A program self-backup method based on on-chip FLASH without re-burning

The program self-backup method using on-chip FLASH with checksum verification and jump instructions addresses the need for reliable backup without re-burning, ensuring system stability and cost-effectiveness in harsh environments.

CN113886145BActive Publication Date: 2024-12-06JIANGSU JITRI INTELLIGENT OPTOELECTRONIC SYST RES INST CO LTD
View PDF 11 Cites 0 Cited by

Patent Information

Application Number
CN202111245677.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-10-26
Publication Date
2024-12-06
Estimated Expiration
2041-10-26

AI Technical Summary

Technical Problem

Existing program backup methods for embedded systems in space or harsh environments require re-burning FLASH, which is prone to failure due to electromagnetic interference and high cost, and are not suitable for systems with limited on-chip RAM resources.

Method used

A program self-backup method using two copies of the program in on-chip FLASH, one as the main and one as the backup, with checksum verification and jump instructions to switch between them without re-burning, utilizing RAM for flag variables and jump pointers.

Benefits of technology

Enables reliable program backup without re-burning FLASH, maintaining system stability and reducing costs, suitable for systems with limited RAM resources and harsh environments.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN113886145B_ABST
    Figure CN113886145B_ABST
Patent Text Reader

Abstract

The present invention discloses a program self-backup method based on on-chip FLASH and without re-burning, comprising the following steps: the original program in the FLASH of a single-chip microcomputer is made into two copies, namely a main program and a backup program; the main program is modified and compiled for error monitoring and switching of main functions and interrupt functions, and a HEX file corresponding to the main program and a HEX file corresponding to the backup program are generated respectively; the code length and checksum information of the main program are stored at a fixed address; the content in the HEX file of the main program, the content in the HEX file of the backup program, and the code length and checksum information of the main program are merged to generate a final HEX file and write it into FLASH. The present invention does not need to re-burn FLASH, does not need an additional boot program, solves the problems of high cost of dual FLASH, need to re-burn FLASH, and risk of burning failure in the prior art, and is suitable for popularization and use.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention belongs to the technical field of embedded application, and in particular relates to a program self-backup method based on on-chip FLASH and without the need for re-burning. Background Art

[0002] Embedded systems represented by single-chip microcomputers often use FLASH to store programs. When such embedded systems are used in space, electromagnetic radiation or single particle radiation often causes some data in the program code in FLASH to be accidentally modified, resulting in program failure and product failure to work properly. In order to improve reliability, it is often necessary to back up the program code. When one copy of the program is damaged, the backup program can be used to continue working, thereby improving product reliability.

[0003] The Chinese invention patent, application number 201310034458.4, patent name is a dual-boot method based on a single FLASH memory chip, which discloses a method of improving device reliability by using the main and standby BOOT. This method requires the use of an external CPLD device to complete the switching of the main and standby programs. Similarly, the Chinese invention patent, application number 200810101827.6, patent name is a boot program backup method and backup device, and application number 200810212826.9, patent name is a method for updating and repairing the basic input and output system, which also adopts a similar method of using the main and standby BOOT to achieve program backup and switching.

[0004] The Chinese invention patent, application number 201110425886.0, patent name is flash memory program automatic error correction and repair method and device, and application number 202010785872.9, patent name is a method and device for monitoring the logic code of a satellite-borne software program, and application number 202011326245.5, patent name is a program running method and device, electronic equipment, storage medium, all disclose various monitoring and error correction methods, when the program running in the RAM area has an error, it is reloaded from FLASH to achieve program repair. The application conditions of the above patents are that the program code is stored in the FLASH memory, and the program code is copied to the RAM area after startup. The program is executed in the RAM area instead of the FLASH, so the RAM area is needed to run the program. The program code itself is not backed up, but the program code copied to the RAM area is monitored and repaired. In actual applications, many embedded systems have limited on-chip RAM resources, and the program can only run directly in FLASH. The on-chip RAM is only used to store variables when the program is running. The above patents cannot be applied to such embedded systems.

[0005] The Chinese invention patent, application number 201410357317.0, is titled "A method and system for recovering from single-chip microcomputer software errors". It discloses backing up the main program into multiple functions with the same function content, and then packaging them into a new program code. In the new program code, error detection is used to jump into different backup sub-functions, thereby realizing program backup.

[0006] The Chinese invention patent, application number 201210225631.4, is titled "A method for realizing automatic dual-FLASH startup of CPU system". It discloses the use of two FLASH chips to realize program backup. However, the patent uses two FLASH chips, which is costly and requires an additional CPLD for control switching.

[0007] The Chinese invention patent, application number 201510955701.5, is titled "A method for implementing program backup". It discloses that the program backup is implemented in the FLASH inside the microcontroller, but the method requires re-burning the FLASH when switching between the main and standby modes. Because the corresponding entry address of the interrupt of the microcontroller is fixed, and the corresponding interrupt response function address is stored in the interrupt vector table, according to the implementation method of the patent, the FLASH interrupt vector table must be re-burned, and the FLASH cannot be re-burned locally, so all codes and interrupt vector tables must be re-burned. In the aerospace environment, electromagnetic interference is serious and often in extreme high and low temperature conditions. In such a harsh environment, re-burning the FLASH is likely to fail, causing the entire FLASH to fail to work, so this method has certain risks, especially in harsh industrial and aerospace environments.

[0008] In summary, it can be seen that designing a program self-backup method that does not require re-burning has very important practical significance. Summary of the invention

[0009] In order to solve the technical problems existing in the prior art, the object of the present invention is to provide a program self-backup method based on on-chip FLASH and without the need for re-burning.

[0010] In order to achieve the above purpose and the above technical effect, the technical solution adopted by the present invention is:

[0011] The program self-backup method based on the on-chip FLASH and without re-burning includes the following steps:

[0012] Make two copies of the original program in the MCU FLASH, one as the main program and the other as the backup program;

[0013] Modify and compile the error monitoring, main function and interrupt function switching of the main program, and generate the HEX file corresponding to the main program;

[0014] Modify and compile the interrupt function of the backup program to generate the HEX file corresponding to the backup program;

[0015] The code length and checksum information of the main program are stored at a fixed address;

[0016] The contents in the HEX file of the main program, the contents in the HEX file of the backup program, and the code length and checksum information of the main program are combined to generate a final HEX file, and then the final HEX file is used as the final programming file to be burned into the FLASH in the microcontroller and written into the FLASH;

[0017] After power-on, the code is read from the starting address of FLASH and verified to obtain the current code checksum, and the current code checksum is compared and analyzed with the checksum of the main program read from the fixed address to obtain the verification result. The main and backup programs are switched based on the verification result, and the main program or backup program is executed without re-burning FLASH.

[0018] Furthermore, at the main function entry of the main program, the current code checksum is compared and analyzed with the checksum of the main program read from the fixed address. If the two checksums are equal, it means that the main program code has not been modified, and the subsequent code of the main program continues to be executed. If they are not equal, it means that the main program code has been accidentally modified. The backup program start address is jumped to by the jump pointer and jump instruction of the backup program defined in the main program, and the main-backup switching flag variable is marked as the switch-in backup program flag, and the backup program is jumped to realize the switching of the main functions of the main and backup programs; when an interrupt occurs, the main-backup switching flag is first judged in the interrupt function of the main program. If it is the switch-in backup program flag, the address of the backup program changed from the interrupt function to the ordinary function is jumped to by the jump pointer and jump instruction, otherwise the interrupt function of the main program continues to be executed, thereby realizing the switching of the interrupt function in the main and backup programs; the main-backup switching of all codes is realized through the jump control of the above-mentioned main function and interrupt function.

[0019] Furthermore, the modifications to the master program specifically include:

[0020] a. Define the master-slave switching flag variable in the main program, the master-slave switching flag variable is common to the main program and the backup program, and the storage address of the master-slave switching flag variable is assigned to the RAM area that is not modified by the initialization program;

[0021] b. Define the jump pointer of the backup program in the main program, and the jump address of the jump pointer is the starting address of the backup program;

[0022] c. Define the jump pointer of the interrupt program in the main program, and the jump address of the jump pointer corresponds to the compiled address of the function changed from the interrupt function to the normal function in the backup program;

[0023] d. Add CRC check code and backup program jump instruction at the main function entry of the main program. If the check fails, jump to the starting address of the backup program through the jump pointer and jump instruction. Otherwise, continue to execute the main program, thereby realizing the switching of the main function in the main and backup programs;

[0024] e. Add the master-slave switching flag judgment and interrupt program jump instruction at the interrupt function entry of the main program. If it is the backup program switching flag, jump to the address of the backup program changed from the interrupt function to the normal function through the jump pointer and jump instruction of the interrupt function. Otherwise, continue to execute the interrupt function of the main program, thereby realizing the switching of the interrupt function in the master-slave program.

[0025] Furthermore, compiling the main program specifically includes:

[0026] Set the main program code storage start address to the default FLASH start address, compile and link the main program, and generate the HEX file corresponding to the main program.

[0027] Furthermore, modifications to the backup program specifically include:

[0028] Change the interrupt function in the backup program to a normal function.

[0029] Furthermore, compiling the backup program specifically includes:

[0030] Set the backup program code storage start address to the address after the main program storage end address, and select half of the FLASH address as the backup area start address; compile and link the backup program, generate the HEX file corresponding to the backup program, and record the function compilation address that is changed from an interrupt function to a normal function.

[0031] Furthermore, after the backup program code storage end address, two fixed addresses are selected to store the code length and checksum information of the main program.

[0032] Compared with the prior art, the present invention has the following beneficial effects:

[0033] The invention discloses a program self-backup method based on on-chip FLASH and without the need for re-burning, comprising the following steps: making two copies of an original program in the FLASH of a single-chip computer, one of which is used as a main program and the other as a backup program; modifying and compiling the main program to generate a HEX file corresponding to the main program; modifying and compiling the backup program to generate a HEX file corresponding to the backup program; storing the code length and checksum information of the main program at a fixed address; merging the content in the HEX file of the main program, the content in the HEX file of the backup program, and the code length and checksum information of the main program to generate a final HEX file, and then using the final HEX file as the final burning file burned into the FLASH in the single-chip computer and writing it into the FLASH; after power-on, reading and verifying the code from the starting address of the FLASH to obtain the current code checksum, comparing and analyzing the current code checksum with the checksum of the main program read from the fixed address to obtain the check result, making a main-backup program switching judgment according to the check result, and executing the main program or the backup program without re-burning the FLASH. The program self-backup method based on on-chip FLASH and without the need for re-burning provided by the present invention is a program self-backup method in which there is no RAM program running area and the program is stored and run in the FLASH area. There is no need to re-burn FLASH, no additional boot program, no need to make changes to the single-chip computer hardware, and the system has high stability. At the same time, it solves the problems of high cost of using dual FLASH, the need to re-burn FLASH, and the risk of burning failure in the prior art, and is suitable for industrial promotion and use. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 is a flow chart of the present invention;

[0035] Figure 2 It is the interrupt response flow chart of the present invention. DETAILED DESCRIPTION

[0036] The present invention is described in detail below so that the advantages and features of the present invention can be more easily understood by those skilled in the art, thereby making a clearer and more definite definition of the protection scope of the present invention.

[0037] A brief summary of one or more aspects is given below to provide a basic understanding of these aspects. This summary is not an exhaustive overview of all conceived aspects, and is neither intended to identify the key or critical elements of all aspects nor to define the scope of any or all aspects. Its only purpose is to give some concepts of one or more aspects in a simplified form as a prelude to a more detailed description that will be given later.

[0038] like Figure 1-2As shown, the program self-backup method based on the on-chip FLASH and without re-burning includes the following steps:

[0039] Step 1): Make two copies of the original program in the MCU FLASH, one of which is used as the main program and the other as the backup program;

[0040] Step 2): Modify the main program, including: adding error monitoring, main function jump control, interrupt function jump control;

[0041] Step 3): Set the code storage start address of the main program to the default FLASH start address, that is, address 0; compile and link the main program to generate a HEX file corresponding to the main program;

[0042] Step 4): Modify the backup program

[0043] Change the interrupt function in the backup program to a normal function;

[0044] Step 5): Set the backup program code storage start address to the address after the main program storage end address, usually select half of the FLASH address as the backup area start address; compile and link the backup program to generate the HEX file corresponding to the backup program;

[0045] Step 6): After the backup program code storage end address, two fixed addresses are selected to store the main program code length (2 bytes) and checksum (2 bytes) information. This information usually reserves 4 bytes of storage space;

[0046] Step 7): Merge the contents in the main program HEX file, the contents in the backup program HEX file, and the contents in step 6) to generate a final HEX file;

[0047] Step 8): Write the HEX file generated in step 7) into FLASH as the final burning file;

[0048] After power-on restart, execution starts from the starting address 0 of FLASH and will be Figure 1 The process shown in the figure is to perform fault detection and master-slave switching without re-burning FLASH. The interrupt function will also be executed according to Figure 2The process shown is switched together: read the code from the starting address of FLASH and verify it to obtain the current code checksum, compare and analyze the current code checksum with the checksum of the main program read from the fixed address, obtain the verification result, and judge the main and backup programs according to the verification result, and execute the main program or the backup program. The verification results include two types: one is that the two checksums are equal, which means that the main program code has not been modified, and the subsequent code of the main program continues to be executed; the other is that the two checksums are not equal, which means that the main program code has been accidentally modified, and the backup program start address is jumped to by the jump pointer and jump instruction of the backup program defined in the main program, and the main and backup switching flag variable is marked as the switch-in backup program flag, and jumps to the backup program to realize the switching of the main function in the main and backup programs. When an interrupt occurs, the main and backup switching flag is first judged in the interrupt function of the main program. If it is the switch-in backup program flag, the address of the backup program that is changed from the interrupt function to the ordinary function is jumped to by the jump pointer and jump instruction, otherwise the interrupt function of the main program continues to be executed. Other interrupt functions are processed in the same way to realize the switching of interrupt functions in the main and standby programs. Through the jump control of the main function and interrupt function, the switching of all codes in the main and standby programs is realized.

[0049] In step 2), the modification of the main program includes the following steps:

[0050] a. Define the master-slave switching flag variable in the master program. Since the master-slave switching flag variable is shared by the master program and the backup program, the storage address of the master-slave switching flag variable should be assigned to a RAM area that is not modified by the initialization program;

[0051] b. Define the jump pointer of the backup program in the main program, and the jump address of the jump pointer is the starting address of the backup program;

[0052] c. Define the jump pointer of the interrupt program in the main program, and the jump addresses of the jump pointer correspond to the function addresses of the ordinary functions changed from the interrupt function in the backup program;

[0053] d. Add CRC checksum code and backup program jump instruction at the main function entry of the main program. Specifically: read the code length of the main program from the fixed address described in step 6), perform CRC check on this length, compare the calculated checksum with the checksum stored at the fixed address described in step 6), if the two values ​​are equal, it means that the main program code has not been modified, and the subsequent code of the main program can continue to be executed; if they are not equal, it means that the main program code has been accidentally modified, then jump to the backup program starting address through the jump pointer and instruction, and mark the main-backup switching flag variable as switching into the backup program;

[0054] e. Add the master-slave switching flag judgment and interrupt program jump instruction to the interrupt function entry of the main program. If it is the backup program switch flag, jump to the address of the backup program changed from the interrupt function to the normal function through the jump pointer and jump instruction of the interrupt function, otherwise continue to execute the interrupt function of the main program.

[0055] In step 5), set the code storage start address of the backup program to the backup area start address of FLASH, modify the code segment start address in the backup program startup file; compile and link the backup program, generate the HEX file corresponding to the backup program, and record the code storage address corresponding to the change from interrupt function to normal function.

[0056] Example 1

[0057] Take the microcontroller C8051F553 and the FLASH storage address range of 0x0000H to 0x8000H as an example.

[0058] The program self-backup method based on the on-chip FLASH and without re-burning includes the following steps:

[0059] Step 1): Make two copies of the original program, one as the main program and the other as the backup program;

[0060] Step 2): Modify the main program, including: adding error monitoring, main function jump control, interrupt function jump control;

[0061] Step 3): Set the code storage start address of the main program to the default FLASH start address, that is, 0x0000H address; compile and link the main program to generate a HEX file corresponding to the main program, which is named main.hex in this embodiment;

[0062] Step 4): Modify the backup program

[0063] Change the interrupt function in the backup program to a normal function;

[0064] Step 5): Set the backup program code storage start address to the address after the main program storage end address. Usually, half of the FLASH address is selected as the backup area start address. In this embodiment, 0x4000H is selected as the backup program start address, that is, the backup area start address. Compile and link the backup program to generate the HEX file corresponding to the backup program, which is named backup.hex in this embodiment.

[0065] Step 6): After the backup program code storage end address, two fixed addresses are selected to store the main program code length (2 bytes) and checksum (2 bytes) information. This information usually reserves 4 bytes of storage space. In this embodiment, 0x7800H is set to store the main program code length, and 0x7802H is set to store the main program CRC checksum;

[0066] Step 7): Merge the contents in the main program HEX file, the backup program HEX file, and the contents in step 6) to generate a final HEX file, which is named combine.hex in this embodiment;

[0067] Step 8): Write the HEX file generated in step 7) into FLASH as the final burning file;

[0068] After power-on restart, the program starts from the starting address 0x0000H of FLASH and executes according to Figure 1 The process shown in the figure is to perform fault detection and switch between the main and backup programs. There is no need to re-burn FLASH, and the interrupt function will also be executed according to the Figure 2 The processes shown are switched together.

[0069] In step 2), the steps of modifying the main program include:

[0070] a. Define the master-slave switch flag variable in the master program, such as IsBackup. If running in the master program, it is 0, and if switching to the backup program, it is 1. The variable storage address is assigned to a RAM area that is not modified by the initialization program, such as the XDATA area. This area will not be modified when switching to the backup program;

[0071] b. Define the jump pointer of the backup program in the main program, such as *pFlashBackup. The jump address of the jump pointer is the starting address of the backup program 0x4000H;

[0072] c. Define the jump pointer of the interrupt program in the main program, such as *pFlashCanInt. The jump address of the jump pointer corresponds to the function address of the backup program that is changed from an interrupt function to a normal function, such as the address 0x62C2H of the CAN0_INT() function;

[0073] d. Add CRC checksum code and backup program jump instruction at the main function entry of the main program. Specifically: read the code length of the main program from the fixed address 0x7800H described in step 6), perform CRC check on this length, read the code from the starting address 0x0000H of FLASH and perform check, calculate the current code checksum, and compare it with the checksum stored at the fixed address 0x7802H described in step 6). If the two values ​​are equal, it means that the main program code has not been modified and the subsequent code of the main program can continue to be executed. If they are not equal, it means that the main program code has been accidentally modified, then jump to the backup program starting address 0x4000H through the jump pointer and instruction, and mark the main-backup switching flag variable IsBackup as switching into the backup program;

[0074] e. Add the master-slave switch judgment and interrupt program jump instructions to the interrupt function entry of the main program. Specifically: when an interrupt occurs, first judge the master-slave switch flag in the interrupt function of the main program. If it is the switch to the backup program flag, jump to the address of the CAN0_INT() function in the backup program through the jump pointer and jump instruction, otherwise continue to execute the interrupt function of the main program. Other interrupt functions are processed in the same way.

[0075] In step 4), the modification of the backup program includes the following steps:

[0076] Change the interrupt function in the backup program to a normal function. The specific modification method is introduced using the CAN bus interrupt function as an example. The same processing method is used for other interrupt functions:

[0077] For example: void CAN0_INT()interrupt 16{……}

[0078] Modified to: void CAN0_INT(){……}.

[0079] Step 5) includes the following steps:

[0080] a. Set the backup program code storage starting address to the FLASH backup area starting address, which is 0x4000H in this embodiment. Modify the starting address of the code segment in the backup program startup file to 0x4000H. The modification method is as follows:

[0081] In the STARTUP.A51 file, change CSEG AT 0 to CSEG AT 4000H;

[0082] b. Compile and link the backup program to generate a HEX file corresponding to the backup program. In this embodiment, the HEX file generated by the backup program is named backup.hex, and records the code storage address (i.e., the function compilation address) corresponding to the change of the link file from an interrupt function to a normal function, i.e., the address of the above-mentioned void CAN0_INT(){...} function.

[0083] Parts or structures not specifically described in the present invention may adopt existing technologies or existing products and will not be described in detail here.

[0084] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A program self-backup method based on on-chip FLASH without re-burning, characterized in that: The following steps are involved: Make two copies of the original program in the MCU FLASH, one as the main program and the other as the backup program; Modify and compile the main program to generate the HEX file corresponding to the main program; Modify and compile the backup program to generate the HEX file corresponding to the backup program; The modifications to the backup program include: Change the interrupt function in the backup program to a normal function; The code length and checksum information of the main program are stored at a fixed address; The contents in the HEX file of the main program, the contents in the HEX file of the backup program, and the code length and checksum information of the main program are combined to generate a final HEX file, and then the final HEX file is used as the final programming file to be burned into the FLASH in the microcontroller and written into the FLASH; After power-on, the code is read from the starting address of FLASH and verified to obtain the current code checksum. The current code checksum is compared and analyzed with the checksum of the main program read from the fixed address to obtain the verification result. The main and backup programs are switched based on the verification result, and the main program or backup program is executed without re-burning FLASH. Specifically, at the main function entry of the main program, the current code checksum is compared and analyzed with the checksum of the main program read from the fixed address. If the two checksums are equal, it means that the main program code has not been modified, and the subsequent code of the main program continues to be executed. If they are not equal, it means that the main program code has been accidentally modified. The backup program start address is jumped to by the jump pointer and jump instruction of the backup program defined in the main program, and the main-backup switching flag variable is marked as the backup program entry flag, and the backup program is jumped to realize the switching of the main functions of the main and backup programs; when an interrupt occurs, the main-backup switching flag is first judged in the interrupt function of the main program. If it is the backup program entry flag, the address of the backup program changed from the interrupt function to the ordinary function is jumped to by the jump pointer and jump instruction, otherwise the interrupt function of the main program continues to be executed, thereby realizing the switching of the interrupt function in the main and backup programs; the main-backup switching of all codes is realized through the jump control of the above-mentioned main function and interrupt function.

2. The program self-backup method based on on-chip FLASH and without re-writing according to claim 1 is characterized in that: The modifications to the master program include: a. define a master-slave switch flag variable in the master program, the master-slave switch flag variable is common to the master program and the backup program, and assign the storage address of the master-slave switch flag variable to a RAM area that is not modified by the initialization program; b. Define the jump pointer of the backup program in the main program, and the jump address of the jump pointer is the starting address of the backup program; c. Define the jump pointer of the interrupt program in the main program, and the jump address of the jump pointer corresponds to the compiled address of the function changed from the interrupt function to the normal function in the backup program; d. Add CRC check code and backup program jump instruction at the main function entry of the main program. If the check fails, jump to the starting address of the backup program through the jump pointer and jump instruction. Otherwise, continue to execute the main program, thereby realizing the switching of the main function in the main and backup programs; e. Add the master-slave switching flag judgment and interrupt program jump instruction at the interrupt function entry of the main program. If it is the backup program switching flag, jump to the address of the backup program changed from the interrupt function to the normal function through the jump pointer and jump instruction of the interrupt function. Otherwise, continue to execute the interrupt function of the main program, thereby realizing the switching of the interrupt function in the master-slave program.

3. The program self-backup method based on on-chip FLASH and without re-writing according to claim 1 is characterized in that: Compiling the main program specifically includes: Set the main program code storage start address to the default FLASH start address, compile and link the main program, and generate the HEX file corresponding to the main program.

4. The program self-backup method based on on-chip FLASH and without re-writing according to claim 1 is characterized in that: Compiling the backup program specifically includes: Set the backup program code storage start address to the address after the main program storage end address, and select half of the FLASH address as the backup area start address; compile and link the backup program, generate the HEX file corresponding to the backup program, and record the function compilation address that is changed from an interrupt function to a normal function.

5. The program self-backup method based on on-chip FLASH and without re-writing according to claim 1 is characterized in that: After the backup program code storage end address, two fixed addresses are selected to store the code length and checksum information of the main program.

Citation Information

Patent Citations

  • Backup method and apparatus for bootstrap

    CN101256527A

  • Method for updating and repairing basic input and output system

    CN101667128A

  • Method and device for automatically correcting error of program of flash memory and repairing program of flash memory

    CN102402470A

  • Method for automatically starting central processing unit (CPU) system by utilizing double flashes

    CN102736941A

  • Double-booting method based on single FLASH storage chip

    CN103116511A