporting.Premium
  11. Q0206Cover arbitrary type and priority valuesCrosses & scale · MediumRepresent sparse, noncontiguous domains without inventing goals.Premium
  12. Q0207Track a sparse three-way packet crossCrosses & scale · MediumTrack a three-dimensional cross with complete sparse keys.Premium
  13. Q0208Exclude illegal coverage combinationsCrosses & scale · MediumKeep excluded combinations out of both numerator and denominator.Premium
  14. Q0269Cover handshakes and backpressureSampling & ownership · MediumCover completed monitored traffic and purposeful backpressure crosses.Free
  15. Q0132Build accepted-traffic cross coverageSampling & ownership · MediumSample exactly once after acceptance and matching completion.Free
  16. Q1080Trace a Transaction Through a UVM EnvironmentSampling & ownership · EasyTrace monitor publication and independent coverage ownership.Premium
  17. Q0827Track named bin hitsReporting & closure · EasyMaintain a required-bin registry with actionable gap reports.Free
  18. Q0623First action when coverage misses a targetReporting & closure · EasyDiagnose reachability, stimulus, RTL, and model gaps before adding seeds.Free
  19. Q0943Purpose of regression testingReporting & closure · EasyUse repeatable regressions to build trends and quality evidence.Premium
  20. Q0733Make a verification-closure caseReporting & closure · HardDefend closure with independent gates, reviewed waivers, and owned risk.Premium

Coverage strategyTurn verification intent into evidence

Coverage is useful when every goal traces to a requirement or risk, samples a meaningful event, and complements a checker that establishes correctness.

Concept model

Four evidence systems, four different questions

  1. 01Requirementwhat must hold
  2. 02Observationwhat must occur
  3. 03Checkingwas it correct?
  4. 04Coveragewas it seen?
  5. 05Dispositionwhat remains?
EvidenceAnswersCannot prove alone
CodeWhat structure ran?Required scenarios
FunctionalWhat intent occurred?Correct results
CheckerWas behavior correct?Scenario completeness
Code
structure exercisedStatements, branches, conditions, toggles, and states.
Functional
intent observedPlanned values, transitions, scenarios, and combinations.
Checking
behavior judgedAssertions, scoreboards, and reference models detect errors.
Traceability
evidence justifiedEvery goal has a requirement, owner, target, and disposition.

Functional coverage records planned scenarios, code coverage exposes exercised implementation structure, and checkers decide correctness. Traceability explains why the combined evidence supports a requirement.

Strong answer

Code coverage reports exercised implementation structure. Functional coverage reports whether authored behaviors, values, transitions, and combinations were observed. Neither says by itself that the observed result was correct.

Reason it through

  • Code metrics are derived from RTL or other instrumented implementation constructs.
  • Functional goals are written from the specification, architecture, and risk analysis.
  • Assertions, scoreboards, and reference models provide the complementary correctness decision.
Practice a related question in the bank
Strong answer

Every instrumented construct can execute while important data combinations, orderings, error cases, or protocol outcomes remain absent or unchecked.

Reason it through

  • One statement execution can cover many semantically different behaviors only superficially.
  • Code coverage follows the implementation, so a missing requirement may have no RTL structure to measure.
  • Closure also requires checking quality, bug status, exclusions, and evidence for planned scenarios.
Practice a related question in the bank
Strong answer

Each item needs a requirement or risk, measurable observation, sampling definition, checker relationship, target, owner, and policy for unreachable or excluded behavior.

Reason it through

  • Use traceability to prevent goals from becoming disconnected from intent.
  • Define the denominator and sample event before regression data arrives.
  • Record ownership and review criteria so closure decisions are accountable.
Practice a related question in the bank

Structural diagnosticsRead code coverage without overclaiming

Code metrics reveal which implementation structures were exercised. They are powerful gap detectors, but a hit says nothing by itself about whether the behavior was intended or correct.

Concept model

Each metric illuminates a different structural blind spot

MetricA hit meansBlind spot
Statementoperation executeduntaken outcome
Branchdecision outcomemasked operand
ConditionBoolean influencesemantic scenario
Toggle0→1 and 1→0encoded values
FSMstate or arc visitedresult correctness
line hitif (valid) send();true ✓false ○
toggle hitmode[1:0]00→11 ✓01 never seen
Statement
operation ranDoes not prove every decision outcome occurred.
Branch
path selectedSeparates taken and not-taken control outcomes.
Toggle
0↔1 observedBit activity is not the same as semantic field coverage.
FSM
state + arcExposes unseen states and transitions in recognized machines.

A line may execute while one branch outcome never occurs, a condition may be masked, or an FSM transition may remain unseen. Diagnose the specific hole before changing stimulus or excluding logic.

Strong answer

Statement coverage records executed operations, branch coverage records control outcomes, and condition coverage examines Boolean operands or combinations that may be masked inside a taken branch.

Reason it through

  • An if statement can execute while only its true or false outcome is observed.
  • Short-circuit evaluation can prevent a subexpression from being independently exercised.
  • Tool definitions vary, so interpret the exact metric documented by the simulator.
Practice a related question in the bank
Strong answer

A bit may toggle because of reset or incidental traffic without exercising its intended function, while a multi-bit field can have full bit toggles and still miss important encoded values.

Reason it through

  • Independent 0→1 and 1→0 activity does not establish legal field combinations.
  • Reset activity should be distinguished from functional operating modes.
  • Use functional bins and checkers when the semantics live above individual bits.
Practice a related question in the bank
Strong answer

First establish why it is unreachable in the verified configuration. Remove dead logic when appropriate, or use formal evidence and a narrowly reviewed exclusion when the behavior is intentionally impossible.

Reason it through

  • Confirm the hole is not caused by a broken constraint or missing initialization sequence.
  • Scope exclusions to the exact construct and operating modes supported by the evidence.
  • Revisit the waiver when RTL, configuration, or assumptions change.
Practice a related question in the bank

Functional modelMake every bin encode intent

A maintainable covergroup partitions sampled behavior by verification consequence: legal classes, boundaries, transitions, errors, and deliberately excluded space.

Concept model

Classify a stable transaction before incrementing a counter

  1. Sampleaccepted txn
  2. Expressiontxn.opcode
  3. Partitionnormal · error
  4. Counterhit count
Scalarbins idle = {0};
Range arraybins op[] = {[1:4]};
Fixed countbins band[4] = {[0:31]};
Transitionbins wake = (0 => 1);
Wildcardwildcard bins err = {4'b1??1};
Excludedignore_bins rsvd = {[12:15]};
Unsized array
bins name[]Creates fine-grained bins from a finite declared set.
Fixed count
bins name[N]Distributes the value set across at most N bins.
Transition
(a => b)Measures ordered sampled changes, not just endpoint values.
Residue
defaultCollects otherwise unclassified values for model review.

SystemVerilog bins can collect a scalar, range, distributed set, wildcard pattern, or transition. Resolution should follow the specification rather than the raw width of the field.

Strong answer

Partition values by behavioral consequence: normal classes, boundaries, modes, errors, and meaningful transitions. Numeric convenience is secondary to verification intent.

Reason it through

  • Give distinct bins to values whose behavior or risk differs.
  • Group values only when one representative observation provides equivalent evidence.
  • Keep the model readable enough that an unhit bin leads to an actionable test or decision.
Practice a related question in the bank
Strong answer

An unsized bin array expands a finite value set into fine-grained bins, while a fixed-count array distributes that set across the requested bounded number of bins.

Reason it through

  • Empty brackets emphasize per-value or otherwise fine-grained partitioning of the declared set.
  • A numeric size caps report resolution and simulator bookkeeping.
  • When the values do not divide evenly, inspect the language and tool allocation rules at the boundaries.
Practice a related question in the bank
Strong answer

Ignored values are intentionally removed from the coverage goal. Illegal-bin hits identify forbidden observations and are reported, but assertions and protocol checkers should remain the primary correctness mechanism.

Reason it through

  • An ignored value changes the denominator because it is outside the modeled goal.
  • An illegal hit carries diagnostic meaning and should not silently disappear.
  • Do not use illegal bins as the only check for temporal or protocol rules.
Practice a related question in the bank

Combination spaceControl cross coverage before it controls you

Marginal coverpoints can each be complete while required combinations remain unseen. Cross only behaviorally important dimensions and make the legal denominator explicit.

Concept model

A complete row and column can still hide an empty cell

Raw dimensions4 × 8 × 396 binsSemantic groups2 × 4 × 324 bins
Hit
Unhit
Ignored
Forbidden
Raw product
4 × 8 × 3 = 96Small dimensions multiply quickly.
Group first
2 × 4 × 3 = 24Merge values only when behavior is equivalent.
Condition
sample validExclude cycles that are outside the transaction contract.
Disposition
hit / prove / waiveEvery remaining cell needs evidence, not hope.

The matrix distinguishes hit, unhit, ignored, and forbidden combinations. Semantic grouping and justified filtering keep the cross aligned with useful, reachable behavior.

Strong answer

Individual completion proves each value occurred somewhere. A cross reveals whether requirement-critical values occurred together in the same valid sample.

Reason it through

  • Two two-bin coverpoints form four starting cross bins, not two.
  • A complete row and complete column do not guarantee every matrix cell is hit.
  • Cross only relationships whose joint occurrence changes behavior or risk.
Practice a related question in the bank
Strong answer

Cross fewer dimensions, group behaviorally equivalent values before crossing, sample only valid contexts, and remove combinations only when their irrelevance or impossibility is justified.

Reason it through

  • Estimate the bin product before implementing the model.
  • Use semantic abstraction instead of relying on large automatic partitions.
  • Keep an auditable distinction among ignored, forbidden, and unreachable combinations.
Practice a related question in the bank
Strong answer

Confirm the combination is required and reachable first, then inspect constraints, generated transactions, monitor reconstruction, sampling conditions, and bin definitions before adding tests.

Reason it through

  • A legal-looking bin can still contradict protocol or configuration dependencies.
  • A reachable transaction may be generated but never sampled because the acceptance condition is wrong.
  • Random seed volume helps only after the legal model and observation path are correct.
Practice a related question in the bank

Sampling semanticsSample the transaction that actually happened

The best model is still wrong if it observes a drive attempt, stale signal, partial transaction, duplicate publication, or invalid protocol outcome.

Concept model

Move from pins to one owned, stable observation

  1. 0Drivevalid + payload
  2. 1Acceptvalid && ready
  3. 2Observemonitor builds txn
  4. 3Classifychecker result
  5. 4Sampleone coverage hit
Pins
attemptValid without ready may not be an accepted transfer.
Monitor
transactionReconstructs stable protocol meaning from sampled signals.
Checker
classificationSeparates correct results, errors, and expected failures.
Collector
one sampleCounts the intended observation exactly once.

A monitor reconstructs the accepted transaction, publishes it once, and lets independent consumers check and cover it. Coverage tied to a checked outcome can sample a classified result later.

Strong answer

Use an event when the covergroup has one unambiguous stable sampling point. Use manual sampling when a transaction or its classification is assembled later or the correct sample context varies.

Reason it through

  • Manual sampling can pass stable transaction fields directly into a covergroup.
  • An event is concise when every observation shares the same clock and acceptance rule.
  • Whichever style you choose, define reset, validity, and duplicate-sample behavior explicitly.
Practice a related question in the bank
Strong answer

Usually in a dedicated subscriber or collector fed by monitor transactions. Coverage tied to a checked outcome may consume scoreboard-classified observations instead.

Reason it through

  • The monitor owns protocol reconstruction and publishes facts without knowing every consumer.
  • Analysis fanout keeps checking and coverage independently reusable.
  • Protocol violations still need assertions or error coverage even if successful transactions are covered later.
Practice a related question in the bank
Strong answer

The DUT, interface, monitor, and coverage code may act in different scheduling regions. Sampling before the observed transaction is stable can capture the prior or partially updated state.

Reason it through

  • Clocking blocks and monitor architecture can establish a deterministic observation region.
  • Sampling a reconstructed transaction avoids reaching back into changing pin-level signals.
  • Counters should reconcile one accepted transfer with one monitor publication and one intended coverage sample.
Practice a related question in the bank

Regression evidenceMerge runs without losing meaning

Merged coverage is a union of compatible bin hits, not an average of percentages. Reproducible metadata and per-instance visibility keep the result interpretable.

Concept model

Union compatible bin evidence across a regression

RunB0B1B2B3B4
smoke
error
stress
directed
merged
Aggregate type100%looks closedchannel[0]100%completechannel[1]42%hidden gap
Merge
set unionRepeated hits remain one covered goal.
Instance
per_instanceShows whether every replicated channel met its obligation.
Goal
report targetChanges the reported target, not what behavior occurred.
Weight
roll-up influenceChanges aggregation without creating new evidence.

Each run contributes unique hits to the merged database. Model version, design revision, configuration, and instance identity determine whether that union is valid.

Strong answer

It represents the union of compatible bin hits across selected runs. It is meaningful only when the design, model, configuration, and merge policy are controlled.

Reason it through

  • Two 50% runs can merge to anywhere from 50% to 100% depending on overlap.
  • Changing bin definitions or model identity can invalidate a direct union.
  • Archive the exact run manifest so the result can be reproduced and audited.
Practice a related question in the bank
Strong answer

Use instance-level visibility when replicated agents, channels, or interfaces have separate obligations. Aggregate type coverage is sufficient only when instances are truly interchangeable for the goal.

Reason it through

  • A global 100% can hide one idle instance if another instance hits every bin.
  • Per-instance reporting costs database space and closure effort, so enable it deliberately.
  • Tie the choice to whether the plan requires every replica to demonstrate the behavior.
Practice a related question in the bank
Strong answer

Goal sets a reporting target, weight changes contribution to an aggregate score, and auto_bin_max limits automatic partitioning. None of them creates verification evidence.

Reason it through

  • Changing a goal can make a report look complete without adding a single hit.
  • Weight expresses reporting policy and can hide or amplify a component in roll-up.
  • Automatic-bin limits control resolution and should not replace explicit semantic bins.
Practice a related question in the bank

ClosureClose holes with a disposition, not a denominator

Closure means planned reachable goals are hit and every remaining hole has a reviewed explanation supported by stimulus, reachability, model, or specification evidence.

Concept model

Use disagreement between metrics as a diagnostic

CodeFunctionalRead asInvestigate
highhighCheckers + bugsstrong evidence, still review
highlowFunctional gapscenario, model, or sampling
lowhighStructural gapdead RTL or coarse model
lowlowEnvironment gapstimulus or maturity
  1. 1required?
  2. 2reachable?
  3. 3generated?
  4. 4observed?
  5. 5sampled?
  6. 6merged?
Required?
plan + riskConfirm the hole represents real intent before chasing it.
Reachable?
stimulus or proofSeparate a generation problem from impossibility.
Observed?
monitor + sampleTrace the event through reconstruction and coverage.
Reviewed?
owned dispositionNo residual hole disappears without accountable evidence.

Code-high and functional-low points to missing scenarios or models; functional-high and code-low points to unexercised implementation or a coarse model. Neither quadrant replaces checking and bug review.

Strong answer

Planned reachable goals are hit, remaining holes have evidence-backed dispositions, exclusions are reviewed, and complementary checking and bug evidence support the release decision.

Reason it through

  • Closure is scoped to a version, configuration set, plan, and signoff policy.
  • A justified unreachable goal can be closed without fabricating a hit.
  • High coverage does not erase open correctness failures or weak checking.
Practice a related question in the bank
Strong answer

Validate the requirement and bin first, then inspect reachability, constraints, generated traffic, monitor reconstruction, sample conditions, and database merge status in that order.

Reason it through

  • The sequence prevents wasted seed farming when the model itself is wrong.
  • Use a targeted reproducer or formal reachability result to separate generation from impossibility.
  • Record the final fix or disposition with an owner and reviewer.
Practice a related question in the bank
Strong answer

Treat the mismatch as diagnostic evidence. Structural holes suggest unexercised RTL; functional holes suggest missing scenarios, coarse stimulus, or model and sampling gaps. Investigate both before waiving either.

Reason it through

  • Functional-high and code-low can mean the functional model is too coarse or dead implementation remains.
  • Code-high and functional-low can mean required combinations were never modeled or observed.
  • Both-high still requires strong checkers, reviewed bugs, and plan traceability.
Practice a related question in the bank

Continue preparing