Microprocessor Fault Detection via Hardware FAME and Safe Mode

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Traditional fault tolerance configurations for microprocessors incur significant performance and area overheads in detecting and responding to malicious fault attacks, making them inefficient for robust protection against fault attacks.

Innovation Solution

A microprocessor fault detection and response system utilizing a hardware-based fault-attack aware microprocessor extension (FAME) and a software-based trap handler, which detects fault injections, switches the microprocessor to a safe mode, and restores it to a fault-free state using a fault recovery register (FRR), thereby protecting the processor without substantial performance and area overheads.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If traditional fault tolerance configurations are used to detect and respond to malicious fault attacks, then the microprocessor security is improved, but the performance and area overheads increase significantly

Engineering Contradiction:
Improvemicroprocessor securityVSAvoidperformance
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The fault detection and response system is divided into distinct modular components: a fault detection unit that monitors for fault conditions, a safe mode transition mechanism that isolates the fault, and a recovery mechanism that restores normal operation. This segmentation allows each component to be optimized independently, reducing overall overhead while maintaining security.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system performs preliminary actions by pre-configuring safe mode operation and recovery procedures before faults occur. The fault detection unit continuously monitors for fault conditions, and the safe mode transition is pre-established, allowing rapid response without significant performance penalty during normal operation.

Inventive Principle:
Principle #10Preliminary action

2Reliability

If traditional fault tolerance configurations are used to detect and respond to malicious fault attacks, then the microprocessor security is improved, but the area overheads increase significantly

Engineering Contradiction:
Improvemicroprocessor securityVSAvoidarea overhead
Core Design Contradiction:
ReliabilityVSArea of stationary object

Solution Approach 1:

The fault detection unit and safe mode mechanism are designed to be universal components that can protect multiple microprocessor operations and instructions. A single fault detection unit monitors all instruction executions, and the safe mode transition can handle various fault types, reducing the need for redundant specialized hardware and minimizing area overhead.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The system implements self-service through automatic fault detection and self-contained recovery mechanisms. The fault detection unit autonomously identifies fault conditions, automatically transitions to safe mode, and the recovery mechanism restores normal operation without requiring extensive external intervention or additional hardware resources.

Inventive Principle:
Principle #25Self-service

3Reliability

If the microprocessor switches to safe mode and invokes software-based trap handler, then the fault recovery capability is improved, but the execution time increases

Engineering Contradiction:
Improvefault recovery capabilityVSAvoidexecution time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The system uses periodic action by implementing structured recovery phases: fault detection occurs continuously during normal operation, safe mode transition is triggered periodically when faults are detected, and recovery is executed in defined phases. This periodic structure allows the system to maintain normal high-speed execution while efficiently handling fault recovery only when necessary.

Inventive Principle:
Principle #19Periodic action

Solution Approach 2:

The system implements skipping by rapidly transitioning through the fault response sequence: immediate detection triggers rapid safe mode entry, which quickly invokes the trap handler, and then swiftly restores normal operation. This rushed-through approach minimizes the time spent in safe mode while ensuring complete fault recovery, reducing overall execution time loss.

Inventive Principle:
Principle #21Skipping (Rushing through)

Data Source

PatentUS10452493B2Microprocessor fault detection and response system
Publication Date: 2019.10.22 VIRGINIA TECH INTELLECTUAL PROPERTIES INC
  • US10452493B2 patent drawing
  • US10452493B2 patent drawing
  • US10452493B2 patent drawing

AI summary

Aspects disclosed in the detailed description include a microprocessor fault detection and response system. The microprocessor fault detection and response system utilizes a hardware-based fault-attack aware microprocessor extension (FAME) and a software-based trap handler for detecting and responding to a fault injection on a microprocessor. Upon detecting the fault injection, the hardware FAME switches the microprocessor from a normal mode to a safe mode and instructs the microprocessor to invoke the software-based trap handler in the safe mode. The hardware-based FAME provides fault recovery information to the software-based trap handler via a fault recovery register (FRR) for restoring the microprocessor to a fault-free state. By utilizing a combination of the hardware-based FAME and the software-based trap handler, it is possible to effectively protect the microprocessor from malicious fault attacks without significantly increasing performance and area overheads.