Central processing unit design method supporting software code data confidentiality and credibility execution

Abstract

The invention provides a central processing unit design method. Belonging to the field of digital circuits, the method comprises the following steps: 1, processing; dividing and reserving a memory address range for adding a read-only memory and a random access memory in a chip; the instruction in the off-chip memory is limited to be invalid when the content of the on-chip memory is read and written; a private key certificate and a safety kernel code are persistently stored in a chip, memory access is monitored, address monitoring event interruption is triggered, an address monitoring event interruption processing program is written to achieve real-time decryption and encryption of to-be-accessed data, and a to-be-executed instruction is preset to a register and executed from the register.A trusted computing system can be realized on the novel central processing unit, an application program accommodating an encrypted ciphertext form code segment and a data segment can be loaded and executed in real time. Meanwhile, confidential information of the application program is prevented from being accessed unauthorized and leaked to an external bus, and the confidentiality, integrity and authenticity of the confidential information in the application program are protected.

Description

technical field [0001] The present invention relates to the architecture and function design of several central processing units (CPU), which can be used to protect the information security of application software codes and data, mainly belonging to the field of digital circuits. Background technique [0002] Information security has four aspects: equipment security, data security, content security and behavioral security. Data security can be guaranteed through data encryption. For example, the SSL and TLS protocols provide methods to ensure channel data security in network communication. The signature certification of the data content by the data producer can ensure the safety and reliability of the content source. Trusted computing belongs to behavioral security. According to the description of Chinese information security experts in the book "Software Behavior", behavioral security should include: behavioral confidentiality, behavioral integrity, and behavioral authenti...

AI Technical Summary

Problems solved by technology

However, the security technology of these two types of CPUs is difficult to prevent side channel attacks

Attackers can use specially designed hardware to intercept or forge the data exchange between the CPU and the device. Attackers can also modify the operating system to bypass the isolation mechanism of the CPU. Attackers can also use software virtual machines or software virtual CPUs. The method of execution intercepts the memory distribution of code and data in the software and conducts further attacks based on this. These attack methods limit the applicability and effectiveness of the above techniques

[0005] At present, the widely used SSL/TLS protocol can ensure the confidential transmission of data in the communication channel, but the encrypted data transmitted over the network will eventually need to be restored in the application program or browser in the user terminal. If the user terminal application program, browser or operating system is not secure It also cannot guarantee the security of communication data. For example, implanting a malicious plug-in in the browser can steal user communication data. Information security also depends on the security of the executed software. To ensure the trusted execution of applications, common methods Relying on the trust chain formed by checking whether the developer's signature of the application file is credible, and relying on the layer-by-layer verification of the system from initialization to the final execution of the environment established by the application software, but this process and method are too It is cumbersome, there are too many intermediate links, and it is difficult to configure and quickly adapt to changes in the environment. It is only suitable for special scenarios with relatively fixed environments. For example, the upgrade of software and hardware systems will make some security measures that rely too much on the old system go wrong or fail. Too many All intermediate links can become unreliable due to user behavior, and often cannot guarantee the trustworthiness of people who come into contact with the system, such as malicious users or users with bad habits

[0006] At present, many security technologies assume that the users of the applicatio

Images