JP3741535B2 - Nonvolatile storage device - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a technique that is effective when applied to a data storage system in an electrically writable / erasable nonvolatile semiconductor memory and a sector management system of the nonvolatile memory. For example, data can be erased collectively in units of blocks. The present invention relates to a technology that is effective when used for a memory card incorporating a plurality of flash memory chips.
[0002]
[Prior art]
Conventionally, in a memory card incorporating a plurality of flash memory chips, for example, data is stored with a unit such as 528 bytes as one sector. In addition, flash memory has fewer rewritable times than volatile memory such as RAM, so if there is uneven rewriting frequency such as concentration of rewriting on some storage elements, the life of the memory chip will be shortened that much faster. Will end up. Therefore, in the flash memory card, an operation of sector replacement is performed in order to average the rewrite frequency of the storage element.
[0003]
Further, in a flash memory card, a spare sector is provided to replace a defective sector in order to compensate for a sufficient storage capacity even if some sectors are defective in writing. In order to enable replacement and replacement of the sector as described above, the sector is provided with an area for storing management information in addition to the data area, and this management information is read when data is written to the sector. Sector management.
[0004]
[Problems to be solved by the invention]
In the above memory card, if the power is cut off while writing data to the memory chip, the data in the write target sector will be destroyed, so the management information will be lost and the sector will be judged as a bad sector. Will be. In addition, if the sector to be written is replaced, it may be determined as a bad sector even when another sector that is paired with the sector is accessed. As a result, there is a problem in that the memory space of the card becomes discontinuous and the actual storage capacity decreases.
[0005]
An object of the present invention is to provide a data management technique for a memory card that can repair a sector in which data has been destroyed due to power interruption or the like and prevent a reduction in actual storage capacity due to data destruction.
[0006]
The above and other objects and novel features of the present invention will become apparent from the description of the present specification and the accompanying drawings.
[0007]
[Means for Solving the Problems]
Outlines of representative ones of the inventions disclosed in the present application will be described as follows.
[0008]
That is, in the nonvolatile storage device configured to store data in units of sectors including the data storage unit and the management information storage unit, the management information is not read even if the management information of the target sector is read when data is written. If it is not obtained, a replacement sector that is paired with the sector to be written with reference to the management information storage section of all sectors and in which the data of the other sector forming a pair is stored in each data storage section When there is a replacement sector, information indicating the replacement sector is stored in the management information storage unit of the target sector, and data is written to the data storage unit of the replacement sector. It is. As a result, it is possible to repair a defective sector whose data has been destroyed due to, for example, the power being cut off during writing to a normal sector, and to prevent the memory space from becoming discontinuous and prevent a decrease in the actual storage capacity. be able to.
[0009]
Also, if management information cannot be obtained by reading management information of the sector at the time of data writing, it is registered as a bad sector by referring to a predetermined storage area before referring to the management information of all sectors If it is registered as a bad sector, data is written to the alternative sector. As a result, in a storage device having a spare sector, when a bad sector is accessed, writing to the replacement sector can be performed efficiently.
[0010]
Further, when the management information is not obtained even if the management information of the sector is read at the time of data writing and is not registered as a bad sector, the replacement is performed with reference to the management information of all sectors. When the sector does not exist, the sector is registered as a bad sector and a replacement sector is set and data is written to the replacement sector. Further, in this case, when the management information storage unit of the alternative sector has a sector pointer area in which the sector number is entered, information (self sector number) indicating itself is written in the sector pointer area. As a result, even if the data of the sector that has not been replaced is destroyed and becomes a defective sector, the overall storage capacity is not reduced.
[0011]
DETAILED DESCRIPTION OF THE INVENTION
Preferred embodiments of the present invention will be described below with reference to the drawings.
[0012]
FIG. 1 shows an embodiment of a flash memory card to which the present invention is applied.
[0013]
The memory card in this embodiment is not particularly limited, but the n flash memory chips FM1 to FMn, the interface and bus switching with the outside, the formation of the selection signal for each memory chip based on the address signal and the control signal, Controller chip CONT having functions such as ECC code generation and checking, sector number change based on write / read control to flash memory chip based on externally supplied command, management information read from memory chip, etc. A microprocessor MPU for performing the above and a work memory WMM made of DRAM or the like used as a work area of the microprocessor MPU are mounted on the printed wiring board 10 and are entirely molded by resin or the like.
[0014]
The controller chip CONT is composed of a gate array, and is connected to the flash memory chips FM1 to FMn and the working memory WMM via an address & control bus 11 and a data bus 12 formed on the substrate 10, and is connected to an external device. It is connected to an external terminal 13 that is inserted into a card slot such as a personal computer main body. The working memory WMM includes an address conversion table that is referenced when the microprocessor MPU performs conversion from a logical sector number (logical address) to a real sector number (physical address) when accessing the flash memory. It is used to store a bad sector table that is referred to when managing sectors in the flash memory, hold write data, and synthesize write data and management information.
[0015]
FIG. 2 shows a configuration example of sectors configured in the flash memory chips FL1 to FLn and an example of a defective sector table stored in the working memory WMM. Although not particularly limited, as shown in FIG. 2A, each sector has an original data storage unit DATA having a storage capacity of 512 bytes and 16 bytes in which management information of the sector is stored. Management information storage unit MGDT, the management information storage unit MGDT includes an erase count storage area, a sector pointer SP in which the number of a sector to be paired at the time of sector replacement is stored, and an ECC code of data in the data storage unit Includes a first ECC code area in which data is stored, a second ECC code area in which an ECC code of data in the management information storage unit MGDT is stored, and the like.
[0016]
In the memory card of this embodiment, the microprocessor MPU checks the number of erasures in the management information storage unit MGDT of each sector and finds another sector that does not reach the predetermined number when the predetermined number is reached. It is configured to replace a sector. In the sector replacement, two sectors are paired, and as shown in FIG. 3A, the sector numbers of each other are stored in the sector pointer SP of each sector, and the other sector is stored in the data storage unit DATA. This is done by storing data to be stored in. Further, the sector pointer SP of a sector that has not been replaced is stored with its own sector number.
[0017]
In this embodiment, as shown in FIG. 2B, the defective sector table includes a column for registering a replacement sector number (a spare sector number) for each sector number n-1, n, n + 1, n + 2,. The sector is provided, and all “1” data (FFFF in the case of 16 bits) is stored in the sector column that has not been replaced. Therefore, the microprocessor MPU can know whether or not the sector is a bad sector by referring to the bad sector table using the sector number, and if the sector is a bad sector, can know the number of the substitute sector.
[0018]
The defective sector table provided in the working memory WMM can be stored in the flash memory chip when the system is terminated, and read out from the flash memory chip when the system is started, and placed on the working memory WMM. As a result, information on bad sectors can be retained even when the power is cut off, and the work memory WMM composed of a DRAM having a high operation speed may be accessed without accessing a flash memory having a low operation speed during system startup. High-speed processing by a microprocessor becomes possible.
[0019]
Next, referring to the flowchart of FIG. 4 showing the processing procedure by the microprocessor MPU when a data write command is input from the outside to the memory card of the present embodiment, the method of writing data to the sector in the normal state and the power shutdown A method of repairing a sector whose data has been destroyed by the above will be described in detail.
[0020]
When a write command is issued (input) to the memory card from the outside, the microprocessor MPU refers to the address conversion table in the working memory WMM and uses the logical sector number input together with the command as the real sector number of the flash memory. (Procedure P2). Next, the management information of the sector is read using this real sector number to determine whether or not the management information has been read (procedure P3). Here, when the management information can be read, it is determined whether or not there is a sector replacement by looking at the sector pointer SP (procedure P4). Whether or not there is a sector replacement can be determined by whether or not the sector number in the sector pointer matches its own sector number. That is, it is determined that there is no replacement when the sector numbers match, and there is a replacement when they do not match.
[0021]
Next, when there is a replacement, write data is written in the data storage section of the replacement destination sector indicated by the sector pointer (procedure P5). This writing is performed after erasing all data in the sector. For this reason, the microprocessor reads the management information in the management information storage unit from the replacement destination sector before writing, and holds it in the built-in M. Then, the data in the replacement destination sector is erased, the number of times of erasure in the read management information is updated, and the updated management information and write data are synthesized on the work memory WMM, and the flash memory Write to the chip.
[0022]
On the other hand, when it is determined in the procedure P4 that there is no replacement, write data is written to the data storage section of the sector (procedure P6). Also at this time, the management information of the sector is read and the data is erased, and the updated management information and write data are written to the sector after updating the erase count in the read management information.
[0023]
When the management information cannot be read from the sector in the procedure P3, it is checked whether the sector number is registered in the bad sector table (procedure P7). Then, it is determined whether the sector is registered as a bad sector (procedure P8). If it has been registered, write data is written to the substitute sector using the substitute sector number registered in the table (procedure P9). Also in this case, the management information of the replacement sector is read to erase the data, and the updated management information and write data are written to the sector after updating the number of times of erasure in the read management information. .
[0024]
On the other hand, when it is determined in procedure P8 that the sector is not registered as a bad sector, management information is read from all sectors (procedure P10). Then, the sector pointer in the read management information is checked to determine whether or not there is one storing the sector as a replacement destination sector (procedure P11). If there is a sector having a stored sector pointer, the sector number of that sector is entered into the sector pointer of the sector, and write data is written into the data storage area of the replacement sector (procedure P12). Also in this case, the management information of the replacement destination sector is read, the data is erased, the sector pointer in the read management information is updated, and the updated management information and write data are written to the sector. .
[0025]
If the sector holding the sector number in the sector pointer is not found in the procedure P11, the spare sector is regarded as a bad sector and one unused sector is allocated as a substitute sector, and the substitute sector number is assigned. Is registered in the sector column of the bad sector table, and then write data is written to the alternative sector (procedure P13). It should be noted that the sector number in the management information of the alternative sector is set with the sector number. That is, the sector is not replaced for the time being.
[0026]
According to the above procedure, for example, even when the power is shut off during writing and data in the sector (sector j in the figure) is lost as shown in FIG. 3B, the replacement sector ( The sector pointer of i) remains in the sector pointer of i), which can be found by procedures P10 and P11. By writing the number of the replaced sector found in the sector pointer of FIG. ), The sector in which the management information is lost can be repaired as a usable sector. As a result, it is possible to prevent the memory space from becoming discontinuous and prevent a decrease in the actual storage capacity.
[0027]
The invention made by the present inventor has been specifically described based on the embodiments. However, the present invention is not limited to the above embodiments, and various modifications can be made without departing from the scope of the invention. Nor. For example, in the above-described embodiment, the registration column is provided on the bad sector table for all sectors, but the numbers of only the bad sectors and alternative sectors may be registered. In the embodiment, the controller chip and the microprocessor are separately configured. However, it is also possible to configure them as one semiconductor chip by a gate array or an ASIC or the like incorporating a CPU core as a macro cell. Further, in the embodiment, it has been described that the conversion from the logical sector number to the real sector number is performed using the address conversion table in the working memory WMM, but it is also possible for the microprocessor to perform the calculation.
[0028]
In the above description, the case where the invention made by the present inventor is applied to a memory card having a built-in flash memory, which is the field of use behind it, has been described. However, the present invention is not limited to this, and the EEPROM It can also be used for a memory card incorporating a chip or a RAM chip or a memory module in which a plurality of memory chips are mounted on a single board.
[0029]
【The invention's effect】
The effects obtained by the representative ones of the inventions disclosed in the present application will be briefly described as follows.
[0030]
In other words, the present invention has an effect that it is possible to repair a sector in which data has been destroyed due to power interruption or the like, thereby preventing a decrease in the actual storage capacity of the memory card due to data destruction.
[Brief description of the drawings]
FIG. 1 is a block diagram showing an embodiment of a flash memory card to which the present invention is applied.
FIG. 2 is an explanatory diagram illustrating a configuration example of a sector of a flash memory and a configuration example of a bad sector table according to an embodiment.
FIG. 3 is an explanatory diagram showing a state of sector replacement and recovery of a sector whose data has been destroyed;
FIG. 4 is a flowchart illustrating a processing procedure performed by a microprocessor MPU when a data write command is input from the outside to the memory card according to the embodiment.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 10 Printed wiring board 11 Address bus 12 Data bus 13 External terminal MPU Microprocessor CONT Controller chip FM1-FMn Flash memory chip

Claims (5)

In a non-volatile storage device configured to store data in units of sectors including a data storage unit and a management information storage unit, management information can be obtained even if the management information of the target sector is read when data is written. If not, whether there is a replacement sector that is paired with the write target sector with reference to the management information storage section of all sectors and in which each data storage section stores the data of the other sector in the pair When there is a replacement sector, information indicating the replacement sector is stored in the management information storage unit of the target sector, and data is written to the data storage unit of the replacement sector. A non-volatile storage device. If management information is not obtained even when reading the management information of the sector at the time of data writing, it is registered as a bad sector by referring to a predetermined storage area before referring to the management information of all sectors. 2. The non-volatile storage device according to claim 1, wherein if it is registered as a bad sector, data is written to a substitute sector. If the management information is not obtained even when reading the management information of the sector at the time of data writing, and if it is not registered as a bad sector, the replacement sector is referred to by referring to the management information of all sectors. 3. The non-volatile memory device according to claim 1, wherein when the non-existent sector does not exist, the sector is registered as a defective sector, and a replacement sector is set and data is written to the replacement sector. 4. The sector according to claim 1, wherein the sector includes a data storage unit and a management information storage unit, and the management information storage unit includes a sector pointer area in which a sector number of the replacement sector is entered. The non-volatile storage device described. 5. The non-volatile storage according to claim 1, wherein the sector pointer area in the management information storage section of the sector that has not been replaced is stored with its own sector number. apparatus.

Images