A Fault Tolerant Approach to Detect Transient Faults in Microprocessors Based on a Non-Intrusive Reconfigurable Hardware

Abstract

International audienceThis paper presents a non-intrusive hybrid fault detection approach that combines hardware and software techniques to detect transient faults in microprocessors. Such faults have a major influence in microprocessor-based systems, affecting both data and control flow. In order to protect the system, an application-oriented hardware module is automatically generated and reconfigured on the system during runtime. When combined with fault tolerance techniques based on software, this solution offers full system protection against transient faults. A fault injection campaign is performed using a MIPS microprocessor executing a set of applications. HW/SW implementation in a reprogrammable platform shows smaller memory area and execution time overhead when compared to related works. Fault injection results show the efficiency of this method by detecting 100% of faults

Survey of Soft Error Mitigation Techniques Applied to LEON3 Soft Processors on SRAM-Based FPGAs

“…to enable consistency checks, whenever any variable is modified or a procedure is called, to detect data errors. A controlflow duplication technique, explained in [40], was also deployed at the assembly instruction level to further enhance error detection. Finally, configuration memory scrubbing was applied through the instantiation of the Xilinx Soft Error Mitigation (SEM) IP core [41] in the system for detection and correction of configuration bit upsets.…”

Efficient mitigation of data and control flow errors in microprocessors

“…A reconfigurationbased approach is proposed in [15]. This approach requires switching to an application-specific module for every application program.…”

A Framework and Protocol for Dynamic Management of Fault Tolerant Systems in Harsh Environments

“…Most of the works targeting on the radiation fault tolerance propose solutions to Single-Event Effects (SEE) [3], [4], [5], [6], [7], therefore focusing on the types of radiations encountered in space. Although Total Ionizing Dose (TID) effects on low-level devices have been reviewed in literature [8], [9], few works focused on the solutions to provide reliability on higher levels.…”

“…To deal with such environments, many proposals have been employed in the state-of-the-art: (i) hardware enhancement to provide fault tolerance for circuits [6], [7]; (ii) software fault-tolerant techniques [4], [5] and (iii) a mix of different proposals [3]. To the best of the authors' knowledge, there is no such system or framework which can provide the flexibility to combine such different techniques.…”