Information processing device and method
The dual-CPU information processing device with a watchdog mechanism addresses the inefficiencies of conventional failure analysis in embedded systems by collecting fault information efficiently, minimizing performance impact and shortening analysis time.
Patent Information
- Application Number
- JP2020199237
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2020-11-30
- Publication Date
- 2024-11-06
- Estimated Expiration
- 2040-11-30
AI Technical Summary
Conventional failure analysis in embedded systems is laborious and time-consuming, particularly due to the difficulty in determining failure causes when the operating system stalls, and existing methods to detect failures require CPU resources, degrading system performance.
An information processing device utilizing two CPUs, where one CPU provides normal functions and the other collects fault information, with a watchdog mechanism to notify the second CPU of failures via interrupt signals, allowing data collection without allocating resources to non-essential operations.
Enables efficient failure information collection in embedded systems with limited processing capacity, reducing the risk of performance degradation and shortening the analysis time by isolating CPU resources for essential functions.
Smart Images

Figure 0007577985000001 
Figure 0007577985000002
Abstract
Description
[Technical field]
[0001] The present invention relates to an information processing device and method. [Background technology]
[0002] Traditionally, fault analysis of embedded systems has been time-consuming and laborious. For example, one type of fault is an operating system stall (hang-up). When this fault occurs, it is not possible to execute commands for debugging purposes, and it is also not possible to use the debugging functions provided by the operating system, making fault analysis difficult.
[0003] Patent Document 1 discloses a main processing unit that provides functions, and a fault detection unit / monitoring control unit that detects faults and collects information. The main processing unit periodically sends a predetermined signal to the fault detection unit / monitoring control unit during normal operation, and when a certain amount of time has passed since the last information transmission, the main control unit considers a fault in the fault detection unit / monitoring control unit to be an occurrence of a fault. Summary of the Invention [Problem to be solved by the invention]
[0004] However, in order to correctly determine the cause of a fault in the conventional technology, the device status must be periodically sent to the fault detection unit / monitoring control unit, which requires the allocation of CPU resources to this operation. Embedded systems often have lower CPU performance than general PCs, so there is a problem that using this method poses a very high risk of degrading product performance.
[0005] The present invention has been made in consideration of the above problems, and aims to provide an information processing device and method that are capable of collecting information when a fault occurs even in an environment where the processing power of an embedded system is limited. [Means for solving the problem]
[0006] In order to solve the above problems and achieve the object, an information processing device of the present invention includes a function providing unit that provides a normal function by the operation of a first CPU, and a function providing unit that provides a normal function by the operation of a second CPU. , notified by the function providing unit of the first CPU Fault Information The address of the memory area a fault information collection unit that collects the fault information; In The occurrence of a fault The second CPU A notification unit that notifies the fault information collection unit Watch Dog as and the Watch Dog notifies the second CPU of an occurrence of a fault in the function providing unit by sending an interrupt signal when a refresh operation for the Watch Dog, which is periodically performed by the first CPU, is not performed for a certain period of time, thereby notifying the fault information collecting unit of the occurrence of a fault in the function providing unit; The fault information collecting unit The above interrupt signal by Watch Dog According to the above In the function provision department When the occurrence of a failure is notified, the function provision unit The corresponding From memory area The address indicating the fault information The data is read out and stored in a storage unit. Effect of the Invention
[0007] According to the present invention, it is possible to effectively collect information when a failure occurs even in an environment where the processing power of an embedded system is limited. [Brief description of the drawings]
[0008] [Figure 1] FIG. 1 is a diagram illustrating an example of a configuration of an information processing device according to an embodiment. [Diagram 2] FIG. 2 is a diagram illustrating an example of a process flow in which a first CPU notifies a second CPU of an allocation area of a function providing unit. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0009] Hereinafter, embodiments of an information processing apparatus and method will be described in detail with reference to the accompanying drawings.
[0010] (Embodiment) FIG. 1 is a diagram showing an example of the configuration of an information processing device according to an embodiment. FIG. 1 shows the configuration of an information processing device of an embedded system such as a printer. The information processing device 1 has a hardware configuration including a CPU (Central Processing Unit), a RAM (Random Access Memory) 13 which is a volatile memory, a non-volatile memory 23, and a network device 24. The CPU is a plurality of CPUs, for example a first CPU 11 and a second CPU 21. The information processing device 1 also includes a timer 12 and a watch dog 14 as hardware. The timer 12 is used by the first CPU 11.
[0011] 1, the information processing device 1 includes a function providing means 10 and a fault information collecting means 20. The function providing means 10 provides normal functions by software function units (corresponding to "function providing units") 15 such as an OS 151 and various applications operated by a first CPU 11. The fault information collecting means 20 provides a function of collecting fault information by software function units (corresponding to "fault information collecting units") 25 such as an information collection module 252 operated by a second CPU 21. Here, the OS 151 refers to a general-purpose OS (Operating System) / real-time OS incorporated in an embedded system.
[0012] The function providing means 10 corresponds to a processing circuit that provides a printing function in the case of an embedded system of a printer, for example. The fault information collecting means 20 has various components for collecting and storing information when a fault occurs in the function providing means 10.
[0013] In this configuration, the first CPU 11 notifies the second CPU 21 of an allocation area of the RAM 13 to be used by the function providing unit 15. First, a process of notifying the second CPU 21 of an allocation area of the function providing unit 15 from the first CPU 11 will be described.
[0014] Fig. 2 is a diagram showing an example of a process flow in which the first CPU 11 notifies the second CPU 21 of an allocation area of the function providing unit 15. As shown in Fig. 2, when the first CPU 11 starts the OS (S11), it allocates a log area on the RAM 13 (S12), notifies the second CPU 21 of an address d1 of the log area (S13), and further notifies the second CPU 21 of an address d2 of the OS area (S14).
[0015] The second CPU 21 starts the information collecting module 252 (S21), receives the address d1 of the log area transmitted from the first CPU 11 and sets it in the information collecting module 252 (S22), and further receives the address d2 of the OS area transmitted from the first CPU 11 and sets it in the information collecting module 252 (S23). The information collecting module 252 reads data such as the OS from the above-mentioned set address when a failure occurs in the function providing means 10.
[0016] Returning to Fig. 1, the method of collecting data when a failure occurs will be described. The function providing unit 15 reads a periodic signal from the timer 12 by the timer handler 152, and refreshes the watchdog 14 at that timing.
[0017] Here, the Watch Dog 14 is hardware for detecting runaway of the system, and issues a non-maskable interrupt (NMI) after a certain period of time has elapsed. In most cases, the system is restarted when this interrupt is detected. Since the Watch Dog 14 sends an interrupt to the second CPU 21, it may be implemented on the same ASIC as the second CPU 21, or it may be implemented separately and a signal line may be connected to the interrupt terminal of the second CPU 21.
[0018] The function providing unit 15 performs an operation called "refresh" on the Watch Dog 14 at regular intervals. If the function providing unit 15 falls into a fatal abnormal state, a certain period of time will pass without "refreshing." If a certain period of time passes without a refresh operation being performed, the Watch Dog 14, as a "notification unit," notifies the second CPU 21 of a signal (interrupt signal).
[0019] The second CPU 21 monitors the occurrence of a fault by an interrupt signal. When the second CPU 21 detects the interrupt signal, the NMI interrupt handler 251 performs interrupt processing of the information collection module 252. The information collection module 252 reads memory data from a set address (log area or OS area) of the RAM 13 as the interrupt processing. For example, the information collection module 252 reads memory data M1 from an address of the OS area of the RAM 13. Furthermore, the information collection module 252 outputs the read memory data to the network device 24, and transmits the memory data from the network device 24 to a storage unit such as an external server or cloud storage via a connected network. Note that the memory data may be stored in a storage unit such as the non-volatile memory 23 provided inside the information processing device 1, not limited to a storage destination outside the information processing device 1.
[0020] Thereafter, for example, the system is restarted, and an analyst uses the data in the storage unit to analyze the cause of the failure.
[0021] In this manner, in this embodiment, a form has been shown in which separate CPUs are implemented for the function providing means for providing normal functions such as the OS and applications, and the fault information collecting means for collecting fault information. When the OS accesses an illegal memory space or falls into a state called a stall, it enters a state called a "hang-up." However, in this configuration, the debug function operates only when a fault occurs, and reads out the area of data on RAM 13 that corresponds to function providing unit 15. Therefore, there is no need to allocate CPU resources to operations that are not directly related to the normal functions provided by the product. Therefore, the risk of degrading the performance of the product is extremely low.
[0022] In addition, the time it takes to collect fault information can be reduced by limiting the data to be read when a fault occurs to areas such as the OS. If a normal user process (command) accesses an invalid memory space, the OS has a mechanism to detect this and forcibly terminate the process in question. Therefore, it is rare for the user process itself to experience a stall-type fault. On the other hand, the OS can access areas such as hardware registers where erroneous access can have fatal consequences. In other words, it is often the OS that is the software that experiences fatal faults such as stalls. Therefore, when reading information required for fault analysis, the range can be limited to areas such as the OS, making it possible to reduce the time it takes to collect fault information.
[0023] In addition, the read data is stored in the storage unit, so the designer can use this data later for cause analysis. Also, by reading data from the OS area, the cause of the failure can be correctly determined.
[0024] Furthermore, even if a fault occurs in a large-scale embedded system, the time required for fault analysis can be shortened.
[0025] As a result, it is possible to collect information when a failure occurs even in an environment where the processing power of an embedded system is limited.
[0026] Although the embodiments have been described above, the above embodiments are presented as examples and are not intended to limit the scope of the invention. These novel embodiments can be embodied in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments are included in the scope and spirit of the invention, and are included in the scope of the invention and its equivalents described in the claims. [Explanation of symbols]
[0027] 1. Information processing device 10 Function provision means 11 First CPU 12 Timer 13 RAM 14. Watch Dog 15 Software Functionality Unit (corresponding to the "Functional Provision Unit") 20. Means of collecting fault information 21 Second CPU 23 Non-volatile memory 24 Network Devices 25 Software function unit (corresponding to the "fault information collection unit") 151 OS 152 Timer Handler 251 NMI Interrupt Handler 252 Information Collection Module M1 Memory Data [Prior art documents] [Patent documents]
[0028] [Patent Document 1] JP 2017-207903 A
Claims
1. a function providing unit that provides a normal function through the operation of the first CPU; a fault information collection unit that collects, by operation of a second CPU, addresses of memory areas of fault information notified from the function provision unit of the first CPU; a Watch Dog as a notification unit that notifies the fault information collection unit of the second CPU of the occurrence of a fault in the function providing unit; having the Watch Dog notifies the second CPU of an occurrence of a fault in the function providing unit by notifying the fault information collecting unit of an interrupt signal when a refresh operation for the Watch Dog, which is periodically performed by the first CPU, is not performed for a certain period of time; when the fault information collection unit is notified of the occurrence of a fault in the function providing unit by the interrupt signal from the Watch Dog, the fault information collection unit reads data of the address indicating the fault information from the memory area corresponding to the function providing unit and stores the data in a storage unit. Information processing device.
2. the fault information collection unit stores the data in an external storage unit via a network; The information processing device according to claim 1 .
3. the fault information collection unit reads the data from an address of an OS area, which is the memory area notified by the function provision unit after startup; 3. The information processing device according to claim 1 or 2.
4. The first CPU is a CPU of a printer embedded system, The function providing unit provides a printing function. The information processing device according to claim 1 .
5. A method for an embedded system to collect and store normal function failure information, comprising: providing the normal function through operation of a first CPU; a second CPU operates to collect addresses of memory areas of the fault information notified from the first CPU; performing a refresh operation on a Watch Dog from the first CPU at regular intervals; When the refresh operation is not performed by the first CPU for a certain period of time, an interrupt signal is sent from the Watch Dog to the second CPU; When the interrupt signal from the Watch Dog is notified, the second CPU reads data at the address indicating the fault information from the memory area corresponding to the normal function as an interrupt process when a fault occurs in the normal function, and stores the data in a storage unit. method.
Citation Information
Patent Citations
System for collecting failure information
JP2006313406A
Information processing apparatus, memory dump system and memory dump method
JP2011070655A
Processor, method and program
JP2017207903A
Information processing apparatus, control method, and program
JP2020135240A
A Method for Diagnosing Power Supply Failure in a Wireless Communication Device
US20170168537A1
