Security chip design method based on control flow detection and resistant to error injection attack

Abstract

The invention relates to a security chip design method based on control flow detection and resistant to error injection attack. The security chip design method comprises a first step of compiling a standard assembling file which supports a security chip, a second step of dividing the standard assembling file into basic blocks by means of lexical analysis, a third step of generating basic block signature values and forming an assembling file with a control flow detection instruction, a fourth step of translating the assembling file into a target file, generating basic block checking values and inserting basic block checking instructions, and a fifth step of generating an executable file from the target file and a database file link and executing the executable file on hardware. The security chip design method based on control flow detection and resistant to error injection attack can effectively detect skip errors among the basic blocks and inside each basic block and non-skip errors of a command sequence, has the self checking function on the control flow detection instruction which is newly added, and is simple in implementation and low in redundancy code cost by the adoption of the mode of combination of software and hardware.

Description

technical field [0001] The invention relates to a security chip design technology, in particular to a security chip design technology based on control flow detection to resist error injection attacks, so as to achieve the purpose of resisting error injection attacks. Background technique [0002] In the information age, the issue of information security is becoming more and more important. As the core component of information security, security chips that provide encryption, authentication, secure storage and other security services as the main functions are more and more widely used. At the same time, attacks on security chips are also emerging in an endless stream, which makes it urgent to study the protection technology of security chips. [0003] Fault injection attack is a semi-invasive attack, the attack principle is to try to change the environmental parameters (the electrical performance of the chip will change with different voltage, temperature, light, ionizing rad...

AI Technical Summary

Problems solved by technology

[0013] (1) Error detection of jumps inside basic blocks: The existing control flow detection technology mainly focuses on the correctness of jumps between basic blocks, but it is not appropriate for the error jumps of non-jump instructions inside basic blocks. the solution

[0014] (2) Detection of non-jump errors in instruction sequences: The existing control flow detection technology is limited to the error detection of program jumps, and does not take into account the non-jumps caused by errors in the opcode or operand of the instruction itself. sexual error

[0015] (3) Self-verification of new control flow detection instructions: To implement control flow detection, additional control flow detection instructions need to be added, and there is no effective way to detect whether these new control flow detection instructions themselves are wrong

[0016] (4) The contradiction between security and performance: In order to solve the detection blind spots, some methods design some complex detection instruction sequences

However, due to the addition of too many detection instructions, the impact on the performance of the program will also increase, and the detection instruction itself will be attacked and the probability of errors will also increase.

Images