Universal microprocessor and signal processing method thereof

Abstract

The invention discloses a signal processing method of a universal microprocessor. The signal processing method comprises the following steps that S1, a processor processes an instruction in an out-of-order execution and branch prediction mode; S2, the processor performs legal detection on all the instructions, and meanwhile, cache data generated by out-of-order execution and branch prediction is cached into a cache disk; S3, the cache disk sends the cached data to a black area disk, and the black area disk isolates the cached data and records a time node A; S4, the processor sends the legal instruction to a scheduling disk, and the scheduling disk records a time node B of the instruction; and S5, the scheduling disk sends the time node B of the legal instruction to the black area disk, andthe black area disk searches the corresponding node A according to the node B and carries out AB node matching. According to the universal microprocessor and the signal processing method thereof, thephenomenon that cache data of an existing universal microprocessor do not have protection measures and can be illegally read is avoided, and the safety of the universal microprocessor is improved.

Description

technical field [0001] The invention relates to the technical field of microprocessor production, in particular to a general-purpose microprocessor and a signal processing method thereof. Background technique [0002] In the working process of the microprocessor, in order to improve the instruction processing efficiency of the processor, out-of-order execution and branch prediction are usually used. Out-of-order execution means that the processor decides to execute instructions in an order determined by the availability of input data. , instead of processing in the order specified by the program, the method of branch prediction is to predict the instructions that the processor will execute in advance. Although this method improves the work efficiency of the processor, it may become a means of exploiting kernel vulnerabilities . [0003] For new processors with predictive execution capabilities, in the actual processor execution process, the memory loading link required for ...

AI Technical Summary

Problems solved by technology

[0002] In the working process of the microprocessor, in order to improve the instruction processing efficiency of the processor, out-of-order execution and branch prediction are usually used. Out-of-order execution means that the processor decides to execute instructions in an order determined by the availability of input data. , instead of processing in the order specified by the program, the method of branch prediction is to predict the instructions that the processor will execute in advance. Although this method improves the work efficiency of the processor, it may become a means of exploiting kernel vulnerabilities

[0003] For new processors with predictive execution capabilities, in the actual processor execution process, the memory loading link required for subsequent instructions does not depend on whether the preceding instructions can be executed normally, and the link from memory to cache loading does not verify access Whether the memory is legal and valid, even if the illegal instruction is abnormal, its subsequent instructions cannot be executed, but the memory data required by the subsequent instructions of the illegal instruction has been loaded into the processor cache, which results in subsequent instructions even if loaded without permission The memory data will also be loaded into the processor cache, because the processor checks whether the address is legal at the link of the cache to the register, and the processor branch prediction is only to complete the loading of the memory to the processor cache. Subsequent instructions are not actually executed, so his illegal access will not trigger an exception, therefore, there are security holes in this general-purpose microprocessor

[0004] And when installing a general-purpose microprocessor, it is necessary to weld various precision electronic components or circuit boards on the processor substrate. Due to the high temperature during laser welding, it is easy to affect the substrate or other components, so welding wire is generally used Welding, in this process, if the distance traveled by the welding wire at one time is too large, it is easy to cause a large range of solder joints and cover other components, which may cause short circuits or mutual influence between the components, and the distance traveled by the welding wire is too short and easy Causes missing soldering, virtual soldering, etc., resulting in unstable component welding and increasing the risk of component falling off during use. Therefore, ensuring the stability of the welding wire travel distance, that is, the length is very important for the entire welding process

[0005] In order to solve the above problems, the inventor proposed a general-purpose microprocessor and its signal processing method, which has the advantages of high safety and can ensure the stability of the length of the welding wire, and reduces the overheating caused by the unstable length of the welding wire during the welding process. Soldering, missing soldering and virtual soldering problems, improving the quality of microprocessor chip soldering and installation

Images