Skip to main navigation Skip to search Skip to main content

An Automated and Effective Approach for SBST Generation Targeting RISC-V CPUs

  • Endri Kaja
  • , Nicolas Gerlin
  • , Jad Al Halabi
  • , Ares Tahiraga
  • , Sebastian Prebeck
  • , Dominik Stoffel
  • , Wolfgang Kunz
  • , Wolfgang Ecker
  • Infineon Technologies AG
  • University of Kaiserslautern

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Scopus citations

Abstract

The trend toward scaling and more complex fabrication techniques in digital circuit design often leads to numerous faults within Integrated Circuits (ICs). To address these vulnerabilities, various Design-for-Test (DFT) techniques have been developed for thorough testing during IC manufacturing. However, integrating these DFT infrastructures introduces significant overhead in both area and performance. To overcome these challenges, especially for testing processor cores, Software-Based Self Test (SBST) has emerged as a promising alternative. This paper proposes a novel, automated, and efficient approach for generating SBST tailored for RISC-V processor cores. By combining formal verification with fault simulation, the number of required assertions is reduced by eliminating faults detected by existing test patterns. In addition to SBST generation, our approach introduces a novel Program Flow Checking (PFC) technique, ensuring adherence to ISO 26262 standards and providing a high Fault Detection Rate (FDR). Experimental results show that various RISC-V processor components achieve Fault Coverage (FC) greater than 91%, with the PFC providing a 100% FDR for most components. The methodology is fully automated, utilizing Model-Driven Architecture (MDA) principles.

Original languageEnglish
Title of host publication37th IEEE International Symposium on Defect and Fault Tolerance in VLSI and Nanotechnology Systems, DFT 2024
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9798350366884
DOIs
StatePublished - 2024
Event37th IEEE International Symposium on Defect and Fault Tolerance in VLSI and Nanotechnology Systems, DFT 2024 - Didcot, United Kingdom
Duration: 8 Oct 202410 Oct 2024

Publication series

NameProceedings - IEEE International Symposium on Defect and Fault Tolerance in VLSI and Nanotechnology Systems, DFT
ISSN (Print)2576-1501
ISSN (Electronic)2765-933X

Conference

Conference37th IEEE International Symposium on Defect and Fault Tolerance in VLSI and Nanotechnology Systems, DFT 2024
Country/TerritoryUnited Kingdom
CityDidcot
Period8/10/2410/10/24

Keywords

  • Fault Simulation
  • Model-Driven
  • PFC
  • RISC-V
  • SBST

Fingerprint

Dive into the research topics of 'An Automated and Effective Approach for SBST Generation Targeting RISC-V CPUs'. Together they form a unique fingerprint.

Cite this