Skip to content
STIMSMITH

design under verification (DUV)

Concept WIKI v1 · 5/26/2026

In the provided evidence, a design under verification (DUV) is the target design being simulated in a simulation-based processor-verification flow. Stimuli are generated and applied to the processor inputs, functional coverage is monitored, and coverage feedback from the DUV simulation can be used to adjust stimulus-generation constraints.

Overview

A design under verification (DUV) is the design being exercised by a verification process. In the cited processor-verification work, the DUV appears in a simulation-based flow where generated stimuli are applied to processor inputs and the resulting functional coverage is monitored to assess verification completeness. [C1]

Role in simulation-based processor verification

The evidence describes a verification loop with three main elements:

  1. Stimulus generation — stimuli are generated using pseudorandom generators (PRGs). [C1]
  2. Application to the DUV — generated stimuli are applied to the processor inputs; stimuli may be bit vectors on input ports or programs loaded into program memory. [C2]
  3. Coverage monitoring — achieved coverage of the processor functionality is monitored to determine verification completeness. [C1]

Coverage-feedback use

The cited work proposes dynamically altering PRG constraints using a recurrent neural network. The recurrent neural network receives coverage feedback from the simulation of the design under verification, and this feedback is used to influence subsequent stimulus generation. [C3]

Example context from the evidence

For demonstration, the cited study used processors provided by Codasip because their coverage state spaces were reasonably large and varied across processor kinds. The reported experimental outcome was faster coverage closure and the isolation of a small set of high-coverage stimuli suitable for regression tests. [C4]

CITATIONS

4 sources
4 citations
[1] C1: In simulation-based processor verification, stimuli are generated with pseudorandom generators, applied to processor inputs, and coverage of processor functionality is monitored to determine verification completeness. Automation of Processor Verification Using Recurrent Neural Networks
[2] C2: Stimuli in the described flow can be represented as bit vectors applied to processor input ports or as programs loaded directly into program memory. Automation of Processor Verification Using Recurrent Neural Networks
[3] C3: The cited technique dynamically alters PRG constraints using a recurrent neural network that receives coverage feedback from simulation of the design under verification. Automation of Processor Verification Using Recurrent Neural Networks
[4] C4: The cited experiments used Codasip processors for demonstration and reported faster coverage closure plus identification of a small high-coverage stimulus set usable for regression tests. Automation of Processor Verification Using Recurrent Neural Networks