Logic simulation verification regression testing system and method based on random use cases

Abstract

The invention discloses a logic simulation verification regression testing system and method based on random use cases, and relates to the technical field of logic simulation testing. The method comprises the following steps that a random use case lift is created, and regression testing is started after testing parameters are set; a testing recording module records the execution condition of a single round; when storage space of a hard disk is larger than a single use case storage space threshold, a use case execution module executes the single use cases, and only essential information is carried in execution of the first time; after all the single use cases in the random use case list are executed, a seed replacement module judges whether a random seed replacement condition is met or not, if yes, random seeds are replaced, regression testing of the next round is carried out, and if not, regression testing is finished. The contradictions of the daily-increasing number of use cases and the limited storage resources of the hard disk can be solved; efficient and fast locating is achieved; the coverage range of verification space is automatically widened; meanwhile, the execution time and occupied resources of the single use cases in the regression testing process can be reduced.

Description

technical field [0001] The invention relates to the technical field of logic simulation testing, in particular to a logic simulation verification regression testing system and method based on random use cases. Background technique [0002] The scale of chip design and FPGA (Field Programmable Gate Array, Field Programmable Logic Gate Array) logic continues to increase. In order to ensure the quality of the design, it is necessary to rely on as many effective use cases as possible to verify its quality. As the number of use cases increases, more effective methods are required for regression test management of use cases. In regression testing, if all use cases are executed by humans one by one, the efficiency of regression testing will be very low, so it is necessary to use automated tools and scripts to manage and improve the efficiency of regression testing. [0003] In addition, as chip and FPGA verification technologies become increasingly mature, random verification has ...

AI Technical Summary

Problems solved by technology

However, when the number of use cases is huge, the lack of resources will lead to reduced efficiency of regression testing, or even abnormal interruption

In addition, the test coverage of this method is limited, and the purpose of automatically expanding the coverage of functions cannot be achieved; it cannot guarantee that the regression test can run efficiently and quickly in the case of limited operating resources; and it cannot automatically screen effective information for developers to efficiently locate problems. Reinvesting energy to restore the problem site is time-consuming, labor-intensive and resource-intensive

Images