JP4502689B2 - Data storage device and initialization method thereof - Google Patents

Description

  The present invention relates to a data storage device for storing computer data, such as an IC memory, and more particularly to a data recording method in the data storage device.

  Conventionally, in this type of data storage device, FAT is often used as a file system. When using the FAT, the storage area in the data storage device is divided into two areas, a file management area and an actual data storage area. In the file management area, directory entries indicating the location of actual data managed in the form of files and FAT (file allocation table) information are recorded, and the computer refers to these pieces of information to obtain actual data. A file recorded in the storage area can be accessed.

  In addition, as described in Patent Document 1 below, in a conventional data storage device, a table for recording an address at which a defective area exists is prepared in case a physical defect occurs in the storage area. It is prepared at a specific position in the storage area. When the data storage device finds a defective storage area, it writes the information in this table. The data storage device refers to this table when writing new data according to a command from the computer, thereby preventing data from being written to the missing area.

Japanese Patent No. 3226042

  However, in the above-described conventional data storage device, the directory entry and FAT information in the file management area are rewritten whenever a file is written or erased from the computer. For this reason, there is a problem that access concentrates on a specific area called a file management area, and physical loss tends to occur in such area. Further, in the technique described in Patent Document 1, since it is necessary to prepare a table for recording the address of the missing area, it is difficult to efficiently use the storage area in the data storage device. Such a problem is not limited to the IC memory, but is common to other data storage devices such as a hard disk.

  The present invention has been made in view of such problems, and an object thereof is to suppress concentration of access to a specific area in a data storage device and to efficiently use the entire storage area in the data storage device. And

In order to achieve the above object, the data storage device of the present invention is configured as follows. That is, a data storage device that stores data of a computer, a storage unit that stores the data, and a control unit that controls writing and reading of data to and from the storage unit according to instructions from the computer The storage unit includes a storage area partitioned into a data block in which an entity of data is recorded and an empty area block in which an entity of data is not recorded, and the data block is a location of another data block. A data management information storage area for recording an address is integrally held with the data block, and the empty area block has an empty area management information storage area for recording a location address of another empty area block. the free area blocks are integrally holds and, in free area management information storage area of the air-region block, the spatial region Bed Information that can acquire the size of the empty area block is recorded, and the location address recorded in the empty area management information storage area of the empty area block and the information that can acquire the size of the empty area block are stored in the storage area. The gist is that it is set so as to avoid a missing area in the part.

  According to the data storage device of the present invention, each data block recorded in the storage unit is linked in a chain manner by the location address recorded in the data management information storage area. Therefore, the target data can be accessed by following the location address. By adopting such a configuration, it is possible to manage the recording position of data without using management information recorded in a specific area such as a directory entry or FAT information, thereby suppressing concentration of access to the specific area. can do. In the present invention, the empty area is also managed by the same structure as the data block. Therefore, even when there are a plurality of empty areas, the empty areas are linked to each other, so that the position of the empty area can be easily grasped.

  In the empty area management information storage area of the empty area block, information capable of acquiring the size of the empty area block may be further recorded. In this way, the size of the empty area can be easily grasped. The information capable of acquiring the size of the empty area block is, for example, information indicating the size of the empty area block itself or information indicating the end address of the empty area block. The end address may be specified by an absolute address, or may be specified by a relative address such as an offset address.

  In such a configuration, the location address of the empty area block in the empty area management information storage area and the information from which the size of the empty area block can be acquired avoid the missing area existing in the storage unit. It is good also as what is set to. This makes it possible to write and read data avoiding the missing area without preparing a separate table for storing address information of the missing area. Therefore, the entire recording area in the data storage device can be used efficiently.

  In the data management information storage area of the data block, type information indicating the type of the data block may be further recorded. As the type information, for example, a parameter indicating any one of a file, a directory (folder), and a volume can be recorded. By doing so, it becomes possible to easily grasp the type of data recorded in the storage area.

  When a parameter representing a directory is recorded as the type information in the data management information storage area of the data block, a file or directory existing in the directory is further stored in any area of the data block. The location address of the stored data block may be recorded. With such a configuration, it is possible to construct a data block link for each directory.

  The control unit, when receiving a data write command from the computer, arranges a new data block in which the data is recorded in any area where the empty area block exists, and manages data of the data block Record the location address of another data block in the information storage area, and record the location address of the new data block in relation to the location address recorded in the data management information storage area of the other data block. Can be. In this way, data can be written to the storage unit.

  Further, the control unit further records the location address of the empty area block in the empty area management information storage area when the location address of the empty area block changes due to the arrangement of the new data block. The location address after the change may be recorded in the empty area management information storage area of the empty area block. By doing so, the link state of the empty area in the storage unit can be managed.

  The present invention can also be configured as an initialization method for a data storage device that stores computer data. That is, a step of recognizing a missing area in a data storage device, a step of segmenting an empty area in the data storage device based on the missing area, and a predetermined area of each segmented empty area, This is an initialization method including a step of recording information that can acquire the size of the empty area and the location address of another segmented empty area.

  Each empty area segmented by the initialization method corresponds to the above-described empty area block. According to such an initialization method, empty area blocks can be arranged in advance so as to avoid missing areas before using the data storage device. As a method for recognizing a missing area in a data storage device, for example, writing or erasing data in the entire area in the data storage device and detecting a bit that does not change can be recognized. it can. Further, it may be recognized by referring to the information regarding the missing area provided by the memory manufacturer of the memory built in the data storage device.

  The data storage device of the present invention can be applied to a single auxiliary storage device such as an IC memory or a hard disk. The present invention can also be applied to the entire computer system including the computer main body and the auxiliary storage device. Further, the present invention can be configured not only as the above-described data storage device but also as, for example, a data recording method in the data storage device or a computer program for storing data. The computer program may be recorded on a computer-readable recording medium. As the recording medium, for example, various media such as a flexible disk, a CD-ROM, a DVD-ROM, a magneto-optical disk, a memory card, and a hard disk can be used.

Hereinafter, embodiments of the present invention will be described in the following order based on examples.
A. Data storage device configuration:
B. How to record data:
C. File access processing:
(C1) Read processing:
(C2) Erasing process:
(C3) Write processing:
D. Other examples:

A. Data storage device configuration:
FIG. 1 is an explanatory diagram showing a configuration of a data storage device 100 as an embodiment of the present invention. The data storage device 100 is an auxiliary storage device for storing data of a personal computer (hereinafter simply referred to as “computer”), and includes a USB connector 110 connected to a USB port of the computer, a USB device controller 120, The control unit 130 includes a flash memory 140 that stores data in a nonvolatile manner.

  The USB device controller 120 mediates transmission and reception of data and commands between the computer and the data storage device 100 by performing communication conforming to the USB (Universal Serious Bus) standard with a USB host controller provided in the computer. In this embodiment, the communication with the computer is performed based on the USB standard as described above. However, the communication may be performed based on another standard such as the ATA standard or the IEEE 1394 standard.

  The control unit 130 controls writing and reading of data with respect to the flash memory 140 according to a command received from the computer via the USB device controller 120. The control unit 130 includes a CPU 131, a RAM 132, and a ROM 133, and a program for controlling the operation of the data storage device 100 is recorded in the ROM 133. The CPU 131 controls data writing and reading by executing this program while using the RAM 132 as a work area.

  The flash memory 140 is a NAND flash memory, and in this embodiment, has a storage capacity of 32 Mbytes. The flash memory 140 includes a 512-byte data register for temporarily storing data to be written and read, and 2048 block areas. One block area is further divided into areas of 32 pages and has a storage capacity of 512 bytes per page.

B. How to record data:
FIG. 2 is an explanatory diagram showing an image of the address space in the flash memory 140. In the present embodiment, as shown in FIG. 2, the control unit 130 handles the address space in the flash memory 140 in a ring shape. That is, when one address is added to the end address MX, the address after the addition becomes the head address 0.

  FIG. 3 is an explanatory diagram showing the structure of data stored in this address space. The data storage device 100 of this embodiment stores all data in the address space in the format of “data block” shown in FIG. This data block is a word different from the “block area” shown in FIG.

  The data block integrally includes a fixed-length management information storage area MA and a variable-length data storage area DA. For example, when receiving a file write command from the computer, the control unit 130 records the file in the data storage area DA and records predetermined information described later in the management information storage area MA. A data block composed of these pieces of information is arranged in the address space.

  In the management information storage area MA, the forward link address Lf and the backward link address Lb, the type information Fa indicating the type of the data block, and the size Sd of the entire data block are recorded. These pieces of information are all fixed length information. The end address of the data block may be recorded instead of the size Sd. In this case, the end address may be recorded as an absolute address, or may be recorded as an offset address from the head address of the data block.

  The leading address of the data block existing before the data block is recorded in the forward link address Lf, and the leading address of the data block existing after the data block is recorded in the backward link address Lb. “Before” and “after” do not mean before and after the recording position in the address space, but the order in which the data blocks are linked by these link addresses.

  FIG. 4 is an explanatory diagram showing an example of the link state of the data block by the link addresses Lf and Lb. For example, “a3” that is the head address of the data block 3 is recorded in the forward link address Lf of the data block 2 shown in FIG. 4, and “a1” that is the head address of the data block 1 is recorded in the backward link address Lb. Is recorded. Therefore, in the example shown in FIG. 4, each data block is linked in the order of data block 1, data block 2, and data block 3. The control unit 130 can access a desired data block by moving the access position in the address space with reference to such link addresses Lf and Lb.

  The head address a2 of the data block 2 in FIG. 4 is referred to by the forward link address Lf of the data block 1, and further referred to by the backward link address Lb of the data block 3. That is, each data block is linked bidirectionally. As described above, if the individual data blocks are linked in both directions, the searchability of the data is improved, and the validity of the link can be easily verified.

  In the type information Fa shown in FIG. 3, a parameter indicating the type of data block is recorded. Types of data blocks include volume blocks, file blocks, directory blocks, and empty area blocks, and parameters representing each are “fv”, “ff”, “fd”, and “fn”.

  5 to 8 show the data structure for each type of data block. FIG. 5 is an explanatory diagram showing the data structure of the volume block. The volume block is a data block for storing volume information. “Fv” is recorded in the type information Fa, and a volume label, a volume serial number, and the like are recorded in the data storage area DA.

  FIG. 6 is an explanatory diagram showing the data structure of the file block. A file block is a data block for storing a file. In the type information, “ff” is recorded, and in the data storage area DA, the file entity, file name, creation date and time, and the like are recorded.

  FIG. 7 is an explanatory diagram showing the data structure of the directory block. The directory block is a data block for storing directory information. “Fd” is recorded in the type information, and the directory name and the date and time of creation thereof are recorded in the data storage area DA. In the data storage area DA, a second front link address Lf2 and a second rear link address Lb2 are recorded. In these link addresses Lf2 and Lb2, the head addresses of data blocks storing files and subdirectories existing in the directory are recorded. If no file or subdirectory exists in the directory represented by the directory block, both the second front link address Lf2 and the second rear link address Lb2 have the head address of the directory block. To be recorded.

  FIG. 8 is an explanatory diagram showing the data structure of the empty area block. The empty area block is a data block indicating that the area where the data block exists is an empty area. In the type information, “fn” is recorded, and in the initial state, Nil data (for example, “00”) indicating that it is empty is recorded in the data storage area DA. Thus, the data block in which “fn” is recorded in the type information corresponds to the “empty area block” described in the claims.

  FIG. 9 is an explanatory diagram showing the initial state of the address space when there is no missing area in the flash memory 140. When there is no missing area in the flash memory 140, there is one volume block and one empty area block in the address space in the initial state. The forward link address Lf of the volume block refers to the leading address a1 of the empty area block, and the backward link address Lb also refers to the leading address a1 of the empty area block. This is because the address space is formed in a ring shape as described above. Similarly, the forward link address Lf and the backward link address Lb of the empty area block both refer to the head address a0 of the volume block. In this embodiment, the address a0 indicates the beginning of the physical address of the flash memory 140, but may take an arbitrary address value.

  FIG. 10 is an explanatory diagram showing the initial state of the address space when there is a missing area in the flash memory 140. The defective region refers to a region where physical damage has occurred in the memory cells constituting the flash memory 140. FIG. 10 shows a case where there are two missing regions in the initial state. In such a case, as shown in the figure, there are three empty area blocks in the initial state (empty area blocks 1 to 3).

  The forward link address Lf of the empty area block 1 indicates the start address a3 of the empty area block 2, and the backward link address Lb indicates the start address a0 of the volume block. The forward link address Lf of the empty area block 2 indicates the start address a5 of the empty area block 3, and the backward link address Lb indicates the start address a1 of the empty area block 1. The forward link address Lf of the empty area block 3 indicates the start address a0 of the volume block, and the backward link address Lb indicates the start address a3 of the empty area block 2. That is, the empty area blocks 1 to 3 are linked to each other, and the front end (empty area block 1) and the end (empty area block 3) are set to be linked to the volume block.

  The size Sd of the empty area block 1 and the empty area block 2 is set so as to avoid the missing area. For example, in the size Sd of the empty area block 1, a value obtained by further subtracting the size of the missing area 1 from the difference between the value of the address a3 and the value of the address a1 is recorded.

  In order to avoid the missing area and arrange the empty area block as shown in FIG. 10, the control unit 130 performs the following initialization process. First, processing for recognizing a missing area existing in the address space is performed. In this recognition processing, for example, all bits in the address space are rewritten from “0” to “1”, or “1” to “0” to detect the bits that do not change, and the areas where the bits thus detected exist Can be taken as a defective region. In addition, for example, it can be performed by a method of referring to information on a defect area provided by a chip manufacturer that manufactures the flash memory 140. Usually, the flash memory 140 is shipped with such information written in a predetermined area in advance. Therefore, the control unit 130 can recognize the missing area by reading out such information from the flash memory 140.

  Next, a volume block is arranged at the head address, and the entire address space is segmented by this volume block and the missing area recognized by the above-described method. Then, a process for converting each segmented address space into an empty area block is performed. In such processing, the management information storage area MA in which “fn” is recorded in the attribute information is arranged from the head address of each address space. In this management information storage area MA, the size Sd of each segmented address space is recorded, and further, the forward link address Lf and the backward link address Lb are set so that the respective empty area blocks are linked to each other. . Through the processing as described above, the control unit 130 can initialize the address space.

  FIG. 11 is an explanatory diagram showing a specific example in which the above-described various data blocks are arranged in the address space in a ring shape for the sake of convenience when there is one missing region. The innermost ring represents the arrangement of empty area blocks, and the outer circumference represents the arrangement of data blocks existing in the root directory. Further, the outer circumference and the outermost circumference represent the arrangement of data blocks existing in the subdirectory. In the address space, discontinuous front links are indicated by arrows, and rear links are omitted. FIG. 12 shows a directory structure realized by the link state of each data block shown in FIG. 11 in a tree shape.

  In the example shown in FIG. 11, a volume block fv exists at the head address of the address space, and this volume block fv is linked to a file block f1 existing in the root directory. In the root directory, there are file blocks f2 and f8 and directory blocks d1 and d3 in addition to the file block f1. Of these, the file block f8 is stored across two areas so as to avoid a missing area. File block f3, f4, f5, f9 and directory block d2 are linked to the directory block d1. Among these, file blocks f6 and f7 are further linked to the directory block d2. On the other hand, the directory block d3 indicates that the file or directory is not linked and is an empty directory. The directory block d3 is located at the end of a link constituted by a file block and a directory block. Therefore, the empty block n1 is linked to the directory block d3. The empty area block n1 is linked to the empty area block n2, and the empty area block n2 is finally linked to the volume block fv.

  The data recording method in the present embodiment has been described above. In this embodiment, as described above, all data is recorded by the data block shown in FIG. Even with such a recording method, files can be recorded in a hierarchical structure using directories (see FIGS. 11 and 12). Therefore, the computer can handle the data storage device 100 of this embodiment by the same file access method as that of the conventional file system that adopts the directory structure.

C. File access processing:
FIG. 13 is a flowchart of file access processing executed by the control unit 130 in response to a command from the computer. When the control unit 130 receives a command related to file access from the computer (step S100), the control unit 130 determines the type of the command (step S110).

(C1) Read processing:
When a read command is received from the computer (step S110: “read”), the control unit 130 receives the file name of the file to be read from the computer (step S120). Then, a search is made for a data block in which the received file name is recorded and “ff” is recorded as type information (step S130). This search is performed by tracing the address referenced by the forward link address Lf in order from the volume block. For example, in FIG. 11, when the target file is recorded in the file block f2, the search is performed in the order of the volume block fv, the file block f1, and the file block f2. At this time, not only the forward link address Lf but also the backward link address Lb may be traced simultaneously to perform a search from both directions.

  In step S120, when a path name instead of a file name is received from the computer, the target file is searched after sequentially searching the directories included in the path. For example, when the path name “dir1 / file1” is received, the directory block in which the directory name “dir1” is recorded is searched first, and then the second forward link address Lf2 of the directory block is traced. The file block in which the file name “file1” is recorded is searched.

  When the target file is searched in this way, the control unit 130 transmits the file recorded in the data storage area DA of the searched data block to the computer via the USB device controller 120 (step S140). If the file is not found, an error message indicating that the file does not exist is sent to the computer.

(C2) Erasing process:
If an erasure command is received from the computer in step S100 (step S110: “Erase”), the control unit 130 receives the file name of the file to be erased from the computer (step S150). Then, the file block in which the file name is recorded is searched (step S160). The search method is the same as that for reading the file described above.

  When the target file block is searched, first, the control unit 130 rewrites the type information of the searched file block from “ff” to “fn” (step S170). By doing so, since the data block having the file attribute is converted to the empty area attribute, the file is formally deleted and a new empty area block is generated. At this time, if the entire data storage area DA of the newly generated empty area block is rewritten to Nil data, the file contents can be completely erased. However, in this embodiment, such rewriting is not performed to simplify the processing.

  Next, the control unit 130 performs a link address change process associated with the deletion of the file block (step S180). FIG. 14 is an explanatory diagram showing a specific example of such change processing. FIG. 14 shows a state in which three file blocks (file blocks 1 to 3) and one empty area block (empty area block 1) are linked. Of these, the case of deleting file block 2 is shown. Think. First, the control unit 130 rewrites the type information Fa of the file block 2 from “ff” to “fn”, thereby converting the type into an empty area block. Next, a process for linking the empty area block 2 thus generated after the empty area block 1 is performed. Specifically, first, the forward link address Lf of the empty area block 1 is rewritten to the head address a4 of the empty area block 2. For the backward link address Lb of the empty area block 1, the address value is maintained as it is because there is no change in the position where the file block 3 exists. Then, the forward link address Lf of the empty area block 2 is set so as to refer to the head address a0 of the volume block, and the backward link address Lb is set so as to refer to the starting address a6 of the empty area block 1.

  Next, the control unit 130 performs processing for changing the link addresses of the file block 1 and the file block 3 that have referred to the file block 2. Specifically, the forward link address Lf of the file block 1 is rewritten from the head address a4 of the file block 2 to the head address a5 of the file block 3. Further, the backward link address Lb of the file block 3 is rewritten from the head address of the file block 2 to the head address a3 of the file block 1. With the above processing, the change of the link address of each data block accompanying the deletion of the file block 2 is completed. In the figure, a “*” mark is attached to the link address changed by the above processing.

  In the above-described erasure process, the case of erasing a file has been described, but the same process is performed when a directory is erased. When a directory is deleted, other files and directories linked by the second forward link address Lf2 and the second backward link address Lb2 recorded in the data storage area DA are also erased collectively. Also good. Further, when files and subdirectories exist in the directory, the directory may not be deleted until all of them are deleted.

(C3) Write processing:
Next, the writing process will be described. When the write instruction is received in Step S100 (Step S110: “Write”), the control unit 130 receives a file to be written from the computer (Step S190). Then, an empty area block is searched by following the forward link address Lf of each data block (step S200). Such a search is performed by searching for a data block in which “fn” is recorded in the attribute information Fa. If no empty area block is found, an error message indicating that no empty area exists is transmitted to the computer, and the file access process is terminated.

  Next, the control unit 130 writes the received file to the searched empty area block (step S210). FIG. 15 is an explanatory diagram showing a specific example of such write processing. FIG. 15 shows a case where a new file block 2 is written in the address space where the file block 1 and the empty area block 1 exist.

  First, when the empty area block 1 is searched in step S200, the control unit 130 writes the file received in step S190 to the empty area block 1. Specifically, the file received in step S190 is stored in the data storage area DA, and the file block 2 in which “ff” is recorded for the attribute information Fa is shown in the empty area block as shown in the lower part of the figure. Write from 1 head address. Then, the management information storage area MA of the empty area block 1 overwritten by the writing of the file block 2 is moved to the head of the remaining empty area. In this embodiment, a new file block is written from the start address of the empty area block. However, the new file block may be written from the end of the empty area block.

  Next, the control unit 130 performs processing for changing the link address of each data block accompanying the writing of the file block 2. Specifically, first, the leading address a5 of the empty area block 1 is written to the forward link address Lf of the file block 2, and the leading address a3 of the file block 1 is written to the backward link address Lb. Next, the start address a0 of the volume block is written to the forward link address Lf of the empty area block 1, and the start address a4 of the newly written file block 2 is written to the backward link address Lb. A new size is written to the size Sd of the empty area block 1. For the file block 1, since the front link address Lf and the rear link address Lb are not changed, the link address is not changed. With the above processing, the writing process of the new file block 2 is completed.

  If the size of the file received in step S190 is larger than the size of the empty area block searched in step S200 (more precisely, the size of the data storage area DA of the empty area block), In step S200, another empty area block exceeding the size of the file may be searched. In addition, when an empty area block is not searched by such a search, a file may be written using a plurality of empty area blocks. In such a case, the file received in step S190 is divided, and the fragment of the divided file is stored in each empty area block. At this time, for the type information of each empty area block, a parameter (for example, “fg”) indicating that a file fragment is stored in the data block is recorded. By doing so, it is possible to write a file across a plurality of empty area blocks.

  When the file writing process in step S210 is completed, the control unit 130 performs verification by comparing the written file and the file received from the computer in units of bits (step S220). “Verify” is a process for determining whether or not a file is normally written. This verify process may be executed simultaneously with the write process in step S210. When it is determined that the file has been normally written by the verification (step S230: Yes), the control unit 130 ends the file access process. If it is determined that the file has not been written normally (step S230: No), it is determined that a defect has occurred in the written area, and this missing area is used in subsequent file access processing. In order to prevent this, the loss avoidance process is performed after the writing in step S210 is restored (step S240).

  FIG. 16 is an explanatory diagram showing the concept of the loss avoidance process. It is assumed that a defect area of 3 bytes is detected in the empty area block 1 shown in the upper part of FIG. 16 by the verification in step S220. Then, as shown in the lower part of FIG. 16, the control unit 130 divides the empty area block 1 into two empty area blocks (empty area block 1 and empty area block 2) so as to avoid the missing area of 3 bytes. Divide into Then, the link addresses of the empty area block 1 and the newly generated empty area block 2 are set so as to refer to each other, and the size Sd of the empty area block 1 is set so as to avoid the missing area. In the case of FIG. 15, the difference between the head address a5 of the empty area block 2 and the head address a4 of the empty area block 1 is obtained, and the value obtained by subtracting the size of the missing area (3 bytes) from this difference is the empty area. It becomes the size Sd of block 1. By performing such a defect avoidance process, the defect area is excluded from the link structure of the empty area block, so that it is possible to prevent the defect area from being used in the subsequent file access process. .

  After executing the loss avoidance process in step S240, the control unit 130 searches for another empty area block by returning the process to step S200. Then, the above-described series of write processing is executed again. By doing so, the file can be written while avoiding the missing area.

  By the way, the flash memory 140 of this embodiment is a NAND flash memory as described above. Therefore, in actual file access to the flash memory 140, it is necessary to erase data in units of block areas shown in FIG. 1 and to read / write data in units of pages. Further, as is well known, in a NAND flash memory, already written data cannot be directly overwritten. Therefore, the control unit 130 rewrites the link address and data described with reference to FIGS.

(1) The contents of the block area including the part to be rewritten are copied to the RAM 132.
(2) The block area (1) is erased from the flash memory 140.
(3) Rewrite the target portion in the data copied to the RAM 132.
(4) Data is written back from the RAM 132 to the flash memory 140.

  The control unit 130 performs the above processing for all the parts to be rewritten. By doing so, data recording to the flash memory 140 can be realized. In this embodiment, data writing to the empty area block is also performed by such a method. This is because data writing to an empty area block in which data other than “Nil” data is written in the data storage area DA, such as an empty area block generated when a file is erased, is substantially data Is to rewrite. Of course, it is also possible to directly write data in units of pages to an area constituted only by “Nil” data, regardless of the method described above.

  The configuration and processing of the data storage device 100 as one embodiment of the present invention have been described above. According to the data storage device 100 of the present embodiment, each data block recorded in the flash memory 140 is linked in a chain manner by the forward link address Lf and the backward link address Lb recorded in the management information storage area MA. Will be. Therefore, the target data can be accessed by following these link addresses. With such a configuration, the data recording position can be managed without using management information secured in a specific area such as a directory entry or FAT information. Therefore, the concentration of access to a specific area can be suppressed, and the possibility that a defect occurs in the flash memory 140 can be suppressed.

  Further, according to the present embodiment, by changing the link address of each data block, access to the missing area can be easily avoided. Therefore, it is not necessary to prepare a table for recording addresses where a missing area exists as in the prior art, and the entire storage area of the flash memory 140 can be used efficiently. Conventionally, when a defective area occurs, it is necessary to secure a replacement area in advance in the storage area to replace the area. According to the data recording method according to the present embodiment, however, It is not necessary to secure an alternative area. Furthermore, conventionally, for example, when a one-bit missing area exists, the block area or the entire sector including the bit cannot be used. However, if the link address is managed in bit units in this embodiment, such restrictions are eliminated, and the entire storage area can be used very efficiently.

D. Other examples:
(1) In the above embodiment, as shown in FIG. 2, the address space is handled in a ring shape. However, the present invention is not limited to this mode. For example, when the volume block is not arranged other than the head of the physical address of the flash memory 140, a normal address space where the head address and the end address are not continuous may be employed. In such a case, referring to FIG. 10 as an example, the reference destination of the forward link address Lf of the empty area block 3 is the end address mx, and the reference destination of the backward link address Lb of the volume block is the address a0.

(2) In the above embodiment, as shown in FIG. 9, the volume block is arranged at the head of the address space. However, for example, when there is a defect in the head area, the volume block may be arranged at another arbitrary address in the initialization process described above. In such a case, the address value of the physical address where the volume block exists is recorded at a predetermined address such as the end address of the flash memory 140 during the initialization process. The control unit 130 can easily access the volume block by using the address value thus recorded as a pointer to the volume block.

  The address value of the physical address where the volume block exists may be stored in the RAM 131 instead of the flash memory 140. In such a case, at the first access after the data storage device 100 is powered on, the control unit 130 sequentially searches the physical address recorded in the volume block from the top address based on the value of the attribute information Fa. The retrieved physical address is stored in the RAM 131. By using the address value stored in the RAM 131 in this way, the control unit 130 can quickly access the second and subsequent volume blocks.

(3) In the file erasing process described with reference to FIG. 12, it is also possible to perform a process of further joining the physically adjacent empty area blocks after step S170. In such processing, first, it is determined whether or not the data blocks before and after physically adjacent to the empty area block generated in step S170 are empty area blocks. If it is an empty area block, the management information storage area MA of the empty area block that physically exists later is erased, and the management information storage area MA of the empty area block that physically exists earlier is deleted. Record the new size after the merge. The link addresses Lf and Lb are set so as to refer to other empty area blocks. As described above, by performing the process of combining the empty area blocks, it is possible to prevent the file from being fragmented and recorded.

  In addition to the above-described file erasure timing, such a vacant area block combining process is performed at an arbitrary timing such as when there is no access to the data storage device 100 from a computer for a certain period of time. It is good also as what searches and executes. This makes it possible to efficiently suppress file fragmentation.

(4) In the above embodiment, as shown in FIGS. 11 and 14, the empty area block is arranged at the end of the link of the file block or the directory block. However, the empty area block may form a completely independent link with other data blocks except the volume block. In such a case, a link address indicating a link to the empty area block is recorded in the data storage area DA of the volume block. In addition, the end of the link of the file block or directory block refers to a volume block, not an empty area block. By doing so, it is possible to efficiently search for an empty area block without following a file block or a directory block.

  As mentioned above, although various embodiment of this invention was described, it cannot be overemphasized that this invention is not limited to such a form, and can take a various structure in the range which does not deviate from the meaning. For example, a function realized by software may be realized by hardware. Further, the flash memory is not limited to the NAND type, and may be a NOR type or an AND type. In the embodiment, the case where the present invention is applied to an auxiliary recording / rewriting device using a flash memory has been described. However, the present invention may be applied to an auxiliary storage device including a magnetic disk or a magneto-optical disk.

2 is an explanatory diagram showing a configuration of a data storage device 100. FIG. It is explanatory drawing which shows the image of address space. It is explanatory drawing which shows the structure of the data stored in address space. It is explanatory drawing which shows the example of the link state of a data block. It is explanatory drawing which shows the data structure of a volume block. It is explanatory drawing which shows the data structure of a file block. It is explanatory drawing which shows the data structure of a directory block. It is explanatory drawing which shows the data structure of an empty area block. It is explanatory drawing which shows the initial state of address space when there is no defect area | region. It is explanatory drawing which shows the initial state of address space when there exists a defect | deletion area | region. It is explanatory drawing which shows the specific example by which the various data block mentioned above is arrange | positioned in the address space in the case of one missing area for convenience. It is explanatory drawing which shows the directory tree implement | achieved by the link state of each data block shown in FIG. It is a flowchart of a file access process. It is explanatory drawing which shows the specific example of the change process of a link address accompanying the deletion of a file block. It is explanatory drawing which shows the specific example of the writing process of a file. It is explanatory drawing which shows the concept of a defect | deletion avoidance process.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 ... Data storage device 110 ... USB connector 120 ... USB device controller 130 ... Control part 131 ... CPU
132 ... RAM
133 ... ROM
140 ... Flash memory MA ... Management information storage area DA ... Data storage area

Claims (7)

A data storage device for storing computer data,
A storage unit for storing the data;
A control unit for controlling writing and reading of data to and from the storage unit in response to an instruction from the computer;
The storage unit includes a storage area partitioned into data blocks in which data entities are recorded and empty area blocks in which data entities are not recorded,
The data block integrally holds a data management information storage area for recording a location address of another data block with the data block,
The empty area block integrally holds an empty area management information storage area for recording a location address of another empty area block with the empty area block ,
In the empty area management information storage area of the empty area block, information capable of acquiring the size of the empty area block is recorded,
The location address recorded in the empty area management information storage area of the empty area block and the information capable of acquiring the size of the empty area block are set so as to avoid the missing area existing in the storage unit. Data storage device. The data storage device according to claim 1,
A data storage device in which type information indicating the type of the data block is further recorded in the data management information storage area of the data block. A data storage device according to claim 2 , comprising:
A data storage device in which a parameter representing one of a file, a directory, and a volume is recorded as the type information in the data management information storage area. A data storage device according to claim 3 ,
When a parameter representing a directory is recorded as the type information in the data management information storage area of the data block, a file or directory existing in the directory is further stored in any area of the data block. A data storage device in which the location address of a stored data block is recorded. A data storage device according to any one of claims 1 to 4 ,
When the control unit receives a data write command from the computer,
A new data block in which the data is recorded is arranged in any area where the empty area block exists, and the location address of another data block is recorded in the data management information storage area of the data block, A data storage device for recording the location address of the new data block with respect to the location address recorded in the data management information storage area of the other data block. The data storage device according to claim 5 ,
In addition, when the location address of the empty area block fluctuates due to the arrangement of the new data block, the control unit further stores another empty space where the location address of the empty area block is recorded in the empty area management information storage area. A data storage device for recording the changed location address in an empty area management information storage area of an area block.   The data storage device according to any one of claims 1 to 6, further comprising:
  A missing area detecting means for detecting a missing area in the storage unit;
  Dividing means for dividing the empty area block so as to avoid the detected missing area;
  For each of the empty area management information storage areas of the divided empty area blocks, a defect address that records a location address that mutually refers to the empty area block and information that can acquire the size of the empty area block is recorded. Avoidance processing means
  A data storage device comprising:

Images