Design Method and System of Fault Recording Replay Framework Based on Coredump
By designing a fault recording and replay framework based on coredump, the problem of complex fault analysis in distributed multi-tasking devices is solved. It realizes simplified Coredump multi-point continuous acquisition and distributed fault tracing, and provides detailed fault diagnosis and analysis support.
Patent Information
- Application Number
- CN202211000022.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-08-19
- Publication Date
- 2026-11-17
- Estimated Expiration
- 2042-08-19
AI Technical Summary
In distributed, multi-tasking embedded devices, fault analysis and diagnosis are complex, time-consuming, and labor-intensive. Existing technologies struggle to effectively locate problems, especially when application logs in distributed, multi-tasking applications are numerous and complex.
Design a fault recording and replay framework based on coredump. By setting up collection points, filtering, managing, compressing, transmitting and analyzing coredump data, it realizes fault diagnosis of distributed multi-task applications, including collection settings, collection point triggering, data processing, data export and graphical playback.
It achieves streamlined Coredump multi-point continuous acquisition, can completely save task running status data, accurately identify fault execution paths, support distributed multi-task fault tracking and backtracking, and provide detailed fault diagnosis and analysis.
Smart Images

Figure CN116107848B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of embedded operating system technology, and more specifically, to a design method and system for a fault recording and replay framework based on coreedump. Background Technology
[0002] Embedded operating systems and related application development are typically built on cross-compilation environments, where the compilation and runtime environments are separate. This leads to complexity in fault analysis and diagnosis. Especially in distributed, heterogeneous environments composed of multiple embedded devices, applications are often distributed and multi-tasking, making it difficult to locate problems using only single-point debugging methods. While system logs can be analyzed, application logs from distributed multi-tasking applications are usually complex and require manual, step-by-step analysis, which is time-consuming and labor-intensive.
[0003] A core dump, also known as a kernel dump, is a memory snapshot of a process or task at the moment of a sudden crash. When an exception occurs in a program and is not caught within the process, the operating system dumps and saves the process or task's memory, register state, and execution stack information into a core file. Core files are binary files and can be opened and analyzed using tools such as gdb, elfdump, objdump, or, under Windows, windebug, and under Solaris, mdb. It's important to note that core dumps only save the state of the system at a single point in time and cannot dynamically display the overall system status. Summary of the Invention
[0004] To address the shortcomings of existing technologies, this invention provides a design method and system for a fault recording and replay framework based on coreedump.
[0005] According to the present invention, a design method and system for a fault recording and replay framework based on coredump are provided, the scheme of which is as follows:
[0006] Firstly, a design method for a fault recording and replay framework based on coredump is provided, the method comprising:
[0007] Data collection setup steps: The user sets the name of the task to be monitored;
[0008] Collection point triggering steps: Collection points include event monitoring points or hook-triggered collection points provided by the operating system, and active triggering interfaces provided by the framework, which are actively triggered by the user; these two types of collection points record task runtime data;
[0009] Data collection and filtering steps: Filter the data collection points according to the monitoring task name;
[0010] Data collection and management steps: Collect complete Coredump data and simplified Coredump data, and process them;
[0011] Data compression and storage steps: The collected simplified Coredump data file and the complete Coredump data file are packaged and compressed together, and saved to the system's permanent storage medium;
[0012] Data export steps: After the data is packaged and compressed, data transfer will be performed automatically, attempting to send it to the host machine;
[0013] Data import steps: During the data export steps, data is automatically received when data is automatically transmitted over the network, or manually imported when the user exports data.
[0014] Data analysis steps: Merge and analyze data imported from multiple target machines to achieve fault diagnosis analysis for distributed multi-tasking applications;
[0015] Data playback steps: Organize the execution paths of all detected tasks and the execution paths that caused the failures, and display them in a graphical way. Map the sequence of collection points to the path diagram, support single-step playback, and show the running status of each task that caused the failure.
[0016] Preferably, the data collection and management steps include:
[0017] Collection point judgment sub-step: Determine whether the collection point is caused by a fault, in order to determine whether to collect and store simplified or complete Coredump data;
[0018] Complete Coredump data acquisition steps: If the acquisition point is caused by a fault, a complete Coredump data acquisition will be performed, and the acquired data will be saved as a separate file;
[0019] Simplified Coredump data acquisition steps: For typical acquisition points, simplified Coredump data acquisition will be performed. Simplified Coredump data acquisition only saves the Coredump file header information, task runtime data, and register information. The data acquired each time is saved as an independent file.
[0020] Post-collection processing steps: If it is a simplified Coredump data collection, the collection will be completed and the system will wait for the next collection; if it is a complete Coredump data collection, the collection will end, the system's exception handling process will be entered, and data compression will be started.
[0021] Preferably, the saved individual files need to be timestamped.
[0022] Preferably, in the simplified Coredump data acquisition step, the method for acquiring task runtime data includes: saving the stack data of the task to which the acquisition point belongs in real time, and then parsing it together with the complete Coredump data when it is subsequently exported to the host machine to display the runtime data of multiple tasks.
[0023] Preferably, the data analysis step includes:
[0024] Task function execution data analysis sub-step: Based on the simplified Coredump file sequence sorted by timestamp, analyze each file one by one with the complete Coredump data acquisition module to obtain the task function execution data of all acquisition point sequences;
[0025] Fault point related data analysis sub-step: Based on the analysis of the complete Coredump file, obtain the task function call path and the function at the call point that caused the fault;
[0026] Fault execution path analysis sub-step: Combining the task function execution data of all collected point sequences and the fault point related data, analyze the execution paths of all detected tasks and the execution paths that caused the fault;
[0027] Data playback steps: sort out the execution paths of all detected tasks and the execution paths that caused the failures, and display them. Map the collection point sequence to the path graph, support single-step playback, and show the running status of each task that caused the failure.
[0028] Secondly, a design system for a fault recording and replay framework based on coredump is provided, the system comprising:
[0029] Data acquisition settings module: Users can set the names of the tasks to be monitored;
[0030] Collection point triggering module: Collection points include event monitoring points or hook-triggered collection points provided by the operating system, and active triggering interfaces provided by the framework, which are actively triggered by the user; these two types of collection points record task runtime data;
[0031] Data Acquisition and Filtering Module: Filters data acquisition points based on the monitoring task name;
[0032] Data Acquisition and Management Module: Acquires complete and simplified Coredump data and processes them;
[0033] Data compression and storage module: Packages and compresses the collected simplified Coredump data file and the complete Coredump data file into a single package and saves it to the system's permanent storage medium;
[0034] Data export module: After the data is packaged and compressed, it will automatically perform data transmission and attempt to send it to the host machine;
[0035] Data import module: In the data export module, data is automatically received when data is automatically transmitted over the network, or manually imported when the user exports data.
[0036] Data Analysis Module: Merges and analyzes data imported from multiple target machines to enable fault diagnosis and analysis of distributed multi-tasking applications;
[0037] Data replay module: It sorts out the execution paths of all detected tasks and the execution paths that caused the failures, and displays them in a graphical way. It maps the sequence of collection points to the path graph, supports single-step replay, and shows the running status of each task that caused the failure.
[0038] Preferably, the data acquisition and management module includes:
[0039] Collection point judgment submodule: Determines whether the collection point is caused by a fault, in order to determine whether to collect simplified or complete Coredump data;
[0040] Complete Coredump data acquisition module: If the acquisition point is caused by a fault, a complete Coredump data acquisition will be performed, and the acquired data will be saved as a separate file;
[0041] Simplified Coredump data acquisition module: For typical acquisition points, simplified Coredump data acquisition will be performed. Simplified Coredump data acquisition only saves the Coredump file header information, task runtime data, and register information. The data acquired each time is saved as an independent file.
[0042] Post-collection processing module: If it is a simplified Coredump data collection, it will complete one collection and continue to wait for the next collection; if it is a complete Coredump data collection, it will end the collection, enter the system's exception handling process, and start data compression.
[0043] Preferably, the saved individual files need to be timestamped.
[0044] Preferably, in the simplified Coredump data acquisition module, the method for acquiring task runtime data includes: saving the stack data of the task to which the acquisition point belongs in real time, and then parsing it together with the complete Coredump data when it is exported to the host machine to display the runtime data of multiple tasks.
[0045] Preferably, the data analysis module includes:
[0046] Task function execution data analysis submodule: Based on the simplified Coredump file sequence sorted by timestamp, and combined with the complete Coredump data acquisition module, the task function execution data of all acquisition point sequences are analyzed one by one to obtain the task function execution data of all acquisition point sequences;
[0047] Fault Point Related Data Analysis Submodule: Based on the analysis of the complete Coredump file, it obtains the call path of the task function that caused the fault and the function at the call point.
[0048] The fault execution path analysis submodule combines the task function execution data of all collected point sequences with fault-related data to analyze the execution paths of all detected tasks and the execution paths that caused the faults.
[0049] Data replay module: sorts out the execution paths of all detected tasks and the execution paths that caused the failures, and displays them. It maps the sequence of collection points to the path graph, supports single-step replay, and shows the running status of each task that caused the failure.
[0050] Compared with the prior art, the present invention has the following beneficial effects:
[0051] 1. A streamlined Coredump multi-point continuous acquisition mechanism is implemented, which can completely save the running status data of a specified task throughout its lifecycle. At the same time, this invention separates the streamlined Coredump from the operating system exception handling, and designs a complete system running status data saving mechanism on a domestic operating system. While ensuring the completeness of the status data, duplicate data is filtered out, thus realizing a streamlined Coredump multi-point continuous acquisition mechanism;
[0052] 2. Design and implement a fault tracing and backtracking framework supporting distributed multi-tasking. Based on the simplified Coredump multi-point continuous acquisition mechanism described above, this invention designs and implements a complete fault tracing and backtracking framework for distributed multi-tasking applications. It includes functions such as acquisition settings, acquisition filtering, acquisition management, data compression, data transmission management, data analysis, and data playback. It can completely and accurately trace the execution path of the fault, replay the relevant data of the task function execution at key points of the path, and display detailed system operation data related to the fault point; thus providing powerful assistance to users in fault diagnosis and analysis. Attached Figure Description
[0053] Other features, objects, and advantages of the present invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:
[0054] Figure 1 Design of a fault recording and replay framework;
[0055] Figure 2This outlines the workflow for the data acquisition and management module.
[0056] Figure 3 Design a graphical interface for data playback. Detailed Implementation
[0057] The present invention will now be described in detail with reference to specific embodiments. These embodiments will help those skilled in the art to further understand the present invention, but do not limit the invention in any way. It should be noted that those skilled in the art can make several changes and improvements without departing from the concept of the present invention. These all fall within the protection scope of the present invention.
[0058] This invention provides a design method for a fault recording replay framework based on coreedump, referring to... Figure 1 As shown, the functional design of this fault recording replay framework involves a target machine and a host machine. The target machine is an embedded platform running a domestic operating system, such as the Ruihua OS + Loongson 3A3000 platform; the host machine is a general-purpose platform running a domestic desktop operating system, such as the Kylin OS + Phytium 2000 platform. The method specifically includes the following steps:
[0059] Data collection setup steps: The user sets the name of the task to be monitored.
[0060] Data collection point triggering steps: Data collection points are divided into two main categories. The first is data collection points triggered by event monitoring points or hooks provided by the operating system, such as task state switching, signal transmission and reception, etc. The second is actively triggered interfaces provided by the framework, which are triggered by the user, for example, by instrumenting at function entry and exit points. These two types of data collection points can record task runtime data relatively completely, facilitating fault diagnosis and backtracking.
[0061] Data collection and filtering steps: The data collection points are filtered according to the monitoring task names set by the user, thereby reducing the amount of data collected.
[0062] Data acquisition and management steps: This is the core module of the fault recording and replay framework on the target machine, and its working logic design refers to... Figure 2 As shown, it includes the following sub-functional modules:
[0063] The sampling point determination sub-step involves determining the type of sampling point. The core issue is whether the sampling point is caused by a fault, in order to determine whether the collected data is simplified or complete Coredump data.
[0064] Complete Coredump data collection steps: If the collection point is caused by a fault, a complete Coredump data collection will be performed, including Coredump file header information, system data segments, code segments, system stack, register information, etc. The collected data will be saved as a separate file. In addition, for subsequent analysis of the fault execution path, the saved file needs to be timestamped.
[0065] Simplified Coredump Data Acquisition Steps: For typical acquisition points, simplified Coredump data acquisition will be performed. This simplified Coredump data acquisition method is a customized implementation of this invention, saving only the Coredump file header information, task runtime data, and register information. Each acquired data entry is saved as a separate file. Additionally, to facilitate subsequent analysis of fault execution paths, the saved files need to be timestamped. The method for acquiring task runtime data is as follows: The stack data of the task at the acquisition point is saved in real-time. When exported to the host machine, this data is combined with the complete Coredump data for parsing, allowing the display of runtime data from multiple tasks.
[0066] Post-collection processing steps: If it is a simplified Coredump data collection, the collection will be completed and the system will wait for the next collection; if it is a complete Coredump data collection, the collection will end, the system's exception handling process will be entered, and data compression will be started.
[0067] Data compression and storage steps: Compress all collected simplified Coredump data files and complete Coredump data files (a complete data collection should include N simplified Coredump data files and one complete Coredump data file) into a single package and save it to a permanent storage medium available in the system, such as a hard drive.
[0068] Data export steps: Basic network output functionality is provided. After data packaging is complete, data transfer will be automatically initiated, attempting to send it to the host machine. It's important to consider the possibility of network failure in case of an malfunction. In this case, automatic transfer will not occur, and the user will need to manually export the data, such as by copying it to a USB drive.
[0069] Data import steps: It supports automatic data reception when the data export module automatically transmits data over the network, and also supports manual import when users export data themselves.
[0070] Data analysis steps: This is the core module of the fault recording and replay framework on the host machine. It supports the merging and analysis of data imported from multiple target machines, thereby enabling fault diagnosis and analysis of distributed multi-tasking applications. It includes the following sub-functional modules:
[0071] Task function execution data analysis sub-step: Based on the simplified Coredump file sequence sorted by timestamp, and combined with the complete Coredump data acquisition module, we can analyze each one to obtain the task function execution data of all acquisition point sequences, including code location, function call path to that point, local variable information, register information, etc.
[0072] Fault-related data analysis sub-steps: Based on the analysis of the complete Coredump file, the call path of the task function that caused the fault and the function at the call point can be obtained, including code location, global variables, local variables, register information, etc.
[0073] Fault execution path analysis sub-steps: By combining the task function execution data of all the above-mentioned collection point sequences and the fault point related data, the execution paths of all detected tasks and the execution paths that caused the faults can be analyzed.
[0074] Data playback steps: After identifying the execution paths of all monitored tasks and the execution paths that caused the faults, the data is displayed graphically. The sequence of data collection points is mapped to the path graph, and single-step playback is supported. This provides a clear view of the execution status of each task that caused the fault, facilitating fault diagnosis. (See the data playback graphical interface design diagram for reference.) Figure 3 As shown.
[0075] The invention will now be described through implementation examples.
[0076] 1. The embedded platform running a domestic operating system adopts the Ruihua operating system + Loongson 3A3000 platform; the host machine running a domestic desktop operating system adopts the Kylin operating system + Phytium 2000 platform.
[0077] 2. Data collection point triggering is implemented in two ways: First, data collection points are triggered by event monitoring points or hooks provided by the operating system, such as task state switching, signal transmission and reception. Second, data collection is triggered by instrumenting at the entry and exit points of distributed application functions, with the instrumented program actively triggering data collection by calling the data collection interface provided by the framework.
[0078] 3. Streamlining Coredump data acquisition requires saving Coredump file header information, task runtime data, register information, etc. The Coredump file header information should be filled in according to the Coredump standard specification; task runtime data can be saved manually or obtained from the task stack through the operating system interface and then parsed on the host machine using tools such as gdb; register information can be saved by reading the values of key registers one by one using assembly language. The collected data should be saved as a separate file, and the filename should include a timestamp.
[0079] 4. Complete Coredump data acquisition requires saving the Coredump file header information, system data segment, code segment, system stack, register information, etc. The Coredump file header information should be filled in according to the Coredump standard specification; the system data segment, code segment, and system stack should have their corresponding memory addresses and sizes obtained through the operating system interface and then saved; register information can be obtained through the operating system interface or by reading and saving each register value individually through assembly language. The acquired data should be saved as a separate file, and the filename should include a timestamp.
[0080] 5. A complete data acquisition should include N simplified Coredump data files and one complete Coredump data file. These can be packaged together, compressed using the highly portable zlib library, and then saved to a permanent storage medium available in the system, such as a hard drive.
[0081] 6. Data analysis is implemented on the host platform. By combining the task function execution data of all collection point sequences and the relevant data of the fault points, the execution paths of all detected tasks and the execution paths that caused the faults can be identified.
[0082] 7. The data playback interface can be implemented using Java or QT.
[0083] This invention provides a design method and system for a fault recording and replay framework based on coredump, 1) implementing a simplified coredump multi-point continuous acquisition mechanism. Coredump can completely save the program and system running status data at a certain moment, but it will save a lot of duplicate data in the case of multiple acquisitions. Coredump is usually used in the exception handling function of the system, and is a system exception handling mechanism. This invention separates it from exception handling and designs a complete system running status data saving mechanism on a domestic operating system. While ensuring the integrity of the status data, it filters out duplicate data and realizes a simplified coredump multi-point continuous acquisition mechanism.
[0084] 2) Design and implement a fault tracing and backtracking framework supporting distributed multi-tasking. Based on the simplified Coredump multi-point continuous acquisition mechanism described above, this invention designs and implements a complete fault tracing and backtracking framework for distributed multi-tasking applications. It includes functions such as acquisition settings, acquisition filtering, acquisition management, data compression, data transmission management, data analysis, and data playback. It can completely and accurately trace the execution path of the fault, replay the relevant data of the task function execution at key points of the path, and display detailed system operation data related to the fault point; thus providing powerful assistance to users in fault diagnosis and analysis.
[0085] Those skilled in the art will understand that, besides implementing the system and its various devices, modules, and units provided by this invention in the form of purely computer-readable program code, the same functions can be achieved entirely through logical programming of the method steps, making the system and its various devices, modules, and units of this invention function in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers. Therefore, the system and its various devices, modules, and units provided by this invention can be considered as a hardware component, and the devices, modules, and units included therein for implementing various functions can also be considered as structures within the hardware component; alternatively, the devices, modules, and units for implementing various functions can be considered as both software modules implementing the method and structures within the hardware component.
[0086] Specific embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the specific embodiments described above, and those skilled in the art can make various changes or modifications within the scope of the claims, which do not affect the essence of the present invention. Unless otherwise specified, the embodiments and features described in this application can be arbitrarily combined with each other.
Claims
1. A design method for a fault recording and replay framework based on Coredump, characterized in that, include: Data collection setup steps: Obtain the monitoring task name set by the user; Collection point triggering steps: Collection points include event monitoring points or hook-triggered collection points provided by the operating system, and active triggering interfaces provided by the framework, which are actively triggered by the user. Data during task execution is recorded using these two types of data collection points; Data collection and filtering steps: Filter the data collection points according to the monitoring task name; Data collection and management steps: Collect complete Coredump data and simplified Coredump data, and process them; Data compression and storage steps: The collected simplified Coredump data file and the complete Coredump data file are packaged and compressed together, and saved to the system's permanent storage medium; Data export steps: After the data is packaged and compressed, perform data transmission and attempt to send it to the host machine; Data import steps: During the data export steps, data is automatically received when data is automatically transmitted over the network, or manually imported when the user exports data. Data analysis steps: Merge and analyze data imported from multiple target machines to achieve fault diagnosis analysis for distributed multi-tasking applications; Data playback steps: sort out the execution paths of all detected tasks and the execution paths that caused the failure, and display them in a graphical way. Map the sequence of collection points to the path diagram, support single-step playback, and show the running status of each task that caused the failure. The data collection and management steps include: Collection point determination sub-step: Determine whether the collection point is caused by a fault, in order to determine whether to collect simplified Coredump data or complete Coredump data; Complete Coredump data acquisition steps: If the acquisition point is caused by a fault, a complete Coredump data acquisition will be performed, and the acquired data will be saved as a separate file; Simplified Coredump data acquisition steps: For typical acquisition points, simplified Coredump data acquisition will be performed. Simplified Coredump data acquisition only saves the Coredump file header information, task runtime data, and register information. The data acquired each time is saved as an independent file. Post-collection processing steps: If it is a simplified Coredump data collection, the collection will be completed and the system will wait for the next collection; if it is a complete Coredump data collection, the collection will end, the system's exception handling process will be entered, and data compression will be started.
2. The design method of the fault recording and replay framework based on Coredump according to claim 1, characterized in that, Add a timestamp to the saved individual files.
3. The design method of the fault recording and replay framework based on Coredump according to claim 1, characterized in that, In the simplified Coredump data collection steps, the method for collecting task runtime data includes: saving the stack data of the task to which the collection point belongs in real time, and then parsing it together with the complete Coredump data when it is exported to the host machine to display the runtime data of multiple tasks.
4. The design method of the fault recording and replay framework based on Coredump according to claim 1, characterized in that, The data analysis steps include: Task function execution data analysis sub-step: Based on the simplified Coredump file sequence sorted by timestamp, and combined with the complete Coredump data acquisition steps, analyze each step to obtain the task function execution data of all acquisition point sequences; Fault point related data analysis sub-step: Based on the analysis of the complete Coredump file, obtain the task function call path and the function at the call point that caused the fault; Fault execution path analysis sub-step: Combining the task function execution data of all collected point sequences and the fault point related data, analyze the execution paths of all detected tasks and the execution paths that caused the fault; Data playback steps: sort out the execution paths of all detected tasks and the execution paths that caused the failures, and display them. Map the collection point sequence to the path graph, support single-step playback, and show the running status of each task that caused the failure.
5. A design system for a fault recording and replay framework based on Coredump, characterized in that, include: Data Acquisition Settings Module: Retrieves the monitoring task name set by the user; Collection point triggering module: Collection points include event monitoring points or hook-triggered collection points provided by the operating system, and active triggering interfaces provided by the framework, which are actively triggered by the user. Data during task execution is recorded using these two types of data collection points; Data Acquisition and Filtering Module: Filters data acquisition points based on the monitoring task name; Data Acquisition and Management Module: Acquires complete and simplified Coredump data, and processes them. Data compression and storage module: Packages and compresses the collected simplified Coredump data file and the complete Coredump data file into a single package and saves it to the system's permanent storage medium; Data export module: After the data is packaged and compressed, it performs data transmission and attempts to send it to the host machine; Data import module: In the data export module, data is automatically received when data is automatically transmitted over the network, or manually imported when the user exports data. Data Analysis Module: Merges and analyzes data imported from multiple target machines to enable fault diagnosis and analysis of distributed multi-tasking applications; Data replay module: It sorts out the execution paths of all detected tasks and the execution paths that caused the failures, and displays them in a graphical way. It maps the sequence of collection points to the path graph, supports single-step replay, and shows the running status of each task that caused the failure. The data acquisition and management module includes: Collection point determination submodule: Determines whether the collection point is caused by a fault, in order to determine whether to collect simplified or complete Coredump data; Complete Coredump data acquisition module: If the acquisition point is caused by a fault, a complete Coredump data acquisition will be performed, and the acquired data will be saved as a separate file; Simplified Coredump data acquisition module: For typical acquisition points, simplified Coredump data acquisition will be performed. Simplified Coredump data acquisition only saves the Coredump file header information, task runtime data, and register information. The data acquired each time is saved as an independent file. Post-collection processing module: If it is a simplified Coredump data collection, it will complete one collection and continue to wait for the next collection; if it is a complete Coredump data collection, it will end the collection, enter the system's exception handling process, and start data compression.
6. The design system for the fault recording and replay framework based on Coredump according to claim 5, characterized in that, Add a timestamp to the saved individual files.
7. The design system for the fault recording and replay framework based on Coredump according to claim 5, characterized in that, In the simplified Coredump data acquisition module, the method for acquiring task runtime data includes: saving the stack data of the task to which the acquisition point belongs in real time, and then parsing it together with the complete Coredump data when it is exported to the host machine to display the runtime data of multiple tasks.
8. The design system for the fault recording and replay framework based on Coredump according to claim 5, characterized in that, The data analysis module includes: Task function execution data analysis submodule: Based on the simplified Coredump file sequence sorted by timestamp, and combined with the complete Coredump data acquisition module, the task function execution data of all acquisition point sequences are analyzed one by one to obtain the task function execution data of all acquisition point sequences; Fault Point Related Data Analysis Submodule: Based on the analysis of the complete Coredump file, it obtains the call path of the task function that caused the fault and the function at the call point. The fault execution path analysis submodule combines the task function execution data of all collected point sequences with fault-related data to analyze the execution paths of all detected tasks and the execution paths that caused the faults. Data replay module: sorts out the execution paths of all detected tasks and the execution paths that caused the failures, and displays them. It maps the sequence of collection points to the path graph, supports single-step replay, and shows the running status of each task that caused the failure.
Citation Information
Patent Citations
Record playback technology-based embedded software debugging device and method
CN102129408A
Concurrent program reappearance debugging method based on snapshot sequence running
CN104461876A
Equipment state monitoring method and monitoring system based on edge cloud collaboration and storage medium
CN112947290A
