Skip to content
STIMSMITH

Stimulus Generation

Technique WIKI v1 · 5/26/2026

Stimulus generation in simulation-based processor verification is the production of inputs—such as port-level bit vectors or executable programs—to exercise a processor design and measure functional coverage. Evidence describes a PRG-based approach in which generated stimuli are applied to processor inputs, coverage is monitored, and recurrent neural networks can dynamically alter PRG constraints using coverage feedback to reach coverage closure faster and identify compact high-coverage regression stimuli.

Overview

In simulation-based processor verification, stimulus generation refers to generating inputs for a processor under verification and using the resulting functional coverage to assess verification completeness. The cited evidence describes the prevailing approach as generating stimuli with pseudorandom generators, applying those stimuli to processor inputs, and monitoring achieved coverage of processor functionality. [C1]

Stimulus forms

Stimuli may be represented in different forms. The evidence identifies two examples: bit vectors applied to processor input ports, and programs loaded directly into program memory. [C2]

Coverage-guided generation

The evidence describes a technique that dynamically alters constraints for a pseudorandom generator using a recurrent neural network. In this approach, the recurrent neural network receives coverage feedback from simulation of the design under verification, and that feedback is used to guide subsequent stimulus generation. [C3]

Demonstrated use and reported effects

The technique was demonstrated on processors provided by Codasip, with the stated rationale that their coverage state spaces are reasonably large and differ across processor kinds. The source also states that the presented techniques are widely applicable. [C4]

Reported experimental results indicate two outcomes: coverage closure was achieved much sooner, and a small set of high-coverage stimuli could be isolated for use in regression tests. [C5]

LINKED ENTITIES

1 links
Coverage Feedback uses_feedback

CITATIONS

5 sources
5 citations
[1] In simulation-based processor verification, stimuli are generated using pseudorandom generators, applied to processor inputs, and coverage is monitored to determine verification completeness. Automation of Processor Verification Using Recurrent Neural Networks
[2] Stimuli 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] A proposed technique dynamically alters pseudorandom-generator constraints via a recurrent neural network that receives coverage feedback from simulation of the design under verification. Automation of Processor Verification Using Recurrent Neural Networks
[4] The technique was demonstrated using Codasip processors, whose coverage state spaces are described as reasonably large and different across processor kinds, and the paper states the techniques are widely applicable. Automation of Processor Verification Using Recurrent Neural Networks
[5] Experimental results reported faster coverage closure and isolation of a small set of high-coverage stimuli suitable for regression tests. Automation of Processor Verification Using Recurrent Neural Networks