KR102517251B1 - Embedded system, patch server, and method for updating the embedded system - Google Patents

Abstract

An update method of an embedded system according to an embodiment of the present invention includes transmitting a channel join message requesting joining of a multicast channel for transmitting a patch stream to a patch server. step; receiving a patch stream through the multicast channel and storing a patch file required for system update; and performing a runtime patch without rebooting by a patch module configured from the stored patch file.

Description

Embedded system, patch server, and update method of the embedded system

The present invention relates to an embedded system, a patch server, and a method for updating the embedded system, and more particularly, to an embedded system capable of performing a system update without stopping the system, a patch server, and a method for updating the embedded system. will be.

An embedded system is a computer system that performs only special functions by embedding software for operating the system in hardware. Unlike a personal computer (PC), the embedded system has specific requirements and performs only predefined tasks. The personal computer has an operating system (OS) embedded in a mass storage device such as a hard disk. On the other hand, in the embedded system, the operating system and application programs are stored in the form of images in non-volatile memory (e.g., ROM, flash memory, etc.) Disk) is created, and then the operating system and application programs are configured and run on the RAM disk.

The embedded system can be updated as needed, and the system operator connects to a server providing updated software for updating and checks the software version according to the system type of the specific embedded system to determine the existence and application of the updated software. System update may proceed by determining whether or not.

In addition, in the embedded system, the operating software is managed as a single image stored in non-volatile memory. When updating the system, if a specific module or application within the operating software needs to be changed, a new single image must be created and the embedded system must be rebooted to restore the changed software. be able to operate. In this case, the operation of the embedded system is inevitably stopped, resulting in quality degradation in a system that must be operated 24 hours a day.

An object of the present invention is to provide an embedded system capable of performing an update without system interruption, a patch server, and a method for updating the embedded system.

The technical problems to be achieved in the present invention are not limited to the technical problems mentioned above, and other technical problems not mentioned will be clearly understood by those skilled in the art from the description below. You will be able to.

In order to solve the above technical problem, a method for updating an embedded system according to an embodiment of the present invention includes a channel join message requesting subscription of a multicast channel for transmitting a patch stream ( channel join message) to the patch server; receiving a patch stream through the multicast channel and storing a patch file required for system update; and performing a runtime patch without rebooting by a patch module configured from the stored patch file.

Depending on the embodiment, the multicast channel may be established for each system and patch version.

According to an embodiment, the channel join message includes system identification information and patch version information of the embedded system, and the multicast channel is the latest version after the current version according to the patch version information for the system according to the system identification information. It may be a channel corresponding to the patch of the version.

According to an embodiment, the patch stream may include patch target system information on the type of system on which the patch can be installed and whether the system version is compatible, version information of the patch file, the total size of the patch file, and the index of the patch file. can include

According to an embodiment, the storing of the patch file may include determining whether the patch file can be installed in the embedded system from the patch target system information; If the patch file can be installed in the embedded system, comparing version information of the patch file with patch version information of a patch currently installed in the embedded system, determining whether a patch needs to be installed; and if the version of the received patch file is higher than the patch version of the patch currently installed in the embedded system, determining whether the patch file has already been received based on the index of the patch file.

According to embodiments, the performing of the runtime patch may include modifying a code region or data region of an operating system or a specific program being executed by the patch module in a pin-point manner.

According to an embodiment, the method may further include receiving a patch stream from an L3 switch and storing the patch file when the patch server is inoperable.

An update method of an embedded system according to another embodiment of the present invention includes system identification information and patch version information of the embedded system, and requests subscription to a multicast channel for transmitting a patch stream. Receiving a channel join message; allocating to the embedded system a multicast channel corresponding to a patch of a latest version after the current version according to the patch version information for the system according to the system identification information; and transmitting a patch stream to the embedded system through the multicast channel so as to perform a runtime patch without rebooting.

According to the update method of an embedded system according to an embodiment of the present invention configured as described above, system update is possible without the need to replace or reboot the entire software by patching in a pinpoint method.

In addition, the time required to update the entire system can be shortened, the risk of modifying the system image can be reduced, and even if the patch application fails, the patch for that area is intensively retried without updating the entire system again. can do.

In addition, the patch server operates a multicast channel and transmits a patch stream so that multiple embedded systems can be patched at the same time.

Lastly, even when the patch server fails, patching can be completed normally by providing patch streams from other switches.

The effects obtainable in the present invention are not limited to the effects mentioned above, and other effects not mentioned can be clearly understood by those skilled in the art from the description below. will be.

1 is a flowchart illustrating an update method of an embedded system according to an embodiment of the present invention.
FIG. 2 is a diagram showing an example of the patch stream shown in FIG. 1 .
FIG. 3 is a diagram for explaining a method of applying a patch in FIG. 1 .
FIG. 4 is a diagram of an example in which a patch is applied using the method of FIG. 3 .
5 is a flowchart illustrating an update method of an embedded system according to another embodiment of the present invention.

Hereinafter, at least one embodiment related to the present invention will be described in more detail with reference to the drawings. The suffixes "module" and "unit" for components used in the following description are given or used together in consideration of ease of writing the specification, and do not have meanings or roles that are distinct from each other by themselves.

1 is a flowchart illustrating an update method of an embedded system according to an embodiment of the present invention. FIG. 2 is a diagram showing an example of the patch stream shown in FIG. 1 . FIG. 3 is a diagram for explaining a method of applying a patch in FIG. 1 . FIG. 4 is a diagram of an example in which a patch is applied using the method of FIG. 3 .

Referring to FIG. 1 , the first embedded system 10 and the second embedded system 20 are computer systems that perform only special functions by embedding software for operating the system in hardware. The first embedded system 10 and the second embedded system 20 may be connected to the switch A 30 through a network, respectively.

Switch A 30 and switch B 40 may perform routing between the embedded systems 10 and 20 and the patch server 50 . Switch A 30 and switch B 40 may be implemented as L3 switches, but the scope of the present invention is not limited thereto.

The patch server 50 is a server that provides a patch stream for updating embedded systems (eg, 10 and 20), and establishes a multi-cast channel for each system and patch version. and transmit the corresponding patch stream in a multicast method.

The first embedded system 10 may transmit a channel join message to the patch server 50 through the switch A 30 and the switch B 40 (S110). The channel join message is a message requesting multicast channel subscription, and may include system identification information of the first embedded system 10 and patch version information of a patch currently installed in the first embedded system 10 . In addition, the channel join message may include an index of the received patch stream if there is a patch stream received in the first embedded system 10 .

The first embedded system 10 may generate a channel join message at intervals preset by the system manager (eg, every 3 days), but the scope of the present invention is not limited thereto.

The patch server 50 may generate a query message based on the received channel join message and transmit it to the first embedded system 10 through the switch A 30 and the switch B 40 (S120). . The query message is a response message to a request for multicast channel subscription.

The patch server 50 checks the system identification information and the patch version information of the first embedded system 10 to determine whether a patch of the latest version after the current version according to the patch version information exists for the system according to the system identification information. can When a patch of the latest version requiring update exists, the patch server 50 generates a query message including information indicating that it has a patch file along with its own address information and transmits it to the first embedded system 10 .

In addition, the patch server 50 may permit the first embedded system 10 to access a multicast channel corresponding to the system identification information and the patch of the latest version. To this end, the query message includes the multicast Information for access to the channel may be included. Accordingly, the first embedded system 10 may receive the query message and access the multicast channel.

The second embedded system 20 may transmit the channel join message to the patch server 50 through the switch A 30 and the switch B 40 (S130).

The patch server 50 may generate a query message based on the received channel join message and transmit it to the second embedded system 20 through the switch A 30 and the switch B 40 (S140).

There is no substantial difference between the operations of steps S130 and S140 except for the change from the first embedded system 10 to the second embedded system 20, so description thereof will be omitted. However, it will be described on the assumption that the first embedded system 10 and the second embedded system 20 subscribe to the same multicast channel and receive a patch stream. The explanation will focus on application.

The first embedded system 10 and the second embedded system 20 transmit the first patch stream transmitted from the patch server 50 to the multicast channel to which the first embedded system 10 and the second embedded system 20 subscribe. It can be received (S150).

In FIG. 2 , the patch stream 100 may include a header and a body.

The header includes a magic code indicating the start of the patch stream 100, patch target system information on the type of system in which the patch can be installed and whether the system version is compatible, patch file version information, and patch upload. ) date, total size of the patch file, index of the patch file, error checking information (eg, CRC), and the like.

The body may include at least a part of a patch file for updating the embedded systems 10 and 20 . The body may include a plurality of binary files (binary#1 to binary#n (n is an integer of 2 or more)) of a single functional unit. The patch file may configure the patch module 400 of FIG. 4 .

The first embedded system 10 can determine whether a patch file can be installed in the first embedded system 10 from patch target system information of a patch stream transmitted by the patch server 50 through a multicast channel.

If it is a patch file that cannot be installed in the first embedded system 10, the first embedded system 10 may ignore the corresponding patch stream.

However, in the case of a patch file that can be installed in the first embedded system 10, the first embedded system 10 compares the version information of the patch file with the patch version information of the patch currently installed in the first embedded system 10. , it can determine whether patch installation is necessary.

If the version of the received patch file is the same as or lower than the patch version of the patch currently installed in the first embedded system 10, the first embedded system 10 may ignore the corresponding patch stream.

However, if the version of the received patch file is higher than the patch version of the patch currently installed in the first embedded system 10, the first embedded system 10 determines whether the patch file has already been received based on the index of the patch file. can judge

If the index of the patch file already stored is the same as the index of the received patch file, the first embedded system 10 may ignore the corresponding patch stream.

However, if the index of the patch file already stored and the index of the received patch file are not the same, the first embedded system 10 transfers the corresponding patch stream to a non-volatile memory (e.g., the flash memory 200 of FIG. 4). can be stored in

In the embodiment of FIG. 1 , it is assumed that the patch server 50 divides the patch file into four times and transmits the patch file. can detect the start of

In addition, the first embedded system 10 may determine whether all patch files have been received based on the size of all files and the indexes of patch files received so far. For example, if the size of the entire file is 100 MByte and the capacity indicated by the index of patch files received so far is 50 MByte, the first embedded system 10 may determine that all patch files have not been received.

The first embedded system 10 and the second embedded system 20 are transmitted by the patch server 50 through the multicast channel to which the first embedded system 10 and the second embedded system 20 subscribe. And a fourth patch stream can be received (S152, S154, S156).

The first embedded system 10 may store patch files included in the patch stream 100 through the same operation as in step S150 in steps S152, S154, and S156.

In step S156, the first embedded system 10 determines whether all patch files have been received based on the size of the entire file and the index of the patch files received so far, and detects the end of the patch files since four patch files have been received. can do.

If it is determined that all patch files have been received, the first embedded system 10 can apply the corresponding patch files (S160).

Referring to FIG. 3 , the first embedded system 10 stores patch files in the flash memory 200, which is a non-volatile memory, and the received patch files configure the patch module 400. The patch file may operate as the patch module 400 through a separate execution process, but the scope of the present invention is not limited thereto.

The patch module 400 may obtain an address (logical address or physical address) of an area requiring a patch through interworking with the first embedded system 10 to perform a patch operation.

The flash memory 200 may include a network operating system 300, which is an image of the first embedded system 10, in addition to a patch file. That is, the flash memory 200 is a component of the upper system in which the first embedded system 10 operates, and the first embedded system 10 is stored in the flash memory 200 as the NOS 300 in the form of an image. , After creating a RAM disk (RAM Disk, 600) at the same time as booting, it is a system in which an operating system and application programs are configured and driven on the RAM disk (600).

In addition, the first embedded system 10 may access hardware (eg, 200) of the upper system through a kernel 500. An operation of storing the patch file in the flash memory 200 by the first embedded system 10 described above may also be performed through the kernel 500 .

The patch module 400 pin-points the code area or data area of a specific program of the first embedded system 10 by accessing the hardware (eg, 200) of the upper system through the kernel 500. way can be modified.

In addition, the patch module 400 may modify a code area or data area of an operating system or a specific program running on the RAM disk 600 in a pinpoint manner.

As such, when the patch module 400 immediately corrects and applies the corresponding code without rebooting in a pinpoint manner while the kernel 500 and the RAM disk 600 are normally operating, it is called a runtime patch.

For runtime patching, the patch module 400 may monitor whether a region subject to code modification is being accessed, and as a result of the monitoring, the code modification is performed only when the corresponding region is not accessed and there is no possibility of access during the patching process. can

In FIG. 4 , when the patch module 400 is executed, the patch module 400 acquires memory addresses (0x100002fc to 0x10000318) for areas requiring patches from the first embedded system 10, and obtains memory addresses (0x100002fc to 0x10000318). ), it is possible to proceed with runtime patching by modifying in a pinpoint manner. The pinpoint method refers to a method of replacing a code in a corresponding area by performing an erase and program operation only in an area requiring a patch.

For example, the code stored at the memory address (0x10000310) by the runtime patch may be replaced with “11 0a 00 a0” from “38 0a 00 01”.

When the code modification of the corresponding area is completed, the first embedded system 10 may perform a build operation to change the modified area into an executable file. The first embedded system 10 may execute the built file to verify normal operation, and if normal operation is confirmed as a result of the verification, step S160 may be completed.

However, if it does not operate normally as a result of the verification, the first embedded system 10 may request the corresponding patch file from the patch server 50 again, and re-receive the patch stream and apply the patch to the corresponding area again. there is.

In the existing update method of the embedded system, the NOS 300 in the form of an image was modified or replaced and rebooting was essential, so the provision of the service had to be stopped.

However, according to the update method of an embedded system according to an embodiment of the present invention, system update is possible without the need to replace or reboot the entire software by patching in a pinpoint method.

In addition, the time required to update the entire system can be reduced, and the risk of modifying the NOS 300 can be reduced.

In addition, even if the patch application fails, the patch for the corresponding area can be intensively retried without updating the entire system again.

After the application of the patch is completed, the first embedded system 10 may transmit a channel leave message through switch A 30 and switch B 40 (S170). The channel leave message is a message requesting cancellation of a multicast channel for receiving a patch stream.

Upon receiving the channel live message, the patch server 50 may cancel the multicast channel for the first embedded system 10 and transmit the patch stream to the second embedded system 20 (S180).

According to the update method of an embedded system according to an embodiment of the present invention, a patch server operates a multicast channel and transmits a patch stream, so that multiple embedded systems can be patched at the same time.

In addition, patching is performed automatically without intervention of the system manager, so that the convenience of the system manager can be increased along with rapid system upgrade.

5 is a flowchart illustrating an update method of an embedded system according to another embodiment of the present invention.

Referring to FIGS. 1 and 5 , it is assumed that the patch server 50 is unable to transmit data such as query messages and patch streams for some reason (eg, server self-check or error, communication failure, etc.).

The first embedded system 10 may transmit the channel join message to the patch server 50 through the switch A 30 and the switch B 40 (S210). However, since the patch server 50 cannot transmit data, the switch B 40 does not receive the query message.

The switch B 40 may store some patch files for each system and patch version. That is, the switch (B) 40 may store the patch file under the control of the patch server 50 or, in the process of transferring the patch stream between the patch server 50 and the switch A 30, switch B 40 may use its own judgment. can also be saved by

The switch B 40 checks the system identification information and patch version information of the first embedded system 10 included in the channel join message, and for the system according to the system identification information, the latest version after the current version according to the patch version information. You can determine if you are saving the version of the patch.

In addition, the switch B 40 may determine whether the first embedded system 10 stores a necessary patch by checking the index of the patch stream received in the channel join message.

When the first embedded system 10 stores the required patch, the switch B 40 generates a query message including information indicating that it has the patch file along with its own address information, and sends it to the first embedded system 10. It is transmitted (S220).

The second embedded system 20 may transmit the channel join message to the patch server 50 through the switch A 30 and the switch B 40 (S230).

Similarly to step S220, switch B 40 generates a query message based on the received channel join message when the first embedded system 10 stores a necessary patch, and sends the second embedded system 10 through switch A 30. It can be transmitted to the embedded system 20 (S240).

The first embedded system 10 and the second embedded system 20 may receive the first patch stream from the switch B 40 (S250).

Since each of the subsequent steps (S252 to S280) is substantially the same as each of the steps S152 to S180 of FIG. 1, detailed descriptions will be omitted to avoid duplication of description.

According to the method for updating an embedded system according to another embodiment of the present invention, patching can be normally completed by providing a patch stream from another switch even when a patch server fails.

The method described above can be implemented as computer readable codes on a computer readable recording medium. Computer-readable recording media includes all types of recording media in which data that can be decoded by a computer system is stored. For example, there may be read only memory (ROM), random access memory (RAM), magnetic tape, magnetic disk, flash memory, optical data storage device, and the like. In addition, the computer-readable recording medium may be distributed to computer systems connected through a computer communication network, and stored and executed as readable codes in a distributed manner.

In addition, although the above has been described with reference to preferred embodiments of the present invention, those skilled in the art can make the present invention within the scope not departing from the spirit and scope of the present invention described in the claims below. It will be appreciated that various modifications and variations may be made.

Claims (18)

Transmitting a channel join message requesting subscription to a multicast channel for transmitting a patch stream to a patch server;
receiving a patch stream through the multicast channel and storing a patch file required for system update; and
Performing a runtime patch without rebooting by a patch module configured from the stored patch file;
The step of performing the runtime patch,
Modifying a code area or data area of an operating system or a specific program being executed by the patch module in a pin-point manner;
The step of correcting by the pin-point method,
It monitors whether the corresponding area, which is the target of code modification, is being accessed, and based on the result of the monitoring, the code for the corresponding area is modified only when the corresponding area is not accessed and there is no possibility of access during the patching process. Update method of an embedded system that proceeds. According to claim 1,
The method of updating an embedded system in which the multicast channel is established for each system and patch version. According to claim 2,
The channel join message includes system identification information and patch version information of the embedded system,
The multicast channel is a channel corresponding to a patch of a latest version after the current version according to the patch version information for the system according to the system identification information. According to claim 1,
The patch stream,
A method for updating an embedded system, including information about a system type to which a patch can be installed and system versions to be patched, information about the version of the patch file, total size of the patch file, and an index of the patch file. According to claim 4,
The step of storing the patch file is,
determining whether the patch file can be installed in the embedded system from the patch target system information;
If the patch file can be installed in the embedded system, comparing version information of the patch file with patch version information of a patch currently installed in the embedded system, determining whether a patch needs to be installed; and
If the version of the received patch file is higher than the patch version of the patch currently installed in the embedded system, determining whether the patch file has already been received based on the index of the patch file. method. delete According to claim 1,
The method of updating an embedded system further comprising receiving a patch stream from an L3 switch and storing the patch file when the patch server is inoperable. Receiving a channel join message including system identification information and patch version information of an embedded system and requesting subscription to a multicast channel for transmitting a patch stream;
allocating to the embedded system a multicast channel corresponding to a patch of a latest version after the current version according to the patch version information for the system according to the system identification information; and
Transmitting a patch stream to the embedded system through the multicast channel so that runtime patching can be performed without rebooting;
The embedded system,
Modifying a code area or data area of an operating system or a specific program being executed by a patch module in a pin-point manner to perform the transmitted runtime patch;
It monitors whether the corresponding area, which is the target of code modification, is being accessed, and based on the result of the monitoring, the code for the corresponding area is modified only when the corresponding area is not accessed and there is no possibility of access during the patching process. Update method of an embedded system that proceeds. According to claim 8,
The method of updating an embedded system in which the multicast channel is established for each system and patch version. According to claim 8,
The patch stream,
The update method of the embedded system, which includes information about the type of system on which the patch can be installed and system version compatibility, the version information of the patch file, the total size of the patch file, and the index of the patch file. A patch file necessary for system update is transmitted by sending a channel join message requesting subscription to a multicast channel for transmitting a patch stream to a patch server and receiving a patch stream through the multicast channel. is stored, and a runtime patch is performed without rebooting by a patch module configured from the stored patch file,
The runtime patch,
The code area or data area of an operating system or a specific program being executed by the patch module is modified in a pin-point manner,
monitoring whether the area subject to code modification is being accessed;
Based on the result of the monitoring, the embedded system proceeds with code modification for the corresponding region only when the corresponding region is not accessed and there is no possibility of access during patching. According to claim 11,
The multicast channel is established for each system and patch version. According to claim 12,
The channel join message includes system identification information and patch version information of the embedded system,
The multicast channel is a channel corresponding to a patch of a latest version after the current version according to the patch version information for the system according to the system identification information. According to claim 11,
The patch stream,
An embedded system including patch target system information on the type of system to which the patch can be installed and whether the system version is compatible, version information of the patch file, total size of the patch file, and an index of the patch file. According to claim 14,
Determining whether the patch file can be installed in the embedded system from the patch target system information;
In the case of a patch file that can be installed in the embedded system, version information of the patch file is compared with patch version information of a patch currently installed in the embedded system to determine whether the patch needs to be installed;
If the version of the received patch file is higher than the version of the patch currently installed in the embedded system, the embedded system determines whether the patch file has already been received based on the index of the patch file. According to claim 11,
An embedded system that modifies a code area or data area of an operating system or a specific program being executed by the patch module in a pin-point manner. According to claim 11,
An embedded system configured to receive a patch stream from an L3 switch and store the patch file when the patch server is inoperable. A system that includes system identification information and patch version information of an embedded system, receives a channel join message requesting subscription to a multicast channel for transmitting a patch stream, and receives a system according to the system identification information , a multicast channel corresponding to a patch of the latest version after the current version according to the patch version information is allocated to the embedded system, and a runtime patch is performed without rebooting. Sending a patch stream to the embedded system through,
The embedded system,
Modifying a code area or data area of an operating system or a specific program being executed by a patch module in a pin-point manner to perform the transmitted runtime patch;
It monitors whether the corresponding area, which is the target of code modification, is being accessed, and based on the result of the monitoring, the code for the corresponding area is modified only when the corresponding area is not accessed and there is no possibility of access during the patching process. The patch server that runs the .

Images