Method for resolving fault of chip failure

Abstract

The invention discloses a method for solving chip failures, which includes: a master CPU regularly initiates a test message request for service function detection to the slave CPU by using the idle time of the slave CPU, and notifies the slave CPU to perform a normality test of a designated channel or a voice channel; After the slave CPU receives the test message request, it calls the program module normally running in it to perform the loopback test to judge the correctness of the test result; after the slave CPU test is completed, it replies a test response message to the master CPU according to the test result; the master CPU according to The test response message is processed accordingly. If the test result is correct, no change will be made to the channel allocation of the slave CPU; if the test result is wrong or the waiting for the test response message times out, the master CPU will allocate priority to all session channels of the slave CPU. level down. By adopting this method, hidden faults can be found in advance, ensuring the reliability and stability of the entire system.

Description

technical field [0001] The invention relates to a method for solving chip failure and improving the working reliability and stability of the whole system. The method has great application value especially in a communication system with multiple CPU chips. Background technique [0002] In a computer system, there is always a certain hardware failure rate in the hardware, especially during the operation of the system, the memory or registers of the CPU system are affected by various factors such as the surrounding environment, which can easily lead to failure. If the software in the system Failure to consider and avoid this abnormal situation may cause the decline of system performance indicators, and even lead to the paralysis of the entire system. For example, a fault in the memory segment storing the program may lead to the following two situations: one situation is that some program instructions are rewritten, and the program runs away abnormally, causing the CPU to reset....

AI Technical Summary

Problems solved by technology

Usually in this case, due to memory failure, the slave CPU can simply shake hands with the master CPU, and the master CPU still thinks that the slave CPU is working normally, and keeps assigning sessions to the slave CPU, but the slave CPU can no longer get the correct However, the system cannot detect the fault, which leads to the decline of the overall performance index of the system

Images