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

Microprocessor architecture level soft error susceptibility assessment method

A microprocessor and architecture technology, applied in the detection of faulty computer hardware, function inspection, etc., can solve the problem of inaccurate evaluation results of soft error susceptibility, and achieve the effect of improving accuracy

Inactive Publication Date: 2019-06-25
JIANGNAN UNIV
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, un-ACE bits are also included in logic mask instructions, and soft errors occurring on these un-ACE bits will not affect the execution results of the program, so the soft error susceptibility evaluation results are not accurate enough

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
  • Microprocessor architecture level soft error susceptibility assessment method
  • Microprocessor architecture level soft error susceptibility assessment method
  • Microprocessor architecture level soft error susceptibility assessment method

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment 1

[0032] according to figure 1 and figure 2 As shown, the present invention provides a method for assessing soft error susceptibility at the microprocessor architecture level. An analysis module, the method comprising the steps of:

[0033] Step 1: Select a simulator, initialize the simulator through the parameter configuration module, and simulate the function and performance of a specific microprocessor;

[0034] Step 2: judging whether the analyzed component is an instruction queue, a register update unit, a load store queue, or a non-storage component of a functional unit;

[0035] Step 3: The simulator loads the program in the test program module. When the program is running, the command analysis module is used to analyze the command executed in the analyzed component. According to whether the command execution result will affect the subsequent execution of the program, the command Divided into ACE instructions and un-ACE instructions, among them, logical masking instru...

specific Embodiment 2

[0039] According to the different structural characteristics of the microprocessor simulated by the simulator, this method can be applied to the soft error susceptibility analysis of microprocessors with different pipeline structures.

specific Embodiment 3

[0040] The soft error susceptibility analysis process of the present invention comprises the following steps:

[0041] Step 1, simulator initialization configuration: select a simulator, such as the simplesim-ARM simulator in SimpleScalar that targets the ARM instruction set, and initialize the configuration file of the simulator to simulate the complete functions of the microprocessor. The instruction set used in it is ARMv7.

[0042] Step 2, whether it is a non-storage component: judge whether the analyzed component is a non-storage component such as an instruction queue, a register update unit, a load-store queue, a functional unit, etc., and jump to step 3; otherwise, jump to step 6;

[0043] Step 3, instruction analysis module: the ACE bit refers to the bit that must be guaranteed to be correct in order to ensure that the execution process does not go wrong during the running of the program. Correspondingly, the un-ACE bit means that even if a soft error occurs on this bit...

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 relates to a microprocessor architecture level soft error susceptibility assessment method. According to the invention, soft error susceptibility evaluation is carried out on a componentof a microprocessor from two aspects of a storage component and a non-storage component. For a non-storage component, firstly, an instruction analysis method is adopted to classify ACE instructions and un-ACE instructions. In the process of instruction analysis, the logic shielding instruction is divided into ACE instructions. However, the logic shielding instruction still comprises un-ACE bit, and a program execution result is not influenced when a soft error occurs. Therefore, more un-ACE bits are found out through logic analysis. For the storage part, instruction analysis is firstly carried out, and a logic correction module is adopted, so that more un-ACE bits are found out. Meanwhile, life cycle analysis is performed on the storage component, so that soft error susceptibility evaluation is performed on the storage component. The invention provides a more accurate soft error susceptibility assessment method by adopting a logic correction module.

Description

technical field [0001] The invention relates to a soft error susceptibility evaluation method at the architecture level of a microprocessor, and belongs to the technical field of microprocessor simulation and soft error susceptibility analysis of microprocessor components. Background technique [0002] For a long time, the development of microprocessors has been driven by both integrated circuit technology and computer architecture technology. As the semiconductor manufacturing process enters the nanometer era, the accompanying soft errors caused by radiation and interference reduce the reliability of microprocessors, seriously limiting the development and application of modern microprocessors. Therefore, evaluating the soft error susceptibility of microprocessor components at the early stage of design is helpful to provide design reference for designers. [0003] Mukherjee et al. proposed to use the architecture-level soft error susceptibility factor (AVF) to describe the ...

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 Applications(China)
IPC IPC(8): G06F11/22G06F11/26
Inventor 顾晓峰高苗虞致国
Owner JIANGNAN UNIV