Configurable partitioning method and system for embedded software

Abstract

The invention discloses a configurable partitioning method and system for embedded software. The method comprises the steps of scanning a protection domain boundary in a software package and generating a protection domain boundary list; generating a cross-protection-domain glue code in each independently compiled sub-part; inserting each protection domain setting code into the starting code of thesoftware package; enabling the compiler and the linker to compile and link different guard domain codes. According to the method and the device, the setting and configuration of the protection domainhave extremely high flexibility, the adding cost of the protection domain is greatly reduced, the practicability of the protection domain is greatly improved, the same code can be reused when different protection domain partition schemes are set, and the flexibility of the protection domain is greatly improved.

Description

technical field [0001] The disclosure relates to the Internet of Things, the technical field of embedded systems, and the technical fields of information security, and in particular to a method and system for configurable partitioning of embedded software. Background technique [0002] With the rise of Industry 4.0 and the Internet of Things, the security of embedded software is becoming more and more important. An important aspect of security is the use of hardware protection domains for discrete components so that if one of the protection domains fails or is compromised, the other protection domains are not affected. [0003] Because the existing embedded software seldom considers information security and reliability when designing, the use of hardware protection domains is either ignored, or manually divided when writing programs. The former has no information security at all, while the latter has a fixed and immutable protection domain. For example, a certain part of so...

AI Technical Summary

Problems solved by technology

This leads to the immutability of protection domains: if we replace a processor with better performance, we need to set up more protection domains to improve performance, then we need to rewrite the entire software; if we replace a processor with worse performance, we need Reducing the number of protection domains also requires rewriting the entire software

These two approaches make the protection domain completely unconfigurable and unable to adapt to the diversity of IoT devices;

[0010] (2) The existing protection domain technology can only be optimized at the language virtual machine level, but cannot really touch the core of the problem—that is, the overhead of the system call itself

Once we need to extend the program to other architectures or operating systems, the underlying code related to the establishment of protection domains and communication between protection domains must be completely rewritten, which is very complicated;

[0012] (4) All current protection domain establishment technologies have complex configuration processes, which lack the support of automated tools and require manual operations

The only few methods require dynamic identification at runtime, which adds a very large execution overhead

Images