JP4863472B2 - Memory management method - Google Patents

Description

  The present invention relates to a management technique for a memory in which a defective area exists.

  As a memory for storing content data such as an application program, a NOR flash memory is used. The NOR type flash memory is a memory free from errors (a memory in which no defective area exists or a memory in which no bit error occurs), and is suitable for storing contents. For example, in a mobile phone, a portable information terminal, a pachinko machine, and a pachislot machine, a NOR flash memory is used for storing content data such as application programs.

  In order to check the validity of the content data stored in the NOR flash memory, it is effective to use a checksum. FIG. 6 shows how user data such as application data is stored in the NOR flash memory. In order to check the validity of the user data, data is simply read from all physical addresses, and the checksum of the user data is calculated. Then, the validity is judged by comparing with a correct checksum calculated in advance.

  A method for checking ROM data using a checksum is also disclosed in Patent Document 1 below. Patent Document 2 below discloses a technique for rewriting data without changing the checksum recorded in the ROM.

JP 2002-108723 A JP 2002-351749 A

  As described above, in the NOR flash memory, the data correctness check using the checksum is effective, and the content memory can be supplied stably. However, the NOR flash memory has an excellent feature that there is no error, but has a problem that the unit price of the memory is high.

  For example, even if it is considered for storing contents of a pachinko machine or a pachislot machine, the number of devices is so large that the influence of the memory unit price is large. Therefore, if the memory unit price can be reduced, the product supply cost can be significantly reduced.

  Therefore, an option of using a NAND flash memory that is less expensive than a NOR flash memory is conceivable. The NAND flash memory is an erroneous memory (a memory in which a defective area exists or a memory in which a bit error occurs), but realizes a large capacity and a low cost.

  FIG. 7 shows a state in which user data is stored in the NAND flash memory. As shown in the drawing, a defective area (INVALID BLOCK or BAD BLOCK) is mixed in the storage area of the NAND flash memory. The defective area is an area that cannot be used when the device is shipped. Since data cannot be recorded in the defective area, the data is stored in another alternative area. The substitution information indicating the correspondence between the defective area and the substitution area is stored in the memory as error control data as shown in the figure.

  The problem here is checking data validity. As described above, in the NOR flash memory, the same checksum can be calculated for all the solids by simple reading of all physical addresses. However, in the case of a NAND flash memory, data that could not be stored in the defective area is stored in the alternative area, and therefore the data storage location differs from device to device. For example, when the checksum is calculated in consideration of the stored address, the checksum changes depending on the device if the occurrence location of the defective area differs depending on the device.

  In the NAND flash memory, since there is a variation in defective area for each device, data is usually read by converting a logical address and a physical address by some method. For this reason, the contents cannot be confirmed by a method in which all physical addresses are accessed by the same method as a memory having no error and a checksum is calculated for the read data.

  In the first place, since the data value is indefinite for the defective area, the checksum is different for each device. Therefore, in any case, the data validity cannot be checked even if the same method as that for an error-free memory is used as it is.

  In view of the above problems, an object of the present invention is to provide a management technique for checking the validity of data when using an erroneous memory.

  In order to solve the above-described problem, the invention according to claim 1 is a method for managing a memory in which user data is stored, the step of acquiring defective area information from the memory, and based on the acquired defective area information. A step of creating error management information including replacement information of a defective area of the memory, a step of extracting unique information from the error management information and storing it in a predetermined storage means, and a step of storing user data in the memory And storing the error management information in the memory.

  According to a second aspect of the present invention, in the memory management method according to the first aspect, the unique information includes a checksum calculated from the error management information.

  According to a third aspect of the present invention, in the memory management method according to the first aspect, the unique information includes a hash value calculated from the error management information.

  According to a fourth aspect of the present invention, in the memory management method according to the first aspect, the unique information includes the number of blocks in the defective area of the memory.

  According to a fifth aspect of the present invention, in the memory management method according to the first aspect, the unique information includes address information of a defective area of the memory.

  A sixth aspect of the present invention is the memory management method according to any one of the first to fifth aspects, wherein the memory includes a NAND flash memory.

  According to a seventh aspect of the present invention, in the memory management method according to any one of the first to sixth aspects, the predetermined storage unit includes the memory.

  The invention according to claim 8 is a method for managing a memory in which a defective area in which user data is stored is present, wherein specific information is extracted from the stored user data using a predetermined algorithm and stored in a predetermined storage means. A step of storing, a step of acquiring defective area information from the memory, a step of creating error management information including replacement information of a defective area of the memory based on the acquired defective area information, and user data in the memory , Storing the error management information in the memory, and reading out the data of the replaced area based on the replacement information for the defective area, thereby emulating the reading process for the NOR flash memory The process of reading all user data from the memory and the predetermined altitude from the read user data. Extracting the information using the same algorithm as the rhythm, and comparing with the unique information stored in the storage means to determine the validity of the stored user data. To do.

  According to a ninth aspect of the present invention, in the memory management method according to the eighth aspect, the specific information includes a checksum calculated from the error management information.

  According to a tenth aspect of the present invention, in the memory management method according to the eighth aspect, the unique information includes a hash value calculated from the error management information.

  The memory management method of the present invention creates error management information based on the defective area information acquired from the memory, extracts unique information from the error management information, and stores it in a predetermined storage means. Thus, different unique information is stored for each device of the memory in which the user data is stored, and a means for confirming that the user data is correctly recorded based on the alternative information can be provided.

  Further, the memory management method of the present invention reads and holds all user data from a memory in which a defective area exists (for example, a NAND flash memory) by emulating a read process for a NOR flash memory. The validity of the user data is determined by comparing with the unique information. Thereby, it is possible to check the validity of the user data by the same method used in the NOR type flash memory.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a diagram showing a NAND flash memory 1 (hereinafter referred to as a memory 1) and a writing device 2 for writing data into the memory 1.

  For example, the memory 1 is used for writing content data of a pachinko machine or a pachislot machine. As described above, the memory 1 is a NAND flash memory, and includes a normal area 11 where data can be written and a defective area 12 where data cannot be written, as shown in FIG. It is out.

  The writing device 2 is given user data 51 created in a separate process. When the user data 51 is recorded in the memory 1 by the writing device 2, the error control data 52 is recorded together with the user data 51, as shown in FIG. The error control data 52 includes replacement information indicating which replacement area each defective area is replaced with. Further, the error control data 52 includes information (ECC and the like) necessary for correcting bit errors. Further, the error control data 52 may record the number of blocks in the defective area, address information (address list) of the defective area, and the like.

  In the example described above, the user data 51 is content data of a pachinko machine or a pachislot machine. The user data 51 is created in a separate process. This user data 51 is common to all memories that are mass-produced. That is, the user data 51 is stored in the NAND flash memory one after another, thereby creating a content storage memory.

  For this common user data 51, unique information is calculated in a further step. The unique information is, for example, a checksum or a hash value. This unique information is stored in the unique information storage unit 3.

  The writing device 2 is a device that stores user data 51 in a memory 1 that is a NAND flash memory. Hereinafter, a data writing procedure for the memory 1 by the writing device 2 will be described with reference to the flowchart of FIG.

  First, when the writing device 2 is connected to a new memory 1, the writing device 2 accesses the memory 1 and acquires defective area information of the memory 1 (step S1). That is, the writing device 2 acquires the addresses of all defective areas and the addresses of areas where each defective area is replaced. Thereby, the number of blocks in the defective area can be obtained.

  Next, the writing device 2 creates error control data 52 based on the defective area information acquired in step S1 (step S2). The error control data 52 includes alternative information. The replacement information is information indicating a correspondence relationship between the defective area and the replacement area. The error control data 52 stores information necessary for correcting bit errors. Further, the error control data 52 may store the number of blocks in the defective area and the address information of the defective area.

  Next, the writing device 2 calculates unique information of the error control data 52 created in step S2 (step S3). Alternatively, the writing device 2 obtains unique information from the information included in the error control data 52 created in step S2.

  As the unique information, for example, a checksum can be used. FIG. 3 is a diagram illustrating specific information using a checksum. First, as described above, for the user data 51, the checksum of the user data 51 is calculated in a separate process in advance. Then, the checksum of the user data 51 and the number of bytes of the user data 51 are stored in the unique information storage unit 3 as unique information.

  In the example shown in the figure, the checksum (8 digits) of the user data 51 calculated by the CRC calculation method is displayed in hexadecimal.

  This checksum is the same as the checksum calculated when it is assumed that the user data 51 is stored in an error-free memory (for example, a NOR flash memory). That is, as shown in FIG. 6, when user data is stored in order from the top address of a NOR flash memory in which no defective area exists, the stored data is calculated by sequentially reading all stored physical addresses. Same as checksum. Note that the checksum including the blank may be calculated.

  Since the user data 51 is data common to all devices, the unique information for the user data 51 is created in a separate process in advance as described above and stored in the unique information storage unit 3. In step S3, the writing device 2 calculates a checksum based on the error control data 52 that is different data depending on the device. Then, the calculated checksum and the number of blocks in the defective area are stored in the unique information storage unit 3 as unique information (step S4).

  In the example shown in the figure, the 8-digit checksum of the error control data 52 is displayed in hexadecimal. Further, instead of the number of blocks in the defective area, an address list of the defective area may be stored as unique information. Note that the number of defective area blocks and the address list of the defective area may be stored in the specific information storage unit 3 as specific information without being included in the error control data 52.

  Thus, for the user data 51, a common checksum, that is, the same value as the checksum calculated when stored in an error-free memory is calculated and held, and the NAND in which a defective area exists is stored. The unique information is also stored for the error control data 52 that is different for each device in the type flash memory.

  As the unique information, a hash value can be used. FIG. 4 is a diagram showing unique information using the hash value of SHA1. First, as described above, the hash value of the user data 51 is calculated from the user data 51 in a separate process in advance. Then, the hash value of the user data 51 and the number of bytes of the user data 51 are stored in the unique information storage unit 3 as unique information. Note that MD5 may be used to calculate the hash value.

  As in the case of the checksum, since the user data 51 is common to all devices, the unique information for the user data 51 is created in advance in a separate process as described above and stored in the unique information storage unit 3. Has been. In step S3, the writing device 2 calculates a hash value based on the error control data 52 that is different data depending on the device. Then, the calculated hash value and the number of blocks in the defective area are stored in the unique information storage unit 3 as unique information (step S4). Alternatively, instead of the number of blocks in the defective area, the address list of the defective area may be stored as unique information.

  In step S4, the storage unit for storing the unique information may be in the storage area of the memory 1 or outside the memory 1 as in this embodiment.

  After storing the unique information of the error control data 52 in the unique information storage unit 3 in step S4, the writing device 2 writes the user data 51 and the error control data 52 in the memory 1 (step S5). In this way, the memory 1 in which the user data 51 and the error control data 52 are written is in the state shown in FIG. In FIG. 1B, the defective area 12 is mixed only in the writing area of the user data 51. Of course, the defective area 12 is also present in the writing area of the error control data 52. is there.

  As described above, in the present embodiment, for the user data 51 that is data common to all devices, a common checksum and hash value are calculated, and the data size (number of bytes) of the user data 51 is set. Stored as unique information. Thereby, it can be used as information for checking the validity of the user data 51.

  Further, in the present embodiment, a checksum and a hash value are calculated for each error control data 52, which is different data in each device, and held as unique information. Here, the error control data 52 includes defect area replacement information. Therefore, by using this unique information, it is possible to provide information for confirming that data is properly stored based on the substitute information. In addition, since the number of blocks and address information of the defective area are held as the unique information, it is possible to confirm the state in which the defective area is correctly replaced and maintained for the erroneous NAND flash memory. is there.

  FIG. 5 is a diagram showing a system for checking the validity of the memory 1 managed in the above procedure. This system includes a host computer 5 and a memory device 10 including a memory 1. The memory device 10 includes a memory 1 and a controller 4 that controls access to the memory 1.

  The controller 4 can set the NOR interface mode as the operation mode. When the controller 4 is set in the NOR interface mode, when data is continuously read from the memory 1 for all physical addresses, information on the alternative area is read for the defective area. Read data is transferred to the host computer 5. That is, the host computer 5 can input read data by the same procedure and interface as when the user data 51 is stored in the NOR flash memory.

  The host computer 5 calculates a checksum or hash value for the user data 51 read from the memory device 10 using the same algorithm as described in the above embodiment. The validity of the data can be checked by comparing the checksum or hash value of the user data 51 held in the unique information storage unit 3 with the calculated value.

  As described above, according to the present embodiment, the user data stored in the NAND flash memory is also checked for data validity by using the same unique information as the user data stored in the NOR flash memory. Can be done.

  In the above embodiment, the algorithm for calculating the unique information from the user data 51 and the error control data 52 has been described using the CRC calculation method, SHA1, MD5, etc. as an example, but it is used for calculating the unique information. The algorithm is not particularly limited.

  The management method of the memory 1 described in the present embodiment can be used to check the validity of the data stored in the memory 1. Therefore, when the data stored in the memory 1 is read, it is possible to detect falsification of the data by referring to the unique information stored in the unique information storage unit 3. When tampering is detected, it is effective to control so that the system does not operate. Further, if parameters related to the user data 51 are also inserted in the error control data 52, it is possible to effectively detect falsification of the user data 51.

It is a figure which shows the memory and writing device which concern on this Embodiment. It is a write-in flowchart of user data and error control data. It is a figure which shows the checksum as specific information. It is a figure which shows the hash value as specific information. It is a figure which shows the system which checks the correctness of user data. It is a figure which shows the storage form of the user data with respect to a NOR type flash memory. It is a figure which shows the storage form of the user data with respect to NAND type flash memory.

Explanation of symbols

1 Memory (NAND flash memory)
11 Normal area 12 Defective area 51 User data 52 Error control data

Claims (10)

A method for managing memory in which user data is stored,
Obtaining defective area information from the memory;
Creating error management information including replacement information of the defective area of the memory based on the acquired defective area information;
Extracting unique information from the error management information and storing it in a predetermined storage means;
Storing user data in the memory;
Storing the error management information in the memory;
A memory management method comprising: The memory management method according to claim 1,
The specific information is
A checksum calculated from the error management information,
A memory management method. The memory management method according to claim 1,
The specific information is
A hash value calculated from the error management information,
A memory management method. The memory management method according to claim 1,
The specific information is
The number of blocks in the defective area of the memory,
A memory management method. The memory management method according to claim 1,
The specific information is
Address information of a defective area of the memory,
A memory management method. The memory management method according to any one of claims 1 to 5,
The memory is
NAND flash memory,
A memory management method. The memory management method according to any one of claims 1 to 6,
The predetermined storage means is
The memory,
A memory management method. A method for managing a memory in which a defective area in which user data is stored exists.
Extracting specific information from user data to be stored using a predetermined algorithm and storing it in a predetermined storage means;
Obtaining defective area information from the memory;
Creating error management information including replacement information of the defective area of the memory based on the acquired defective area information;
Storing user data in the memory;
Storing the error management information in the memory;
For the defective area, the process of reading all user data from the memory by emulating the reading process for the NOR type flash memory by reading the data of the replaced area based on the replacement information;
A step of extracting the information from the read user data using the same algorithm as the predetermined algorithm and comparing the specific information stored in the storage means to determine the validity of the stored user data When,
A memory management method comprising: The memory management method according to claim 8.
The specific information is
A checksum calculated from the error management information,
A memory management method. The memory management method according to claim 8.
The specific information is
A hash value calculated from the error management information,
A memory management method.

Images