JP3978720B2 - Data storage method - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an improvement in a method for enabling data to be protected even when there is a write interruption by using a part of a storage area to maintain a state during writing in a data storage device.
[0002]
[Prior art]
In portable information terminals and mobile phones, flash memory is often used to store various information such as a telephone directory and a schedule. Furthermore, in order to manage each information, it is common to use some kind of file system. In a flash memory or the like, a byte that is a storage unit is configured by a plurality of sets of bits that are the minimum storage unit, and a page configured by a plurality of bytes (tens of thousands of bytes) is collectively handled as the minimum erase unit. .
[0003]
For example, when the FAT file system (ISO / ICE9293) is applied as a management mode of information storage, the file management information and data are stored in separate pages, so the power supply is cut off when either one is written. And file system consistency could not be maintained, and data loss could occur.
[0004]
In particular, in a portable information terminal and a mobile phone, there is a higher risk that power supply may be inadvertently cut off due to a drop impact or the like than a general stationary electronic device. For this reason, some measure is required to protect information from interruption of writing due to power supply being cut off.
[0005]
Conventionally, there is a technique in which file management information and data are collated at the time of recovery after a sudden power failure, and if there is a contradiction, it is corrected. However, in the case of this prior art, since the page in the middle of writing is not known, it is necessary to make all pages subject to collation, and it takes a long time to correct, and since there is no information before writing, all information can be corrected reliably. Not necessarily.
[0006]
Data from the host, addresses accessed from the host, and data are written to multiple flash memories in the external storage device to automatically perform data recovery operation in the event of a power failure and reduce data recovery time The physical sector consisting of two components of the physical sector management area with a data invalid flag that determines whether or not the physical sector has an error or whether invalid data is stored, and the address where the error occurred A method has been proposed in which a flash memory management area having an error occurrence physical sector address management area and a flash memory management area including a flash memory number, a status table, the number of flash memory erasures, and a status invalid flag is provided in the flash memory (Japanese Patent Laid-Open 11-272569).
[0007]
Also, if the power interruption occurs after erasing a block, which is the data erasure unit in the flash memory, the FAT and directory area data are not updated despite the block being erased, and the power supply is restarted. In order to cope with the problem that an erased file (block file) is read when an attempt is made to read the updated file, an operation history storage area is provided with an identification code indicating a processing content such as a writing process. When the writing process or the like is normally completed, the identification code is deleted, and then the identification code remaining in the operation history storage area is analyzed and stored in the corresponding data area or the like. A method for restoring existing data has been proposed (Japanese Patent Laid-Open No. 2001-84180).
[0008]
Note that a series of memories to be stored in the information storage / reproduction device is a simple method for partially changing the management information for managing the storage information of the information storage / reproduction device using a batch erasure type memory such as a flash memory. The information is divided into predetermined unit frames, and the storage information of the divided unit frames and the management information provided corresponding to the storage information are stored so as to coexist in the block that is a batch erasure unit, and the management information When the management code becomes a predetermined specific code, the management information can be added to the management code for managing the storage information as a predetermined code that makes this management information equivalent to the initial state or a function equivalent to invalidity or invalidity. A method for enabling management information to be changed without an erasing operation has also been proposed (Japanese Patent Laid-Open No. 2000-67857).
[0009]
[Problems to be solved by the invention]
Any of the above-described proposed techniques is suitable for an information storage / reproduction device using a batch erasure type memory such as a flash memory. However, when the data write operation to the data area is interrupted, recovery is performed. The procedure is still cumbersome and in some cases the process may take time.
[0010]
SUMMARY OF THE INVENTION An object of the present invention is to provide a data storage method that can be reliably restored in a shorter time with a simple procedure even if the data write operation to the data area is interrupted.
[0011]
[Means for Solving the Problems]
In order to solve the above-described problems, the data storage method according to the present invention has the following features.
[0012]
(1) A data area for storing data as main information in the same management unit area under the condition that a management unit area is an area having a predetermined capacity larger than a predetermined minimum storage unit on the information storage medium A management information area for storing management information relating to the data, and a state holding unit for holding storage status information indicating an invalid state, a temporary write state, and a valid state of the data in the management information area ; Preceding order holding for holding information indicating the writing order of the management unit area written earlier than the corresponding writing order in the series of writing order of each management unit area set on the information storage medium parts and corresponding to the first state storage unit comprising a data position holding unit for holding the data location information indicating a location of data, the state holding unit and Information sets the second state storage unit comprising a row rank holding unit, when the write operation of the data to recover from being interrupted for the data region, which is held in the state holding unit of the management information area, A data storage method for restoring data interrupted by writing using information held in a preceding rank holding unit and a data position holding unit .
[0013]
(2) The data storage method according to (1), wherein a flash memory is applied as the information storage medium.
[0014]
(3) The data storage according to (1), wherein the state holding unit holds information indicating which of the three states, ie, the invalid state, the temporary write state, and the valid state, can be identified. Method.
[0015]
(4) The data storage method according to (1) , wherein writing of information in the first state storage unit and the second state storage unit can be performed independently .
[0016]
(5) The information representing the three states held in the state holding unit is represented by a combination of the presence / absence state of one 1-byte information and the presence / absence state of the other 1-byte information in the 2-byte information. data storage method (1) to be so.
[0017]
(6) When recovery is performed after the data write operation to the data area is interrupted, each management unit area in which information indicating the invalid state is held in the state holding unit of the management information area The management unit area corresponding to the final rank of the data validation operation when the interruption occurs in the sequence rank of the unit area is searched, and each management information corresponding to the management information corresponding to the final rank is used as a starting point. (3) The data storage method according to (3) , wherein the data in each corresponding management unit area is recovered by retroactively tracking the sequence order based on information held in the preceding rank holding unit of the area .
[0019]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, preferred embodiments of a data storage method according to the present invention will be described in detail with reference to the drawings.
[0020]
FIG. 1 is a conceptual diagram showing a configuration of a flash memory as an example of an information storage medium applied to the present invention. The flash memory 1 itself is a general one. In the example shown in the figure, a plurality of sets of bits, which are the minimum storage unit, constitute a byte, which is a storage unit, and page 2 is composed of several tens of thousands of bytes, and this page 2 is collectively handled as the minimum erase unit. That is, in this example, page 2 forms a management unit area. An arbitrary byte can be written once after erasing the page. When the byte once written is changed, it is necessary to erase the page once. Any byte in the flash memory 1 can be read at any time.
[0021]
FIG. 2 is a diagram showing a setting state of the storage area classification of each page of the flash memory of FIG. In this embodiment, each page 2 of the flash memory 1 is divided into a data area 3 and a management information area 4 as shown in the figure. The data area 3 is an area for storing data that is main information such as images, sounds, and other data that is originally intended to be stored in the flash memory. The management information area 4 is an area for storing management information relating to data stored in the data area 3, such as the position of each page, as will be described in detail later.
[0022]
FIG. 3 is a conceptual diagram showing the configuration of the management information area 4 of FIG. A first state storage unit 5 and a second state storage unit 6 are allocated to the management information area 4. The first state storage unit 5 includes a state holding unit 7 for holding information indicating the storage state of data, and a data position holding unit (in the drawing) for holding data position information indicating the location of the corresponding data. 8) and information indicating the rank (page position) of the management unit area preceding the rank (self page) to which the self belongs in the series of each management unit area (page). For this purpose, a preceding rank holding unit (indicated as “preceding page position” in the figure) 9 is included. The state holding unit 7 holds one of three states of “invalid state (I)”, “temporary writing state (T)”, and “valid state (V)”.
[0023]
The second state storage unit 6 includes a state holding unit 10 similar to the state holding unit 7 and a preceding rank holding unit (indicated as “preceding page position” in the figure) 11 similar to the preceding rank holding unit 9. This state holding unit 10 also holds any one of three states of “invalid state (I)”, “temporary writing state (T)”, and “valid state (V)”.
In the initial state, the state holding unit 7 of the first state storage unit 5 and the state holding unit 10 of the second state storage unit 6 hold the “invalid state (I)”.
[0024]
In the present embodiment, the information in the state holding units 7 and 10 is changed from the “invalid state (I)” to the “provisional writing state (T)” or from the “temporary writing state (T)” to the “valid state (V)”. Need to change. This means that 2 bytes are assigned to the status holding unit. If both bytes are not written with data, it is “invalid (I)”, and if only one byte is written, it is “tentatively written (T)”. If data is written in both bytes, it can be easily realized by expressing the “valid state (V)”.
[0025]
By writing information “temporary writing state (T)” indicating that writing is in progress to the state holding units 7 and 10, even when writing is interrupted, it is possible to return to the state before writing. Become. Furthermore, when writing data that needs to be written in a lump like data and its management information, as described in detail later with reference to FIGS. To remember. As a result, even when writing is interrupted while the state holding unit is being changed, a series of pages can be searched and the recovery process can be performed.
[0026]
Next, the operation of this embodiment will be described. When data is written to the flash memory 1, a plurality of sets of a block A which is a part of the data and a position B of the data are given. These are, for example, file index information and file content data, and if the writing is interrupted due to a power failure or the like when only one of them is written, data consistency is lost. In the present embodiment, a procedure for writing these data as a group is provided.
[0027]
FIG. 4 is a flowchart showing a procedure for writing data into the flash memory 1. In writing data, the flash memory is first searched to check whether data at the same position has already been stored (4-1).
[0028]
FIG. 5 shows details of the search procedure (4-1) in FIG. In the search procedure (4-1) for the same part, the management information area 4 of each page 2 of the flash memory is read (5-1), and the state holding unit 7 of the first state storage unit 5 reads “invalid state (I)”. (5-2) and the state holding unit 10 of the second state storage unit 6 is “invalid state (I)” (5-3), and the data position of the first state storage unit 5 A page having the same value as the position B of data to be written 8 is searched (5-4). By the above procedure, a page in which data at the same position is stored is searched.
[0029]
As described above, when a page storing data at the same position is not found by the procedure (4-1) described with reference to FIG. 5, a search for an empty page is performed next (4-2).
[0030]
FIG. 6 shows details of the empty page search (4-2) procedure in FIG. In the empty page search procedure (4-2), the management information area 4 of each page 2 and 2 of the flash memory is read (6-1), and the information in the state holding unit 7 of the first state storage unit 5 is “invalid. A page in the “state (I)” is searched (6-2). Thereby, a free page is searched.
[0031]
Next, writing to an empty page is performed (4-3). In the empty page writing procedure, block data A is written in the data area 3, the data position B to be written is stored in the data position 8 of the first status storage unit 5, and the page written immediately before in the preceding page number 9 is stored. The number is stored, and finally the “temporary writing state (T)” is stored in the state holding unit 7. However, if it is the first block in one writing procedure, nothing is written in the preceding page number 9.
[0032]
FIG. 7 shows an example of the state of the management information area 4 after the execution of the procedure for the empty page writing procedure (4-3).
[0033]
Next, returning to FIG. 4, the procedure when a page storing data at the same position is found in the search procedure (4-1) for the same portion will be described. In this case, the page is invalidated next (4-4). In page invalidation, the number of the page written immediately before is stored in the preceding page number 11 of the second state storage unit of the page searched in the search procedure (4-1) for the same part, and “ “Temporary writing state (T)” is stored. However, if it is the first block in one writing procedure, nothing is written in the preceding page number 11. After the page invalidation procedure (4-4), the empty page search procedure (4-2) and the empty page writing procedure (4-3) are performed.
[0034]
FIG. 8 shows an example of the state of the management information area 4 after executing these page invalidation procedures (4-4).
[0035]
After the above procedure is executed for all the given blocks, the page validation procedure (4-5) is executed next. In the page validation procedure, starting from the first written page, each page is validated in the written order. In the validation of the page, the status holding units (7, 10) of the first status storage unit 5 and the second status storage unit 6 are referred to, and if this is the “provisional write status (T)”, “valid” Change to "Status (V)".
[0036]
As an example, FIG. 9 shows an example of the state of the management information area 4 after the page validation procedure (4-5) is executed from the state of FIG.
[0037]
Finally, the page erase procedure (4-6) is executed.
In this procedure, the state holding unit 10 of the second state storage unit 6 erases all pages in the “valid state (V)”.
[0038]
Next, the data reading procedure will be described. FIG. 10 is a flowchart showing a data reading procedure. In reading data, a page in which a block at a target position is written is searched from the flash memory. In the reading procedure, the management information area 4 of each page of the flash memory is read (10-1), the state holding unit 10 of the second state storage unit 6 is “invalid state (I)” (10-2), A page in which the state holding unit 7 of the first state storage unit 5 is in the “valid state (V)” (10-3) and the data position 8 of the first state storage unit 5 is the position of the target block (10-4). As a result, the page in which the data of the target block is stored is searched, and data is read from the data area 3 of this page.
[0039]
In a device incorporating a flash memory, if the power is turned off while data is being written to the flash memory, data may be partially written. In this case, a data recovery procedure is executed in order to recover the state before the data is written.
[0040]
The data recovery procedure is shown in the flowchart of FIG.
In data recovery, first, the information of the state holding units (7, 10) of the first state storage unit 5 and the second state storage unit 6 is referred to, and the page in which this information is “provisional write state (T)” (11-1). If no page is found, the procedure proceeds to the empty page recovery procedure (11-4). If found, it means that writing has not been completed, and it is necessary to return to the state before starting writing. For this purpose, the last page search (11-2) is performed. The last page search is a procedure for searching the last written page in the writing process, and can be easily realized by examining the management information area 4 of each page. When the last page is retrieved, a series of pages that were being written in the data writing procedure are retrieved and restored (11-3). In this procedure, the previous page number (9, 11) in the first state storage unit 5 or the second state storage unit 6 of each page is started from the last page searched by the last page search (11-2). ), The pages are processed in the reverse order of writing, that is, the sequence rank (page number) is retroactively based on the information held in the preceding rank holding section of each corresponding management information area. Run by doing.
[0041]
FIG. 12 is a flowchart showing details of the procedure for retroactively retrieving the sequence rank (page number) from the last page searched in this way.
First, the management information area 4 of the current page (in the beginning of the procedure, indicates the last page) is referenced (12-1). When the state holding unit 10 of the second state storage unit 6 is “invalid state (I)” (12-2), the preceding page number 11 of the second state storage unit 6 is stored as the next page number. (12-3). Further, the data written in the data area 3 of the current page is written again by the same procedure as the writing procedure (12-4). When the state holding unit 10 of the second state storage unit 6 is not “invalid state (I)”, the preceding page number 9 of the first state storage unit 5 is stored as the next page number (12-5). ). Further, the current page is deleted (12-6).
[0042]
In the final page search (11-2), when a page to be restored is not found, a free page is restored (11-4). Recovering a free page is a procedure for recovering a page that has been written but has not been erased due to a power failure during erasing. In the free page recovery (11-4), the management information area 4 of each page is referred to, and the state holding unit 10 of the second state storage unit 6 deletes the page in the “valid state (V)”.
[0043]
According to the present embodiment, as described above, even when the power is turned off during execution of the write procedure, the write procedure can be completed or the state before the write procedure is executed can be restored.
[0044]
To summarize the operation and effect of the above-described embodiment of the present invention, first, even if the write procedure is interrupted due to the power being cut off, it is possible to return to the state before the start of the write procedure. is there. As a result, a state in which data consistency is lost such that the file name is written but the actual data is not written does not occur. The reason is that a management information area having two state storage units is provided, writing is performed while leaving the management information before writing, and based on the preceding page number which is information held in the preceding rank holding unit This is because even if this operation is interrupted during the writing of the state holding unit, the previous page number is traced so that the page being written can be retrieved retroactively.
[0045]
Second, the present system can be used even in a storage device such as a flash memory which can be written only once after erasure. The reason is that by providing the first and second state storage units in the management information area, writing to the same place is performed only once, and writing is performed without overwriting the same area. Because it can.
[0046]
The configuration and operation of the preferred embodiment of the data storage method according to the present invention have been described in detail above. However, such an embodiment is merely an example of the present invention and does not limit the present invention. Those skilled in the art will readily understand that various modifications and changes can be made according to a specific application without departing from the gist of the present invention.
[0047]
【The invention's effect】
As described above, according to the data storage method of the present invention, even if the data write operation to the data area is interrupted, the data storage method that can be reliably restored in a short time with a simple procedure. Can be provided.
[Brief description of the drawings]
FIG. 1 is a conceptual diagram showing a configuration of a flash memory as an example of an information storage medium applied to the present invention.
2 is a diagram showing a setting state of a storage area classification of each page of the flash memory of FIG. 1; FIG.
FIG. 3 is a conceptual diagram showing a configuration of a management information area in FIG. 2;
FIG. 4 is a flowchart showing a procedure for writing data to a flash memory.
FIG. 5 is a flowchart showing details of a search procedure in FIG. 4;
6 is a flowchart showing details of an empty page search procedure in FIG. 4;
FIG. 7 is a conceptual diagram illustrating an example of a state of a management information area after a free page writing procedure is executed.
FIG. 8 is a conceptual diagram illustrating an example of a state of a management information area after executing a page invalidation procedure.
9 is a conceptual diagram showing an example of a state of a management information area after executing a page validation procedure from the state of FIG.
FIG. 10 is a flowchart illustrating a data reading procedure.
FIG. 11 is a flowchart showing a data recovery procedure.
FIG. 12 is a flowchart showing details of a procedure for retroactively retrieving the sequence rank from the last page searched.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 Flash memory 2 Page 3 Data area 4 Management information area 5 1st state memory | storage part 6 2nd state memory | storage part 7 State holding part 8 Data position holding part 9 Precedence order holding part 10 State holding part 11 Precedence order holding part

Claims (6)

A data area for storing data as main information in the same management unit area under the condition that the management unit area is an area having a predetermined capacity larger than a predetermined minimum storage unit on the information storage medium, and the data A management information area for storing such management information, and in the management information area, a state holding unit for holding storage status information indicating an invalid state, a temporary write state, and a valid state of data, and the information storage A preceding rank holding unit for holding information indicating the writing rank of the management unit area written before the corresponding writing rank in the series of writing ranks of each management unit area set on the medium, and the corresponding A first state storage unit including a data position holding unit for holding data position information indicating the location of the data to be stored, the state holding unit and the preceding order Set the second state storage unit comprising a holding portion, information writing operation of data to the data area when to recover from being interrupted, which is held in the state holding unit of the management information area, prior ranking A data storage method comprising: recovering data for which writing has been interrupted using information held in a holding unit and a data position holding unit .   The data storage method according to claim 1, wherein a flash memory is applied as the information storage medium. The information that indicates which of the three states, the invalid state, the provisional write state, and the valid state, can be identified in the state holding unit. Data storage method. 2. The data storage method according to claim 1, wherein writing of information in the first state storage unit and the second state storage unit can be performed independently . The information representing the three states held in the state holding unit is represented by a combination of the presence / absence state of one 1-byte information and the presence / absence state of the other 1-byte information in the 2-byte information. The data storage method according to claim 1. When recovery is performed after the data write operation to the data area is interrupted, each management unit area is related to the management unit area in which information indicating the invalid state is held in the state holding unit of the management information area. The management unit area corresponding to the final rank of the data validation operation when the interruption occurs in the sequence rank is searched, and the preceding management information area corresponding to the management information area corresponding to the final rank is used as a starting point. 4. The data storage method according to claim 3, wherein the data in each corresponding management unit area is restored by retroactively tracking the series based on information held in the rank holding unit .

Images