JP2005251045A - Information processor, housing position management method, and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a software update technology capable of contracting a time until completion of software update by suppressing occurrence of overhead in updating software on a nonvolatile memory of a terminal. <P>SOLUTION: A server device 1 includes an information input part for data size analysis 101 for inputting information for the data size analysis for analyzing a data size of stored data stored in the nonvolatile memory of the terminal, a temporary storage location information input part 102 for inputting temporary storage location information indicating a temporary storage location of each stored data, a storage location specifying information generating part 103 for generating storage location specifying information for specifying a storage location of each stored data by arranging a spare area between each of the stored data on the basis of information from the information input part for the data size analysis 101 and the temporary storage location information input part 102, and when the stored data is revised, the storage location specifying information generating part 103 generates the storage location specifying information for instructing to store a revised stored data by using the spare area. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to an information processing apparatus, and more particularly to an embedded information device such as a mobile phone / PDA.

  In recent years, built-in terminals such as mobile phones have become highly functional and have a CPU equivalent to that used in personal computers before, and are equipped with RAM and nonvolatile memory (flash ROM, SRAM with a backup power source, etc.) ) Capacity has also increased dramatically. Along with this, the software installed on the non-volatile memory has become complicated and large-scale. For this reason, maintaining the quality of such large-scale software has become a very big issue in the development of embedded terminals, and there are many examples where defects are found in embedded terminals that are actually on the market. . Since it has become practically difficult to ship software without any defects while complying with the delivery date of the product, means for updating the software at the user's hand after shipment has been realized.

  For example, Japanese Patent Application Laid-Open No. 2001-273147 discloses a technology that provides information distribution processing for generating difference information from upgraded information and its old information and transmitting the difference information to a terminal device. Specifically, the server computer has difference information generation means for generating difference information based on the old information and the upgraded information, having the upgraded information and the old information among the stored information. The terminal device has means for upgrading the information held by the terminal device to the latest information, and when the difference information is transmitted from the server computer, the terminal device is based on the difference information and the information held by the terminal device. Generate information that has been upgraded.

  In addition, with the increasing complexity and scale, the operating system of the installed software has come to be adopted with higher functionality than before, and the file system has been widely used. In the past, methods such as managing only data that was added or modified by the user after shipment using the file system were often used. Recently, however, application programs and some of the basic software are also included in the file system. An increasing number of examples have adopted architectures that are managed below.

A patch command in UNIX (registered trademark) is well known as a conventional example of a system software update method on a file system. The patch command has a function of updating a new version directory structure by inputting a difference file obtained by comparing the old and new directory structures on the file system and applying the contents of the difference file to the old version directory structure. At this time, updating of individual files is performed by using the function of the file system. Specifically, among a plurality of files existing in the directory structure of the old version, only the file in which the difference information is described in the difference file is opened (open), the contents of the file are read (read), and the difference After collating with the information, writing is performed so as to match the new version of the file, and the file is closed. The functions such as open / read / write / close are provided by the file system. By performing this series of operations, the physical device (hard disk drive, flash memory, etc.) on which the file system is mounted is used. The contents are rewritten indirectly. For example, when the file size is changed in the above series of operations, the file size information of the corresponding file stored in the file system is rewritten when the close is performed.
JP 2001-273147 A

  In general, on a physical device, a file system is divided into an area for storing management information of the entire file system, an area for managing a directory structure, an area for storing file data, and the like. When the file update operation as described above is performed, the file data area is rewritten, and at the same time, rewriting occurs in the management area of the file system and the management area of the directory structure. In addition, when the file size is changed, if there is not enough free space on the device, an operation (garbage collection) to secure the free space may occur by rearranging the file data. .

  For this reason, when a plurality of files are updated by an operation on the file system, such as a patch command, even on each intermediate stage in which only updates to some files are completed, Since it is guaranteed that the information is complete, overhead such as rewriting of management information occurs multiple times or garbage collection occurs multiple times, which affects the time until software update is completed.

  One of the objects of the present invention is to solve the above-described problems. The file system is applied when the file system is physically arranged on a nonvolatile memory (open / read / It is to realize a software update technique that can suppress the occurrence of overhead and shorten the time until software update is completed without using a function (such as write / close).

An information processing apparatus according to the present invention includes:
An information processing apparatus that targets a predetermined nonvolatile memory and manages a storage position of data in the nonvolatile memory,
A data size analysis information acquisition unit for acquiring data size analysis information used for analyzing a data size of a plurality of stored data stored in the nonvolatile memory;
A temporary storage position information acquisition unit that acquires temporary storage position information indicating a temporary storage position of each storage data in the nonvolatile memory;
Based on the data size analysis information acquired by the data size analysis information acquisition unit, the data size of each storage data is analyzed, and the data size of each storage data analyzed and the temporary storage position information acquisition unit acquired And a storage location designation information generating section for generating storage location designation information for designating a storage location of each storage data by arranging an empty area based on the storage location information. Information processing device.

  According to the present invention, since an empty area is arranged between a plurality of storage data and storage location designation information for designating the storage location of each storage data is generated, the occurrence of overhead during software update is suppressed, The time until update completion can be shortened.

Embodiment 1 FIG.
FIG. 1 is a diagram illustrating a configuration example of a server apparatus (information processing apparatus) 1 according to the present embodiment. The server device 1 according to the present embodiment has a main role of designating a storage location of a program, file data, etc. in a non-volatile memory of an embedded terminal (hereinafter referred to as a terminal or a terminal device) such as a mobile phone or a PDA. To do.

  FIG. 23 is a diagram illustrating the relationship between the server device 1 and the terminal device 2 according to the present embodiment. The server device 1 and the terminal device 2 are connected via the network 3, and various information is received from the server device 1. It is possible to transmit to the terminal device 2.

  In FIG. 1, a data size analysis information input unit 101 inputs data size analysis information for analyzing the data size of stored data (program, file data, etc.) stored in the terminal. The contents of the data size analysis information will be described later. The data size analysis information input unit 101 is an example of a data size analysis information acquisition unit. The temporary storage position information input unit 102 inputs temporary storage position information indicating the temporary storage position of each storage data in the nonvolatile memory of the terminal. The contents of the temporary storage position information will be described later. The temporary storage position information input unit 102 is an example of a temporary storage position information acquisition unit. Based on the data size analysis information input from the data size analysis information input unit 101 and the temporary storage position information input from the temporary storage position information input unit 102, the storage position designation information generation unit 103 Storage position designation information for designating the storage position of each storage data by generating a predetermined margin area (free area) between them is generated. The storage location designation information writing / reading unit 104 writes the storage location designation information generated by the storage location designation information generation unit 103 to the storage location designation information storage unit 106, and stores the storage location designation information in a predetermined case. The storage location designation information is read from the unit 106. The output unit 105 outputs the storage location designation information generated by the storage location designation information generation unit 103. The storage location designation information storage unit 106 stores the storage location designation information in accordance with the control of the storage location designation information writing / reading unit 104.

  Although not shown, the server device 1 can be realized by a computer having a CPU such as a microprocessor, a semiconductor memory, or a recording unit such as a magnetic disk. A program for realizing the function of each component included in the server device 1 is recorded in the recording means, and the operation of the server device 1 is controlled by the CPU reading these programs, thereby realizing the function of each component. Is also possible.

  FIG. 2 is a diagram showing a typical hardware configuration of the terminal. The terminal includes a CPU, and is connected to communication means for communicating with a network via a bus, input means such as a RAM, a nonvolatile memory (ROM), a key, and display means. The nonvolatile memory stores a boot loader, a kernel, a file system, and a rewriting program. The file system stores an execution program necessary for the function of the terminal.

  Next, the operation of the server device 1 will be described with reference to FIG. In the following description, a case will be described in which the storage location of each stored data is specified when storing the first version stored data in the nonvolatile memory of the terminal.

  First, in step S2201, the data size analysis information input unit 101 inputs data size analysis information (data size analysis information acquisition step). The data size analysis information is information for analyzing the data size of the stored data stored in the terminal. For example, other storage data (boot header, kernel, Rewriting program). FIG. 4 is a diagram illustrating an example of a directory tree input in step S2201. The contents of this directory tree are placed on the file system on the nonvolatile memory. In the figure, the name of the root directory is “/”, and directories dir1 and dir2 are arranged below it. A file file1 is arranged under the directory dir1, and a file 2 and a file 3 are arranged under the directory dir2.

  Next, in step S2202, temporary storage position information is input (temporary storage position information acquisition step). The temporary storage position information is information indicating a temporary storage position of each storage data used for generating the storage position designation information, and is, for example, nonvolatile memory configuration information shown in FIG. In the non-volatile memory configuration information, the start address of each data area mounted in the non-volatile memory is described. It is necessary to provide a sufficient margin so that the data area does not overlap. In the figure, the boot memory, kernel, file system header, directory information, file data, and rewrite program are arranged in the non-volatile memory. For example, the boot header is assigned a hexadecimal address at address 0x00000000, and the file data is assigned at address 0x00400000. It has come to be. The file system header, directory information, and file data are components of the file system and are arranged corresponding to the file system created from the contents of the directory tree.

  Next, in step S2203, storage location specification information is generated based on the information input in step S2201 and the information input in step S2202 (storage location specification information generation step). Although the generation of the storage location designation information will be described later, the storage location designation information generation unit 103 uses the information input in steps S2201 and S2202 as input, and arranges each information at the specified start address. Thus, non-volatile memory image information and non-volatile memory arrangement information, which are storage location designation information, are created. As for the file system, a directory tree is created from the directory tree, and a file system header, directory information, and file data are created. For example, in the case of a CramFS file system that is frequently used as a file system that is mounted on a ROM when a UNIX (registered trademark) operation system is used in an embedded system, it is possible to use a program called mkcramfs. In the case of CramFS, the file system header is called a super block, and the directory information is called an i-node area.

After the storage location designation information (nonvolatile memory image information and nonvolatile memory arrangement information) is generated in step S2203, the storage location designation information is output from the output unit 105 in step S2204, and the storage location designation is performed in step S2205. The information writing / reading unit 104 stores the storage location designation information in the storage location designation information storage unit 106.
FIG. 5 is a diagram illustrating an example of the non-volatile memory image information that is one of the storage location designation information generated by the storage location designation information generation unit 103. In the figure, the shaded area is a margin area (empty area), and no data is arranged, which is the initial state of the nonvolatile memory. For example, if the kernel start address is 0x00100000 and the file system header start address is 0x00200000, which is specified in the non-volatile memory configuration information, the kernel size is smaller than 0x00200000-0x00100000 = 0x100,000 bytes. A margin area is generated on the nonvolatile memory. The file system header and directory information created by the directory tree are also arranged at designated positions.

  File1, file2, and file3 are file data, which are the contents of each file given in the directory tree. In the figure, a margin area is provided between the file system header, directory information, and each file data. The storage location designation information generation unit 103 automatically inserts such a margin area. As a method for determining the size of the marginal area to be automatically inserted, the simplest method is to set the start address of each data at a constant interval.

  FIG. 6 is a diagram illustrating an example of non-volatile memory arrangement information that is one of storage position designation information generated by the storage position designation information generation unit 103. The non-volatile memory arrangement information indicates a result when the storage position designation information generation unit 103 arranges each data on the non-volatile memory based on the non-volatile memory configuration information (temporary storage position information). Basically, it includes non-volatile memory configuration information. For example, since the size of the boot header on the nonvolatile memory is 0x00080042 bytes, the margin area starts from the address. Note that the file system header / directory information / and the margin areas behind file1, file2, and file3 are automatically inserted by the storage location designation information generation unit 103. In the present embodiment, The data interval is increased in increments of 0x00100000 bytes (start addresses are 0x00200000, 0x00300000, 0x00400000, 0x00500000, and 0x00600000, respectively). Therefore, the start address is arranged as described above, and a margin area is automatically arranged.

  Here, the content of the storage location designation information generation procedure (step S2203) will be described with reference to FIG. In the order indicated by the non-volatile memory configuration information (temporary storage location information) input in step S2202 of FIG. 21, each storage data is set as an analysis target in order, and in step S2301, the data size of the target storage data is set in step S2201. It is calculated based on the directory tree, boot header, kernel, and rewrite program (data size analysis information) input in step 1. Next, in step S2302, the calculated data size is added to the start address of the stored data. Next, an address corresponding to the added value obtained in step S2302 is designated as the start address of the margin area (step S2303), and the designated start address of the margin area is written in the nonvolatile memory configuration information (temporary storage position information) ( Step S2304). Next, it is confirmed whether or not the start address of the margin area has been written for all the stored data (step S2305), and the writing of the start address of the margin area has been completed for all the storage data (Yes in step S2305). ) Terminates the process. Note that the nonvolatile memory configuration information (temporary storage location information) in which the start address of the margin area is written for all stored data becomes the nonvolatile memory location information, and this nonvolatile memory location information is converted into image information. It becomes nonvolatile memory image information. On the other hand, if there is storage data for which the start address of the margin area has not been written (No in step S2305), the next storage data is analyzed in the order of the nonvolatile memory configuration information (temporary storage position information). (Step S2306), the process returns to Step S2301.

  The created nonvolatile memory image information is written in the nonvolatile memory on the terminal in the first version, and the terminal is shipped. If a problem occurs after shipment of the terminal, the terminal software on the nonvolatile memory must be corrected. In the present embodiment, it is assumed that the defect correction target is a file existing in a directory tree on the file system. When a problem occurs in a file in the directory tree, first, a modified version of the file on the directory tree is manually created, and then the non-volatile memory image information and the non-volatile memory allocation for the new version of the directory tree. Create information.

  After the new version of the directory tree is created, the new version of the directory tree is input to the data size analysis information input unit 101 as data size analysis information, and the temporary storage location information input unit 102 is different from the first version. Instead of the non-volatile memory configuration information, the non-volatile memory arrangement information created in the first edition is input as temporary storage position information. Since the nonvolatile memory arrangement information created at the time of the first edition is stored in the storage location designation information storage unit 106, the nonvolatile memory created at the time of the first edition of the storage location designation information writing / reading unit 104 is stored. The memory location information may be read from the storage location designation information storage unit 106 and provided to the storage location designation information generation unit 103. In this case, the storage location designation information writing / reading unit 104 functions as a temporary storage location information acquisition unit. The non-volatile memory arrangement information includes file system header / directory information, which is a component of the file system, and start addresses of file data, file1, file2, and file3. When creating a new version of the non-volatile memory image, place it so that these start addresses do not change.

  That is, the storage location designation information generation unit 103 stores each of the new versions based on information from the data size analysis information input unit 101 and the temporary storage location information input unit 102 (or the storage location designation information writing / reading unit 104). Generate storage location specification information that specifies the storage location of each new version of storage data by placing an empty area between the data, but at that time, there is storage data whose data size changes in the new edition due to revision of the storage data The data size in the area (storage candidate area) that combines the area specified as the storage position of the storage data in the storage location specification information of the old version storage data and the free area provided for the storage data. The storage location designation information for designating the storage of the new version of the storage data having changed is generated. For storage data whose data size has not been changed, the storage position and free space of the storage data remain the old version.

  In the past, no extra area was inserted, and the second and subsequent versions created the non-volatile memory image from the non-volatile memory configuration information in the same way as in the first version, so the file size changed. In this case, the address shift occurred in the memory image between the first version and the second version. Such an address shift increases the area that needs to be written to the nonvolatile memory area when writing to the nonvolatile memory area for software update on the terminal. For example, in the present embodiment, when only the file 1 is corrected, the effect of the correction of the file 1 is absorbed in the margin area and does not affect the file 2 and the file 3. However, when the margin area is not provided, the start address of file 2 and file 3 is shifted backward as the size of file 1 increases. As a result, it is necessary to rewrite the memory areas representing file 2 and file 3 in order to cope with the address shift on the terminal, even though file 2 and file 3 are not corrected. On the other hand, in the present embodiment, the insertion of the margin area and the use of the non-volatile memory arrangement information ensure that each data is arranged from the same start address as the old version. Thus, the amount of nonvolatile memory written at the time of software update can be reduced. In the present embodiment, file 2 and file 3 need not be overwritten on the terminal.

  FIG. 7 is an example of the non-volatile memory arrangement information output from the storage location designation information generation unit 103 when the second version is created. FIG. 7 shows an example of non-volatile memory arrangement information when only the file 1 is changed and the size of the file 1 is changed. In this case, since the size of file1 has increased, the size of the margin area behind file1 has decreased. On the other hand, for the other stored data, the start address of each stored data and the start address of the margin area have not changed.

  Thus, in the present embodiment, a data size analysis information acquisition unit that acquires data size analysis information used for analyzing the data size of a plurality of stored data stored in the nonvolatile memory, and a nonvolatile Temporary storage position information acquisition unit for acquiring temporary storage position information indicating the temporary storage position of each storage data in the memory, and data of each storage data based on the data size analysis information acquired by the data size analysis information acquisition unit Analyzing the size and storing each storage data by placing a free area between multiple storage data based on the data size of each storage data analyzed and the temporary storage location information acquired by the temporary storage location information acquisition unit An information processing apparatus having a storage position specification information generation unit that generates storage position specification information for specifying a position has been described.

  Further, in the present embodiment, a case has been described in which storage location designation information that designates an empty area to be placed between each stored data is generated.

  It is also possible to classify a plurality of storage data into a plurality of groups and generate storage location designation information for designating that an empty area is arranged between each group.

  In addition, the data size analysis information acquisition unit acquires data size analysis information used for analyzing the data size of a plurality of new versions of storage data when the storage data is revised, and provisional storage position information The acquisition unit acquires the storage location specification information generated by the storage location specification information generation unit for the old version of the storage data as temporary storage location information, and the storage location specification information generation unit is acquired by the data size analysis information acquisition unit. Analyzing the data size of each new version of the storage data based on the data size analysis information, and analyzing the data size of each storage data of the new version analyzed and the storage location designation information of the old version of the storage data acquired by the temporary storage location information acquisition unit Based on the above, the storage location specification information is generated to specify the storage location of each new version of storage data by allocating an empty area between the multiple versions of storage data. To.

  Further, in the present embodiment, a case has been described in which storage location designation information for designating that a free area is arranged between each new version of storage data is generated.

  It is also possible to classify a plurality of storage data of the new version into a plurality of groups and generate storage location designation information for designating that an empty area is arranged between each group.

  Further, in the present embodiment, the storage location designation information for designating the same storage location as the storage location designated in the storage location designation information of the old version of the storage data for each new version of the storage data is generated.

  As described above, according to the present embodiment, the storage location designation information for designating the storage location of each storage data by generating an empty area between a plurality of storage data is generated, so that the overhead at the time of software update Can be suppressed, and the time to complete the software update can be shortened.

Embodiment 2. FIG.
FIG. 8 is a diagram illustrating an example of the non-volatile memory arrangement information generated by the storage location designation information generation unit 103 when the file size greatly increases in the new version. In this case, the size of file1 has increased to 0x00133332. In this case, the amount of increase does not fit in the margin area behind file1 in the first version, so file1 is moved. That is, file1 is newly arranged at address 0x00700000 within the margin area behind file3. As a result, even if the size of the marginal area is exceeded, the writing position of file2 and file3 is not affected, so that it is possible to reduce the writing amount of the nonvolatile memory in the software update on the terminal. (There is no need to overwrite file2 and file3).

  The configuration of the server device 1 is the same as that shown in FIG.

  As described above, in the present embodiment, when there is storage data whose data size increases in the new version due to revision of the storage data, it is designated as the storage position of the storage data in the storage position designation information of the old version of the storage data. To determine whether it is possible to store the new version of the storage data whose data size has increased in the storage candidate area that combines the existing area and the free space provided for the storage data. As shown in the first embodiment, when the storage location designation information for designating that the storage data of the new version with the increased data size is stored in the storage candidate area is generated and cannot be stored in the storage candidate area, the data size is An empty area (in the example of FIG. 8, an area after file3) in which the increased new version of storage data can be stored is searched, and data is stored in the searched empty area. Generating a storage position specifying information to specify that stores data stored in the new edition size is increased.

  As described above, according to the present embodiment, even if any of the new version of the stored data exceeds the size of the margin area, the writing position of the other stored data is not affected, so the software update on the terminal In this case, it is not necessary to overwrite other stored data, and the write amount of the nonvolatile memory can be reduced.

Embodiment 3 FIG.
In the new version, there may be cases where new files are added.

  FIG. 9 is a diagram showing an example of a new version of the directory tree when a file is newly added. In this example, file 21 is added under the directory dir2.

  In the present embodiment, the order of file arrangement on the memory is basically a lexicographic arrangement of file names. Therefore, if a directory tree as shown in the figure is given in the first version, it will be arranged on the memory in the order of file1, file2, file21, and file3. However, in the second and later versions, in order to reduce the amount of nonvolatile memory written on the terminal, it is desirable to arrange the files in the first version without changing the order as much as possible. Therefore, in the present embodiment, the addition of a new file is performed not on the area where the existing file is already arranged but on the margin area.

  FIG. 10 is a diagram illustrating an example of the nonvolatile memory arrangement information generated by the storage location designation information generation unit 103 when the number of files increases. In the figure, if file 21 is a lexicographic arrangement, it should be placed between file 2 and file 3, but this is placed on a margin area behind file 3. In this manner, the storage location designation information generation unit 103 generates storage location designation information that designates placement in a margin area when a file is newly added.

  The configuration of the server device 1 is the same as that shown in FIG.

  As described above, in the present embodiment, when new storage data is generated by revising the storage data, it is newly generated in the area designated as the empty area in the storage location designation information of the old version of the storage data. Storage location specification information for specifying storage data is generated.

  As described above, according to the present embodiment, in the second and subsequent versions, new files can be arranged without changing the order of the files of the first version as much as possible, and the amount of nonvolatile memory written on the terminal is reduced. It becomes possible.

Embodiment 4 FIG.
By the way, when updating the terminal software on the terminal side, the contents on the nonvolatile memory must be rewritten. The terminal rewrites the nonvolatile memory in accordance with the differential command transmitted from the server device 1 via the network 3. At that time, an image of a unit to be temporarily written in the nonvolatile memory is created on the RAM of the terminal.

  FIG. 11 shows an overall procedure of software rewriting on the terminal. First, the differential command is downloaded via the network (step S1101), and then the nonvolatile memory image is rewritten according to the differential command (step S1102).

  The conventional software update is performed according to the procedure shown in FIG. This is the case, for example, when a patch program frequently used in UNIX (registered trademark) is used. First, the terminal downloads difference information (patch) (step S2001). The difference information describes what correction should be applied to which file. Next, each file included in the difference information is rewritten using the function of the file system (step S2002). For example, when a certain file is to be corrected, the file is opened (open), written (write), and closed (closed). This open / write / close function is provided as a file system function, and is internally converted to a rewrite process to the file system header and directory information area in addition to the file data area. The Therefore, when rewriting a plurality of files, each time each file is modified, rewriting to the file system header and directory information occurs, so the processing efficiency is not good. On the other hand, according to the method of the present embodiment, the file system header or the file system header or the like can be directly rewritten using the difference information between the new version and the old version of the nonvolatile memory image without using the file system function. Rewrite directory information only once.

Hereinafter, the nonvolatile memory image rewriting process will be described.
FIG. 12 is an example of a differential command transmitted to the terminal. In this example, 0x12345678 for 4 bytes is added to the head of file1 in FIG. 6, and thereafter, a command for copying the old version of file1 is shown. As a result, the size of file1 is changed from 0x00424058 to 0x0042405a. Note that the file system header and directory information are actually affected by the change in the size of file1, but this is omitted in this example. The difference command in this example has the following contents.
・ SKIP 0x00400000
Skips the first 0x00400000 bytes of non-volatile memory (does not rewrite)
・ DATA 0x12345678
Write 0x12345678 (from the current 0x00400000 byte) • COPY 0x00400000, 0x00424058
Copy 0x0042000058 bytes from 0x00400000 bytes of the old version of non-volatile memory ・ FLUSH
Actually write the contents so far to nonvolatile memory

  FIG. 13 is a diagram showing how the nonvolatile memory on the terminal changes due to the differential command. As shown in the figure, it is necessary to write data (0x12345678) and copy the area on the nonvolatile memory of the terminal.

  In such a case, the following problems occur in the processing at the terminal. If the COPY (area copy) is performed after the DATA command processing (write 0x12345678) on the nonvolatile memory, the first 4 bytes of the area to be copied are already rewritten to 0x12345678. The desired result is not obtained. In this case, a desirable result can be obtained by processing DATA after performing COPY. However, it is generally difficult to rewrite only by processing on a nonvolatile memory. Therefore, first, the content in the nonvolatile memory is temporarily copied to a RAM (work memory), a desired new version of the memory image is created on the RAM, and then copied to the nonvolatile memory.

  FIG. 14 is a diagram showing how the nonvolatile memory is rewritten on the terminal. As shown in the figure, when executing the differential command, a memory image is once created on the RAM, and copied to the non-volatile memory when the FLUSH command is executed. This solves the problem that the contents of the nonvolatile memory are destroyed by the command execution order. The FLUSH command is inserted at the end of the margin area. Note that the above only explains the necessity of once copying to the RAM, and is not within the scope of the claimed invention.

  By the way, in the embodiments described so far, a margin area is provided for each file. Actually, if there is too much marginal area, the area that can actually be used on the non-volatile memory decreases, which is not preferable. Therefore, it is conceivable to classify a plurality of files into groups and insert a margin area between each group instead of one file at a time. As described above, the FLUSH command is inserted for each unit of the margin area, and triggers writing the memory image created on the terminal RAM so far into the nonvolatile memory.

  However, there are cases where the RAM capacity cannot be increased in the terminal. That is, there is a case where the size of the RAM has to be reduced in order to reduce the cost of the terminal. Therefore, when multiple files are classified into groups and a margin area is inserted for each group, a large number of files are combined, so if the amount of data up to the next margin area becomes too large, the terminal uses it. It is conceivable that the amount of RAM that can be exceeded is exceeded. In this case, the nonvolatile memory cannot be rewritten by the above method.

  Therefore, the storage location designation information generation unit 103 according to the present embodiment takes into account the amount of RAM (work memory) of the terminal in advance, and collects the RAM as a unit for inserting a margin area. It is characterized by not exceeding the upper limit.

  FIG. 15 is an example of the non-volatile memory arrangement information created by the storage location designation information generation unit 103 according to this embodiment. In this case, the size of file1 is 0x0080000, the size of file2 is 0x0060000, and the size of file3 is 0x000f0000. When the upper limit of usable RAM of the terminal is 0x00100000 bytes, the size of file1 and file2 is 0x000d0000, which is within the upper limit of usable RAM. However, when file1, file2, and file3 are combined, the size becomes 0x001c0000, which exceeds the upper limit of the RAM. Therefore, the storage location designation information generation unit 103 puts file1 and file2 together and inserts a margin area behind them. Also, a margin area is inserted after file3.

  The configuration of the server device 1 is the same as that shown in FIG.

  Thus, in this embodiment, when a plurality of stored data is classified as a group, the data size of each group is equal to or smaller than the memory size of the work memory used when storing the new version of stored data in the nonvolatile memory. Thus, the new version of the plurality of storage data is classified into a plurality of groups, and storage location designation information for designating that an empty area is arranged between each group is generated.

  As described above, according to the present embodiment, it is possible to effectively update the terminal software even in a terminal having a small RAM capacity as a work memory.

Embodiment 5 FIG.
By the way, there may be a case where there is a file whose size exceeds the usable size of the RAM. In this case, if the file can be divided, the storage location designation information generation unit 103 inserts a margin area in the divided portion. As a splittable file, a compiled and linked file may be separated into several sections. When a program written in C language or the like is compiler-linked, the compiled file includes an instruction section (sometimes called TEXT), a data section with an initial value, a data section without an initial value (sometimes called BSS), etc. It is often divided into two. In such a case, the storage location designation information generation unit 103 generates storage location designation information that instructs to insert a margin area between these sections.

  FIG. 16 is an example of non-volatile memory arrangement information generated by the storage location designation information generation unit 103 according to the present embodiment. According to this, in file3, the instruction section, the data section with initial value, and the data section without initial value are all 0x000f0000. In this case, assuming that the usable size of the RAM is 0x00100000 bytes, the file 3 alone exceeds this size, so a margin area is inserted for each section.

  The configuration of the server device 1 is the same as that shown in FIG.

  As described above, in the present embodiment, when the data size of any of the storage data of the new edition is equal to or larger than the memory size of the work memory used when storing the storage data of the new edition in the non-volatile memory, Specify that the data stored in the new edition is divided into multiple sections (parts) so that the data size of the section (part) is less than or equal to the memory size of the work memory, and an empty area is placed between each section (part). Generate storage location specification information.

  As described above, according to the present embodiment, it is possible to effectively update the terminal software even in a terminal having a small RAM capacity as a work memory.

Embodiment 6 FIG.
In the present embodiment, a case will be described in which when the terminal software is revised, the server apparatus 1 extracts the difference between the new version and the old version of the program and notifies the terminal apparatus 2 of the extracted difference.

  FIG. 24 is a diagram illustrating a configuration example of the server apparatus 1 according to the present embodiment, and elements 101 to 106 are the same as those in FIG. The difference information generation unit 107 compares the new version and the old version of the storage position designation information stored in the storage position designation information storage unit 106 to extract a difference, and generates the extraction result as difference information. The communication unit 108 transmits the difference information generated by the difference information generation unit 107 to the terminal device 2.

  FIG. 17 is a diagram illustrating an operation procedure of the difference information generation unit 107. The inputs are old and new nonvolatile memory image information and old and new nonvolatile memory arrangement information. Assume that a set of data areas described in the nonvolatile memory arrangement information is S1 and S2, respectively. As shown in the figure, the data area is extracted from S2 in ascending order of address, and is designated as F2. When F2 is not file data (when F2 is either a file system header / directory structure), since the data area corresponding to F2 exists in S1, it is set as F1, and a difference command that is the difference between F2 and F1 is set. Output. When F2 is file data, it may be a new file or an existing file. If file data having the same path as F2 exists in S1, it is set as F1, and a differential command that is the difference between F2 and F1 is output. If the file data having the same path as F2 does not exist in S1, it is a new file, so a differential command for adding F2 is output.

  Note that the simplest method for extracting the differences for each file is to simply output a DATA command based on the contents of the new version, and then output a FLUSH command. As described above, the DATA command is overwritten from the current address on the nonvolatile memory. In this case, the difference is not actually extracted, but there is no problem in rewriting the nonvolatile memory on the terminal. As an improved method, the memory image of the new version is compared with the old version for each fixed size (for example, 4 bytes), the matched part is expressed by the COPY command, and the other part is expressed. A method of expressing using a DATA command is also conceivable. In addition, it is possible to search for the longest match between the new version and the old version, express that part with the COPY command, and express the rest with the DATA command. A generally well-known algorithm for binary difference extraction can also be used. For example, the rsync algorithm described in “Andrew Tridgel and Paul Mackerras, Thersync algorithm, TR-CS-96-05, The Australian National University, June 1996”, the binary difference algorithm, and the diffsync algorithm described in binary 1996 are known.

  Further, the difference command when F2 is a new file can be realized by simply outputting a DATA command based on the contents of the new file and then outputting a FLUSH command.

  As described above, in this embodiment, each time storage data is revised, the storage location designation information of the old version storage data is compared with the storage location designation information of the new version storage data, and the difference between the two is extracted. The difference information indicating the extracted difference is generated. The generation of the difference information is performed by comparing the storage location designation information of the old version storage data and the storage location designation information of the new version storage data for each storage data unit, and extracting the difference between the two.

  As described above, according to the present embodiment, it is possible to supply the terminal device with difference information that can suppress the occurrence of overhead during software update and can shorten the time until software update is completed.

Embodiment 7 FIG.
In the difference information generation unit 107 of the above-described embodiment, the unit of difference extraction is a file. However, if the file is a compiled file and includes a plurality of sections having different functions, difference extraction is performed for each section. A method of outputting a differential command is also conceivable. In general, section information can be obtained from the link map file that is output when compiling and linking, and this information can be used to divide the nonvolatile memory image into sections and extract differences. It is. Here, ELF header, Program header table, and Section header table in the ELF (Executable and Linking Format) format developed by UNIX (registered trademark) System Laboratories are also broadly included.

  The configuration of the server device 1 is the same as that shown in FIG.

  As described above, in this embodiment, the compiled file that can be divided into sections is used as the storage data, and the storage location designation information of the old version of the compiled file and the storage location designation information of the new version of the compiled file are generated. In this case, for each section, the storage location designation information of the old version of the compiled file is compared with the storage location designation information of the new version of the compiled file, and the difference between the two is extracted. Here, the compiled file that can be divided into sections is, for example, a compiled file in an ELF (Executable and Linking Format) format that can be divided into sections.

  As described above, according to the present embodiment, it is possible to supply the terminal device with difference information that can suppress the occurrence of overhead during software update and can shorten the time until software update is completed.

Embodiment 8 FIG.
Also, if the file is a compiled file generated by a compiler and the instruction section of the sections constituting the file can be divided into C function units before compilation, the difference between these parts A method using the unit of extraction as a function is also conceivable. As in the case of sections, it is generally possible to know how far each function corresponds to in the non-volatile memory image from the link map file output at the time of compile linking. It is possible to extract the difference by dividing the memory image into functions.

  FIG. 18 is a diagram showing an example of a file system header according to the present embodiment. In this example, the contents of the file system header in the state of FIG. 6 are shown. The total number of files is 3, which corresponds to the presence of three files, file1, file2, and file3. The pointer to the directory structure is indicated by the number of bytes from the head address of the file system header. In this embodiment, the file system header has a very simple structure of 8 bytes as described above, but it goes without saying that a more complicated structure may be used depending on the implementation of the file system.

  FIG. 19 is a diagram showing an example of a directory structure according to the present embodiment. In this example, the contents of the directory structure in the state of FIG. 6 are shown. In the figure, all pointers are represented by the number of bytes from the head address of the directory structure. In this example, a structure representing a directory and a structure representing a file are mixed. The directory structure consists of the following elements.

-Name (16-byte fixed-length character string)
-Number of lower directories (4 bytes) ... Let this be N.
・ Number of lower files (4 bytes) ... Let this be M.
・ Pointer to lower directory (4 bytes x N)
・ Pointer to lower file (4 bytes x M)
The structure representing a file consists of the following elements.
-Name (16-byte fixed-length character string)
-Pointer to file data (4 bytes)

  For example, in the figure, the top data is data of “/” that is the root directory. “/” Is designated as the name, the number of subordinate directories is 2 in total, dir1 and dir2, and the number of subordinate files is 0 because there is no file immediately below. Subsequently, pointers corresponding to the lower directories dir1 and dir2 are stored. Next, there is dir1 data. This is 1 because the number of lower directories is 0 and the number of lower files is file1. Subsequently, a pointer to a lower file corresponding to file1 is stored. The total data is 152 bytes.

  The configuration of the server device 1 is the same as that shown in FIG.

  As described above, in this embodiment, the compiled data that can be divided for each C language function before compilation is stored data, and the storage location designation information of the old version of the compiled file and the storage location designation information of the new version of the compiled file are used. Is generated, the storage location specification information of the old version of the compiled file is compared with the storage location specification information of the new version of the compiled file for each C function before compilation, and the difference between the two is extracted. .

  As described above, according to the present embodiment, it is possible to supply the terminal device with difference information that can suppress the occurrence of overhead during software update and can shorten the time until software update is completed.

  Here, the features of the server apparatus shown in the first to eighth embodiments will be described below.

The server apparatus shown in the first to eighth embodiments is mounted on a terminal and a nonvolatile memory for storing terminal software, which is a set of one or more files necessary for operating the terminal, on the file system In addition,
When a memory image to be stored is created, a margin area that does not include data is inserted between file data on the memory image.

  The server apparatuses shown in the first to eighth embodiments are characterized in that when a new file is added to the file system due to the revision of the terminal software, the new file is added to the margin area.

  In the server devices shown in the first to eighth embodiments, when a terminal having a RAM rewrites the nonvolatile memory by revising the terminal software, the size of the nonvolatile memory area sandwiched between the margin areas can be used by the terminal A marginal area is inserted so as not to exceed the size of a random RAM.

The server apparatus shown in the first to eighth embodiments is characterized in that a margin area is inserted in the middle of a file having a size exceeding the size of RAM usable by the terminal.
In the server apparatus shown in the first to eighth embodiments, when a file whose size exceeds the RAM size usable by the terminal is composed of a plurality of sections having different functions, a margin area is inserted between the sections. It is characterized by doing.

  The server apparatus shown in the first to eighth embodiments is characterized in that the file format is the ELF format.

  When the server apparatus shown in the first to eighth embodiments generates the difference data between the new version and the old version of the data on the file system stored in the nonvolatile memory of the terminal, the image on the nonvolatile memory of the file system The difference data between the new version and the old version is output.

  The server device shown in the first to eighth embodiments outputs a difference by comparing the new and old processing blocks with a procedure for reading the new version and the old version of the image on the nonvolatile memory of the file system in a certain processing block unit. When performing the procedure, the processing block unit is a file.

  The server apparatus shown in the first to eighth embodiments is characterized in that when the file is a compiled file generated by a compiler and includes a plurality of sections having different functions, the unit of the processing block is a section. To do.

  The server apparatuses shown in the first to eighth embodiments are characterized in that the file format of the compiled file is the ELF format.

  In the server apparatuses shown in the first to eighth embodiments, when a file is a compiled file generated by a compiler and can be divided into C language function units before compilation, the unit of a processing block is a function. It is characterized by doing.

FIG. 3 shows a configuration example of a server apparatus according to the first embodiment. FIG. 3 is a diagram illustrating a hardware configuration example of a terminal device according to the first embodiment. FIG. 4 is a diagram showing an example of nonvolatile memory configuration information according to the first embodiment. FIG. 3 is a diagram showing an example of a directory tree according to the first embodiment. FIG. 4 is a diagram showing an example of nonvolatile memory image information according to the first embodiment. FIG. 6 is a diagram showing an example of nonvolatile memory arrangement information according to the first embodiment. FIG. 6 is a diagram showing an example of nonvolatile memory arrangement information when creating a new version according to the first embodiment. The figure which shows the example of the non-volatile memory arrangement | positioning information at the time of the new version preparation concerning Embodiment 2. FIG. FIG. 10 is a diagram showing an example of a new version directory tree according to the third embodiment. FIG. 10 is a diagram illustrating an example of nonvolatile memory arrangement information when a new version is created according to the third embodiment. The figure which shows the example of a whole procedure of the software rewriting in the terminal device which concerns on Embodiment 4. FIG. FIG. 10 is a diagram illustrating an example of a difference command according to the fourth embodiment. The figure which shows the mode of the change of the non-volatile memory on the terminal by the difference command which concerns on Embodiment 4. FIG. The figure which shows the mode of the change of the non-volatile memory on the terminal by the difference command which concerns on Embodiment 4. FIG. The figure which shows the example of the non-volatile memory arrangement | positioning information at the time of the new version preparation concerning Embodiment 4. FIG. FIG. 20 is a diagram showing an example of nonvolatile memory arrangement information when creating a new version according to the fifth embodiment. FIG. 18 is a diagram illustrating a procedure example of difference extraction processing according to the sixth embodiment. FIG. 20 shows an example of a file system header according to the eighth embodiment. FIG. 20 shows an example of a directory structure according to the eighth embodiment. The figure which shows the example of a procedure of the conventional software update. FIG. 6 is a diagram illustrating an operation example of the server apparatus according to the first embodiment. FIG. 6 is a diagram illustrating an operation example of the server apparatus according to the first embodiment. The figure which shows the relationship between a server apparatus and a terminal device. FIG. 10 shows a configuration example of a server apparatus according to a sixth embodiment.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 Server apparatus, 2 Terminal apparatus, 3 Network, 101 Data size analysis information input part, 102 Temporary storage position information input part, 103 Storage position designation information generation part, 104 Storage position designation information writing / reading part, 105 Output part , 106 Storage position designation information storage unit, 107 Difference information generation unit, 108 Communication unit.

Claims (20)

An information processing apparatus that targets a predetermined nonvolatile memory and manages a storage position of data in the nonvolatile memory,
A data size analysis information acquisition unit for acquiring data size analysis information used for analyzing a data size of a plurality of stored data stored in the nonvolatile memory;
A temporary storage position information acquisition unit that acquires temporary storage position information indicating a temporary storage position of each storage data in the nonvolatile memory;
Based on the data size analysis information acquired by the data size analysis information acquisition unit, the data size of each storage data is analyzed, and the data size of each storage data analyzed and the temporary storage position information acquisition unit acquired And a storage location designation information generating section for generating storage location designation information for designating a storage location of each storage data by arranging an empty area based on the storage location information. Information processing device. The storage location designation information generation unit
The information processing apparatus according to claim 1, wherein storage position designation information that designates an empty area to be arranged between each stored data is generated. The storage location designation information generation unit
The information processing apparatus according to claim 1, wherein the storage position designation information for classifying the plurality of stored data into a plurality of groups and designating that an empty area is arranged between the groups is generated. The information acquisition unit for data size analysis is
When the stored data is revised, the data size analysis information used for analyzing the data size of multiple new versions of the stored data is acquired.
The temporary storage position information acquisition unit
The storage location designation information generated by the storage location designation information generation unit for the old version storage data is acquired as temporary storage location information,
The storage location designation information generation unit
Based on the data size analysis information acquired by the data size analysis information acquisition unit, the data size of each storage data of the new version is analyzed, and the data size of each storage data of the analyzed new version and the temporary storage position information acquisition unit are analyzed. Based on the acquired storage location specification information of the old version of storage data, generate storage location specification information that specifies the storage location of each new version of storage data by placing an empty area between the multiple versions of storage data The information processing apparatus according to claim 1. The storage location designation information generation unit
5. The information processing apparatus according to claim 4, wherein storage position designation information for designating an empty area to be arranged between each new version of storage data is generated. The storage location designation information generation unit
5. The information processing apparatus according to claim 4, wherein the storage version designation information for classifying the plurality of storage data of the new version into a plurality of groups and designating that an empty area is arranged between each group is generated. The storage location designation information generation unit
5. The information processing apparatus according to claim 4, wherein the storage position specifying information for specifying the same storage position as the storage position specified in the storage position specification information of the old version of the storage data for each of the new version of the storage data is generated. apparatus. The storage location designation information generation unit
When there is storage data whose data size changes in the new version due to revision of the storage data, it is provided for the area specified as the storage position of the storage data in the storage position specification information of the storage data of the old version and the storage data Determine whether it is possible to store the new version of the storage data whose data size has changed in the storage candidate area combined with the free area
When the storage candidate area can be stored, the storage location designation information for designating that the storage data of the new version whose data size has changed is stored in the storage candidate area is generated,
A storage that searches for a free area that can store a new version of storage data whose data size has changed and stores the new version of the storage data whose data size has changed in the searched free area when the storage candidate area cannot be stored. The information processing apparatus according to claim 4, wherein position information is generated. The storage location designation information generation unit
Storage location that specifies to store the newly generated storage data in the area that is specified as an empty area in the storage location specification information of the old version storage data when new storage data is generated by revision of the storage data The information processing apparatus according to claim 4, wherein the designation information is generated. The storage location designation information generation unit
The plurality of storage data of the new version is classified into a plurality of groups so that the data size of each group is equal to or smaller than the memory size of the work memory used when storing the storage data of the new version in the nonvolatile memory. The information processing apparatus according to claim 6, wherein storage position designation information for designating an empty area to be arranged is generated. The storage location designation information generation unit
When the data size of any of the new version storage data is equal to or larger than the memory size of the work memory used when storing the new version storage data in the non-volatile memory, the data size of each part is the memory of the work memory. 5. The storage location designation information for designating that the storage data of the new edition is divided into a plurality of parts so as to be smaller than the size and an empty area is arranged between the parts. Information processing device. The storage location designation information generation unit
Generates storage location specification information that specifies that an empty area should be placed between sections when the new version of the stored data with a data size larger than the memory size of the work memory is a compiled file that can be divided into sections. The information processing apparatus according to claim 11. The storage location designation information generation unit
When the new version of the stored data having a data size larger than the memory size of the work memory is a compiled file in ELF (Executable and Linking Format) format that can be divided into sections, an empty area is arranged between the sections. 12. The information processing apparatus according to claim 11, wherein the storage location designation information to be designated is generated. The information processing apparatus further includes:
Each time the storage data is revised, the storage location specification information of the old version of the storage data generated by the storage location specification information generation unit is compared with the storage location specification information of the new version of the storage data, and the difference between them is extracted. The information processing apparatus according to claim 4, further comprising a difference information generation unit that generates difference information indicating the extracted difference. The difference information generation unit
The information processing apparatus according to claim 14, wherein the storage location designation information of the old version storage data and the storage location designation information of the new version storage data are compared for each storage data unit, and a difference between the two is extracted. The difference information generation unit
Compiled files that can be divided into sections are stored data, and the storage location specification information generator generates storage location specification information for the old version of the compiled file and storage location specification information for the new version of the compiled file. 15. The information processing according to claim 14, wherein the storage location designation information of the old version of the compiled file is compared with the storage location designation information of the new version of the compiled file for each section, and a difference between the two is extracted. apparatus. The difference information generation unit
ELF (Executable and Linking Format) files that can be divided into sections are stored data, and the storage location specification information generation unit stores storage location specification information of the old version of the compiled file and storage location of the new version of the compiled file. When the specification information is generated, the storage location specification information of the old version of the compiled file is compared with the storage location specification information of the new version of the compiled file, and the difference between the two is extracted for each section. The information processing apparatus according to claim 14. The difference information generation unit
Compiled files that can be divided into C language functions before compilation are stored data, and the storage location specification information generation unit stores storage location specification information for the old version of the compiled file and storage location specification information for the new version of the compiled file. If generated, the storage location specification information of the old version of the compiled file is compared with the storage location specification information of the new version of the compiled file and the difference between the two is extracted for each C function before compilation. The information processing apparatus according to claim 14. A storage location management method for managing a storage location of data in the nonvolatile memory for a predetermined nonvolatile memory,
A data size analysis information acquisition step for acquiring data size analysis information used for analyzing a data size of a plurality of stored data stored in the nonvolatile memory;
A temporary storage position information acquisition step of acquiring temporary storage position information indicating a temporary storage position of each storage data in the nonvolatile memory;
Based on the data size analysis information acquired in the data size analysis information acquisition step, the data size of each storage data is analyzed, and the analyzed data size of each storage data and the temporary storage position information acquisition step are acquired. And a storage location designation information generation step for generating storage location designation information for designating a storage location of each storage data by arranging an empty area based on the storage location information. Storage location management method. A program that targets a predetermined nonvolatile memory and causes a computer to manage the storage location of data in the nonvolatile memory,
Data size analysis information acquisition processing for acquiring data size analysis information used for analyzing the data size of a plurality of stored data stored in the nonvolatile memory;
Temporary storage position information acquisition processing for acquiring temporary storage position information indicating a temporary storage position of each storage data in the nonvolatile memory;
Based on the data size analysis information acquired by the data size analysis information acquisition process, the data size of each storage data is analyzed, and the data size of each storage data analyzed and the temporary storage position information acquisition process are acquired. Based on the storage location information, the computer executes a storage location specification information generation process for generating storage location specification information for allocating an empty area between a plurality of storage data and specifying the storage location of each storage data. Program.

Images