JP2006164017A - Information processor, information processing method, and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To improve a reproduction processing speed regarding a file to be recorded on a medium. <P>SOLUTION: When the file is updated by adding additionally-written data following file data before update (data before additional writing), at first, a reservation area for defragmentation with the same number of clusters as that of the data before additional writing is set, and the additionally-written data are recorded on a continuous area of the cluster following the reservation area for defragmentation. After that, for example, when a processing load or the like of a CPU is reduced to a level below a fixed level, the data before additional writing are recorded again by moving them to the reservation area for defragmentation, and fragments of the update file are eliminated at that stage. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an information processing apparatus corresponding to a storage medium on which data is recorded in units such as files, and a method thereof. The present invention also relates to a program executed by such an information processing apparatus.

  In general, data stored in a medium (storage medium, storage device) is managed in file units (or data units corresponding to files) by a file system (or data management information corresponding to the file system). . In a file system, even if data portions obtained by dividing data forming one file into a plurality of data are physically discretely recorded on a storage medium, these data portions are connected and appropriately 1 It is managed as one file.

However, for a file in which data portions are discretely recorded and managed, for example, when processing such as reproduction is executed, access is required for each of these data portions. In other words, a plurality of accesses are required to execute processing for a single file, resulting in performance degradation such as processing speed.
In particular, in a disk-shaped recording medium such as an HDD (Hard Disk Drive) or various optical disk-shaped recording media, this problem is conspicuous because a seek operation for physically moving the head for data recording / reproduction is involved as the access time. It becomes.

  As a countermeasure for this, for example, it is known to execute a media maintenance process called defragmentation or optimization for all data stored in the medium. As a result of executing such defragmentation processing, physical and logical continuity can be obtained for data as files stored on the medium. As a result, when processing a file, it is necessary to first access the top data, and thereafter, to read data continuously physically and logically. That is, the access operation accompanied with the seek operation is performed only once, and the processing speed is increased accordingly. Incidentally, such media maintenance for defragmentation is generally executed in response to a user operation, as represented by a personal computer or the like.

  In addition, if the configuration for recording moving image information described in Patent Document 1 is applied, data forming one file is continuously written in one collective storage area without being dispersed. Can be stored in the state.

JP 2004-86823 A

An object of the present invention is to improve performance such as data processing speed by recording data forming unit data such as a file with physical continuity on a medium. To do.
As a typical example in which the data forming one file can be fragmented, the update of the file already stored in the medium is continued after the data of the file before update (data to be updated: data before appending). In this case, it is possible to add the new content data (additional data).
In such a file update, if data such as another file has already been recorded even immediately after the recording area of the pre-append data, which is file data already stored on the medium, the pre-append data It is not possible to record additional data in the recording area. In this case, additional data is recorded by searching for an empty area that is physically separated from the recording area of the data before additional recording. As a result, the data forming one file is discretely recorded in the first half of the pre-recording data and the latter half of the additional recording data.
In particular, an object of the present invention is to be as efficient as possible with regard to a configuration for eliminating a fragment of unit data (file) that is updated by additional writing in this way.

In view of the above-described problems, the present invention is configured as an information processing apparatus as follows.
That is, it comprises recording means for recording data on the storage medium, and management means for managing data stored in the storage medium using predetermined unit data.
In addition, when updating the contents of unit data by adding additional data following the update target data that was unit data stored in the storage medium, the update target data in the storage medium can be written. Additional data is recorded in the storage area following the reserved area by the reserved area securing means and the recording means for ensuring a free area having a capacity of only a reserved area for the update target data. As for the additional data recording means to be updated and the unit data after the update, the processing is executed so that the additional data is connected to the update target data and is managed by the management means. Primary update means is provided.
In addition, after the processing of the primary update means is completed, the update target data recording means for recording the update target data in the reserved area by the recording means, and the update target recorded in the reserved area by the update target data recording means It is assumed that the additional write data is connected to the data and formed, and the secondary update means for executing a process for managing by the management means.

  The information processing method includes a recording procedure for recording data on a storage medium, a management procedure for managing data stored in the storage medium using predetermined unit data, and unit data already stored in the storage medium. When the content of the unit data is updated by adding additional data following the update target data, a free area of the storage medium with a capacity sufficient to write the update target data Reserved area securing procedure to be secured as a reserved area for data, additional record data recording procedure to record additional data in a storage area following the reserved area by the recording procedure, and updated unit The data is managed by the management procedure on the assumption that the postscript data is formed after the update target data. A first update procedure for executing the processing for the first time, and an update target data recording procedure for recording the update target data in the reserved area by the recording procedure after the processing as the first update procedure is completed, By this update target data recording procedure, a secondary update is executed to execute the process for managing the update target data recorded in the reserved area following the update target data to be formed by concatenating the data. The procedure is configured to be executed.

  Further, the program is configured such that the information processing apparatus executes the procedure as the information processing method.

According to each of the above configurations, the present invention updates the content of the unit data by adding the additional data to the update target data (data before additional data) that is the unit data that has been stored in the storage medium. The case is assumed.
In addition, when updating the unit data, first, a free space for the update target data is secured as a reserved area on the storage medium, and additional data is added to the storage area that follows the reserved area. To be recorded. In addition, in the steps so far, management is performed so that updated unit data is formed by linking the update target data stored in the storage medium and the additional write data.
Then, the update target data is recorded in the reserved area at a predetermined opportunity and timing after managing the updated unit data as described above. In response to this, the updated unit data is managed as formed by linking the update target data recorded in the reserved area and the additional data. At this stage, the update target data and the additional data are stored in a continuous storage area on the storage medium.
In addition, according to the above, in the present invention, first, the additional data is recorded, and the additional data and the data before update are connected to manage the data as unit data. Yes. At this stage, as the unit data, the data before update and the additional data may be stored discretely on the storage medium, but the unit data update process is once completed. After that, by recording the pre-update data in the reserved area, finally, the unit data after update in which fragmentation has been eliminated is obtained. In other words, two recording operations, ie, recording additional data and recording pre-update data in the reserved area, should be executed as recording operations until the updated unit data is obtained. The operations are executed at completely different opportunities and timings.

In this way, the present invention prevents the update target data and the additional write data from being dispersed on the storage medium for the unit data that is updated so that the additional write data is added after the update target data. , It is possible to obtain a continuously recorded state. As a result, it is possible to improve the processing speed regarding data processing such as reproduction for the unit data.
In addition, two recording operations (recording additional data, recording update target data in the reserved area) are not performed as a series of operations, but are performed at different opportunities and timings. The total processing time and processing load required to obtain updated unit data without fragmentation will be distributed. This means that the processing for updating unit data according to the present invention is efficient.

  Hereinafter, the best mode for carrying out the present invention (hereinafter referred to as an embodiment) will be described. In this embodiment, a case where the configuration of the information processing apparatus according to the present invention is applied to a digital video camera is taken as an example.

FIG. 1 is a block diagram illustrating a configuration example of a digital video camera 1 according to the present embodiment.
In the digital video camera 1 shown in this figure, the optical system unit 2 includes an imaging lens, a diaphragm, and the like, and forms incident light on the photoelectric conversion unit 3 as imaging light. Further, the optical system unit 2 includes a focus adjustment mechanism for focus adjustment, a diaphragm variable mechanism that varies the diaphragm according to the diaphragm value, and the like. Is performed by a drive signal output from the camera function unit 6. The camera function unit 6 is configured to output a required drive signal so as to obtain an appropriate focus state, aperture state, and the like, under the control of the CPU 10.
For example, when an optical zoom function is to be provided, a zoom mechanism for moving the zoom lens is provided in the optical system unit 2 and driving for moving the zoom mechanism in accordance with the control of the CPU 10 in the same manner as described above. What is necessary is just to provide a part. Further, the camera function unit 6 may be configured to provide a strobe light emission function after providing a strobe.

  The photoelectric conversion unit 3 includes, for example, a CCD (Charge Coupled Device) that is a photoelectric conversion element, and performs imaging conversion by photoelectrically converting imaging light incident from the optical system unit 2 and imaged on the light receiving surface. Is output to the video signal processing unit 4. At the time of shooting, for example, an instruction of a shutter speed determined according to the exposure setting result is notified from the CPU 10 to the video signal processing unit 4. The video signal processing unit 4 outputs a scanning timing signal corresponding to the notified shutter speed to the photoelectric conversion unit 3. The photoelectric conversion unit 3 performs scanning according to the scanning timing signal, executes photoelectric conversion processing, and outputs a video signal.

The video signal processing unit 4 performs waveform shaping on the analog video signal (captured image signal) input from the photoelectric conversion unit 3 by performing gain adjustment and sample hold processing, for example, and then performs A / D conversion. By doing so, it is converted into digital video signal data. Then, for the digital video signal obtained by this conversion process, for example, a process for generating display luminance data is performed, and a video signal process for displaying on the display unit 7 is executed. Accordingly, the video signal processing unit 4 can also perform signal processing for a so-called on-screen display so that a character image or the like can be displayed superimposed on the captured image under the control of the CPU 10.
The actual display device employed as the display unit 7 is not particularly limited, but currently, liquid crystal display panels are widely employed.

In addition, the video signal processing unit 4 performs, for example, compression encoding processing by a predetermined method on the digital video signal obtained by converting the analog video signal input from the photoelectric conversion unit 3 to generate compressed video data. Is possible.
The digital video camera of this embodiment also has a still camera function. That is, it is possible to generate a still image data file in a predetermined format as a photograph for the captured image signal. Such image processing is also performed by the video signal processing unit 4.

Further, the video signal processing unit 4 uses an image (video) signal input from the photoelectric conversion unit 3, an AV (Audio Video) data file (AV file) read from a medium to be described later, and the like in a predetermined format. It can be converted into an analog video signal or a digital video signal and output to an external device or the like via the image input / output unit 5.
The image input / output unit 5 can input a video signal of a predetermined method from the outside, and the input video signal can be displayed on the display unit 7 through the processing of the video signal processing unit 4. Further, the video signal processing unit 4 converts the video signal input by the image input / output unit 5 into recording data and transfers it to the media controller 13 in the same manner as the analog video signal input from the photoelectric conversion unit 3. You can also.
Correspondingly, the image input / output unit 5 includes, for example, a video (image) signal output terminal / video signal input terminal according to a predetermined method.

In addition, the digital video camera 1 according to the present embodiment includes an audio processing unit 8 and an audio input / output unit 9 so that audio signals can be input / output.
First, for voice input, a microphone or the like is provided as the voice input / output unit 9, and external voices are collected and converted into voice signals to be inputted. The audio signal input in this manner is output to the audio processing unit 8. The audio processing unit 8 performs audio signal processing such as conversion into compressed audio data encoded by an audio compression encoding method corresponding to compression encoding of a captured image.

The CPU 10 forms an AV file of a predetermined format from the compressed video data for the captured image obtained by the video signal processing unit 4 and the compressed audio data for the collected sound obtained by the audio processing unit 8. Control processing is executed. In this AV file, the playback time axis of the sound output by reproducing the compressed audio data is synchronized with the moving image output by reproducing the compressed video data. The configuration for actually forming the AV file may be a software configuration for digital signal processing obtained by the CPU 10 executing the program, or hardware for forming the AV file. In addition, the CPU 10 may be configured to control the operation of the hardware.
The data as the AV file is transferred to the media controller 13 as recording data, for example, under the control of the CPU 10. The CPU 10 can also transfer a still image data file of a predetermined format as a photographic image generated by the video signal processing unit 4 to the media controller 13 as recording data.

  The audio input / output unit 9 may be configured to include an audio signal input terminal and the like to input an audio signal from an external audio device or the like. The audio processor 8 converts the audio signal input from the audio signal input terminal into a digital audio data file in a predetermined format. The CPU 10 can also transfer data of such a digital audio data file to the media controller 13 as recording data.

  The media controller 13 is configured to be able to execute control processing related to data processing for these media in association with predetermined different types of external media (storage media) and external storage devices (storage devices) in cooperation with the CPU 10. Is done. Data processing for the media here refers to data to be stored in the media, such as media format processing and file write / read processing for file management information (file management information). Any related process.

In the present embodiment, a hard disk drive (HDD) is first connected to the media controller 13. The HDD is a storage device including a magnetic disk as a storage medium, which is called a hard disk as is well known, and currently, a large capacity of a gigabyte class can be obtained at a relatively low cost. As is well known, physical data is read / written from / to a magnetic disk as a storage medium by applying a magnetic field and detecting a magnetic field while tracing a track formed on the magnetic disk by a magnetic head. Realized.
In this case, the HDD may be fixedly incorporated in the digital video camera 1, for example, or may be removable from the digital video camera 1 (host) according to a predetermined standard. It may be said.

Also, in this figure, the media controller 13 is capable of data processing corresponding to a predetermined type of optical disk, magneto-optical disk, and a semiconductor storage device which is a storage device including a semiconductor memory element. .
In the case of dealing with optical discs and magneto-optical discs, it is actually provided with a device as a drive configured to be able to write / read data corresponding to these recording (storage) media, and these devices and media controllers 13 will be connected.
In the case of supporting a semiconductor memory device, the main body of the digital video camera 1 is provided with a slot in which the semiconductor memory device is attached and detached according to the actual standard of the semiconductor memory device. When the semiconductor memory device is properly loaded in the slot, the pin terminal of the semiconductor memory device is connected to the electrode at the connector portion of the slot, and the semiconductor memory device can communicate with the media controller 13. It will be connected as possible.

  As described above, the media controller 13 sends the recording data to the media (storage medium, storage device) connected to the media controller 13 in response to the transfer of the recording data. Is further transferred to the medium selected as the target. In the medium to which the data has been transferred, the data is written and stored in the storage area in accordance with an instruction from the media controller 13 side. In this way, data stored in the medium is stored and managed as a file. Note that management of files stored on the media is performed by a predetermined file system.

For example, when playing back an AV file as playback of a file stored in the medium, the CPU 10 and the media controller 13 access and read the designated AV file. The AV file read in this way is separated into compressed video data and compressed audio data by processing of the CPU 10, for example, and the compressed video data is transferred to the video signal processing unit 4, and the compressed audio signal is processed by the audio processing unit. 8 is handed over.
In this case, the video signal processing unit 4 and the audio processing unit 8 respectively perform necessary reproduction signal processing including demodulation processing on the compressed video data and the compressed audio data transferred as described above. As a result, an image obtained by reproducing the compressed video data is displayed on the display unit 7 and the audio input / output unit 9 has an audio signal obtained by reproducing the compressed audio data in synchronization with the reproduction time of the image. Then, it can be output as audio by the speaker, or can be output from the headphone terminal.

  In addition, for example, an audio data file reproduced from a medium is output as a predetermined format audio signal and audio data to the outside via the audio input / output unit 9 after being subjected to the audio signal processing of the audio processing unit 8. It is also possible. In this case, the audio input / output unit 9 includes an audio output terminal corresponding to a format of a predetermined audio signal and audio data output from the audio processing unit 8.

A CPU (Central Processing Unit) 10 executes various control processes for the digital video camera 1 by executing a program. The ROM 11 stores various setting information used for the CPU 10 to execute processing in addition to various programs executed by the CPU 10. The RAM 12 is used as a work area when the CPU 10 executes processing according to a program, and holds data such as various arithmetic processing results.
The non-volatile memory 12a is formed of a memory element such as a flash memory that does not erase the stored contents even when the power supply is stopped, and data writing / reading is executed under the control of the CPU 10. The The data (information) to be stored in the nonvolatile memory 12a is generally setting information whose contents are appropriately changed, but is not particularly limited, and the actual specifications of the digital video camera 1 are not limited. It is sufficient to store various necessary information according to the above.

  In this case, the operation input unit 15 collectively indicates various operators provided in the digital video camera 1. For example, the operators in the operation input unit 15 include a shutter button operated at the time of taking a picture, an operator for selecting a shooting mode, an operator for increasing / decreasing parameters, and the like. Further, the operation input unit 15 may include a configuration for realizing an input operation as a GUI (Graphical User Interface) using the display screen of the display unit 7, for example.

  The communication unit 16 is a part configured by mounting hardware and software for communicating with an external device by a predetermined data communication method according to the control of the CPU 10. The data communication system supported by the communication unit 16 is not particularly limited regardless of whether it is wired or wireless, and the number of corresponding data communication systems should not be limited. At present, the data communication system can be a network such as Ethernet (trademark), data bus standards such as USB (Universal Serial Bus), IEEE (the Institute of Electrical and Electronic Engineers) 1394, etc. . In the case of wireless, short-range wireless communication between devices such as Bluetooth (trademark) and wireless LAN (Local Area Network) standards such as IEEE802.11a / b / g can be cited.

  The power supply unit 17 supplies operating power to various hardware devices in the digital video camera 1 and includes, for example, a power supply circuit that operates by receiving power supply from a battery or a power adapter.

As described above, in the digital video camera 1 according to the present embodiment, the AV file obtained by imaging / sound collection is mainly stored in the HDD, various optical disk-shaped recording media (including magneto-optical disks), and the semiconductor storage device. It can be stored in a medium (storage medium).
The file stored in the medium as described above is normally managed by a file system in a predetermined format. In this embodiment, the file is managed by a FAT (File Allocation Table) file system. It is supposed to be. As is well known, the FAT file system manages files by a tree-type directory structure, and data writing / reading is performed by a logical minimum data management unit called a cluster. It is said that. A cluster is a unit obtained by collecting a predetermined number of sectors, which are the minimum units of physical data writing / reading on a medium.

FIG. 2 shows a general system configuration of the FAT file system using a hierarchical model.
First, this hierarchical model is roughly divided into a software layer and a hardware layer as a lower layer.
In this case, the software layer corresponds to a program executed by the CPU and software processing realized by various firmware, middleware, and the like in a device serving as a host (in the embodiment, the digital video camera 1) for the medium. It is supposed to be. As shown in the figure, the software layer in this case includes the application 100, the file system 101, and the device driver 102 from the upper layer to the lower layer.
It can be considered that the physical storage area of the medium itself is located in the hardware layer.

  The application 100 has a file recording / playback function, for example, and is compatible with application software that uses media, and makes an access request at the file level to the file system 101.

The file system 101 corresponds to software that realizes a function as a file system. In this embodiment, since the FAT file system is adopted, the software that provides the functions of the file system 101 is also configured corresponding to the FAT file system.
In the file system 101, the access request at the file level from the application 100 is converted into a cluster level that is a data management unit in the FAT file system format, and an access request is made to the device driver 102.

  The device driver 102 corresponds to software for controlling a medium as a device to be controlled. An access request at a cluster level according to the FAT file system format from the file system 101 is recorded and reproduced on the medium 103. The access level to the medium 103 is requested after conversion to the sector level as a unit.

The medium 103 in this case is logically formatted (initialized) according to the FAT file system. In FIG. 1, a storage medium and storage device connected to the media controller 13 such as an HDD, an optical disk-shaped recording medium, and a semiconductor storage device correspond to the medium 103 here. Then, in response to the sector level access request from the device driver 102, the medium 103 reads data from the designated sector address and returns it to the device driver 102. That is, an access response at the sector level is executed.
The device driver 102 receives an access response from the medium 103 at the sector level, that is, receives data in units of sectors, and treats the received data as data in units of clusters and transfers it to the file system 101 (at the cluster level). Access response).
The file system 101 transfers data received from the device driver 102 to the application 100 as file level data. The application 100 executes a required process at the application level according to, for example, an operation input by the user on the data received as a file.

  Further, the FAT file system manages storage files with a tree-type directory structure, and manages files as a cluster unit set. Such file management and data management are realized by providing directory entries and table information called FAT as is well known. The directory entry is information indicating the location of a file or directory (subdirectory) on the medium at the cluster level, and the FAT is information indicating a chain (link or link) at the cluster level that forms the directory or file.

FIG. 3 shows the format structure of the media according to the FAT standard.
Although a plurality of FAT formats are defined, here, the format structure corresponding to each of the FAT16 and FAT32 formats is shown.
First, the format structure corresponding to the FAT 16 shown on the left side of FIG. 3 will be described.
The structure shown in this figure is logical according to LBA (Logical Block Addressing). In other words, the top sector is the top sector (LBA = 0), and the block (sector) number advances downward thereafter.
Further, in the FAT file system, one physical storage area can be divided into a plurality of partitions, but here, for the convenience of explanation, a format structure in the case of one partition is shown.

  First, the head sector represented as LBA = 0 is a boot area called MBR (Master Boot Recorder).

The structure of MBR is shown in FIG.
One sector has a size of 512 bytes. Therefore, the MBR is 512 bytes. In FIG. 4A, for the 512 bytes in the MBR area, the last byte position from the beginning is expressed by assigning a hexadecimal number from 0000h to 01FFh, and then arranged in units of 16 bytes. As shown. In the hexadecimal notation, h indicating the hexadecimal notation is added to the end of the numerical value, such as 0000h or 01FFh.
First, a 446-byte area from byte positions 0000h to 01BDh stores a code (boot code) for booting when used as an OS (Operating System) boot medium. It is an area.

  A 64-byte area from byte positions 01BEh to 01FDh in the MBR is used as a partition table, and stores predetermined information required for each partition at the time of startup. This partition table area is further divided into four areas in units of 16 bytes. These four divided areas are partitioned into partition 1, partition 2, partition 3, and partition 4 in order from the top. It is a corresponding area. Each area corresponding to each of these partitions is an entry area for each partition.

  Also, 55AAh is stored in the final 2-byte area in the MBR as an identifier that the current sector is the MBR.

The partition entry has a structure shown in FIG. First, the most significant byte position 00h is an area for storing a flag indicating whether or not the partition is designated as a startup drive.
In the subsequent 3-byte area at byte positions 01h to 03h, a value expressing the start sector of the partition in terms of CHS (Cylinder / Head / Sector) is stored. A 3-byte area consisting of byte positions 05h to 07h stores a value representing the end sector of the partition in CHS.
In addition, a value indicating the start sector of the partition by LBA is stored in a 4-byte area at byte positions 08h to 0Bh.

  A 4-byte area consisting of byte positions 0Ch to 0Fh stores a value indicating the data size (partition size) of the partition.

  A 1-byte area indicated by the byte position 04h stores a value indicating a type of a platform, a file system, or the like corresponding to the partition, which is also called a system identifier.

Returning to FIG.
An empty area corresponding to a predetermined number of sectors is provided after the head sector as the MBR. Then, an area in units of partitions is formed by a sector area after this empty area.

An area from the first sector to a predetermined number of bytes in one partition is a system area for storing system-related information.
In this system area, an area having a predetermined number of bytes starting from the first sector is a BPB (BIOS Parameter Block / Boot Parameter Block). As described above, necessary data to be used by the block device control program is stored. The information stored in the BPB includes information such as the number of FAT areas, the start sector of the main FAT area, and the number of sectors in the FAT area, which will be described below.

  Following BPB, FAT1 and FAT2 FAT areas are arranged in this order. Normally, one of the FAT1 and FAT2 areas is used as the main FAT area. The other FAT area is generally used as a spare area or backup area of the main FAT area, for example, as a mirror area obtained by copying the contents of the main FAT area.

These FAT areas are areas formed by arranging FAT entries in the order of cluster numbers in the data area. There is a one-to-one correspondence between a FAT entry and a cluster in the data area. In the FAT entry, as information on the cluster, for example, unused or in the corresponding file, according to the directory and file storage results. Information indicating any one of a cluster number, a defective cluster, an EOF (End Of File: last cluster in a file), etc. chained next to the current cluster is stored.
In the FAT16 format, the cluster number is expressed by 2 bytes (16 bits), and accordingly, the size of each FAT entry is also 2 bytes.

  Subsequent to the FAT1 and FAT2 areas, a root directory entry of a predetermined size is arranged. The root directory entry stores directory entries for directories and files in the root directory.

  A sector area lower than the root directory is a data area. Writing / reading data to / from this data area is managed by the FAT file system. Therefore, as shown in the figure, the file system format structure is such that the data area is managed in cluster units composed of a sequence of one or more predetermined number of sectors. As described above, the FAT area is formed with the FAT entries corresponding to all the clusters forming the data area in principle.

Next, a format structure corresponding to FAT32 shown on the right side of FIG. 3 will be described.
Even in the format structure corresponding to FAT32, MBR is arranged in the head sector represented by LBA = 0. Then, an area in units of partitions is arranged subsequent to the empty area of a predetermined number of sectors arranged after the MBR.

As a system area in the partition area of the FAT32 compatible format structure, an FSinfo area is provided after BPB.
The FSinfo is an area for storing predetermined information used to calculate the free space in the partition. The FAT areas of FAT1 and FAT2 are sequentially arranged following FSinfo.

  The FAT area in the case of the FAT32 format is also formed by arranging FAT entries corresponding to the clusters in a one-to-one order in the order of the cluster numbers. The FAT entries include predetermined entries for the corresponding clusters according to the directory and file storage results. Is stored. However, in the FAT32 format, since the cluster number is expressed by 4 bytes (32 bits), the size of each FAT entry is also divided by 4 bytes.

Further, in the FAT32 format structure, the root directory entry area provided in the system area in the FAT16 format structure is omitted.
In the FAT32 format structure, the root directory is placed in the data area. The starting cluster number of the root directory in the data area is indicated by a value stored in a predetermined area (RootClus) in the BPB. When accessing the root directory, the cluster number recognized with reference to the RootClus is accessed. To be. The starting cluster number of the root directory indicated by RootClus is usually 2.

  In this case as well, the data area is an area in which writing / reading of data is managed in cluster units by the FAT32 file system.

FIG. 5 shows the structure of a directory entry defined in the FAT file system. Here, the structure of the directory entry corresponding to FAT32 is shown.
In this case, the directory entry is 32 bytes. In this figure, each byte position from the upper byte to the lower byte is indicated by a value from 0h to 1Fh.
In 32 bytes forming the directory entry, an 8-byte area from the most significant byte position 0h to byte position 7h stores the name of the current file or current directory indicated by the current directory entry.
In the subsequent 3-byte area from the lower byte position 8h to the byte position Ah, an extension corresponding to the file format of the current file is stored.

In the subsequent lower byte position Bh, a value indicating the attribute for the current file / current directory is stored.
Here, the byte position Ch is a reserved area.
A 3-byte area from byte positions Dh to Fh stores a value indicating the creation time of the current file / current directory.
A 2-byte area at byte positions 10h to 11h stores a value indicating the creation date of the current file / current directory.
The 2-byte area at byte positions 12h to 13h stores a value indicating the date (last access date) when the current file / current directory was last accessed.

  In addition, depending on the total of 4 bytes of the 2-byte area by byte positions 14h-15h and 2-byte area by byte positions 1Ah-1Bh, the head position on the medium where the current file / current directory is stored is indicated by the cluster number. To be done. That is, the value of the leading cluster number for the current file / current directory is stored in this 4-byte area. For the 2-byte area at byte positions 14h-15h, the value of the upper 4 bytes for the leading cluster number is stored. For the 2-byte area at byte positions 1Ah-1Bh, the lower 4 bytes of the first cluster number are stored.

The 2-byte area at byte positions 16h to 17h stores a value indicating the time (recording time) when the recording (last update) for the current file / current directory is executed.
The 2-byte area at byte positions 18h to 19h stores a value indicating the date (recording date) when the recording (last update) for the current file / current directory was executed.
A 4-byte area from byte positions 1Ch to 1Fh stores a value indicating the size (capacity) of the current file / current directory.

FIG. 6 shows an example of file management by the FAT file system. In this figure, a case where the data is formatted according to FAT32 is shown.
Here, it is assumed that at least the file A, the file B, the file C, and the file D are stored in the data area, and the directory entries of the files A, B, C, and D are respectively shown in FIG. b) As shown in (c) and (d). As an explanation in this figure, the first cluster number in the directory entry (byte position 14h-15h / 1Ah-1Bh)
Since the information is the main information, there is a portion omitted for information on other areas.

FIG. 6E shows a part of the FAT area, and has the contents of the FAT entry corresponding to the storage results of the files A, B, C, and D. In FIG. 6 (e), only the FAT entry contents corresponding to only the files A, B, C, and D are extracted in consideration of simplicity of illustration and easy understanding. As described above, in FAT32, each FAT entry has a size of 32 bits.
In addition, the FAT entry in the FAT area shown in FIG. 6E is incremented for each digit of the cluster number corresponding to the FAT entry as shown as 00000000h, 00000010h, 00000020h, 00000030h. They are arranged in a matrix of rows and columns indicating + 00h to + 0Fh, which are values of the least significant digit added to the cluster number value indicated by each row. For example, a FAT entry at a position where a row of 00000000h and a column of + 07h intersect corresponds to the cluster number 00000007h.

In the FAT file system, the first cluster represented by the cluster numbers 00000000h and 00000001h and the second cluster following this are reserved areas (RSV). Accordingly, the FAT entries corresponding to the cluster numbers 00000000h and 00000001h are also defined as reserved areas.
As a value stored in the FAT entry, the cluster number to be chained next to the current cluster is represented by 00000002h to 0FFFFFF6h. Further, EOF (last cluster (file end cluster) in a file including the corresponding cluster) is represented by EOF = 0FFFFFFFh. Further, in FIG. 6E, “−” indicates that the cluster is unused, but actually, for example, it is represented by a value of 00000000h.

In order to access the file A, first, the directory entry of the file A shown in FIG. 6A is accessed. For example, the directory entry itself of the file A is accessed as follows.
Taking the case where the file access command is an absolute path as an example, in order to access the directory entry of file A, the root directory is first accessed, and the directory entry is followed from here according to the path. Is done. In the process of tracing a directory entry, a subdirectory of a certain current directory may be traced. As is well known, the parent-child relationship between a parent directory and a child directory is also expressed by the directory entry. The For this purpose, for example, the directory entry on the parent directory side is provided with a name indicating the child directory, and the directory entry of the child directory indicated by the directory entry on the parent directory side has the parent directory. A directory entry storing information indicating this in the name area is provided. As a result, the directory can be traced in both directions from the parent to the child and from the child to the parent.
As a result of tracing the directory entry according to the path as described above, the directory entry of the file A is finally reached as the access destination.

When the directory entry of the file A is accessed, the file A itself is subsequently accessed starting from the directory entry of the file A.
The access to the file A itself is conceptually as follows.
In order to access file A, the beginning (start) cluster number in the directory entry of file A is acquired. This leading cluster number indicates the starting cluster number in which data of file A is stored in the data area. In this case, it is 00000007h as shown in FIG. That is, at this stage, it is recognized that the start cluster number of file A is 00000007h.
Therefore, the file system accesses the FAT entry corresponding to the cluster number 00000007h in the FAT area shown in FIG. 6E, and refers to the value stored in the FAT entry. In this case, 00000008h is stored in the FAT entry corresponding to the cluster number 00000007h. As a result, the file system recognizes that the data of the file A is connected to the cluster data of the cluster number 00000008h following the cluster data of the cluster number 00000007h. Then, when accessing and referring to the FAT entry corresponding to the cluster number 00000008h, since 00000009h is stored here, the data of the file A is further followed by the cluster data of the cluster number 00000008h. It is recognized that the data of the cluster with the number 00000009h is concatenated. Therefore, when referring to the FAT entry with the cluster number 00000009h, since the value indicating EOF is stored here, the file A is recognized to end with the cluster with the cluster number 00000009h.
From this, it is recognized that the file A is formed by concatenating cluster data in the order of cluster numbers 00000007h-00000008h-00000009h as shown in the cluster chain of FIG. 6 (f). The file system is made to access the cluster having the cluster numbers 00000007h-00000008h-00000009h, and access to the file A is thereby performed.

As can be seen from the above description, the data position of the file stored in the data area is indicated by the top cluster number stored in the directory entry and the contents of the FAT entry in the FAT area that is referenced from this start cluster number. The file can be accessed by referring to the information.
It can also be seen that the management of files stored in the data area by the tree-type directory structure is expressed by directory entries.

The storage locations of the remaining files B, C, and D are managed as follows using the file management information including the directory entry and the FAT area.
First, regarding the file B, as shown in the directory entry of FIG. 6B, the cluster number of the first (starting) cluster of the file is 0000000Ah. Therefore, referring to the FAT entry corresponding to the cluster number 0000000Ah in the FAT area shown in FIG. 6E, the cluster number 0000001Fh is shown as chain information indicating the next cluster. Therefore, when the FAT entry is further traced to refer to the FAT entry corresponding to the cluster number 0000001Fh, the cluster number 00000025h is stored as chain information. The FAT entry corresponding to the cluster number 00000025h stores the cluster number 00000031h. The FAT entry corresponding to the cluster number 00000031h stores the cluster number 00000030h. The FAT entry corresponding to the cluster number 00000030h stores a value indicating EOF.
In this way, the file B is connected to the data stored in the cluster according to the order of the cluster numbers 0000000Ah-0000001Fh-00000025h-00000031h-00000030h as shown in FIG. 6 (g) by the directory entry and the FAT area. It is expressed that it is formed by.

  As for the file C, as shown in FIG. 6C, first, in the directory entry of the file C, the cluster number 0000001Bh is shown as the head (start) cluster of the file. Based on this, depending on the contents referred to from the FAT entry of the cluster number 0000001B in the FAT area shown in FIG. It is formed by concatenating data stored in a cluster in the order of 00000012h-00000013h-00000014h-00000003h.

  For the file D, the directory number of the file D shown in FIG. 6D indicates the cluster number 0000002Ch as the head (start) cluster of the file. Based on this, as the contents referred to from the FAT entry of the cluster number 0000002C in the FAT area shown in FIG. It is formed by concatenating data stored in a cluster in the order of 0000002Eh-0000002Fh-00000038h-00000039h-0000003Ah-0000003Bh.

Based on the format contents of the FAT file system described so far, first, the circumstances leading to the present invention will be described with reference to FIG.
FIG. 7A schematically shows an example of file management corresponding to the contents of the FAT area. In this figure, the clusters are arranged by a matrix of column numbers 1 to 4 and row numbers 1 to n. As the correspondence of the cluster order of this figure to the actual cluster order, first, the cluster number is incremented by 1 according to the column number, and the cluster number is incremented by 4 according to the row number. That is, if the representation by [row number-column number] is performed, the leading cluster in FIG. 7 is [1-1], and then [1-2] [1-3] [1-4] [2 -1] [2-2]..., The cluster number is actually incremented one by one.

FIG. 7A shows a state where four files A, B, C, and D have been recorded (stored) on the medium.
The file A is managed as being recorded by a cluster chain in accordance with the cluster order after using 20 clusters having consecutive cluster numbers from cluster [1-1] to [5-4]. As management by the actual FAT file system for the file A, first, the actual cluster number of the cluster [1-1] is stored as the first cluster by the directory entry corresponding to the file A, and is stored in the FAT area (FAT1). , The appropriate chain for the FAT entries corresponding to the clusters [1-1] to [5-3] so that the cluster chains corresponding to the cluster order of the clusters [1-1] to [5-4] are shown. Stores the destination cluster number. Then, a value indicating EOF (End Of File) is stored in the FAT entry corresponding to cluster [5-4].

The file B is managed as recorded in a cluster chain according to the cluster order using eight consecutive clusters of the clusters [6-1] to [7-4].
File C is managed as being recorded by a cluster chain according to this cluster order using 18 consecutive clusters of clusters [8-1] to [12-2].
The file D is managed as recorded in a cluster chain according to this cluster order using 20 consecutive clusters of clusters [12-3] to [17-2].

When updating the file B from the state shown in FIG. 7A, the data of the new contents (additional data) are added to the data of the contents of the file B already recorded on the medium. Consider the case. Here, it is assumed that four clusters are additionally recorded as additional data.
According to the recording state of FIG. 7A, the four clusters after [8-1] following the last cluster [7-4] of the recorded file B as the update source are already used for the file C. Yes. Therefore, the additional data of file B cannot be recorded using the cluster [8-1] or later.

  Therefore, in such a case, the additional data of file B is additionally recorded as described below. In the following description, as shown in FIG. 7A, the data recorded using the clusters [6-1] to [7-4] as the file B before update will be described. This data is referred to as “data before appending (data to be updated)”, and data using 4 clusters to be added after the data before appending is referred to as “additional data”.

According to the recording state of FIG. 7A described above, the cluster [17-3] and subsequent areas following the last cluster [17-2] of the file D are unused areas, that is, free areas. Therefore, the additional data of the file B is recorded using four consecutive clusters of clusters [17-3] to [18-2] as shown in FIG. 7B, for example.
For the FAT area (FAT1), the FAT entry is edited so that the first cluster [17-3] of the additional data is chained with the final cluster [7-4] of the data before additional data in the file B. To be. As a result, the entire file B is managed as being formed by connecting the additional write data immediately after the pre-write data that is the file B before update. When this is viewed as a cluster chain, file B is formed by concatenating 12 clusters in the order of clusters [6-1] to [7-4] → [17-3] to [18-2]. It will be managed in this way.

In this way, the FAT file system manages the data forming the file by the chain structure of the cluster unit. Therefore, when updating the file by adding additional data, the data ( Data before appending) and appending data are managed as one file with a structure in which the data before updating and the appending data are linked even if they are recorded discretely instead of being consecutive in cluster number. Can be done.
In this way, it is an advantage of the FAT file system that data portions that are discrete in terms of cluster numbers can be connected to form one file. However, from another viewpoint, this is as follows. You will have a problem.

  Here, consider an HDD as a medium. The HDD employs a magnetic disk that is one of disk-shaped recording media as the recording medium itself. When the HDD is formatted (initialized) under the FAT file system, as is well known, the storage unit unit as a cluster is obtained by separating a predetermined number of storage unit regions called sectors on the recording track. To be formed. The clusters formed in this way are assigned so that the cluster number is sequentially incremented along the progress of the recording track.

  In such a format, the fact that the cluster numbers are separated without being consecutive means that the physical position on the magnetic disk is also separated without being consecutive on the track. Therefore, in the case where management is performed on the HDD as a file B described above with reference to FIG. 7B on the HDD as being composed of a plurality of data portions with different sector numbers, these data portions are stored on the magnetic disk. It will be physically discretely recorded above.

  The HDD is configured to write / read data to / from a track by a magnetic head. Therefore, in order to access data such as a file stored in the HDD, an operation called seek that physically moves the magnetic head is performed so that the track in which the data is written is accessed. Necessary. Thus, as is well known, the access time for disk media such as HDDs always includes the seek time required for the seek operation.

  When the access time becomes longer, for example, the transfer rate of the file data is lowered, and the performance such as the file reproduction processing speed is lowered. Therefore, it is preferable to shorten the access time as much as possible. However, as described above, in the HDD or the like, the above-described seek time is unavoidable.

On the assumption that this is the case when data processing such as reproduction of file B shown in FIG. 7B is executed, a cluster area that constitutes pre-append data, a cluster area that constitutes append data, Are in a discrete state as recording positions on the HDD. Therefore, in order to execute the data processing for the file B, the access to the cluster area of the data before appending and the access to the cluster area of the appending data are executed, and the access time is accordingly increased. Will take longer.
In particular, in the case where a plurality of data parts constituting a file are discretely recorded in this way, the physical recording position on the recording medium is due to the fact that the cluster numbers between the data parts are considerably separated. May be considerably separated. In this case, the seek time between the cluster areas corresponding to these data portions is considerably long, and the above-described decrease in the file processing speed becomes more remarkable.
Also, for example, a semiconductor memory device using a semiconductor memory element such as a flash memory has a structure in which a file to be processed is discretely recorded as an area unit corresponding to the cluster. It cannot be said that there is no possibility that the performance of the file processing speed deteriorates due to overhead caused by factors such as processing load for obtaining data continuity.
Therefore, from the viewpoint of the file processing speed, the data from the beginning to the end of the data forming one file is recorded using a continuous cluster area (or a unit area equivalent to a cluster). Is preferred.

  Therefore, in the present embodiment, as described below, the data of the update source file that has already been recorded on the medium is used as the data before appending, and the appending data is added after the data before appending. When updating (additional update), it is possible to obtain a state in which the new data portion is recorded in the cluster area following the pre-additional data portion without the pre-additional data portion and the additional-data portion being dispersed at the cluster level. Configure file update processing. Thereby, regarding the data processing of the updated file, the overhead caused by the discrete recording of a plurality of data portions is eliminated, and the data processing speed is improved.

FIG. 8 and FIG. 9 show an example of a procedure for additionally updating a file according to this embodiment.
FIG. 8A shows an example of a file management state before update addition of a file. FIG. 8A shows the same file management state as FIG. 7A described above. Also in this case, the file B is updated by adding additional data that consumes 4 clusters to the file B as the update target, following the data before additional data that is the file data before update. That is, the contents of the file B to be updated are the same as those in FIG. In the description here, an update source file B is formed, and after the update, data is recorded using the data part that is the first half of the file B, that is, clusters [6-1] to [7-4]. The data part becomes the data before appending. Further, data of four clusters having contents to be added subsequent to the data before additional recording becomes additional recording data.

Then, when performing the postscript update for the file B, the process for the file B is executed as follows.
The file B consumes 8 clusters according to the data before additional recording, and consumes 4 clusters according to the additional recording data. Therefore, the file B after the additional recording update can be considered as consuming 12 clusters.
Therefore, in this embodiment, when updating the file B, first, as shown in FIG. 8B, from the recording area of the media, these 12 clusters are continuously unused (free area). This area is searched for and secured as a cluster area (update file area) used for final data recording of file B. In FIG. 8B, it is assumed that all the clusters after the cluster [17-3] following the terminal cluster [17-2] of the file D are not used. Cluster [17-3] is reserved as the leading cluster, and 12 clusters up to cluster [20-2] are reserved as update file areas.

When the update file areas from cluster [17-3] to cluster [20-2] are secured as described above, the head of the update file area is subsequently updated as shown in FIG. An area of 8 sectors continuous from the cluster [17-3] is further secured as a defragmentation reserved area. In this case, the fact that the update file area is set to 8 sectors corresponds to the fact that the data before appending of file B uses 8 sectors.
Further, as described above, the defragmentation reserved area is secured in the update file area, and clusters [19-3] to [20], which are areas for four clusters following the defragmentation reserved area in the same update file area. -2], record additional data of file B.
As a result, the free area as the update file area for 12 clusters is divided into the reserved area for defragmentation by the area for 8 clusters by the continuous clusters [17-3] to [19-2] on the front side and the back side following this This is used as a writing area for additional data of file B by an area for four clusters.

Then, at the stage where the additional data of the file is recorded as shown in FIG. 9 (a), the data before additional recording recorded using the clusters [6-1] to [7-4], and the cluster Directory entries and FAT areas (FAT1, FAT2), etc., so that file B is managed by linking with additional data recorded using [19-3] to [20-2] The file management information consisting of is edited for rewriting.
Therefore, even in the recording state shown in FIG. 9A, although the file B is fragmented (fragmentation), the additional data is logically connected to the data before the additional recording. It is properly managed as an update file.

  In the present embodiment, as the process of update addition of a file, first, the first stage is completed when the process up to FIG. 9A is completed. Then, after the process up to the first stage is completed, the process as the second stage is executed, for example, at a predetermined opportunity and timing. As the timing for executing the second stage, for example, the processing load of the CPU or the processing load of the media drive becomes equal to or less than a predetermined value, so that data writing / reading operations on the medium with a certain level of stability and speed can be obtained. It is conceivable that the operation situation becomes as follows.

As shown in the transition from FIG. 9A to FIG. 9B, the second stage processing is performed as data before appending to the file B recorded in the clusters [6-1] to [7-4]. The data for 8 clusters is copied so as to be copied to the area for 8 clusters of clusters [17-3] to [19-2] reserved as the defragmentation area. At the same time, the clusters [6-1] to [7-4] in which the data before appending of the file B was originally recorded are managed as unused areas. That is, apparently, the data before the additional recording is moved from the clusters [6-1] to [7-4] to the clusters [17-3] to [19-2].
As a result of such movement of the data before additional recording, the data existing on the medium as the file B is included in the data before adding the clusters [17-3] to [19-2] and the cluster [19 −3] to [20-2]. That is, the file B is formed by twelve cluster chains that are consecutive in cluster number from the first cluster [17-3] to the last cluster [20-2].
Corresponding to this, the file management information is also managed so that the file B is formed by a cluster chain of 12 clusters from the first cluster [17-3] to the last cluster [20-2]. The contents will be rewritten.
As a result of the execution of the second stage process, the file B can be recorded on the medium without fragmentation. That is, a state where the file B is defragmented is obtained.

  According to the procedure described above, in the present embodiment, the file to be updated with the addition of the additional data is prevented from being fragmented at the cluster level. As a result, as described above, when executing data processing of one file, it is not necessary to access two or more fragmented data portions, so that the total access time for data processing is reduced accordingly. The data processing speed is also improved.

Further, in the present embodiment, the process as the first stage from FIG. 8A to FIG. 9A until the defragmentation of the append update file is finally completed, The temporal execution timing is divided by the second stage process shown in FIG.
For example, when the process from the first stage to the second stage is executed as one complete process sequence as the file additional update process, the actual data writing process to the media is performed in one sequence process. Thus, it is necessary to execute twice, that is, writing of the additional data shown in FIG. 9A and moving writing of the data before additional data shown in FIG. 9B. Then, in one processing sequence, access for data writing occurs twice, and as a result, the total access time becomes longer, resulting in a longer data processing time required for file update. In addition, the data processing load becomes heavy accordingly. For example, this increases the waiting time until the file update is completed and causes a decrease in the stability of the program execution process. It can lead to lowering.
Therefore, as in the present embodiment, first, if it can be completed up to the first stage where the processing of FIG. 9A is completed, at this stage, as access for data recording, Thus, only once is required for additional data, and accordingly, the data processing time is shortened and the data processing load is reduced. Therefore, for example, the waiting time for completing the file update at this time is shortened, and the user's stress is reduced. Also, the stability of the program execution process can be ensured. In the following, for example, when the processing load is light, the second stage process of FIG. 9B is executed as a background process. It will be completed.

  According to the procedure described with reference to FIG. 8 and FIG. 9 above, the fragment at the cluster level for the append update file is eliminated. However, if only this procedure is executed, the actual data of the file stored in the cluster When viewed at the level of (valid data), discontinuity may occur as data forming a file. In other words, for example, at the hierarchical level of the file system 102 that executes data processing at the cluster level, file fragmentation is eliminated, but at the level of the application 100 that processes actual data as a file, the file data is not valid. Can cause continuity. This point will be described with reference to FIG.

FIG. 10 shows a data area in the medium as a concatenation in units of clusters. The cluster number is incremented from left to right in the figure.
FIG. 10A shows a state in which file A, file B, file C, and file D are recorded. File A is recorded using three clusters that are continuous from the beginning of the data area. File B is recorded using 5 consecutive clusters following the last cluster of file A. File C is recorded using 4 consecutive clusters following the last cluster of file B, and file D is recorded using 23 consecutive clusters following the last cluster of file C. Assume that the cluster after the last cluster of file D is unused (free area).

From the state shown in FIG. 10 (a), assuming that additional update is performed on the file B by adding additional data that consumes two clusters, first, as the present embodiment, FIG. 10 (b) As shown in FIG. 5, an update file area for the file B is secured. In this case, since the pre-append data of file B uses 5 clusters and the add-on data of file B uses 2 clusters, a total of 7 cluster update file areas are secured. Then, 5 clusters from the top of the update file area are secured as defragmentation reserved areas, and additional data is written into the remaining 2 clusters. Here, the first stage as the postscript update process is completed.
Then, as a process of the second stage of the postscript update, by moving the pre-write data from the cluster position shown in FIG. 10B to the defragmentation reserved area as shown in FIG. 10C, When viewed as a cluster unit, a state in which additional data is recorded subsequent to the data before additional recording of the file B, that is, a state in which the file is defragmented is obtained.

However, the data before appending of the file B is originally managed as one complete file before the appending update. For this reason, as shown as file B in FIG. 10A, the final cluster is the end of the actual data except when the size of the actual data as the file B is just an integer multiple of the cluster size. To invalid data that occupies the cluster end.
Then, assuming that such a file is subjected to the additional update of the file as described above as described above, the cluster used by the file B as shown as the file B in FIG. In seven consecutive clusters from a to g, invalid data exists in the cluster e corresponding to the end of the pre-append data. That is, the cluster that forms one file is continuous with no fragment, but an invalid data area is inserted in the middle cluster.

In the FAT file system, basically, with respect to the cluster chain area in which the cluster number is continuously incremented, each cluster is completely filled with actual data except for the cluster that ends in this area. For example, data processing at the application level is generally an algorithm based on this assumption.
From this, by performing additional file update according to the present embodiment, as shown in file B in FIG. 10C, the middle of the cluster area to be linked so that the cluster number is incremented. This results in a format violation state in which invalid data is included in the cluster. When the application executes data processing such as reproduction for the file B having such a structure, there is a possibility that an invalid data of the cluster in the middle is carelessly processed and an appropriate data processing result cannot be obtained.

  In the present embodiment, the recording process for the postscript update data in the recording area of the medium is the same as the procedure described with reference to FIG. Therefore, in actuality, as shown in FIG. 10C, the post-update file is defragmented at the cluster level but includes invalid data in the middle cluster. There is. In addition, in the present embodiment, when data processing such as file reproduction is performed, invalid data of the cluster in the middle is discarded, and an appropriate connection state of the actual data forming the additionally-updated file is obtained. To do. This will be described with reference to FIG. 11, taking the file B of FIG. 10 as an example.

FIG. 11A schematically shows the management state of the file B shown in FIG. 10C at the level of the file system 101. FIG. In the file system, the file B is managed as being formed by concatenating seven clusters from clusters a to g incremented by the cluster number. However, as described with reference to FIG. 10, the cluster e corresponding to the terminal cluster of the pre-append data includes invalid data.
In this embodiment, when executing data processing such as reproduction for a file, for example, at the application level, as shown in FIG. Following the end of valid data of the same cluster e, it is configured to be able to handle file data in which actual data at the cluster start position of cluster f is concatenated.

Subsequently, as a configuration for realizing the file additional update processing according to the present embodiment performed as described above, refer to FIGS. 12 to 16 for the file management configuration under the FAT file system. Will be explained.
First, in the following description, the handling of the FAT area in the present embodiment will be described. As described above, the FAT area can be provided with a plurality of areas, for example, FAT1 and FAT2. In this embodiment, as described above, it is assumed that at least two areas of FAT1 and FAT2 are formed and used as the FAT area at the time of media formatting. In addition, FAT1 is used as a main FAT area (main table information) for managing file data in a cluster chain structure as described above with reference to FIG. The FAT2 is basically a mirror area as a spare for the FAT1 and holds the same contents as the FAT1, but additional information necessary for file management is added depending on the file additional update. It is handled as an area (sub-table information) that should be edited by storing typical contents according to a predetermined rule.

FIG. 12A shows a specific content example of FAT1 before the additional update is performed. In FIG. 12A, the contents corresponding to the data of the file (additional update target file) to be updated are stored in the FAT entry area Ar1 for four clusters corresponding to the cluster numbers 00000011h-00000012h-00000013h-00000014h. Has been. In the stage shown in FIG. 12 (a), the data of the additional update target file is data before the additional update is performed, and thus becomes “data before additional update”.
Here, for each FAT entry (instruction area) forming the FAT entry area Ar1 corresponding to the data before appending, as described above with reference to FIG. 6, a value indicating the chain number of the chain destination is stored. Become. In this case, first, 00000012h is stored in the FAT entry corresponding to the cluster number 00000011h, 00000013h is stored in the FAT entry corresponding to the cluster number 00000012h, and 00000014h is stored in the FAT entry corresponding to the cluster number 00000013h. A value indicating EOF is stored in the FAT entry corresponding to the cluster number 00000014h. In other words, the file to be additionally updated in this case is recorded using four clusters of cluster numbers 00000011h−00000012h−00000013h−00000014h. In addition, since the cluster numbers are linked in the order of increment, no fragment is generated in the additional update target file (data before additional recording) at this stage.

  In addition, as a directory entry for the additional update target file at the stage shown in FIG. 12A, 00000011h is stored as the leading cluster number as shown in FIG. 14A. Although not shown, as the size, the capacity value of effective data actually stored using 4 clusters is stored.

Further, as described above with reference to FIG. 5, one byte (8 bits) as the byte position Ch in the directory entry is a reserved area. As shown in FIG. 15, an area for storing a defragmentation reserved area securing flag is defined.
In FIG. 15, the area of the byte position Ch of the directory entry is shown by a bit string structure of 8 bits from bit position 0 to bit position 7. In this case, 1 bit at bit position 7 is assigned as a bit area for the defragmentation reserved area reservation flag.
Here, when a bit value of 0 is stored as the defragmentation reserved area reservation flag, the cluster managed as the defragmentation reserved area is not secured for the file managed by this directory entry. When a bit value of 1 is stored, it is defined as indicating that a cluster area as a reserved area for defragmentation is secured.
In the reserved area of the directory entry, the bit position other than the bit position 7 may be used as an area for storing the defragmentation reserved area reservation flag.
For convenience of explanation, the remaining 7-bit area from bit position 0 to bit position 6 in the reserved area at byte position Ch of the directory entry is undefined, and stores a bit value of 0 for each. It shall be.

  Returning to FIG. 14A, since the defragmentation reserved area has not yet been secured as the additional update target file before the additional update, 0 is set as the defragmentation reserved area reservation flag. Therefore, 00h is indicated as the reserved area at the byte position Ch.

In this case, with respect to the update target file managed as shown in FIG. 12 (a), the size of additional data using 4 clusters is added to the pre-addition data that is the data of the update target file before the update. It is assumed that the file is updated by additionally connecting data.
As described above, the additional update of the file according to the present embodiment secures the defragmentation reserved area and records the additional data, and the second process of moving the pre-additional data to the defragmentation reserved area. It is assumed that a two-step processing procedure, that is, a step processing, is performed. FIG. 12B shows the contents of FAT1 corresponding to the stage where the processing as the first stage is completed, and FIG. 12C shows the contents of FAT2.

As a process of the first stage of the postscript update, first, an update file area is secured as described with reference to FIG. In the state of FIG. 12A before the update, it is assumed that the cluster areas after the cluster number 0000001Ch are managed as unused (free areas). Therefore, in this case, eight consecutive clusters from the cluster number 0000001Ch to the cluster number 00000023h are secured as update file areas. In FAT1 here, the FAT entry area corresponding to the cluster numbers 0000001Ch to 00000023h is assumed to be the FAT entry area Ar4.
In this case, 8 clusters are secured in the update file area because the data before appending uses 4 clusters and the appending data also uses 4 clusters. This is because, as an area, 8 (= 4 + 4) clusters obtained by adding the number of used clusters are necessary as the update file area Ar2.

Then, as reflected in the contents of FAT1 in FIG. 12B, four clusters that are continuous from the beginning in the update file area are secured as defragmentation reserved areas. In addition, in the update file area, additional data is recorded for four clusters up to the end that is supposed to follow the defragmentation reserved area.
In the FAT1 shown in FIG. 12B, the FAT entry area corresponding to the file to be updated corresponds to the area Ar1 corresponding to the data before appending and the appending data according to the processing result up to the first stage. There will be a corresponding area Ar2 and an area Ar3 corresponding to the defragmentation reserved area.

  After the additional write data is recorded on the medium by the processing up to the first stage in this way, the FAT entry area corresponding to the update target file in FAT1 is defined as shown in FIG. Use FAT entries. That is, the area Ar1 corresponding to the pre-recording data, the area Ar3 corresponding to the post-recording data, and the area Ar4 corresponding to the defragmentation reserved area shown in FIG. 12B are defined as shown in FIG. Will be used.

The FAT entry shown in FIG. 16A has 32 bits (0th to 27th bits) corresponding to the FAT file system format being FAT32.
Under the original basic FAT32 format, the next cluster number to be chained is stored by the lower 28 bits from the 4th bit to the 27th bit in the 32-bit FAT entry. Conventionally, the remaining 4 bits from the upper 0th bit to the 3rd bit are unused, and 0,0,0,0 (0h) is stored as the bit value.
In the present embodiment, the upper 4 bits that have been unused so far are defined as shown in the figure. That is, for example, the least significant third bit in the upper 4 bits is defined as an area for an end cluster instruction flag indicating the end cluster of data before additional recording.
Here, when the bit value of the termination cluster instruction flag is 0, it indicates that the cluster corresponding to the FAT entry is not the termination cluster in the data before the additional recording. It is assumed that the corresponding cluster is a terminal cluster in the data before additional recording.
Also in this case, the area for storing the termination cluster instruction flag in the upper 4 bits of the FAT entry may be a bit position other than the third bit.
Also in this case, it is assumed that areas other than the termination cluster instruction flag in the upper 4 bits of the FAT entry are undefined and each stores a bit value of 0.

Based on the definition of the FAT entry shown in FIG. 16A, the FAT entry area (areas Ar1, Ar2) of FAT1 corresponding to the update target file after the processing up to the first stage shown in FIG. , Ar3) will be described in detail.
First, similarly to the area Ar1 in FIG. 12A, the area Ar1 corresponds to the data before additional recording and includes FAT entries corresponding to the cluster numbers 00000011h-00000012h-00000013h-00000014h.
As the FAT entry corresponding to these clusters forming the pre-append data, first, as the FAT entry corresponding to the first cluster number 00000011h, 0000012h is stored as the chain destination cluster number in the lower 28 bits. Further, since the cluster corresponding to the FAT entry is not the end cluster of the data before the additional recording, the end cluster indication flag of the third bit of the FAT entry is 0, and the upper 4 bits are 0h. Therefore, the whole FAT entry indicates 00000012h.
As the FAT entries corresponding to the cluster numbers 00000012h and 00000013h, 0 is stored for the end cluster instruction flag, and the cluster number of the chain destination is stored by the lower 28 bits. 00000013h and 00000014h are indicated. For the FAT entry corresponding to the cluster number 00000014h, which is the terminal cluster of the pre-append data, 1 is stored for the terminal cluster instruction flag. For the lower 28 bits, a value indicating the head cluster number of additional data is stored as the chain destination cluster number. Therefore, the whole FAT entry indicates 10000020h.

  Next, the area Ar2 corresponding to the additional write data shown in FIG. 12B is formed by FAT entries corresponding to the cluster numbers 00000020h-00000021h-00000022h-00000023h. As a matter of course, there is no end cluster of the data before the additional recording in the clusters forming the additional recording data, and therefore, the end cluster indication flags in these FAT entries store all 0 bit values. Then, the cluster number of the chain destination is stored by the lower 28 bits. As a result, 00000021h, 00000022h, and 00000023h are stored in the FAT entries corresponding to the cluster numbers 00000020h-00000021h-00000022h from the top in the area Ar2. Since the FAT entry corresponding to the last (fourth) cluster number 00000023h in the area Ar2 corresponds to the end cluster of the additionally updated file, a value indicating EOF is stored in this FAT entry. .

Also, in FIG. 12B, the area Ar3 corresponding to the defragmentation reserved area corresponds to the cluster number 0000001Ch-0000001Dh-0000001Eh-0000001Fh, which corresponds to four consecutive clusters immediately before the area Ar2 corresponding to the additional data. It consists of FAT entries.
When attempting to secure a cluster as a defragmentation reserved area, it is necessary to store a value indicating that the FAT entry in the reservation corresponding area Ar4 corresponding to the defragmentation reserved area is not unused. You can think of it as a condition. Therefore, it can be said that any value that does not affect proper file management may be stored in the FAT entry in the reservation corresponding area Ar4 other than indicating that it is not used. However, in the present embodiment, the following contents are stored for the FAT entry in the reservation corresponding area Ar4 in consideration of, for example, the processing efficiency for subsequent update.

That is, the area Ar3 corresponding to the FAT1 defragmentation reserved area is stored in the cluster chain at this time in anticipation that the data before appending is moved and recorded in the second stage process. It is said.
Specifically, in the area Ar3 in this case, the FAT entry with the cluster number 0000001Ch at the head stores the chain destination cluster number 0000001Dh, and the FAT entry with the next cluster number 0000001Dh has the chain destination cluster number 0000001Eh. The chain destination cluster number 0000001Fh is stored in the FAT entry of the next cluster number 0000001Eh. In the area Ar3, the FAT entry with the cluster number 0000001Fh corresponding to the terminal cluster is stored by copying the same value as the FAT entry corresponding to the cluster number 00000014h serving as the terminal cluster of the pre-append data.

In addition, the directory entry of the update target file is also changed in accordance with the completion of the process up to the first stage for the postscript update. If it is the directory entry of the update target file corresponding to FIG. 12B, the contents are changed as shown in FIG. 14A to FIG. 14B.
First, since the processing up to the first stage is completed, the defragmentation reserved area is secured on the medium, so 1 is set for the defragmentation reserved area reservation flag. Therefore, the reserved area at the byte position Ch indicates 01h.
Although not shown here, when the additional recording is completed, the recording time and recording date information corresponding to the update date and time indicate that the processing up to the first stage is completed. It is changed according to. Further, the size is changed so that the actual data as additional write data is added.
In addition, the head cluster number indicates the head cluster of the data before append writing, but this cluster does not change before the append update and after the processing up to the first stage for append update is completed. Similarly to FIG. 14A, 00000011h is stored, and as a result, there is no change.

  Assume that the first cluster number of the directory entry shown in FIG. 14B is referred to, for example, in order to reproduce the update target file in the state where the processing up to the first stage has been completed. Then, since 00000011h is indicated as the first cluster number, the FAT entry corresponding to the cluster number 00000011h in FAT1 is accessed. That is, the FAT entry corresponding to the first cluster of the pre-append data in the update target file is accessed. Then, by following the cluster chain described in the FAT entry, the update target file is formed by concatenating clusters in the order of cluster number 00000011h-00000012h-00000013h-00000014h-00000020h-00000021h-00000022h-00000023h. Will be managed as being. Even in the state where the process up to the first stage of the file additional update is completed, the update target file is managed so that the pre-append data and the additional data are appropriately linked at the cluster level. Become.

FIG.12 (c) has shown the content of FAT2 corresponding to the said FIG.12 (b). That is, for the file to be updated before being updated as shown in FIG. 12A, the FAT2 after the processing up to the first stage (defragment reserved area recording and additional data recording) is completed. It becomes contents.
As described above, the basic use of FAT2 is a mirror area as a spare of FAT1 which is the main FAT area. As a general rule, the contents of FAT1 are copied, but updated and updated. With respect to the file, information having predetermined contents different from the chain number of the chain destination is stored in the FAT entry at the required position. Prior to the description of FIG. 12 (c), this will be described with reference to FIGS. 16 (b) and 16 (c).

FIG. 16B shows the definition contents of the FAT entry corresponding to the first cluster of the data before appending as the FAT entry in FAT2 at the stage where the processing up to the first stage is completed. As shown in this figure, in FAT2, if the FAT entry corresponds to the first cluster of the data before appending data, it is originally stored in the lower 28-bit area in which the chain destination cluster number is to be stored. On the other hand, the head cluster number of the defragmentation reserved area currently reserved for the update target file is stored.
In addition, a bit value of 1 is set as the end cluster instruction flag in the FAT entry corresponding to the first cluster of the data before appending on the FAT1 side when the processing up to the first stage is completed. Therefore, the FAT entry having the definition shown in FIG. 16B is used for the FAT2 side FAT entry corresponding to the same cluster number as the FAT1 side FAT entry in which 1 is set for the termination cluster instruction flag. It can also be defined as follows.

  FIG. 16C shows the definition contents of the FAT entry corresponding to the end cluster of the data before appending as the FAT entry in FAT2 at the stage where the processing up to the first stage is completed. In FAT2, when the FAT entry corresponds to the end cluster of the data before appending data, the size of the effective data actually stored in the area of the end cluster of the data before appending as shown in the figure Is stored. In this figure, the effective data size is expressed using all 32 bits forming the FAT entry, but the effective data size is expressed using a bit area of less than 32 bits at a predetermined bit position. You may make it do.

The FAT entry contents of the areas (areas Ar1, Ar2, Ar3) related to the update target file in FAT2 shown in FIG.
First, at this stage, the area Ar1 corresponding to the data before additional recording is a set of FAT entries corresponding to the cluster numbers 00000011h-00000012h-00000013h-00000014h, similarly to the area Ar1 shown in FIG.
Then, in the area Ar1 of FIG. 12C, the first cluster number of the defragmentation reserved area described with reference to FIG. 16B is stored as the FAT entry of the cluster number 00000011h corresponding to the first cluster of the data before additional recording. To be. In this case, since the cluster number of the first cluster in the defragmentation reserved area is 0000001Ch, this value is stored in the FAT entry of the cluster number 00000011h in FIG.
In the area Ar1 in FIG. 12C, the FAT entry corresponding to the cluster numbers 00000012h to 00000013h following the cluster number 00000011h is the content obtained by copying the values of the cluster numbers 00000012h to 00000013h in the FAT1.
Then, in the area Ar1 in FIG. 12C, the FAT entry corresponding to the cluster number 00000013h that is the terminal is the effective data that is actually stored in the terminal cluster of the data before appending as described with reference to FIG. Stores a value indicating the size of data.

  In FIG. 12C, the contents of the FAT entry in the area Ar3 shown in FIG. 12B are copied as they are for the area Ar2 corresponding to the additional write data.

  In the area Ar3 corresponding to the defragmentation reserved area in FIG. 12C, first, the contents of FAT1 shown in FIG. 12B are copied for the cluster numbers 0000001Ch-0000001Dh-0000001Eh from the top. . For the fourth cluster number 0000001Fh that is the end, 1 is set for the end cluster instruction flag in the FAT entry on the FAT1 side in FIG. In response to this, as shown in FIG. 12C, the value indicating the size of the effective data in the terminal cluster of the pre-append data shown in FIG. 16C is stored on the FAT2 side.

As described above, the FAT entry area corresponding to the update target file has the contents shown in FIG. 12B, so that even before the process up to the first stage is completed, before the addition at the cluster level. The connection between data and additional data is managed appropriately. However, simply managing the connection between the pre-append data and the post-append data at the cluster level may cause invalid data of the end cluster of the pre-append data to be included in the file data as described in FIG. This causes inconvenience.
In the present embodiment, as a countermeasure against this, as shown in FIG. 12C, the effective data size of the terminal cluster of the data before appending is stored in the FAT entry of the appropriate cluster in FAT2. .
That is, in the process of referring to FAT1 to execute file playback as processing of the file system 101, when referring to a FAT entry in which a bit value of 1 is set for the termination cluster indication flag, the same position on the FAT2 side is Reference is made to the FAT entry. That is, when referring to the FAT entry A in FIG. 12B, the FAT entry B in FIG. 12C is also referred to. Then, since the FAT entry B indicates the effective data size of the terminal cluster of the pre-append data, information on the effective data size is passed to the application 100. In the application 100, when processing the data stored in the end cluster of the pre-append data, the data up to the size indicated by the information on the effective data size is handled as the effective data and the processing is executed. Thereby, for example, as shown in FIG. 11, file data not including invalid data can be obtained.

Then, after the processing up to the first stage shown in FIGS. 12B and 12C is performed, the process as the second stage of moving and recording the pre-append data to the defragmentation reserved area is performed. The contents of FAT1 and FAT2 at the stage of termination are shown in FIGS. 13 (a) and 13 (b), respectively. Further, according to the processing as the second stage, the directory entry of the update target file is changed from the content shown in FIG. 14B to the content shown in FIG.
First, regarding the directory entry of the update target file, the value stored in the first cluster number is changed from 00000011h to 0000001Ch. In addition, as a result of moving the data before appending to the defragmentation reserved area, there is no state where the defragmentation reserved area is secured on the medium, so the defragmentation reserved area securing flag is changed to a bit value from 1 to 0. Will do.
Note that the recording time, recording date, size, and the like may be the same as the directory entry at the stage where the first stage in FIG. 14B is completed.

Then, as shown in FIG. 13A, the contents of the FAT1 after the end of the second stage processing first form an area Ar1 where the data before appending was managed in the stage until the end of the first stage. Each FAT entry stores a value indicating that it is unused. As a result, the four clusters with cluster numbers 00000011h-00000012h-00000013h-00000014h are managed as unused (free) areas.
As for the area Ar3, the four consecutive clusters having the cluster numbers 0000001Ch-0000001Dh-0000001Eh-0000001Fh, which are managed as the defragmentation reserved areas, are managed as the areas where the data before additional recording is recorded.

  In this area Ar3, first, a value indicating the cluster number of the chain destination is stored in each FAT entry corresponding to the cluster number 0000001Ch-0000001Dh-0000001Eh from the top to the third. In addition, the FAT entry corresponding to the cluster number 0000001Fh that terminates in the area Ar3 is set to 1 for the termination cluster instruction flag in accordance with the definition shown in FIG. In addition, the lower 28 bits indicate the cluster number that is the head of the additional data as the cluster number of the chain destination. As a result, 10000020h is stored in the FAT entry corresponding to the cluster number 0000001Fh.

  Further, the area Ar2 in FIG. 13A corresponds to the additionally recorded data in the same manner as the area Ar2 at the stage where the processing up to the first stage shown in FIG. Accordingly, the chain entry cluster number is stored in the FAT entry corresponding to the cluster number 00000020h-00000021h-00000022h, and the value indicating EOF is stored in the FAT entry corresponding to the cluster number 00000023h. In addition, the termination cluster indication flags in these FAT entries are all 0 bit values.

Here, when comparing FIG. 13A and FIG. 12B as the contents of FAT1 related to the update target file, first, FIG. 13A corresponds to the data before appending, and FIG. In a), all the FAT entries in the area Ar3 corresponding to the defragmentation reserved area have the same contents. In other words, although the function of the area Ar3 changes from managing the defragmentation reserved area to managing additional data, the stored contents are not changed. This means that the FAT entry area, which had previously managed the defragmentation reserved area, will not be rewritten when editing FAT1 in the second stage of processing. It will be.
Also, the FAT entry in the area Ar2 corresponding to the additional write data is not changed between FIG. 13A and FIG.
In FIG. 12B and FIG. 13A, only the area Ar1 is changed in content. As described above, the area Ar1 is rewritten from a state in which the cluster number indicating the cluster chain of the additional write data is stored to a state in which a value indicating unused is stored.
Thus, in the present embodiment, consideration is given to making the FAT1 editing process associated with the second stage process as efficient as possible.

Also, FAT2 after the end of the second stage processing is shown in FIG.
First, also in FAT2, a value indicating that the area Ar1 corresponding to the data before additional recording is unused in each FAT entry in the stage up to the end of the first stage is stored. That is, this area is a copy of the contents of FAT1 shown in FIG.

Also in FAT2, the area Ar3 corresponding to the defragmentation reserved area in the stage up to the end of the first stage corresponds to the data before appending according to FAT1. For this reason, in the area Ar3, the FAT entry corresponding to the cluster number 0000001Ch-0000001Dh-0000001Eh from the beginning to the third stores the same value as the FAT entry at the same position in FIG. Is done. However, for the FAT entry corresponding to the fourth terminal cluster number 0000001Fh, the cluster number 0000001Fh corresponds to the terminal cluster of the pre-append data (the FAT entry with the terminal cluster instruction flag = 1). According to the definition described in (c), a value indicating the effective data size (Size) in the terminal cluster of the data before appending is stored.
Further, the area Ar2 shown in FIG. 13B corresponds to the additional write data as in FIG. 13A, and the contents of the FAT entry in the area Ar2 on the FAT1 side are copied.
Again, when comparing the contents of FAT2 related to the update target file with FIG. 13B and FIG. 12C, the area Ar1 corresponding to the data before the additional recording in the stage up to the end of the first stage. For the other areas Ar2 and Ar3, the FAT entry contents are not changed.

  After the processing in the second stage is completed in this way, the update target file is connected in accordance with the ascending order of the cluster numbers using these eight consecutive clusters from the cluster numbers 0000001Ch to 00000023h. Thus, it is managed as being formed. Also in this case, when referring to the FAT entry in which the termination cluster instruction flag = 1 is set in FAT1, the valid data size can be recognized by referring to the FAT entry of FAT2 at the same position. In other words, as in the case where the processing up to the first stage is finished, the data processing can be appropriately executed so that invalid data is not included in the middle of the file data.

The flowchart in FIG. 17 illustrates processing that is executed by the CPU 10 in order to realize the first-stage processing for additionally updating a file. For confirmation, the first stage process is a process for securing a defragmentation reserved area and recording additional data.
As described above, the processing shown in this figure is performed by programs such as an OS (Operating System) and a predetermined application 100 stored by the CPU 10 installed in the ROM 11 (or a medium such as an HDD). This is realized by execution, and media control such as data processing for media is processing under the FAT file system hierarchical structure shown in FIG. This also applies to the flowcharts described below.

In the file additional update, first, as shown in step S101, the current file additional update recognizes the size tksz of the additional data to be additionally recorded following the file data before the update. In the next step S102, the size tmsz of the free area in the end cluster is recognized among the clusters storing the pre-append data that is the file data before the update.
As described above, if the size of valid data as a file is not an integer multiple of the cluster size, the end of valid data is located in the middle of the cluster in the end cluster in the cluster that stores the file. The valid data end to the cluster end are filled with invalid data. The size of this invalid data is the size tmsz of the free area here. Accordingly, in calculating the size tmsz of the free area, for example, first, a remainder obtained by dividing the file data before update (data before appending) by the cluster size is obtained. This remainder value is the effective data size in the terminal cluster of the data before additional recording. Therefore, if the remainder value is subtracted from the cluster size, the size tmsz of the free area can be obtained.

In step S103, it is determined whether or not tksz> tmsz is satisfied with respect to the size tksz and the free capacity size tmsz of the additional data recognized in steps S101 and S102.
Here, if the size tksz of the additional recording data is less than the free capacity size tmsz and a negative determination result is obtained, it is possible to record all of the additional recording data this time using the free capacity of the end cluster of the data before additional recording. It will be. Therefore, in this case, the process proceeds to step S112.
In step S112, in the terminal cluster of file data (data before additional recording) serving as an update source, processing for recording additional recording data is performed so as to be linked to the end of valid data. For this purpose, for example, the end cluster of the data before appending is read from the medium and expanded in the work RAM, and the appending data is connected on the work RAM following the end of the valid data of the data before appending. Update the data contents. Then, a process for writing the data updated in this way to the medium as the data of the end cluster of the original pre-additional data is executed. Further, according to the processing result, for example, necessary information such as the size and update date and time in the directory entry for the update target file is also changed.

  On the other hand, when a negative determination result is obtained in step S103 that tksz> tmsz is established, additional data cannot be additionally recorded on the terminal cluster of the data before additional data as described above. In this case, the process proceeds to step S104 and subsequent steps. The processing from step S104 to step S111 corresponds to the first stage processing for additional writing update according to the present embodiment.

  In step S104, first, the number of clusters numcr required as a defragmentation reserved area is calculated. For this purpose, the size of the update source file data (data before appending) is divided by the cluster size. And when the remainder does not occur as a result of the division (that is, when the size tksz of the additional data is an integral multiple of the cluster size), the division value is set as the number of clusters numcr, but when the remainder occurs, The value obtained by adding 1 to the division value is set as the number of clusters numcr.

In the next step S105, a process for securing an update file area is executed. For this purpose, first, the number of clusters necessary for the update file area is obtained. The number of clusters in the update file area is obtained by the sum of the number of clusters used by the data before appending, that is, the number of clusters numrc in the reserved area for defragmentation and the number of clusters necessary for storing the appending data. The number of clusters necessary for storing the additional write data can be obtained by dividing the size tksz of the additional write data by the cluster size in accordance with step S104.
When the number of clusters in the update file area is obtained in this way, the contents of the FAT1 at the current stage before executing the postscript update are referred to, and this number of clusters continuously becomes an unused area. Searches for available space. The free space thus found is set as an update file area, thereby ensuring the area.

In the next step S106, the FAT entry corresponding to the end cluster of the pre-append data in the FAT area (FAT1, FAT2) is edited as follows.
First, for the FAT entry of the end cluster of the pre-append data in FAT1, 1 is set for the end cluster instruction flag.
Further, for the FAT entry of the end cluster of the pre-append data in the same FAT1, the start cluster number of the append data is stored as the lower 28-bit chain destination cluster number. The top cluster of the additional data is the cluster number numcr + 1 of the reserved area for defragmentation from the top cluster in the cluster group forming the update file area secured in step S105. The cluster number of this cluster may be stored as the first cluster number of the additional data.

  In step S106, the effective data size (Size) of the end cluster of the data before appending is stored for the FAT entry of the end cluster of the data before appending in FAT2. For this purpose, first, an effective data size Size is acquired. The method of calculating the effective data size Size is as described in the previous step S102. Further, the effective data size obtained in the process of calculating the free space size tmsz in step S102 is retained, and the retained data is read to obtain the effective data size in this step S106. It is good also as processing. Then, the value of the effective data size Size obtained in this way is written and stored in the FAT entry of the terminal cluster of the data before appending in FAT2.

In the next step S107, a process for securing a defragmentation reserved area on the medium is executed.
For this purpose, first, in the cluster group forming the update file area assumed in the previous step S105, continuous clusters corresponding to the number of clusters numrc from the first cluster are changed to the cluster group for the defragmentation reserved area. Determine as. In the example of FIG. 12, the cluster numbers 0000001Ch-0000001Dh-0000001Eh-0000001Fh are determined as the cluster group for the defragmentation reserved area. For each FAT entry of FAT1 corresponding to these cluster groups, the cluster number incremented by 1 with respect to the current cluster is stored as the chain number of the chain destination. As a result of this processing, the FAT1 FAT entry corresponding to the final cluster in the defragmentation reserved area indicates the first cluster of additional data following the defragmentation reserved area as the chain destination cluster. At this stage, 1 is not set for the termination cluster instruction flag in the FAT1 FAT entry corresponding to the final cluster in the defragmentation reserved area.

  In the next step S108, the FAT entry corresponding to the first cluster of the pre-append data is edited. As this processing, editing is performed on the first cluster of the data before additional data in FAT2. As described with reference to FIG. 16B, the FAT entry corresponding to the leading cluster of the pre-append data in FAT2 is a FAT entry that stores the leading cluster in the defragmentation reserved area. In step S108, the leading cluster number of the defragmentation reserved area is stored for this FAT entry.

In step S109, the FAT entry corresponding to the terminal cluster of the defragmentation reserved area is edited. In the example of FIG. 12, the FAT entry corresponds to the cluster number 0000001Fh.
As the processing of step S109, first, for the FAT entry on the FAT1 side, 1 is set for the termination cluster instruction flag. For the FAT2 side FAT entry, the effective data size Size of the end cluster of the pre-append data is stored.

  In the next step S110, a process for writing additional data after the end cluster of the defragmentation reserved area and storing it in the medium is executed. In step S110, the FAT entry (FAT1, FAT2) corresponding to the cluster storing the additional write data is edited in accordance with the write process of the additional write data. That is, the FAT entry corresponding to the cluster other than the terminal cluster in the FAT entry group of the cluster storing the additional write data, the chain destination is set so that the cluster number incremented by 1 with respect to the current cluster becomes the chain destination. Stores the cluster number. A value indicating EOF is stored in the end cluster of the postscript data.

In step S111, editing for changing required contents in the directory entry of the additional update target file is executed according to the current update result.
First, 1 is set for the defragmentation reserved area reservation flag. Also, the size is changed to a value obtained by adding the additional data, and information corresponding to the update date is updated.

Next, a process that is executed by the CPU 10 in order to realize the process of additionally writing a file as the second stage will be described with reference to FIG. In the second stage process, after the process up to the first stage is completed, the pre-append data of the update target file is moved to the cluster area reserved as the defragmentation reserved area on the medium. By performing the recording, it becomes a process for eliminating the fragment of the additional update target file.
As the second stage process, first, in step S201, the process waits for the timing to start executing the pre-append data movement process, which is the second stage process. As described above, the condition for satisfying the timing for executing the second stage processing is that the processing load on the CPU 10 is below a certain level, and data writing / reading to / from the medium can be executed stably and at high speed. Can be mentioned. If a positive determination result is obtained in step S201, the processes in and after step S202 are executed.

In step S202, one required directory entry is accessed according to a predetermined order, and in the next step S203, it is determined whether or not 1 is set for the defragmentation reserved area reservation flag in the accessed directory entry. To do. Here, if a negative determination result is obtained assuming that 0 is set for the defragmentation reserved area reservation flag, the defragmentation reserved area is not secured as a file corresponding to the directory entry, and accordingly This means that the file management state is not in the stage where the postscript update process up to the first stage is completed. Therefore, in this case, the processing shown in this figure is once ended, and the processing returns to step S201. If a positive determination result is obtained in step S201, the process proceeds to step S202 again to access a directory entry that has not yet been accessed. In this manner, the directory entries are sequentially accessed until a directory entry in which 1 is set for the defragmentation reserved area reservation flag is searched in steps S201 → S202 → S203.
Then, in step S203, if a positive determination result is obtained because 1 is set for the defragmentation reserved area in the directory entry accessed this time, the file corresponding to the directory entry is first updated as an additional update. The process up to the stage has only been completed, and the data before additional recording and the additional recording data are still fragmented. Therefore, the processing from step S204 is executed as processing for defragmenting the update target file.

  In step S204, the FAT entry of FAT2 corresponding to the first cluster of the data before appending is accessed. The FAT entry stores the top cluster number of the defragmentation reserved area. In step S204, the top cluster number of the defragmentation reserved area is acquired.

In the next step S205, the pre-append data is recorded in the cluster area that has been reserved as a defragmentation reserved area on the medium.
For this purpose, first, the data before additional recording managed as being recorded in the same cluster as the data file before update is read from the medium and held in the work RAM. Then, the cluster indicated by the cluster number acquired in step S204 is set as the start cluster, and thereafter, the additional data held in the work RAM is sequentially filled in the cluster with the number incremented by one. Go. That is, the process of step S205 is a process of copying data recorded in the cluster area at the same position as before the update to the cluster area managed as the defragmentation reserved area.

  In step S206, the directory entry is edited in response to the completion of the process in step S205. In editing the directory entry, first, the head cluster number is changed to indicate the head cluster of the data before appending copied in step S205 (the head cluster of the defragmentation reserved area before processing in step S205). . Further, 0 is set (reset) for the defragmentation reserved area reservation flag.

In step S207, the FAT area is edited according to the processing result in step S205. As the editing of the FAT area, for both FAT1 and FAT2, before the processing of step S205, a value indicating unused is stored for each of the FAT entries corresponding to the cluster area in which the data before the additional recording is recorded. To do.
As a result of the processing so far, as the contents of the FAT area (FAT1, FAT2), for example, as shown as a transition from 12 (a) (b) to FIG. 13 (a) (b), as the update target file, On the media, the fragment of the pre-append data and post-append data is resolved, and a state in which data is continuously recorded in cluster units is obtained, and the update target consisting of the pre-append data and the append data is also obtained by the FAT file system. It is possible to obtain a state where the file is properly managed as being formed by a cluster chain that is continuous in cluster number.

Next, with reference to a flowchart of FIG. 19, a process for reproducing a file corresponding to the additional update file according to the present embodiment will be described.
For example, if an instruction to reproduce a file is obtained, first, in step S301, a directory entry of a file designated as a reproduction target (reproduction target file) is accessed and the contents thereof are referred to. Then, as the processing of the next step S302, the first cluster of the file to be reproduced is accessed according to the first cluster number stored in the directory entry, and data reading in this cluster unit is executed. The data of one cluster size read in this way is held in, for example, a work RAM.

In the next step S303, the FAT entry of FAT1 corresponding to the read execution cluster is accessed. The “read execution cluster” here refers to a cluster in which data is finally read by accessing the medium at the stage of executing step S303 at the present time. Accordingly, in step S303, which is executed first after step S302, the read execution cluster is the first cluster of the file to be reproduced.
When the FAT entry of FAT1 corresponding to the read execution cluster is accessed, the chain number of the chain destination stored here is referred to.

  In step S304, it is determined whether or not the referenced cluster number of the chain destination is a value indicating EOF. Here, if a negative determination result is obtained, there will be a cluster to be chained after the read execution cluster. In this case, the process proceeds to step S305 and subsequent steps.

In step S305, the termination cluster instruction flag area in the FAT1 FAT entry corresponding to the read execution cluster is referred to, and in step S306, whether or not the bit value of the referred termination cluster instruction flag is 1 is referred to. .
Here, assuming that the bit value of the end cluster instruction flag is 0 instead of 1, the result of negative determination in step S306 is that the current read execution cluster is the end cluster (EOF) of the additional data. In other words, it is not the end cluster of the data part before the additional recording. This means that the current read execution cluster is always filled with valid data for the cluster size, and there is no invalid data.
Therefore, in this case, in the processing of step S307, the data for one cluster size (read data) read this time as the read execution cluster (read data) is used as the upper processing layer (for example, the application 100) with the same one cluster size as the effective data. To the next level.

On the other hand, if a positive determination result is obtained in step S306, the current read execution cluster is the terminal cluster of the data before appending. In this case, in step S308, the FAT2 FAT entry corresponding to the current read execution cluster is referred to recognize the effective data size (Size) stored in the end cluster of the data before appending.
In step S309, of the data for one cluster read this time as the read execution cluster, the upper data only indicated by the effective data size (Size) is transferred to the upper processing layer as the effective data.

After the processing in step S307 or step S309 is completed, in step S310, the chain of the chain destination number referenced in the previous step S303 is accessed, and data reading is executed in units of clusters. To return to. When the process reaches step S303 via step S310, the cluster from which data has been read by the process of the last step S310 becomes the read execution cluster in step S303.
In the course of the process described so far, the process from step S308 to S309 is executed in response to a positive determination result in step S306, and as described with reference to FIG. For example, at the application level, a file can be formed as file data in which invalid data does not exist in the middle of the file.

As the processes in steps S303 to S310 are repeated, finally, EOF is detected in step S304, and a positive determination result is obtained. In this case, the process proceeds to step S311 to recognize the effective data size Size_eof of the read execution cluster (file end cluster) at this time. For this purpose, for example, the size of the file indicated in the directory entry is divided by the cluster size, and the remainder is obtained as Size_eof.
In step S312, among the data for one cluster read this time as the read execution cluster, the upper data only indicated by the size (Size_eof) is transferred to the upper processing layer as valid data, and the file reproduction process is terminated. .

  For reference, in the file system hierarchical model shown in FIG. 2, a configuration example showing the internal functions of the file system 101 in more detail is shown in FIG. In this figure, as in FIG. 2, the entire system hierarchy model including the application 100, the file system 101, the device driver 102, and the medium 103 is shown.

  As shown in this figure, when the processing executed by software as the file system 101 is viewed as a functional block, it can be considered that the file system 101 includes the recording control unit 200 and the media control unit 210.

  The recording control unit 200 executes various control processes related to recording (writing) and reading of data with respect to a medium, and includes a directory entry control unit 201, a FAT control unit 202, and a cluster control unit 203. Further, the media control unit 210 includes a position calculation unit 211 as a functional part capable of exchanging data with the device driver 102 in order for the file system to control the media.

The directory entry control unit 201 in the recording control unit 200 performs various necessary control processes for the directory entry, such as creating, deleting, and updating the directory entry in accordance with the data processing results such as writing and erasing files on the medium. Execute. The FAT control unit 202 executes various necessary control processes for the FAT area, such as rewriting a predetermined FAT entry in the FAT area in accordance with the file data processing result for the medium. The cluster control unit 203 executes various control processes for causing processes in the file system to be executed at the cluster level.
For example, if data to be written to the medium is transferred from the application 100 as file-based processing, the cluster control unit 203 converts the file level data into data at the cluster level. Then, it instructs the position calculation unit 211 of the media control unit 210 to write the data at the cluster level to the medium.
The position calculation unit 211 calculates the position (cluster number) of the cluster to which the file data is to be written from the unused area of the media recognized by referring to the contents of the FAT area managed by the FAT control unit 202. To do. Then, the cluster unit data to be written is transferred to the device driver 102 and the cluster number to which this data is to be written is instructed. The device driver 102 finally converts the instructed cluster number into an address of a physical sector in a storage area on the medium, and executes data writing to the medium at the sector level.

At the same time, the file system rewrites (updates) the directory entry and the FAT area so that the file written and stored in the medium as described above is properly managed on the FAT file system. .
For example, first, the directory entry control unit 201 creates a directory entry for the file to be recorded this time. At this time, in cooperation with the FAT control unit 202, a cluster in which a file is to be recorded is determined from the free area in the FAT area. Along with this, the value stored in the starting cluster in the directory entry is also determined. Further, in response to the determination of the cluster on which the file is to be recorded, the FAT control unit 202 sets the FAT entry of the required cluster number so that the contents such as the cluster chain for the file are indicated in the FAT area. Rewrite the value.
Then, the directory entry created in this way and the rewritten contents of the FAT area are processed by the cluster control unit 203 and the media control unit 210 (position calculation unit 211) in the same manner as described above, so that the predetermined area of the media Write to and store.

  17 and 18 and the file playback process shown in FIG. 19 are mainly performed for data recording on the medium, directory entry and FAT area recorded on the medium. Editing. Therefore, the processing shown in these figures is mainly executed by the file system in response to an instruction from the file recording / playback application 100.

As described above, according to the present embodiment, in the file additional update, first, as a first step process, defragmentation is performed on the area immediately before the empty area where the additional write data is to be recorded. Defragmentation is resolved by recording the pre-addition data so that it is moved to the defragmentation reservation area at the timing after the additional write data is recorded as a second stage process. Like to do. The file management example as the present embodiment for the additional update target file corresponding to this processing has been described with reference to FIGS. 12 to 16. FIG. 12 shows how the actual file management is performed. The configuration described with reference to FIG.
Further, although the FAT file system is adopted for file management in the embodiment, the present invention can be applied to other file systems such as HSF (Hierarchical File System). There are no particular restrictions on the file system or file management standard.
In addition, the postscript update process of the present invention is particularly effective when a file with a high data rate, such as a moving image file of various formats, is considered in light of recent circumstances, The format of the file to be subjected to the additional writing update according to the present invention should not be particularly limited.
The present invention can also be applied to unit data in a format that cannot be handled as a file.

Further, the effect of improving the data processing speed due to the elimination of the fragment of the additional update target file based on the present invention is, for example, among those connected to the media controller 13 in FIG. This is remarkably obtained on a medium of a type in which the relative position between the head and the data recording surface is physically moved during access, such as a magneto-optical disk. However, for example, for semiconductor memory devices, file data defragmentation at the recording area unit level equivalent to a cluster causes overhead such as heavy address conversion processing during data processing such as file reproduction. In view of the possibility that the data processing speed may decrease, the present invention can be applied to a case where a file is updated for a medium other than the type accessed by the head.
As for the relationship between the media and the host side device, for example, in addition to media connected using physical connection means such as cables and connectors, it is connected to the host wirelessly via Bluetooth, wireless LAN (Local Area Network), etc. It is also possible to target media.
Further, the information processing apparatus of the present invention should not be limited to the digital video camera described in the embodiment, and includes, for example, a digital still camera, a reserved recording device for a television broadcast program, and a personal computer. The present invention can be applied to any device capable of data processing such as data writing / reading corresponding to various storage media, such as various information processing devices conforming thereto.

It is a block diagram which shows the structural example of the digital video camera of embodiment of this invention. It is a figure which shows the system configuration | structure of a FAT file system by a hierarchical model. It is a figure which shows the format structure of the medium in a FAT file system. It is a figure which shows the structure of MBR in a FAT file system. It is a figure which shows the structure of the directory entry in a FAT file system. It is a figure which shows the example of management of the file recording position by a FAT file system. It is a figure which shows typically the processing procedure of the additional write update of the file by the file management content by a FAT file system. It is a figure which shows typically the process procedure (1st step) of the additional writing update of the file as this Embodiment by the file management content by a FAT file system. It is a figure which shows typically the processing procedure (2nd step) of the additional writing update of a file as this Embodiment by the file management content by a FAT file system. FIG. 10 is a diagram for explaining the reason why invalid data is included in the midway position of file data when a file is additionally written and updated according to the present embodiment. It is a figure which shows typically the procedure for setting it as the file data structure except an invalid data about the file of the postscript update object of this Embodiment. It is a figure which shows the example of the content of the FAT area | region corresponding to the postscript update process (it complete | finishes the 1st step) of a file as this Embodiment. It is a figure which shows the example of the content of the FAT area | region corresponding to the postscript update process (2nd step completion | finish) of a file as this Embodiment. It is a figure which shows the example of the content about the directory entry of the additional recording update object file corresponding to progress of the additional recording update process of the file of this Embodiment. It is a figure which shows the example of a definition of the reservation area reservation flag for defragmentation in embodiment. It is a figure which shows the structure of the FAT entry defined corresponding to the postscript update process of the file of embodiment. It is a flowchart which shows the process for implement | achieving the postscript update (1st step) of a file as this Embodiment. It is a flowchart which shows the process for implement | achieving additional update (2nd step) of the file as this Embodiment. It is a flowchart which shows the process for implement | achieving the file reproduction | regeneration processing as this Embodiment. It is a figure which shows the processing function in the hierarchy of a file system.

Explanation of symbols

  1 digital video camera, 2 optical system unit, 3 photoelectric conversion unit, 4 video signal processing unit, 5 image input / output unit, 6 camera function unit, 7 display unit, 8 audio processing unit, 9 audio input / output unit, 10 CPU, 11 ROM, 12 RAM, 12a nonvolatile memory, 13 media controller, 15 operation input unit, 16 communication unit, 100 application, 101 file system, 102 device driver, 200 recording control unit, 201 directory entry, 202 FAT control unit, 203 Cluster control unit, 210 media control unit, 211 position calculation unit

Claims (7)

Recording means for recording data on a storage medium;
Management means for managing data stored in the storage medium using predetermined unit data;
The update target data in the storage medium can be written when the content of the unit data is updated by adding additional data following the update target data that has been stored in the storage medium. A reserved area securing means for securing a free area having a sufficient capacity as a reserved area for the update target data;
Additional recording data recording means for recording the additional recording data in a storage area following the reserved area by the recording means;
A primary update unit that executes processing for managing the unit data after the update so that the additional data is connected to the update target data to be formed by the management unit.
Update target data recording means for recording the update target data in the reserved area by the recording means after the processing of the primary update means is completed;
The update target data recording means executes processing for managing by the management means on the assumption that the additional data is formed after the update target data recorded in the reserved area. Secondary update means;
An information processing apparatus comprising: The recording means is
The data is recorded by a predetermined minimum recording data unit, and
The management means is
The unit data is managed as being formed by concatenating one or more minimum recording data units,
The primary update means and the secondary update means are:
Executing the process for managing by the management means as the additional data is connected to the end of the effective data in the minimum recording data unit constituting the update target data;
The information processing apparatus according to claim 1. The management means is
A main table formed by a set of instruction areas corresponding to each of the minimum recording data units on the storage medium, and storing information indicating the minimum recording data unit to be connected next in the instruction area By providing information, the unit data is managed as being formed by concatenation of one or more minimum recording data units, and sub-table information having the same structure as the main table information is provided.
The primary update means and the secondary update means are:
Stores the size of valid data in the last minimum recording data unit constituting the update target data for a predetermined instruction area determined corresponding to the management content of the main table information corresponding to the update target data,
The information processing apparatus according to claim 2. The recording means is
The data is recorded by a predetermined minimum recording data unit, and
The management means is
A main table formed by a set of instruction areas corresponding to each of the minimum recording data units on the storage medium, and storing information indicating the minimum recording data unit to be connected next in the instruction area By providing information, the unit data is managed as being formed by concatenation of one or more minimum recording data units, and sub-table information having the same structure as the main table information is provided.
The reserved area securing means is
For the main table information, the management means is controlled so that necessary information for storing the indication area of the minimum recording data unit corresponding to the reserved area as an unused area is stored. ,
As for the sub table information, information indicating the position of the reserved area is stored with respect to a predetermined instruction area determined corresponding to the management content of the main table information corresponding to the update target data. Controlling the management means,
The information processing apparatus according to claim 1. The management means is
It is provided for each unit data, and is provided with entry information storing predetermined information to be a starting point for accessing one unit data.
Whether the reserved area for the update target data in the corresponding unit data is secured in accordance with the processing result of the primary update means and the secondary update means, stored in a predetermined position in the structure of the entry information. Further comprising an instruction information management means for changing the reservation state instruction information indicating whether or not
The information processing apparatus according to claim 1. A recording procedure for recording data on a storage medium;
A management procedure for managing data stored in the storage medium using predetermined unit data;
The update target data in the storage medium can be written when the content of the unit data is updated by adding additional data following the update target data that has been stored in the storage medium. A reserved area securing procedure for ensuring that a free area with as much capacity as a reserved area for the update target data;
The additional recording data recording procedure for recording the additional recording data in the storage area following the reserved area by the recording procedure;
A first update procedure for executing a process for managing the unit data after the update as the data to be managed by the management procedure, as the additional data is connected to the update target data.
An update target data recording procedure for recording the update target data in the reserved area by the recording procedure after the processing as the first update procedure is completed;
As a result of the update target data recording procedure, the update target data recorded in the reserved area is followed by the additional write data, and the process for managing by the management procedure is executed. A second update procedure;
The information processing method characterized by performing. A recording procedure for recording data on a storage medium;
A management procedure for managing data stored in the storage medium using predetermined unit data;
The update target data in the storage medium can be written when the content of the unit data is updated by adding additional data following the update target data that has been stored in the storage medium. A reserved area securing procedure for ensuring that a free area with as much capacity as a reserved area for the update target data;
The additional recording data recording procedure for recording the additional recording data in the storage area following the reserved area by the recording procedure;
A first update procedure for executing a process for managing the unit data after the update as the data to be managed by the management procedure, as the additional data is connected to the update target data.
An update target data recording procedure for recording the update target data in the reserved area by the recording procedure after the processing as the first update procedure is completed;
As a result of the update target data recording procedure, the update target data recorded in the reserved area is followed by the additional write data, and the process for managing by the management procedure is executed. A second update procedure;
A program characterized by causing an information processing apparatus to execute.

Images