68 results about "Crash" patented technology

The invention provides a system operation lagging construction method and device, equipment, a medium and a program product. Relates to the technical field of system performance testing. The method comprises the steps that the CPU thread number and a watchdog threshold value of a current system are obtained; on the basis of the CPU thread number and the watchdog threshold value, the duration of the target thread executing busy waiting operation is calculated through a preset formula; initializing a spin lock of a kernel mode, and creating target threads of which the number is the same as that of the CPU threads, so that the target threads execute preset processing logic to construct system operation jamming; the preset processing logic comprises the following steps: periodically obtaining a spin lock; responding to the situation that the spin lock is not obtained, and continuing to execute the spin lock obtaining operation; and in response to obtaining the spin lock, executing a busy waiting operation. According to the method, CPU resources are continuously occupied through kernel spin lock competition, and a real system jamming scene is simulated. The busy waiting operation duration is dynamically calculated, it is ensured that the single-thread lock holding time is slightly lower than a watchdog threshold value, and crash of a system triggering unrecoverable is avoided.

The invention discloses an optimization sensitive test system and method for an AI compiler, and belongs to the technical field of software testing. The method mainly comprises the following steps: firstly, automatically extracting an IR sub-graph mode capable of triggering specific optimization Pass from an existing test of a compiler through dynamic instrumentation; secondly, through a strategy of multiplexing compatible nodes or creating new nodes, the optimization modes are intelligently synthesized into a diversified seed calculation graph context, and a large number of test cases capable of effectively triggering compiler optimization are generated; and finally, efficiently revealing the defects of the AI compiler in the optimization stage by taking crash detection and reasoning consistency comparison as test predictions. The defect that an existing random generation or grammar generation method is insufficient in test coverage rate in the optimization stage is overcome, the deep optimization defect in the AI compiler can be automatically and efficiently detected, and the method has the high error detection rate and good universality.

The invention provides a method and system for analyzing system crash, and the method for analyzing the system crash comprises the following steps: firstly, obtaining a parameter group through a system interface; then, marking a parameter group; then, when the system crashes, grabbing the Ramdump information and the useful information in the parameter group; thirdly, storing the Ramdump information and the useful information into a specified partition; analyzing data in the Ramdump information by using the useful information so as to find clues of system crash; in this way, the Ramdump data can be directly grabbed and analyzed without additionally storing and keeping symbol files of various versions and without an analysis function provided by additional external software.

Systems and methods include an Information Handling System (IHS) that is adapted to provide crash handling that is particularized to each phase of the IHS boot sequence. A boot sequence is initiated that includes two or more phases. Prior to each phase of the boot sequence, a crash handler is initiated that is configured to collect state information for IHS hardware configured during that respective phase of the boot sequence. Upon detecting a boot sequence crash by the IHS, the crash handler operating during the phase of the boot sequence of the crash stores hardware state information it has collected to a shared memory location.

The application discloses a file naming method and an electronic device, relates to the software field, and is used for helping to locate a crashed process by naming a crash report file as the name of the crashed process. The file naming method is applied to a Crashpad run by the electronic device, the Crashpad comprises a Crashpad processing end and a Crashpad client, and the method comprises the following steps: in response to a crash of a target process, the Crashpad processing end acquires a command line of the target process, the command line comprising the name of the target process; the Crashpad processing end analyzes the command line to obtain the name of the target process; the Crashpad processing end sends a target string to the Crashpad client; and the Crashpad client converts a crash report into a crash report file named by the target string, the target string comprising the name of the target process.

This application discloses a program running method and apparatus. The method and apparatus determine a target process to be allocated to a lightweight program from a preset program process, wherein each preset program process has independent running resources; the target process's running resources uniquely run the lightweight program with the allocated relationship. By limiting the number of preset program processes and ensuring that the corresponding target process exclusively occupies running resources during the lightweight program's operation, process-level isolation between different lightweight programs is achieved. A crash in a single lightweight program only affects its own process and will not cause abnormalities in the host app or other lightweight programs, thus ensuring overall operational stability. Each process uniquely runs its corresponding lightweight program, avoiding conflicts caused by multiple instances sharing process resources and improving the independence and reliability of lightweight program operation.

Systems and methods for in-system, self-recovery (ISSR) in a system-on-a-chip (SoC) are disclosed for detecting and recovering from a bit corruption event that has caused an SoC subsystem to crash. If a bit flip event is detected through observation of a subsystem crash, ISSR steps are taken in an attempt to correct the issue. If the ISSR steps are successful, then the system is not halted and mission mode operations continue. If the ISSR steps are unsuccessful, then the system is halted and an return material authorization (RMA) can then be issued for return of the failed part to the OEM. Crash events and self-recovery attempts preferably are logged by the ISSR system and provided to the OEM.

The application discloses a kind of methods for monitoring crash log, equipment, device and storage medium, the method comprises: if the application program of running exists exception is monitored, the exception data corresponding to the application program is obtained;Abnormal data is filtered and handled based on the exception type corresponding to the exception type, and the filtered exception data is stored. By quickly investigating the exception of application program, so that the developer more timely, accurate solution crash problem.

The embodiment of the invention provides an application monitoring method and device, equipment, a storage medium and a program product, and relates to the field of big data. The method comprises the following steps: in response to a captured memory exception event when a to-be-monitored application runs, acquiring memory stack information corresponding to the memory exception event; wherein the memory exception event comprises at least one of a memory leak event, a memory depletion event and an exceptional memory allocation event; the memory stack information, the crash event data and the operation behavior data are subjected to association analysis, an association analysis result is obtained, and the association analysis result comprises associated transaction identifiers and corresponding memory stack information within a preset duration before the crash triggering time; and according to the correlation analysis result, identifying an abnormal reason of the to-be-monitored application. According to the method, the abnormal problem identification accuracy of the application can be effectively improved.

The invention discloses a global monitoring and collecting method and device for Android application crash information and a medium, belongs to the technical field of mobile terminal technologies, and is used for solving the problems that existing Android terminal crash information monitoring is not comprehensive, specific reasons of crash are difficult to locate, use experience of a user is not facilitated, and user experience is poor. And the processing complexity of a developer is increased. The method comprises the following steps: performing data monitoring processing on crash information of a Java layer by customizing a default uncaptured abnormal processor of a system, and collecting and obtaining Java layer crash structured data; the method comprises the following steps: intercepting key crash signals of a Native layer through a global signal processor, and asynchronously acquiring and obtaining Native layer crash related data; performing structured assembly of related data packets on the Java layer crash structured data and the Native layer crash related data, and performing asynchronous transmission to obtain a crash report; and transmitting the crash report to a server through the asynchronous queue to obtain crash report information.

The application discloses a kind of fuzzy test method and process based on deep learning model.First, different formats of Linux application are collected, and the application collected and seed file with the same format are saved;Second, the variation model under fixed variation strategy is constructed based on value-oriented DQN algorithm and strategy-oriented TRPO algorithm, and the improved Thompson sampling algorithm is used to construct Havoc random variation model, to output the effective variation operator and variation strategy chain information of test sample;Then, the AFL source code is modified, the two proposed variation models are embedded, and the interaction between the model and AFL is realized using shared memory;Finally, for the AFL embedded with the model obtained in the last step, an automated execution script is written and its execution time is set, and the afl-fuzz, afl-showmap and afl-cov instructions are executed to train the embedded two models and perform automated fuzz testing, and according to the fuzz testing execution result, the code coverage, the number of crashes, the use of variation operator and other results are visualized.

The embodiment of the invention provides a memory repeated release detection method and device, electronic equipment and a storage medium, and the method comprises the steps: extracting abnormal features related to memory repeated release from a target file, and updating an abnormal feature template based on the extracted abnormal features; the target file is a core dump file generated by system kernel crash; the abnormal feature template comprises abnormal entries, and each abnormal entry comprises an abnormal feature tag used for describing related information of the abnormal features; based on the abnormal feature template, risk scanning is conducted on the kernel source code, a risk calling list is generated, the risk calling list comprises risk calling points in the kernel source code, and the risk calling points comprise memory allocation function calling points and / or memory release function calling points; the risk calling point hits the abnormal feature tag; and according to the risk calling list, carrying out instrumentation on the risk calling points, and carrying out verification in the target environment. According to the method, the memory repeated release vulnerability of the system kernel can be automatically detected, and the quick response requirement can be met.

The invention discloses an MCI (Module Control Interface) interface abstraction service which is added between a module driver and an application, provides a unified and standardized operation set upwards, abstracts interfaces and functional processes of different modules downwards, and hides interface differences of different SDKs (Software Development Kits). Meanwhile, through a specific software hierarchical architecture design, collapse or abnormity of a module driving layer is limited in the module, and the whole application process cannot be collapsed through recovery of a monitoring and restarting mechanism. Moreover, when a new module is replaced, only the MCI interface needs to be adapted, and the application does not need to modify the existing service logic.

Embodiments of the present application disclose a system startup method, device and equipment for preventing system crash, and a storage medium. The method comprises: starting the system to enter a kernel boot program according to a user's startup operation, and checking whether a first check code and a second check code are consistent by the kernel boot program; if not, booting into a repair program to repair the system by the repair program; generating a first random code, replacing the first check code and the second check code with the first random code after the repair is completed, and restarting the system; if consistent, booting for normal startup, generating a second random code, and replacing the first check code with the second random code; and after the normal startup is successful, replacing the second check code with the second random code. Embodiments of the present application adjust the first check code and the second check code according to whether the system is successfully started normally or repaired normally, so that when the system abnormally restarts, the system can automatically identify the abnormality and start the repair program to repair, effectively avoiding system crash.

Example implementations described herein are directed to an application crash testing platform that tests provided applications through a machine learning algorithm. Multiple instances of the application are spawned on a cloud-based platform, wherein the machine learning algorithm provides input to each of the instances until an incident causing the application to become inoperable is reached.

This invention provides an interactive software debugging and automatic repair method based on a large language model, addressing the problems of traditional software debugging relying on manual operation and being time-consuming, as well as the low repair accuracy caused by the lack of a dynamic operating environment in existing static automatic program repair technologies. The method first triggers a program exception by executing a reproduction command and captures the original crash stack information; then, it uses a large model to extract a simplified core function call sequence; subsequently, it constructs a closed-loop debugging mechanism, controlling the large model to generate hypotheses about the root cause of the defect and executing debugger probe commands, iteratively revising the hypotheses based on the dynamic information obtained by the debugger until the defective code is accurately located; finally, it combines the debugging context with the large model to complete root cause analysis and automatically generate and verify executable repair patches. This invention alleviates the "illusion" of large model inference by continuously capturing the underlying dynamic operating state, significantly improving the automation and accuracy of defect repair.

According to some embodiments, systems and methods are provided including receiving a crash dump file; extracting a call stack from the received crash dump file, wherein the call stack includes one or more functions, the functions having ordered positions in the call stack; converting the extracted call stack to natural language sentences; converting the natural languages sentences to a first call stack matrix; receiving the first call stack matrix and a second call stack matrix at a crash model, wherein the crash model is a Siamese neural network model; determining a similarity score for the first call stack matrix and the second call stack matrix; and determining whether the first call stack matrix and the second call stack matrix represent duplicate crashes based on the similarity score. Numerous other aspects are provided.

The embodiment of the invention provides an application crash diagnosis method and device, electronic equipment and a storage medium, and relates to the technical field of computers, target data of at least two modes when an application crashes are obtained, and the target data are used for representing the running condition when the application crashes; performing feature extraction on the target data of the at least two modes to obtain target feature vectors of the at least two modes; fusing the target feature vectors of the at least two modes to obtain a fused feature vector; and performing diagnosis based on the fused feature vector to determine a crash reason when the application program crashes, so that the application program crash can be automatically diagnosed, and the diagnosis efficiency of the application program crash can be improved. And the diagnosis is carried out by using the fused feature vector obtained by carrying out feature extraction and fusion on the target data of the at least two modes, so that the accuracy of diagnosis of the application crash can be improved.

The invention discloses a method, a device and equipment for judging a first crash event of a stream processing task, a medium and a product. The method comprises the following steps: determining current event attribute information of a current crash event; acquiring a crash event set in the current time period from a local memory; if the crash event set is a non-empty set, performing identification information matching on the event core identification information of the current crash event and event core identification information of a first historical crash event stored in the crash event set to obtain an identification information matching result; if the identification information matching result is that the event core identification information of each first historical crash event is not matched with the event core identification information of the current crash event, calling a plurality of pre-constructed Bloom filters to sequentially match the event core identification information of the current crash event to obtain a matching result, and according to the matching result, determining that the event core identification information of each first historical crash event is not matched with the event core identification information of the current crash event. And judging whether the current crash event is a first crash event or not to obtain a first crash event judgment result.

The invention provides a fault memory positioning method and computing equipment, relates to the technical field of computing equipment, and is used for quickly and accurately positioning a fault memory. The method comprises the steps of obtaining first machine inspection architecture MCA fault information and CXL fault information in response to crash of computing equipment; the first MCA fault information is information for recording hardware faults of the computing equipment, and specifically comprises a host physical address of a fault memory in the computing equipment. The CXL fault information comprises the equipment physical address of the fault memory of the equipment. And converting a host physical address in the first MCA fault information into an equipment physical address to obtain second MCA fault information. And matching the equipment physical address in the second MCA fault information with the equipment physical address in the CXL fault information to obtain a first matching result. Wherein the first matching result is used for indicating memories to which the equipment physical addresses in the second MCA fault information and the CXL fault information point. And based on the first matching result, determining a fault memory unit causing downtime in the CXL equipment.

This invention discloses a method and apparatus for verifying patch target functions based on symbol offset verification. The method includes the following steps: obtaining a patch file and extracting the name of the target function to be replaced, symbol offset, original function copy, and build identifier; obtaining the target program and parsing its symbol table and build identifier; selecting a verification path based on the availability of target function symbols: if a corresponding symbol exists, matching by function name and obtaining the actual offset, comparing it with the symbol offset in the patch; if no corresponding symbol exists, using the symbol offset in the patch as a reference, reading binary data of a predetermined length at the corresponding position in the target program, and comparing it with the corresponding data in the original function copy; outputting a verification report based on the comparison results; if all comparisons pass, patch installation is allowed; otherwise, execution is stopped and an alarm is issued. This invention can prevent system errors or crashes caused by accidental patch modifications.

The present disclosure describes a computer system for generating test code. According to an embodiment, the computer system includes one or more memories and one or more processors communicatively coupled to the one or more memories. The one or more processors, individually or collectively, parse software code to determine a function in the software code, generate a prompt based on the function, generate, using a machine learning model and based on the prompt, test code for testing the function, and execute the test code to generate a crash.

The invention is applicable to the technical field of computers, and provides a log recording method and device, electronic equipment and a storage medium, and the log recording method comprises the following steps: monitoring the running condition of target software, writing a preset number of logs recently generated by the target software into a snapshot cache region, and when monitoring that the target software crashes, recording the preset number of logs in the snapshot cache region; and calling the persistent interface to record a preset number of logs recently generated by the target software from the snapshot cache region to the log file, so that the logs recently generated by the target software are not lost, and a complete log when the target software crashes is obtained.

The present disclosure describes a computer system for generating test code. According to an embodiment, the computer system includes one or more memories and one or more processors communicatively coupled to the one or more memories. The one or more processors, individually or collectively, parse software code to determine a function in the software code, generate a prompt based on the function, generate, using a machine learning model and based on the prompt, test code for testing the function, and execute the test code to generate a crash.

A method of repairing crashed applications includes detecting a crash in an application operating in a host computing device. The application is migrated to a remote computer server. The remote computer server provisions computing resources to the application, while the application is resident in the remote computer server. Resumed operation of the application is executed, using the provisioned computing resources, in the remote computer server. Execution results are generated from the application, in the remote computer server. The generated execution results are migrated from the application back to the host computing device.

The invention discloses a method, a device and equipment for detecting and verifying compatibility of a opened software package and a storage medium, and relates to the technical field of software package compatibility detection.The method comprises the steps of extracting and analyzing metadata of the opened package to generate software package dependency data; the scanning system performs static dependency analysis according to the software package dependency data to generate static dependency data; performing a dynamic behavior test to generate dynamic test data; and generating a compatibility report according to the static dependency data and the dynamic test data. Static analysis is carried out on the open package metadata layer, quantitative evaluation is carried out from the static dependency satisfaction degree and the API change degree, dynamic function consistency verification is combined, and dynamic testing is carried out from the whole life cycle of installation and operation, version iteration and system migration. Compatibility verification of system calling hierarchy comprehensive coverage is carried out on the open software package in a local system environment, an upgraded target system environment and a version updating environment, and the problems of system crash and the like caused by compatibility risks are avoided.

The application discloses a kind of component running method and system of low code platform, belong to electric digital data processing technical field, the method includes the following steps: for component setting sandbox isolation environment;Load the component, and establish second window for component;To the component is style isolation, scope isolation and code isolation;In the second window, the component is run;Based on proxy mode, the component is monitored.Through sandbox, independent second window, style isolation, scope isolation, code isolation and monitoring in running etc. multiple isolation modes, component is run in low code platform, improve the reuse rate of component, external component is isolated from main application, guarantee the running performance and stability of main application, avoid error due to component and lead to main application error or crash;Ensure that component does not contaminate main application.

The disclosure provides a fault analysis method and an electronic device, and relates to the technical field of communication. The method comprises the following steps: generating a rule library according to historical fault information corresponding to the system type of a target system kernel; in the case that a BMC detects a kernel crash state, reading data from a log source in a streaming manner and storing the data to a circular buffer; obtaining a first log line from the circular buffer according to the reading order of at least one log line and obtaining a preset key information set; in the case that the first log line includes any key information in the preset key information set, obtaining a second rule mode corresponding to the key information of the first log line; according to the key information of the first log line, traversing a first rule group corresponding to the second rule mode in the rule library to obtain a target rule corresponding to the first log line; and according to the target rule, obtaining a fault analysis result corresponding to the fault of the target system kernel. The fault analysis efficiency and accuracy can be improved.

The application discloses a kind of based on virtualization dual-domain security fault handling method, device, equipment and medium, it is related to virtual system fault handling technical field, including: in security domain and non-security domain respectively deploying security virtual machine and non-security virtual machine and build the kernel module dependency relationship tree of non-security virtual machine While obtaining the resource occupation information of each kernel module;When kernel module in non-security virtual machine crashes, according to dependency relationship tree and resource occupation information, whether to execute fault degradation operation is evaluated and decided;If degradation is executed, the process related to fault module is terminated, unloads fault module and releases the system resource occupied by it;Final collection evaluation decision result and complete operation record report.It realizes the intelligent evaluation and elaboration processing of kernel module crash fault, limits the impact of fault in the minimum range, avoids single module crash to cause kernel panic or system crash, significantly improves the stability, observability and maintainability of operating system.

The invention discloses a kernel module log redirection method, electronic equipment, a storage medium and a product, and relates to the technical field of computers.According to the kernel module log redirection method, dynamic recognition and on-demand redirection of log paths are achieved in a kernel mode, logs can be flexibly switched among different output targets, and the efficiency is improved. And the problem of tight coupling of log output and system logs in a traditional scheme is avoided. After log content carries effective path information, a storage position can be autonomously determined, log isolation between modules is achieved, and mutual interference of logs of different subsystems is prevented. And after redirection is completed, default output is automatically recovered, the complexity of manual configuration and file operation is reduced, and the log processing automation level is improved. Therefore, the independence, high reliability and efficient management of log output are realized, and the risk of system crash and performance loss is remarkably reduced.