Unlock instant, AI-driven research and patent intelligence for your innovation.

A SDC Fragile Instruction Identification Method Based on Error Propagation Analysis

An instruction, fragile technology, applied in the computer field, can solve problems such as high time and space cost, error injection, state explosion, etc., to avoid limitations, injection errors, and high accuracy.

Active Publication Date: 2016-08-24
SOUTHEAST UNIV
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, some errors in its simulation will not occur in reality, so the accuracy rate is low
And symbolic execution leads to state explosion, which makes the time and space cost huge
[0014] To sum up, the advantage of the dynamic method is that the SDC fragile instructions obtained are accurate, but the cost of error injection is high; the implementation of the static method is simple, but the accuracy is low

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A SDC Fragile Instruction Identification Method Based on Error Propagation Analysis
  • A SDC Fragile Instruction Identification Method Based on Error Propagation Analysis
  • A SDC Fragile Instruction Identification Method Based on Error Propagation Analysis

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0050] by figure 1 The sum program in is taken as an example to illustrate the process of the recognition algorithm. First, find out the key instructions among all instructions. According to the definition of key instructions, node 18 is a key instruction, and other instructions are all non-key instructions. Fault injection is performed on node 18 by modifying one bit of the destination operand, and the result is found to be wrong, and node 18 is added to the vulnerable key instruction set of SDC; the second step is to perform error injection and inference on non-critical instructions. Use the equivalent class injection method to sample the above non-critical instructions, add them to the injection schedule, and modify the destination operand of node 1 from 0 to 1, and the final result changes from 1 to 2, resulting in SDC, and then run inference algorithm. Starting from node 1, sequentially compare the written data of nodes 2-18 with the written data of the same node durin...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a Silent Data Corruption (SDC) vulnerable instruction recognition method based on error propagation analysis. The SDC vulnerable instruction recognition method comprises the steps of performing error injection on key instructions and verifying the SDC vulnerability of all the key instructions first, and then performing SDC vulnerability judgment on non-key instructions by use of an inference method. The SDC vulnerable instruction recognition method is capable of remarkably reducing the number of errors injected while guaranteeing relatively high accuracy rate and coverage rate.

Description

technical field [0001] The invention relates to soft error detection in the computer field, in particular to an SDC fragile instruction identification method based on error propagation analysis. Background technique [0002] The soft error caused by Single Event Upset is a transient fault phenomenon in semiconductor circuits, which is usually induced by electronic noise such as high-energy particle irradiation in the external environment, voltage disturbance, and electromagnetic interference. Although soft errors will not damage the internal hardware circuit, they can affect the normal operation of the program by changing the state of the processor or the value of the storage unit, and even cause abnormal operation or loss of control of the satellite. In recent years, the number of chip-integrated transistors has increased exponentially. While the performance has been greatly improved, the processor has become more sensitive to single-event upsets, and the overall soft error...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06F11/36
Inventor 汪芸马骏驰李凯
Owner SOUTHEAST UNIV