JP2010026931A - Host device and data recording method for the host device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a host device, capable of restoring data based on information storage in a storage device even when writing is not normally terminated, and performing efficient writing control. <P>SOLUTION: In the host device 1 which successively and continuously records data to a storage device 10 which performs data management by a file system, file specification information for specifying a data of a writing object is formed, and a recording start position for recording the data is determined. A part of file management information including at least the file specification information and information for the recording start position is formed before starting the recording of the data, and update of the file management information is not performed in the storage device 10 during recording of the data, and the update of the file management information is performed after completion of the recording of the data. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a host device and a data recording method thereof. For example, a host device that records real-time data such as video recording in an OTP (One Time Programmable) memory device or a write-once memory device, and a data recording method thereof It is what is used.

  In the storage device, data is managed by a file system. Therefore, when data is continuously recorded sequentially in the storage device, the data writing is completed when the file management information of the file system is written. The contents of the file management information include recorded data position information, data size, recorded date, and the like.

  If a power interruption occurs during the writing of the file management information, the file management information is lost, and therefore all data from the start of writing is lost. In particular, in a portable device such as a video recording device, a case where the user drops the device and the battery is removed is assumed.

  In the conventional method, an operation of periodically updating file management information is performed during data recording in order to minimize the loss of data.

  However, since the OTP memory device and the write-once memory device have the following limitations, a problem may occur if the file management information is updated many times to write to the same address.

(1) Locations that have already been written cannot be updated.

(2) There is a limit to the number of updates.

(3) Each time the file management information is updated, the storage capacity decreases.

(4) The lifetime of the device is reduced.

  Even in a normal memory device, there is a problem that if the file management information is written, the total writing performance is lowered.

Further, as a prior art related to the present invention, for example, Patent Document 1 describes a file allocation table (FAT) management method by a memory controller for a memory card.
JP 2003-187203 A

  According to the present invention, it is possible to provide a host device capable of recovering data based on information stored in a storage device and performing efficient write control even when writing is not normally completed. For the purpose.

  A host device according to an embodiment of the present invention creates file specifying information for specifying data to be written in a host device that sequentially records data in a storage device that performs data management by a file system. And determining a recording start position for recording the data, creating a part of file management information including at least the file specifying information and the recording start position information before starting the recording of the data, The storage device is provided with a control unit that does not update the file management information during recording of the data, and updates the file management information after the data recording is completed.

  According to another embodiment of the present invention, there is provided a data recording method for a host device in which the data is sequentially and continuously recorded in a storage device for managing data by a file system. Obtaining from the storage device information for determining whether or not recording of the storage device has ended abnormally, and when detecting abnormal end, from the directory management information of the storage device, the latest recording start date and time without an erasure mark The step of acquiring the recording start position information as the last recorded file of the set file, the step of acquiring the recording final position information of the data that has been correctly written last from the storage device, and the recording start Determination of the recording position used when recording the data from the position to the final recording position By identifying the recorded data location using repeated stages, the connection information of the data is restored, and after the data location is identified, the total data length is calculated, the directory management information is restored, and the recorded data is restored. And the step of performing.

  According to the present invention, it is possible to provide a host device capable of recovering data based on information stored in a storage device and performing efficient write control even when writing is not normally completed. It is possible.

  Embodiments of the present invention will be described below with reference to the drawings. In the description, common parts are denoted by common reference symbols throughout the drawings.

  First, a host device according to an embodiment of the present invention will be described.

[1] Hardware Configuration of Host Device and Storage Device FIG. 1 is a block diagram illustrating a configuration of a host device according to the embodiment and a storage device in which data is stored by the host device.

  The host device 1 includes hardware and software for accessing the storage device 10. The host device 1 includes a CPU (Central Processing Unit) 2 as a control unit, a hard disk 3, a memory 4, and a card interface (card I / F) 5. The storage device 10 includes, for example, a memory card, an OTP memory device, a write-once memory device, and the like.

  The CPU 2 controls the operation of the entire host device 1. The hard disk 3 is a memory that stores a control program used by the CPU 2. The memory 4 is a volatile memory that is used as a work area of the CPU 2 and stores a control program and various tables.

  For example, when receiving power supply, the CPU 2 loads firmware (control program) stored in the hard disk 3 onto the memory 4 and executes predetermined processing. For example, various tables are created on the memory 4, and a write command, a read command, and an erase command are issued to the storage device 10.

  The storage device 10 operates upon receiving power supply when connected to the host device 1, and performs processing in accordance with access from the host device 1. The storage device 10 includes a semiconductor storage device, for example, a nonvolatile semiconductor storage device such as a NAND flash memory (hereinafter referred to as a flash memory) 11, and a controller 12.

  The controller 12 has a function of managing the physical state inside the flash memory 11 (where the physical address includes what logical sector address data is included, or which block is in the erased state). The controller 12 includes a central processing unit (CPU) 13, a read-only memory (ROM) 14, a random access memory (RAM) 15, a buffer 16, a host interface (host I / F) 17, and a memory interface (memory I / F). F) 18 is mounted.

  The CPU 13 controls the operation of the entire storage device 10. The ROM 14 is a memory that stores a control program used by the CPU 13. The RAM 15 is a volatile memory that is used as a work area for the CPU 13 and stores control programs and various tables. For example, when the storage device 10 is supplied with power, the CPU 13 loads the firmware (control program) stored in the ROM 14 onto the RAM 15 and executes predetermined processing to thereby store various tables. Created on the RAM 15, receives a write command, a read command, and an erase command from the host device 1, executes access to a corresponding area on the NAND flash memory 11, and controls data transfer processing through the buffer 16. .

  The memory interface 18 performs interface processing between the controller 12 and the flash memory 11. The host interface 17 performs interface processing between the controller 12 and the host device 1. The card interface 5 in the host device 1 performs interface processing between the host device 1 and the controller 12.

  The buffer 16 temporarily stores a certain amount of data (for example, for one page) or writes data read from the flash memory 11 when writing data sent from the host device 1 into the flash memory 11. When sending the data to the host device 1, a certain amount of data is temporarily stored.

  The flash memory 11 may be a binary memory that stores 1-bit information in one memory cell, or may be a multi-level memory that stores information of 2 bits or more in one memory cell.

[2] Example of Data Recorded in Storage Device FIG. 2 shows an example of data when the host device 1 performs video recording on the storage device 10.

  In FIG. 2, the video format is designated by the header 21 arranged at the head of the data, and the format of the subsequent data is determined. In the example shown in FIG. 2, the case where the image data 22 and the sound data 23 are recorded alternately is shown. Since video is usually managed in units of frames, it becomes periodic data. In this case, the image data 22 and the audio data 23 are managed in units of frames, and are paired as one frame data of the same time.

  When such data is recorded in the storage device, it is divided and recorded in a size called a cluster 24 which is a management unit of the file system. The data area of the storage device is divided and managed in units of clusters, and each cluster is designated by a cluster number assigned in order.

[3] File System [3-1] File Structure FIG. 3 is a diagram showing a file structure in the storage device 10. There is one subdirectory 26 under the root directory 25, and one file 27 is recorded in the subdirectory 26.

[3-2] Storage Area Map in Storage Device Next, FIG. 4 shows an example in which the file shown in FIG. 3 is managed and recorded by the file allocation table (FAT) file system. FIG. 4 shows a map 28 in the storage area of the storage device 10 managed by the FAT file system. In the root directory 25, the subdirectory name and the cluster number “0002h” indicating the cluster position of the subdirectory are recorded as the subdirectory information. Thereby, the location of the directory entry 29 in which the subdirectory information is recorded can be known. In the directory entry 29, the file name, creation date, data start cluster number, data size, and the like of the recorded file are recorded.

  The data is written in the order of data 1, data 2, data 3, etc. These data 30 correspond to each cluster 24 shown in FIG.

  The map 28 includes a FAT table (1) 31 and a FAT table (2) 32. In the FAT table (2) 32, the same data as the FAT table (1) 31 is recorded as a backup.

  The map 33 shows an enlarged part of the FAT table (1) 31 so as to correspond to the storage area of the map 28. The cluster number indicating the recording location of the first data (1) indicates “1234h”. In the FAT table map 33 corresponding to the data (1), the cluster number “1235h” in which the following data (2) is recorded is recorded. Similarly, the cluster number “1237h” of the following data (3) is recorded in the FAT table corresponding to the data (2). In this way, the data connection can be understood by following the FAT table. In the FAT table corresponding to the last data, “FFFFh” is recorded, indicating that the data ends here. As described above, the file is managed by the FAT tables 31 and 32 and the directory entry 29.

[3-3] Update of FAT Table in FAT File System Next, consider a case where the user drops the video camera during recording from the host device (for example, video camera in this case) 1 to the storage device 10. If the video camera is dropped, the battery may come off and writing may stop during recording. In that case, if the file information is not recorded in the FAT table and the directory entry, the recorded data is lost. For this reason, in order to determine the recording data, the host device conventionally updates the FAT from time to time during recording, that is, updates the FAT table and directory entry.

  FIG. 5 is a diagram showing a state of FAT update. As shown in FIG. 5, while writing data to the cluster 24, the FAT table (1) 31, the FAT table (2) 32, and a part of the directory entry 29 are written. Here, these three writes are collectively referred to as FAT update. Regarding the FAT update cycle, there are a method of inserting in a specific data length unit and a method of inserting at regular intervals by a system timer.

[4] Recording Flow as Comparative Example FIGS. 6A and 6B show a recording flow by a host device (for example, a recording device) 1 as a comparative example. FIG. 6A shows a procedure for preparing in advance a state where recording is possible. A directory for recording data is determined, and directory data is created (step S1). In order to determine a place to record data, an empty cluster is searched (step S2), and a cluster to start recording is determined (step S3). Then, the cluster for starting this recording is recorded as directory information (step S4). At this stage, the file name, start cluster number, and creation date / time are recorded in the directory information. At this time, it is important that the latest date and time is reached as will be described later. Thus, the recording preparation in the host device is completed.

  6B shows a processing procedure for writing the cluster data 24 shown in FIG. 2 successively sent to the storage device 10 to the storage device according to the file system while the recording button of the host device 1 is pressed. Show.

  First, data is written to the start cluster described in the directory entry 29 (step S11). Subsequently, it is determined whether or not to perform FAT update (step S12), and when performing FAT update, FET update is performed (step S13). On the other hand, when the FAT update is not performed or after the FET update is performed in step S13, it is determined whether or not the recording is finished (step S14). Here, when the recording is not finished, that is, when data is further written, a free cluster is searched and a storage location (cluster) to be written next is determined (step S15). Then, it returns to step S11. When the recording is finished in step S14, the FAT is updated (step S16), and the recording is finished. That is, these processes are repeated until the recording button is released or until the recording ends.

  Note that the determination of the cluster in step S15 may depend on the write characteristics of the storage device. For example, in the case of a storage device using a NAND flash memory, the write performance is affected by the number of empty clusters in the erase block, so a cluster that matches the required performance is selected.

[5] Recording Flow of Embodiment Hereinafter, a recording procedure by the host device of this embodiment will be described.

  FIG. 7 is a flowchart showing a procedure from the start to the end of recording by the host device of this embodiment. Compared with the flow shown in FIG. 6B, it is unnecessary to determine whether or not to perform the FAT update in step S12 and to perform the FAT update in step S13.

  In this case, as described above, when a trouble such as the power being cut off occurs during recording, the state of the FAT table and the directory entry in the storage device 10 is the state where the “recording preparation” shown in FIG. 6A is completed. It has become. The host device 1 needs to have means as shown in FIG. 8 in order to recover the data whose recording has been interrupted.

  FIG. 8 shows a procedure of “restoration processing” for repairing a recorded file that has not ended normally due to a power failure or the like during recording.

  Here, the storage device 10 includes means for indicating to the host device 1 that an abnormal end has occurred, and means for indicating the cluster number at which writing was correctly completed last.

  The host device 1 initializes the storage device 10 after the power is turned on again, and starts “repair processing”. First, the abnormal end is determined (step S21). If the FAT table corresponding to the last written cluster number is already used, it is not subject to recovery. However, it is preferable not to make a determination based on this alone, but to make a more accurate determination in combination with other determinations. For example, the end process of the storage device is defined so that it can be determined whether or not the end process has been performed at the time of reinitialization. The host device that performs the termination process can determine that the data recovery process is necessary when the termination process is not performed. There is also a method of holding a flag indicating that data is being recorded in the storage device and reading and confirming it later. For the cluster numbers returned by the recording device and correctly written, writing in the FAT table and writing with a data length shorter than the cluster size must be excluded.

  When the host device 1 detects the abnormal end by the above-described means (step S22), the host device 1 searches the directory information (step S23). In this directory information search, the update date is searched to find a file with the latest date and without an erasure mark. This can be done by the “recording preparation” process shown in FIG. 6A or by adding the latest date and time before starting recording. From the directory information, the cluster number where recording is started is acquired (step S24). As for the end cluster number, the cluster number that has been correctly written last is determined as the end cluster number by means of the storage device 10 (step S25).

  Next, an empty cluster is sequentially searched from the start cluster number until it matches the end cluster number (steps S26 to S29). In this case, the algorithm for searching for an empty cluster in step S15 and the algorithm for searching for an empty cluster in step S28 used at the time of recording must be the same. The empty clusters found until the end cluster number is matched are sequentially connected by a cluster chain and recorded in a temporary FAT table created on the system RAM (steps S28 and S29). At this stage, the FAT table of the storage device 10 is not yet updated. If it matches the end cluster number, “FFFFh” indicating the end of the file is recorded in the FAT table corresponding to the end cluster number.

  Next, the host device 1 performs data analysis of the recorded data stored in the storage device 10. That is, it is confirmed whether or not the recording format defined by the header 21 shown in FIG. 2 has been correctly created (step S30). If the recording format is broken, it is preferable to repair the recording format. Specifically, it is preferable to change the last cluster, change the file size, or change the cluster chain (steps S31 to S33). At this time, if the device that reproduces the data can ignore the last broken data, there is no problem even if the last recorded data does not end correctly.

  FIG. 9 shows an example in which a halfway cluster is damaged in the recorded data. The cluster 34 is a damaged cluster. For this reason, a part of the audio data 23 is lost, and the audio data 23 is deleted. When the recording data is managed in units of frames, the audio data 23 and the image data 22 are paired, and the image data 22 is also deleted at the same time. As a result, the two clusters 35 are set as unused clusters, and the cluster 36 is set as an end cluster. Since the frame data is a short time of about 30 ms, even if one frame is deleted, it is a very small amount based on the whole data and does not have a great influence.

  Returning to FIG. 8 again, the procedure from step S30 to S34 will be described below. FIG. 10 shows the results of data analysis and data restoration in steps S30 to S34. Based on the data analysis in step S30, the end cluster number and the overall file size are changed (steps S31 and S32). Therefore, “FFFFh” is recorded in the FAT table corresponding to the end cluster 36 of the FAT table temporarily recorded on the RAM. Further, “0000h” is recorded in the FAT table corresponding to the unused cluster 35 and registered as an unused area. Further, empty clusters are sequentially connected by a cluster chain (step S33). The directory information found by the directory information search in step S23 is temporarily copied onto the RAM, the file size is changed as shown in FIG. 10, and the date is further changed.

  Next, in step S34, the FAT update is executed, and the FAT table and directory information temporarily created on the RAM as described above are written in the storage device 10. This completes the recovery of the recorded data.

  As described above, according to the embodiment of the present invention, the file management information is not updated during recording, and the file management information is updated after recording. By having a means for recovering data based on this, it is possible to provide a host device that can perform efficient write control.

  In the present embodiment, the total write performance is improved by not updating the file information. Further, the last recorded data can be recovered without updating the file information. Further, efficient write control can be performed on an OTP (One Time Programmable) memory device or a write-once memory device, and real-time data recording such as video can be performed. Also, depending on the playback device, malfunction may occur when an incomplete file is played back. In order to prevent this, the data recorded after being divided in time can be corrected to the correct format by deleting the last incomplete part.

  The embodiment described above is not the only embodiment, and various embodiments can be formed by changing the configuration or adding various configurations.

  The present invention can take the following embodiments.

(1) In a host device that sequentially and sequentially records data to a storage device that manages data by a file system,
Creating file specifying information for specifying data to be written, and determining a recording start position for recording the data;
A part of the file management information including at least the file specifying information and the recording start position information is created before starting the recording of the data,
A host device comprising: a control unit that updates the file management information after the data recording is completed, without updating the file management information in the storage device during the recording of the data.

(2) The file specifying information is directory management information for specifying a directory hierarchy of a file recorded by a file name and position information where the content is stored,
Record the file name, recording start position information, recording start date and time as the directory management information of the data to be created before starting recording,
The host according to (1), wherein data connection information indicating that data recorded at different positions is one sequential data and a data length of the recorded data are recorded as the file management information. machine.

(3) The file allocation table (FAT) system is used as the file system, the directory management information indicates a root directory and a directory entry, and the data connection information indicates a file allocation table. The listed host device.

(4) In a data recording method of a host device that sequentially and sequentially records data in a storage device that manages data by a file system,
Obtaining information from the storage device to determine whether recording of data to the storage device has ended abnormally; and
A step of acquiring recording start position information from the directory management information of the storage device as a file last recorded last time from a directory management information of the storage device when a file with no erasure mark and the latest recording start date and time is set; ,
Obtaining the recording final position information of the data that has been correctly written last from the storage device;
By identifying the data position recorded by repeatedly using the recording position determining means used at the time of recording the data from the recording start position to the recording final position, the connection information of the data is restored, and the data position After the identification is completed, calculate the total data length, recover the directory management information, recover the recorded data,
A data recording method for a host device, comprising:

(5) After the data recovery, the data recorded in the time-divided format is deleted part of the last recorded data incompletely recorded, and the connection information of the data is corrected accordingly. The data recording method of the host device according to (4), wherein the total data length of the corrected data is calculated and the directory management information is corrected.

It is a block diagram which shows the structure of the host apparatus and storage device of embodiment of this invention. It is a figure which shows the example of data in the case of video-recording to a storage device by the host apparatus of embodiment. It is a figure which shows the file structure of the storage device in embodiment. It is a figure which shows the map in the storage area of the storage device managed by the FAT file system in the embodiment. It is a figure which shows the mode of FAT update. It is a figure which shows the recording flow by the host apparatus as a comparative example. It is a figure which shows the recording flow by the host apparatus as a comparative example. It is a figure which shows the flow from the recording start by the host apparatus of embodiment to completion | finish. It is a flowchart which shows the procedure of the "restoration process" which repairs the recording file which was not complete | finished normally in embodiment. It is a figure which shows the example of data in which the cluster in the middle was damaged in recording data. It is a figure which shows the data sequence which implemented data analysis and data restoration.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Host device, 2 ... CPU (control part), 3 ... Hard disk, 4 ... Memory, 5 ... Card interface (card I / F), 10 ... Storage device, 11 ... NAND type flash memory, 12 ... Controller, 13 ... CPU: 14 ROM (Read-Only memory), 15 RAM (Random Access Memory), 16 Buffer, 17 Host interface (Host I / F) 18 Memory interface (Memory I / F), 21 Header 22 ... Image data, 23 ... Audio data, 24 ... Cluster, 25 ... Root directory, 26 ... Subdirectory, 27 ... File, 28 ... Map, 29 ... Directory entry, 30 ... Data, 31 ... FAT table (1), 32 ... FAT table (2), 33 ... Map, 34-36 ... Cluster.

Claims (5)

In a host device that records data sequentially and continuously to a storage device that manages data by a file system,
Creating file specifying information for specifying data to be written, and determining a recording start position for recording the data;
A part of the file management information including at least the file specifying information and the recording start position information is created before starting the recording of the data,
A host device comprising: a control unit that updates the file management information after the data recording is completed, without updating the file management information in the storage device during the recording of the data. The file specifying information is directory management information for specifying a directory hierarchy of a file recorded by a file name and position information where the contents are stored,
Record the file name, recording start position information, recording start date and time as the directory management information of the data to be created before starting recording,
The host apparatus according to claim 1, wherein the file management information includes data connection information indicating that data recorded at different positions is one sequential data, and a data length of the recorded data. .   3. The host device according to claim 2, wherein a file allocation table (FAT) system is used as the file system, the directory management information indicates a root directory and a directory entry, and the data connection information indicates a file allocation table. . In a data recording method for a host device that sequentially and sequentially records data to a storage device that performs data management by a file system,
Obtaining information from the storage device to determine whether recording of data to the storage device has ended abnormally; and
A step of acquiring recording start position information from the directory management information of the storage device as a file last recorded last time from a directory management information of the storage device when a file with no erasure mark and the latest recording start date and time is set; ,
Obtaining the recording final position information of the data that has been correctly written last from the storage device;
By identifying the data position recorded by repeatedly using the recording position determining means used at the time of recording the data from the recording start position to the recording final position, the connection information of the data is restored, and the data position After the identification is completed, calculate the total data length, recover the directory management information, recover the recorded data,
A data recording method for a host device, comprising:   After the data recovery, for the data recorded in a time-divided format, a part of the last recorded data recorded incompletely is deleted, and the connection information of the data is corrected accordingly. 5. The data recording method for a host device according to claim 4, wherein the directory management information is corrected by calculating the total data length of the data.

Images