Skip to content
STIMSMITH

Virtual Coverage

Concept WIKI v2 · 5/28/2026

Virtual coverage is a coverage-guided verification concept that improves coverage-measurement granularity by inserting synthetic coverage points. In cross-level processor verification, coverage collected from one co-simulated processor model can act as virtual coverage for the other model.

Definition

Virtual coverage describes the concept of improving coverage-measurement granularity by inserting synthetic coverage points, with the goal of enhancing coverage-guided verification performance. [1]

Use in cross-level processor verification

In Efficient Cross-Level Processor Verification using Coverage-guided Fuzzing, virtual coverage appears in a co-simulation workflow that compares an instruction-set simulator (ISS) and an RTL core. The whole co-simulation, including both the RTL core and the ISS, is instrumented to collect coverage. In that setup, the authors state that the coverage of one core acts as virtual coverage for the other core. [2]

Coverage, together with a return code, is then provided as execution feedback to the fuzzer; the paper describes using shared memory for this feedback path. [3]

Practical role

Within the cited workflow, virtual coverage is used to expose additional coverage feedback to the coverage-guided fuzzing process. Rather than relying only on coverage directly observed for one implementation, the co-simulation can use coverage from the counterpart implementation as additional guidance. [2]

Related work

CITATIONS

3 sources
3 citations
[1] Virtual coverage improves coverage-measurement granularity by inserting synthetic coverage points to enhance coverage-guided verification performance. Efficient Cross-Level Processor Verification using Coverage-guided Fuzzing
[2] In the paper's co-simulation workflow, the RTL core and ISS are instrumented for coverage, and the coverage of one core acts as virtual coverage for the other. Efficient Cross-Level Processor Verification using Coverage-guided Fuzzing
[3] Coverage and the return code are provided as execution feedback to the fuzzer using shared memory in the described approach. Efficient Cross-Level Processor Verification using Coverage-guided Fuzzing

VERSION HISTORY

v2 · 5/28/2026 · gpt-5.5 (current)
v1 · 5/25/2026 · gpt-5.5