Fault diagnosis and recovery method and device for heterogeneous system

Abstract

The invention provides a fault diagnosis and recovery method and device for a heterogeneous system. The method includes the steps that in the heterogeneous system formed by an X86 and an FPGA, logic area division is performed on the FPGA according to the physical position and the logic relevancy, and fault classification judgment is preset in the FPGA for fault recognition; when a first access breaks down, a second access is started to be used as a backup link in the X86, a third access feeds the fault category back to the FPGA according to divided logic areas, and fault diagnosis is performed on the first access in the mode that the X86 sends fault diagnosis instructions to an ARM; in the FPGA, the broken-down logic area is located, then the ARM sends out recovery instructions, and recovery is performed on the broken-down logic area through the third access so as to complete elimination of faults of the heterogeneous system. The fault diagnosis and recovery method and device for the heterogeneous system can achieve the communication control link between the backup FPGA and the X86, obtain fault diagnosis information, eliminate accidental faults, and improve the fault debugging capacity and the stable operation capacity of the system.

Description

technical field [0001] The present invention relates to the technical field of heterogeneous systems, and more specifically, to a method and device for fault diagnosis and recovery of heterogeneous systems. Background technique [0002] Heterogeneous systems are regarded as the third era of computer processors after single-core and multi-core, which can realize collaborative computing and mutual acceleration between computing units of different types of instruction sets and architectures, thereby breaking through the bottleneck of CPU development. Effectively solve the problems of energy consumption and system scalability. [0003] In the heterogeneous system that combines the X86 system and the FPGA (Field Programmable Gate Array, field programmable logic gate array) system, the X86 system can play a major role in controlling the heterogeneous system and maintain the normal operation of the entire heterogeneous system ; and the FPGA system can improve the performance of he...

AI Technical Summary

Problems solved by technology

In other words, it is necessary to consume a large cost for control channel redundancy

Even so, it cannot guarantee the stability and control of the FPGA system

Because the control data will always arrive at the same control point, when the FPGA system has a serious global failure, it may cause all redundant links to fail, resulting in failure of the entire control path of the FPGA system

[0006] In the heterogeneous system that combines the X86 system and the FPGA system, once the control path fails, there will be problems in the control and maintenance of the FPGA system by the X86 system, so that the entire FPGA system will lose its ability to continue to work normally and cannot continue Obtaining control data from the X86 system makes troubleshooting and analysis of faults difficult, thus increasing the cost of debugging problems

Images