19 results about "Code coverage" patented technology

The application discloses a kind of multi-peripheral register interaction protocol perception's firmware fuzz testing method and equipment, it is related to firmware test technical field, method includes: in the process of carrying out fuzzy test to firmware, determine the multiple execution rounds of the firmware that triggers new code coverage;Multiple execution rounds each corresponding peripheral register access sequence is obtained;According to multiple peripheral register access sequence, frequent co-occurrence input stream sub-sequence is mined, wherein each input stream corresponds to a peripheral register;Select target sub-sequence in frequent co-occurrence input stream sub-sequence, simultaneously carry out variation to multiple input streams in target sub-sequence;Based on the input stream after variation, execute firmware, carry out fuzzy test to firmware.The application improves the code coverage of firmware fuzzy test while guaranteeing the efficiency of fuzzy test, and then increases the possibility of discovering more crashes and potential security vulnerabilities.

A method for testing code coverage of a software program executed by a machine in an industrial system, comprising: obtaining a control flow model of a first version of the software program as a first control flow model; obtaining a control flow model of a second version of the software program as a second control flow model, wherein the second version is a modified version of the first version; comparing the first control flow model with the second control flow model, to identify one or more code changes of the second version compared to the first version; preparing a set of codes to be tested based on the identified one or more code changes; preparing test data for testing the coverage of the set of codes to be tested; compiling the second version into an assembly code or a binary machine code to be executed by a runtime environment of the machine; adding a check point into the assembly code or the binary machine code at a position corresponding to each code in the set of codes to be tested, wherein the check point is an instruction added into the assembly code or the binary machine code that enables the runtime environment to indicate, when the instruction of the check point is executed, that the code in the set of codes to be tested corresponding to the check point is executed; starting the execution of the assembly code or the binary machine code in the runtime environment using the test data; and determining whether each code in the set of codes to be tested is covered according to the indication from the runtime environment.

The application provides a unit test case automatic generation method and device based on preferred iteration feedback, which comprises the following steps: setting a target module identifier and a target code coverage rate of a test case to be generated; extracting source code of the target module and necessary reference code; encapsulating the source code of the target module, the necessary reference code, a project status, a requirement description and feedback information into a template prompt word, inputting the template prompt word into an AI model to generate a unit test case; performing iteration legality verification and / or correction on the generated unit test case until the unit test case passes the compilation; executing the test case that passes the compilation to obtain a current code coverage rate and un-covered code information; judging whether the code coverage rate of the unit test case that passes the compilation is greater than a cached historical highest coverage rate; and judging whether the code coverage rate of the highest coverage rate test case obtained is greater than the target code coverage rate. The application can efficiently generate a unit test case with high quality, high coverage rate and operability.

The application provides a multi-dimensional layered hardware Trojan evaluation method for an IP soft core, constructs a Trojan multi-dimensional feature library, performs function integrity analysis to identify a first module in which a Trojan may exist, obtains equivalent RTL level code for a module in which no anomaly is found, takes each module of the RTL level code as a node and takes the calling relationship between each module as a weight to establish an information flow graph, compares the information flow graph with the Trojan multi-dimensional feature library to obtain a second module in which a Trojan may exist, analyzes the information flow graph, and takes a module in which the calling is less than a set threshold, a module with an isolated node and a module adjacent to the isolated node in the information flow graph as a third module in which a Trojan may exist, analyzes the first module, the second module and the third module, calculates code coverage, signal flip rate, signal and clock delay anomaly, power consumption anomaly and layout blocking anomaly, and comprehensively evaluates to locate a module in which a Trojan exists in the to-be-detected IP soft core.

This disclosure provides a fuzzing method and system. The fuzzing method includes: performing fuzzing tests on the object under test using test programs based on test operations. Each round of test operations includes: determining a first indicator for each interface and a second indicator for each test program based on the code coverage of the previous round of test operations, wherein the first indicator is based on the probability that the interface will be tested, and the second indicator is based on the probability that the test program will be tested; selecting a target interface from the interfaces based on the first indicator; selecting a target test program from the test programs used to test the target interface based on the second indicator; and performing fuzzing tests on the target interface using the target test program to obtain the result of the current round of test operations.

The application discloses an Internet of Things firmware testing method, device, equipment and medium. When a target firmware startup instruction is received, the target firmware is loaded and a target pointer of the target firmware pointing to a direct memory access (DMA) memory area is detected, the target pointer including a source address pointer and a destination address pointer; when the target pointer of the target firmware pointing to the DMA memory area is detected, a target DMA transmission descriptor is determined according to the target pointer and a target DMA input channel is constructed; a pointing value of the destination address pointer is obtained according to the target DMA transmission descriptor; the pointing value of the destination address pointer is stored through the target DMA input channel, and the target firmware is run according to the pointing value; the running process of the target firmware is monitored, and a test result is generated according to a running result. The method solves the problem that the existing firmware hosting technology cannot simulate DMA input, expands the types of simulatable embedded devices and the code coverage of firmware testing, and thus improves the security of the embedded device.

This application discloses a test case recommendation method, system, device, storage medium, and program product, relating to the field of software testing technology. The method includes: collecting multimodal data; inputting the multimodal data into a pre-trained multi-task learning model to obtain a test case set and the corresponding code coverage of the test case set. The multi-task learning model is constructed and optimized by performing feature extraction and feature fusion on the multimodal data through an end-to-end training strategy. This scheme employs a multi-task learning model to simultaneously optimize both test case recommendation and code coverage prediction tasks, improving the performance of each task. Furthermore, the training of the task learning model adopts an end-to-end training strategy, which not only enables the model to process multiple types of input data simultaneously within a single framework and automatically adjusts parameters during training to optimize all objective functions, but also simplifies the model structure and reduces training costs.

This application discloses a precise testing method, apparatus, device, and medium based on code change and test case mapping, relating to the field of software testing technology. The method includes: executing test cases and collecting their execution traces using code coverage tools, recording the signatures of all covered code methods and their coverage data; storing the mapping relationship between test case identifiers, code method signatures, and coverage; obtaining a list of changed files through a version control system when a code repository commits; performing method-level structural comparisons for each changed file; processing all changed files in parallel and merging them to obtain a final set of changed methods composed of affected method signatures; comparing the set of changed methods with an established test case-code method mapping database; retrieving all test cases covering any method in the set of changed methods, and generating a recommended test case list. Therefore, this application can achieve a regression testing scheme with high test quality and execution efficiency.

The application discloses an Android application testing method based on Monte Carlo tree search and deep reinforcement learning, and belongs to the technical field of software testing.The method adopts deep reinforcement learning to perform adaptive learning on a testing strategy, a fine-grained state representation mode is designed for an application program interface state, a new reward function is used during exploration, state changes caused by interface jumps can be rewarded, and fine-grained control position and text changes in the interface can also be rewarded, so that the learned testing strategy is more comprehensive and detailed, the Monte Carlo tree search method is used to optimize the testing strategy, potential states are provided with opportunities for long-term exploration, and local optimization is avoided.The method can continuously reach some application program states that are difficult to traverse previously, high-code-coverage Android application program testing is realized, and the code coverage and fault detection rate performance of Android application testing are improved.

This invention discloses a method and system for evaluating software testing hint injection attacks based on coverage differential analysis. The method includes receiving source code fragments and performing structured parsing to construct baseline hints for test code generation tasks; selecting corresponding adversarial injection templates from an adversarial vector library according to a preset attack intent, and injecting the adversarial injection templates into non-executable regions of the source code using a hint synthesis engine to generate adversarial hints; inputting the baseline hints and adversarial hints into a large language model to generate baseline test code and adversarial test code, respectively; running the two sets of test code in an isolated execution sandbox and collecting code coverage metrics generated during test execution; performing differential analysis on the coverage metrics of the two sets of test code, calculating the robustness score of the large language model under hint injection attack conditions, and identifying potential attack success patterns. This invention improves the accuracy and reliability of detection.

An agent-based network protocol fuzzing method is provided. The method constructs a knowledge base through static analysis, information retrieval and code instrumentation. A coverage-driven seed priority management and mutation strategy allocation mechanism is designed. The global state of the system under test (SUT) is monitored in real time, and an alarm is given when the set threshold is exceeded. The agent combines the knowledge base, call chain and source code context, and global state indicators to call a large language model (LLM) in steps, and guide the test strategy. Finally, the method automatically analyzes the vulnerabilities and generates a report in combination with process signals, logs and other information. The method significantly improves the vulnerability triggering efficiency and code coverage, and reduces the test blindness and the need for human intervention.

This invention relates to a test case recommendation method, apparatus, storage medium, and electronic device. The method classifies the popularity level of each function based on the number of times each function is called in the first test code, determines the association level between each test case and each function based on code coverage information between each test case and each function, and then automatically generates recommended test cases that best meet the user's testing needs based on the popularity level information and association level information of each function in the second test code, as well as the popularity level information and association level information of the input recommended test cases. This method can effectively improve software testing efficiency.

The application discloses a model-oriented generation and variation method for a deep learning compiler, and comprises the following steps: step 1, constructing an ONNX operator rule library and performing initialization configuration; step 2, constructing a computation graph DAG based on the operator rule library; step 3, constructing a constraint model based on the generated DAG; and step 4, performing oriented generation and model variation based on the constraint model constructed in step 3. The application can improve the code coverage rate and defect discovery efficiency of the compiler under the premise of guaranteeing the semantic correctness and structural legality of the model.

The application provides a code coverage closed-loop test optimization method. The application focuses on changed code through an incremental coverage calculation engine, solves the resource waste problem of full statistical calculation, a closed-loop optimization engine drives test generation and verification with coverage gaps, forms a self-evolution feedback loop, an intelligent scene recommendation engine quantifies priority based on multi-dimensional scoring, and converts subjective decision-making into objective standards. The three engines work together to realize accurate placement of test resources to changed areas and core scenes, reduce runtime overhead, improve defect detection rate, and establish an efficient and controllable automated test paradigm.

The application discloses a user page testing method and device and related products, and the method comprises the following steps: obtaining target page information of a target user page in a target project, wherein the target page information comprises target metadata randomly generated based on historical interface data of multiple interfaces of target code of the target user page; then writing the target metadata into a global variable window of the target code, wherein the global variable window comprises variables corresponding to the multiple interfaces respectively; running the target code to obtain code coverage information representing a probability of successfully executed code in multiple lines of code; after the running of the target code is completed, if the code coverage information indicates that the probability of the successfully executed code increases, performing screenshot on a page after the running of the target code is completed to obtain a screenshot result; and finally generating a target test case according to the target metadata, the code coverage information and the screenshot result, wherein the target test case is used for testing the target user page. In this way, the testing efficiency and the testing accuracy are improved.

The application discloses a data transmission protocol constraint perception firmware fuzzing method and related equipment, and relates to the technical field of firmware testing, which comprises the following steps: performing reverse assembly processing on a binary file of firmware, constructing a control flow graph and a taint propagation graph, and generating constraints in the form of SMT-LIB2; solving feasible solutions of MMIO peripheral register input values when the constraints are met and feasible solutions when the constraints are not met, respectively; in the fuzzing test phase, when it is determined that the MMIO peripheral register corresponding to the current input stream to be mutated exists a series of candidate values recorded by a structured data file, the data bit corresponding to the MMIO peripheral register in the input stream to be mutated is assigned to any candidate value. The application can avoid the performance overhead of repeatedly performing code analysis and constraint solving in the dynamic fuzzing test phase, realize fine-grained mutation, improve input diversity, and thus improve the code coverage rate of the fuzzing test and the vulnerability discovery capability.

This invention relates to a method and apparatus for fuzzing kernel-driven virtual devices based on a large language model. The method includes: extracting key device interaction information from the source code of the target driver using static analysis, and generating a driver context knowledge base by structurally indexing the kernel source code tree in the entire operating system kernel; under the constraints of the driver context knowledge base and predefined modeling specifications, iteratively guiding the large language model to generate a QEMU virtual device model using the key device interaction information, and performing initialization interactions to obtain a fully initialized virtual device model; in the QEMU environment, using the Syzkaller fuzzing framework to perform fuzzing tests on the fully initialized virtual device model against the system call sequence generated by the target driver, and collecting potential defect information under code coverage feedback and memory safety detection conditions. This method effectively resolves the core contradiction between the difficulty of simultaneously achieving automated scalability and high model fidelity.

The application belongs to the technical field of software testing, and specifically discloses a software regression test case priority sequencing method considering defect distribution, which comprises the following steps: obtaining a historical defect data set of software, dividing the historical defect data set into a training set and a verification set, inputting a machine learning classification algorithm, constructing a model, and obtaining performance evaluation results; extracting a current version metric element of the software under test to input the model to obtain a defect prediction probability; extracting dynamic code coverage information of the test case of the software under test, weighting and integrating the defect prediction probability and the metric element, and constructing a linear function as a priority sequencing objective function; sequencing through a sequencing strategy to obtain a test case priority sequencing result; and the method has the following advantages: considering defect distribution, metric elements, code coverage and other factors for linear weighting, solving the problem that existing methods do not consider defect distribution and reduce defect detection rate, or rely on prior knowledge when multiple factors are weighted, leading to unstable sequencing results, and improving the performance of the software test case priority model.

This application provides a method and storage medium for evaluating R&D work efficiency, relating to the field of data processing technology, for accurately evaluating the efficiency of R&D work. The method includes: in response to an evaluation instruction, acquiring multiple target evaluation indicators and target R&D coefficients; the R&D tasks include: deleting, modifying, and / or adding target functions of the target software. The multiple target evaluation indicators include: target code coverage, unit test case effectiveness, and technical debt repair rate. Based on the multiple target evaluation indicators, an initial indicator score is determined for each target evaluation indicator. Based on the multiple initial indicator scores and the target R&D coefficients, an R&D evaluation result is determined, which is used to indicate the efficiency of the target R&D team in completing the R&D tasks.