Kernel code protection method based on Harvard architecture

Abstract

The invention belongs to the field of computer security, and provides a kernel code protection method by using the virtualized Harvard memory architecture by combining the research of virtualization technology in order to overcome the shortage of the kernel code protection of the prior operating system. The method realizes the virtualized Harvard memory architecture on a computer based on the vonNeumann architecture by a kernel patch form and separates an instruction operation and a data operation in a memory area of the kernel code so as to basically prevent the probability of distortion forthe kernel code. The invention can well prevent the threat of the kernel grade Rootkit on the kernel code, record information associated with illegal operations while ensuring the integrality of thekernel code, and provide clue for analyzing attacks; in the aspect of performance, the method fully utilizes an instruction TLB (Translation Lookaside Buffer, namely branch translation buffer or pagetable buffer) and data TLB without remarkably influencing the performance of the operating system.

Description

technical field [0001] The invention belongs to the field of computer security, in particular to a method for protecting kernel codes in a Linux operating system. Background technique [0002] The computer operating system manages hardware resources downwards and provides interfaces for application programs upwards, playing an important role in the computer system. In order to make better use of underlying hardware resources and provide better services for upper-layer software, operating systems are becoming increasingly complex, the most direct manifestation of which is the increasing complexity of operating system codes. For example, the kernel codes of Linux and Windows XP, the two most popular operating systems, are 4.3 million lines and 40 million lines respectively. There must be some loopholes in such complicated operating system kernel codes, and the resulting security risks have seriously affected the security of computer systems. [0003] The traditional way to p...

AI Technical Summary

Problems solved by technology

In order to improve the parallelism of the CPU executing instructions, the five processes of the pipeline should be able to run without interfering with each other, but the two stages of fetching and accessing memory must access the memory, and the memory can only be used at a certain time. a memory access

Images