JPH11110983A - Controlling method for the number of times of erasing of flash memory and data processing device using it - Google Patents

Abstract

PROBLEM TO BE SOLVED: To realize a controlling method for the number of times of erasing of a flash memory in which a control information region is preserved even when power failure is caused during rewriting data, a control information region can be reduced, and the situation in which a control sector reaches the pre scribed number of times of erasing quickly can be evaded. SOLUTION: When a controlled object data sector is not yet erased, in all data of first to (n) bites of a control sector, $7F is written in a first bite B1 of the control sector in first erasing of an object data sector with $ FF, and a D7 bit is made '0'. D6-D1 bits of the first bite B1 are made '0' every erasing of 2nd to 7th of an object data sector, $00 is written in the first bite B1 in erasing of 8th in an object data sector, and a D0 bit of the first bite B1 is made '0'. If the control sectors are (n) bites in all, $01 is written in (n)th bites Bn in erasing of (8×n-1)th of an object data sector, a D1 bit of (n)th bite Bn is made '0'. $00 is written in (n)th bites Bn in erasing of (8×n)th of an object data sector, a D0 bit of (n)th bite Bn is made '0'.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for managing the number of erasures of a flash memory having a limited number of erasures.

[0002]

2. Description of the Related Art A flash memory is a non-volatile memory, and has a limited number of erasures. Erasing is performed for each specific unit byte (hereinafter "sector"). Then, by erasing, the data in the sector all have a value of “1” in bit units, and the write operation is performed with “1” for the bit value.
To “0” or “1” to “1”, or
"0" to "0" are allowed. Alternatively, the data in the sector becomes all "0" in bit units by being erased, and the write operation is performed with "0" for the bit value.
To “1” or “1” to “1”, or
"0" to "0" are allowed.

In a computer system using the above-mentioned flash memory, in a system where importance is placed on reliability, the number of erasures of the flash memory is managed. Conventionally, a method of managing the number of erasures as shown in FIGS. 6, 7, and 8 has been performed on a flash memory.

FIG. 6 is a diagram showing a method of mixing a data area and a management information area in the same sector A. In the method shown in FIG. 6, after erasing the sector A, the incremented erasure count information is recorded in the management information area.

FIG. 7 is a diagram showing a method of adding one management sector B to one data sector A. The data sector A has a data area and a management information area, and the sector B has only a management information area. When erasing sector A, first, sector B is erased and incremented erasure count information is recorded in the management information area of sector B. Then, after the erasure of the sector A, the erasure count information recorded in the sector B is recorded in the management information area.

FIG. 8 is a diagram showing a method of adding one management sector to a plurality of data sectors. Sector A, sector B, and sector C each have a data area and a management information area, and sector D includes sector A, sector B, and sector C.
And a management information area of sector D.

In FIG. 8, when erasing sector A, first, sector D is erased and incremented erasure count information is recorded in management information area A of sector D. Then, the incremented erase count information is recorded in the management information area D of the sector D.

After that, after erasing the sector A, the erasure count information of the management information area A recorded in the sector D is recorded in the management information area. In this method, after erasing sector D, or
A power failure occurs before recording the incremented erasure count information in the management information area, and the erasure count information cannot be recorded. Also, after erasing the sector A, or before recording the incremented erasure count information in the management information area. Even if a power failure occurs and the erase count information cannot be recorded,
The information on the number of times of erasure of the sectors A and D is stored. The method shown in FIG. 8 also alleviates the problem that the management information area for the data area becomes too large.

[0009] In addition, in FIG. 7, a management information area is formed only in sector B without forming a management information area in sector A, and in FIG. 8, a management information area of sector A, sector B, and sector C is formed. Is not formed, and a management information area is formed only in the sector D.

[0010]

However, the above-described methods of managing the number of times of erasure as shown in FIGS. 6, 7, and 8 have the following problems. The method shown in FIG. 6 is a method in which a data area and a management information area are mixed in the same sector. After the sector A is erased, incremented erasure count information is recorded in the management information area. According to this method, if a power failure occurs and the erase count information cannot be recorded after erasing the sector A or before recording the incremented erase count information in the management information area, the erase count for the sector A becomes unknown. Problem.

The method shown in FIG. 7 is a method of adding one management sector to one data sector. Sector A has a data area and a management information area, and sector B has only a management information area. Have. When erasing sector A,
First, the sector B is erased, and the incremented erase count information is recorded in the management information area of the sector B. Then, after the erasure of the sector A, the erasure count information recorded in the sector B is recorded in the management information area.

According to this method, after a power failure occurs and the erase count information cannot be recorded after erasing the sector B or before recording the incremented erase count information in the management information area, and after erasing the sector A, Alternatively, even if a power failure occurs and the erase count information cannot be recorded before the incremented erase count information is recorded in the management information area, the erase count information of the sector A and the sector B is stored, but the management of the data area is performed. There is a problem that the information area becomes too large.

The method shown in FIG. 8 is a method of adding one management sector to a plurality of data sectors. Sectors A, B, and C have a data area and a management information area. D has a management information area of sector A, sector B and sector C and a management information area of sector D.

When erasing sector A, first, sector D is erased, incremented erase count information is recorded in management information area A of sector D, and incremented erase count information is written in management information area D of sector D. Record. After that, after erasing the sector A, the erasure count information of the management information area A recorded in the sector D is recorded in the management information area.

In this method, after a power failure occurs and the erase count information cannot be recorded after erasing the sector D or before recording the incremented erase count information in the management information area, and after erasing the sector A, Alternatively, even if a power failure occurs before the incremented erase count information is recorded in the management information area and the erase count information cannot be recorded, the erase count information of the sectors A and D is stored.
Note that the problem that the management information area for the data area becomes too large is also alleviated.

However, this method has a problem that the sector D reaches the specified number of times of erasure earlier than the sectors A, B and C.

In FIG. 7, a management information area is not formed in sector A and a management information area is formed only in sector B. In FIG. 8, management information is not formed in sector A, sector B and sector C. As for the system having management information only in the sector D, the problem shown in FIGS. 6, 7 and 8 is that the management information area is not saved when a power failure occurs during data rewriting. On the other hand, the problem that the management information area is enlarged and the problem that the management sector reaches the specified number of times of erasure earlier than the data sector similarly occur.

As described above, in the conventional method, the problem that the management information area is not preserved when a power failure occurs during data rewriting, the problem that the management information area becomes large relative to the data area, and that the management sector is smaller than the data sector. There has been a problem that the specified number of erasures is reached quickly.

An object of the present invention is to save the management information area even when a power failure occurs during data rewriting, to reduce the management information area with respect to the data area, and to reduce the number of erasures by the management sector faster than the data sector. It is an object of the present invention to realize a flash memory erasure count management method and a data processing device using the same, which can avoid the situation where the number of flash memory reaches the limit.

[0020]

[Means for Solving the Problems]

(1) In order to achieve the above object, the present invention is configured as follows. That is, the number of erasures is limited, and erasing is performed for each sector which is a specific unit byte. By erasing, all the data in the sector are set to “1” in bit units, and the writing operation is performed in accordance with the bit value. "1" for
In the method of managing the number of times of erasing of a flash memory which is a nonvolatile memory that allows “0” to “0”, “1” to “1”, and “0” to “0”, a single sector is managed for a plurality of sectors. Assigned as a sector, this management sector,
Each of the plurality of sectors to be managed is divided into areas respectively corresponding to the plurality of sectors to be managed, and each time a management target sector is erased, “0” is sequentially set to consecutive bits in a plurality of bytes of the area corresponding to the management target sector in the management sector. By writing "" and counting the number of erasures, the life of the flash memory having a limited number of erasures is managed.

(2) Further, the number of erasures is limited, and erasing is performed for each sector which is a specific unit byte. By erasing, the data in the sector becomes all "0" in bit units, and the data is written. Operation is “0” for the bit value
In the method of managing the number of erasures of a flash memory which is a non-volatile memory which allows “1” to “1”, “1” to “1”, and “0” to “0”, a single sector is managed for a plurality of sectors. Assigned as a sector, this management sector,
Each of the plurality of sectors to be managed is divided into areas respectively corresponding to the plurality of sectors to be managed, and each time a management target sector is erased, “0” is sequentially set to consecutive bits in a plurality of bytes of the area corresponding to the management target sector in the management sector. By writing "" and counting the number of erasures, the life of the flash memory having a limited number of erasures is managed.

(3) Preferably, the above (1) or (2)
In the above, one management sector is allocated to a cluster which is a group of a plurality of sectors, and a specific address of an erasure frequency counter for the number of sectors for managing the number of erasures provided in the management sector is periodically assigned. Then, the data at the address is compared with predetermined specific data, and it is determined whether or not the number of erasures has reached a specified number of erasures.

(4) Preferably, in the above (3), a cluster information area indicating the state of the cluster is provided, and erase information of each sector in the cluster and information on the entire cluster are recorded in the cluster information area. By doing so, the information in the cluster is obtained only by the read operation of the cluster information area.

(5) Preferably, the above (1),
In (2) or (3), by writing a specific data pattern at a specific address of the management sector,
By detecting that the specific data pattern has been rewritten, it is detected that the management sector has been initialized.

(6) The MPU, the main storage means, and the number of erasures are limited, and erasing is performed for each sector, which is a specific unit byte. All become "1", and the write operation is performed on a flash memory, which is a nonvolatile memory that allows "1" to "0", "1" to "1", and "0" to "0" for bit values. In the data processing device, the flash memory has a management sector which is a single sector allocated to the plurality of sectors, and the management sector is located in an area corresponding to each of the plurality of sectors to be managed. Each time a sector to be managed is erased, “0” is sequentially written to consecutive bits in a plurality of bytes of an area corresponding to the sector to be managed in the managed sector to control the number of erases. By going to cement, to manage the life of the flash memory with limited number of erasures.

(7) The MPU, the main storage means, and the number of erasures are limited, and erasure is performed for each sector which is a specific unit byte. All become "0", and the write operation is performed on a flash memory, which is a nonvolatile memory that allows "0" to "1", "1" to "1", and "0" to "0" for bit values. In the data processing device, the flash memory has a management sector which is a single sector allocated to the plurality of sectors, and the management sector is located in an area corresponding to each of the plurality of sectors to be managed. Each time a sector to be managed is erased, “0” is sequentially written to consecutive bits in a plurality of bytes of an area corresponding to the sector to be managed in the managed sector to control the number of erases. By going to cement, to manage the life of the flash memory with limited number of erasures.

(8) Preferably, the above (6) or (7)
, The flash memory has one management sector assigned to a cluster which is a group of a plurality of sectors, and has a number of erase counters equal to the number of sectors for managing the number of erases in the management sector. By periodically reading a specific address of the erase counter, comparing data at that address with predetermined specific data, and determining whether the number of erases has reached a specified number of erases,
Manages the life of a flash memory with a limited number of erases.

(9) Preferably, in the above (8), the flash memory includes a cluster information area indicating a state of the cluster, and the cluster information area includes erase information of each sector in the cluster, By recording information on the whole, information in the cluster is obtained only by the read operation of the cluster information area.

(10) Preferably, the above (6),
In (7) or (8), by writing a specific data pattern at a specific address of the management sector,
By detecting that the specific data pattern has been rewritten, it is detected that the management sector has been initialized.

In the flash memory erase number management method and the data processing apparatus using the same according to the present invention, the write operation of the flash memory is performed by changing the bit value from "1" to "0" or from "1" to "1". Or from “0” to “0” is possible, and from the least significant byte to the most significant byte in a plurality of bytes allocated to the management target sector in the management sector, and from the least significant byte in the byte data. From the lower bit (hereinafter “LSB”) to the most significant bit (hereinafter “MSB”), or “0” to the next bit where “0” is written from the MSB to the LSB for erasure of the management target sector Writing and erasing are counted.

As a result, even when a power failure occurs during data rewriting, the management information area is preserved, the management information area for the data area is minimized, and the management sector reaches the specified erase count earlier than the data sector. It is possible to implement a flash memory erasure count management method that does not have any.

[0032]

Next, embodiments of the present invention will be described with reference to the accompanying drawings. FIG. 1 is a diagram showing a flash memory erasure count management method according to an embodiment of the present invention. In FIG. 1, each sector (specific unit byte) in the flash memory manages a data sector for storing normal data and the number of erases of a plurality of data sectors.
It is divided into one management sector, and shows a change in data in the management sector due to an increase in the number of erasures in that case.

The management sector is divided into areas corresponding to a plurality of data sectors to be managed, ie, a first byte B1 to an n-th byte Bn. In this flash memory, the value of all the data in sector 1 is changed to "1" in bit units by erasing, and the writing operation is performed by changing the bit value from "1" to "0" or
"1" to "1" or "0" to "0" are permitted.

If the data sector to be managed has never been erased after the start of the management of the number of erasures, the data from the first byte to the n-th byte of the management sector are all $ FF. When the first erasure is performed on the data sector to be managed, the D7 bit of the first byte B1 of the management sector is set to “0”.
$ 7F is written to the first byte B1.

Next, when the data sector to be managed is erased for the second time, D6 of the first byte B1 of the management sector is deleted.
To set the bit to “0”, $ 3F is written to the first byte B1. Every time the third to seventh erasure is performed on the data sector to be managed, D5 to D5 of the first byte B1 of the management sector are used.
The D1 bit is set to “0”. When the data sector to be managed is erased for the eighth time, the first sector of the management sector is deleted.
$ 00 is written to the first byte B1 to set the D0 bit of the byte B1 to "0".

Next, when the data sector to be managed is erased for the ninth time, D7 of the second byte B2 of the management sector is deleted.
$ 7F is written to the second byte B2 to set the bit to "0".

If the management sector is n bytes in total, and the (8 × n-1) -th erasure is performed on the data sector to be managed, the D1 of the nth byte Bn of the management sector is obtained.
$ 01 is written to the n-th byte Bn to set the bit to "0". When the data sector to be managed is erased for the 8 × n times, the data sector D0 of the n-th byte Bn of the management sector is erased.
$ 00 is written to the n-th byte Bn to set the bit to "0". In this way, every time the data sector to be managed is erased, the data sector is counted by a method of increasing “0” of the data in the management sector.

That is, from the least significant byte to the most significant byte in the plurality of bytes allocated to the management target sector in the management sector, and from the least significant bit (LSB) to the most significant bit (MSB) or MSB in the byte data. From LSB
Then, “0” is written to the next bit in which “0” is written for erasure of the management target sector, and the number of erasures is counted.

As a result, even when a power failure occurs during data rewriting, the management information area is preserved, the management information area can be reduced with respect to the data area, and the management sector reaches the specified erase count earlier than the data sector. Thus, it is possible to realize a method of managing the number of times of erasing of the flash memory, which can avoid the situation of being erased.

That is, according to the embodiment of the present invention, every time the management target sector is erased, “0” is sequentially set to consecutive bits in a plurality of bytes of the area corresponding to the management target sector in the management sector. Is written and the number of erases is counted. The data update in the management sector is 1-byte data write, and the data update in the management sector does not require the erase operation of the management sector as in the related art. Therefore, there is no problem that the power failure occurs and the erase count information cannot be recorded and the erase count for the data sector becomes unknown.

Further, since the management sector can have management information of a plurality of data sectors, the problem that the management information area for the data area becomes too large does not occur. Further, since the management sector does not require an erasing operation, there is no problem that the management sector reaches the specified number of times of erasure earlier than the data sector.

FIG. 2 shows a configuration in which the present invention is applied to a flash memory having a data capacity of 32 sectors and 16 M bits. In FIG. 2, the data capacity per sector is 64 Kbytes, and 100,000 times is the specified number of erasures. Sectors SA0 to SA6
(Cluster) is a data sector for storing instruction codes and data, and sector SA7 is a management sector for storing management information such as the number of data sector erasures shown in FIG.

Similarly, sectors SA8 to SA14
(Cluster) is a data sector for storing instruction codes and data, and sector SA15 is a management sector for storing management information such as the number of times of erasure of the data sector.

The sectors SA16 to SA22
(Cluster) is a data sector for storing instruction codes and data, and sector SA23 is a management sector for storing management information such as the number of times the data sector has been erased. Further, the sectors SA24 to SA30 (cluster) are data sectors for storing instruction codes and data, and the sector SA31 is a management sector for storing management information such as the number of erasures of the data sector.

FIG. 3 shows the configuration of the management sector SA7 shown in FIG. 2. The SA0 erase count information area (erase count counter) of the relative address $ 0000 to $ 1FFF contains the erase of the data sector SA0. Stores frequency information. Similarly, relative address $ 2000 to $ 3FFFF
The SA1 erase count information area stores the erase count information of the data sector SA1.

Also, the relative address $ 4000 to $ 5F
The FF SA2 erase count information area includes the data sector SA.
2 is stored. And the relative address ＄
The SA3 erase count information area of 6000 to $ 7FFF stores the erase count information of the data sector SA3. The SA4 erase count information area of the relative address $ 8000 to $ 9FFF stores the erase count information of the data sector SA4. Then, relative address {from A000}
The SA5 erase count information area of the BFFF stores the erase count information of the data sector SA5.

Further, the relative address $ C000 to $ DF
The FF SA6 erase count information area includes the data sector SA.
6 is stored. Relative address $ E000
The empty area SAE (cluster information area) from $ FFFF to #FFFF has a specific data pattern written therein, and indicates that this management sector has not been erased. In other words, when this management sector is erased, it is no longer the specific data pattern previously written, so whether or not this management sector has been erased is determined by whether or not the specific data pattern has been written. can do. Each erase count information area has a capacity of 8 kbytes, and manages the erase count by the erase count management method shown in FIG.

In the method of managing the number of erasures shown in FIG. 1, if there is a capacity of 8 kbytes, counting can be performed 8 × 1024 × 8 = 65536 times. The number of erasures specified in the flash memory device is 100,000. However, if the number of erasures is 50,000 by providing a margin for securing reliability, 65536 is a sufficient count.

As described above, the method of managing the number of times of erasing shown in FIG. 1 is applied to the usage configuration of the flash memory shown in FIGS.
It is possible to make 5% a data area (for example,
Since one management sector SA7 is provided for the data sectors SA0 to SA6, (SA0 to 7) / (SA0
-SA8) = 0.875). Therefore, it is possible to reduce the capacity of the management sector as compared with a configuration in which one data sector has one management sector (50% of the flash memory capacity is a data sector) as in the conventional example shown in FIG. Yes, a low-cost system can be constructed.

FIG. 4 is a diagram showing a method for checking whether or not the number of erasures stored in the erasure number information area SA has reached a prescribed number of erasures in one embodiment of the present invention. The capacity of the erasure count information area SA is 64 kbytes, and the prescribed erasure count is 50,000 times. When the erasing frequency management method shown in FIG. 1 is used, the erasing frequency exceeds 50,000 times when the data of the relative address $ 186B becomes $ 7F.

That is, when the data at the relative address $ 186B is $ FF, the number of erases has not reached 50,000 times. If the data of the relative address $ 186B is other than $ FF, it can be determined that the number of erasures has reached 50,000.

Thus, the monitoring task reads the data at this address at a specific cycle and checks whether the read data is $ FF or any other data, so that the number of erases of the management target sector has reached the specified number of erases. Can be confirmed. As a result, when the number of erasures of the management target sector has reached the specified number of erasures, it becomes possible to perform processing such as moving data to a replacement sector.

FIG. 5 is a diagram for comparing the embodiment of the present invention and the prior art with respect to the possibility of data loss during a power failure. In FIG. 5, 2 m is used to erase the data sector.
s, 2 μs per byte is required for writing data, and 131 ms is required for one sector.

Since the number of erasures of the data sector is managed, in the prior art, when rewriting the data sector, it takes 2 μs to erase the data sector, 131 ms to write the data sector, and 2 μs to erase the management sector. It takes 131 ms to write the management information to the sector.

On the other hand, in the erase count management method according to the present invention, 2 μs is used for erasing the data sector, 131 ms is used for writing the data sector, and 2 μs is used for writing the management sector.
s is required. In the case of the present invention, the erasing of the management sector is not necessary, and the number of erasures of the data sector is a method of increasing “0” of the data in the management sector. Compared to technology,
It has been greatly shortened.

In the prior art, the management information is updated by 1
It takes 33 ms, and if a power failure occurs during this time, the information of the management sector disappears or becomes incomplete and cannot be used.

On the other hand, in the erase count management method according to the present invention, the update of the management information requires only 2 μs. Therefore, even if a power failure occurs during this time, it takes several hundred μs until the power is turned off from the power failure warning interrupt. Since it is necessary, the information of the management sector does not become incomplete information.

As another embodiment of the present invention, the MPU
And a main storage device, and uses a flash memory managed by the above-described erase count management method. According to this data processing device, even when a power failure occurs during data rewriting, the management information area can be saved, and the management information area can be reduced with respect to the data area. Since such a situation can be avoided, a highly reliable data processing device for processed data and the like can be realized.

In the flash memory of the above-described example, the data in the sector is changed to "1" in bit units by erasing, and the writing operation is changed from "1" to "0" for the bit value. Or "1" to "1" or "0" to "0" is permitted, but the present invention is also applicable to other examples of flash memory.

That is, as a result of erasure, the data in the sector all have a value of “0” in bit units, and the write operation is performed by changing the bit value from “0” to “1” or
The present invention is also applicable to a flash memory in which “1” to “1” or “0” to “0” is permitted.

[0061]

Since the present invention is configured as described above, it has the following effects. In a method for managing the number of times of erasing of a flash memory having a limited number of times of erasing and a data processing device using the same, in the byte data from the least significant byte to the most significant byte among a plurality of bytes allocated to a management target sector in a management sector Then, the least significant bit (L
SB) to the most significant bit (MSB) or “0” for erasure of the management target sector from the MSB to the LSB
"0" is written to the next bit in which is written, and the number of erasures is counted to manage the number of erasures of the flash memory. Therefore, even when a power failure occurs during data rewriting, the management information area is preserved, the management information area for the data area is minimized, and the management sector does not reach the specified erase count earlier than the data sector. It is possible to realize a method of managing the number of times of erasing and a data processing device using the method.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating a method of managing the number of erasures of a flash memory according to an embodiment of the present invention.

FIG. 2 is an explanatory diagram when the present invention is applied to a 32 sector, 16 Mbit flash memory.

FIG. 3 is a schematic configuration diagram of a management sector SA7 in FIG. 2;

FIG. 4 is a diagram illustrating a method of checking whether the number of erasures stored in an erasure number information area SA has reached a specified number.

FIG. 5 is a diagram comparing an embodiment of the present invention and a conventional technique with respect to a possibility of data loss at the time of a power failure.

FIG. 6 is a diagram showing an example of a conventional technique.

FIG. 7 is a diagram showing another example of the related art.

FIG. 8 is a diagram showing still another example of the related art.

[Explanation of symbols]

B1 to Bn 1st byte to nth byte SA Erase count information area SA0 to SA6, SA8 to SA14 Data sector SA16 to SA22, SA24 to SA30 Data sector SA7, SA15, SA23, SA31 Management sector

Claims (10)

[Claims] 1. The number of erasures is limited, and erasure is performed for each sector which is a specific unit byte. By erasing, all data in a sector become "1" in bit units.
The write operation is performed by changing the bit value from “1” to “0”,
In a method of managing the number of erasures of a flash memory which is a nonvolatile memory that allows “1” to “1” and “0” to “0”, a single sector is assigned as a management sector to a plurality of sectors. The management sector is divided into areas respectively corresponding to the plurality of sectors to be managed, and each time the management target sector is erased, consecutive bits in a plurality of bytes of the area corresponding to the management target sector in the management sector are erased. A method of managing the number of erasures of a flash memory, wherein the number of erasures is counted by sequentially writing "0" to the flash memory to limit the life of a flash memory having a limited number of erasures. 2. The number of erasures is limited, and erasing is performed for each sector which is a specific unit byte. By erasing, all data in the sector become "0" in bit units.
The write operation is performed by changing the bit value from “0” to “1”,
In a method of managing the number of erasures of a flash memory which is a nonvolatile memory that allows “1” to “1” and “0” to “0”, a single sector is assigned as a management sector to a plurality of sectors. The management sector is divided into areas corresponding to the plurality of sectors to be managed, and each time the management target sector is erased, consecutive bits in a plurality of bytes of the area corresponding to the management target sector in the management sector are erased. A method of managing the number of erasures of a flash memory, wherein the number of erasures is counted by sequentially writing "0" to the flash memory to limit the life of a flash memory having a limited number of erasures. 3. The method according to claim 1, wherein one management sector is assigned to a cluster which is a group of a plurality of sectors.
A specific address of the erase counter corresponding to the number of sectors for managing the number of erases provided in this management sector is periodically read, and the data at that address is compared with predetermined specific data. A management method for managing the number of erasures of a flash memory, wherein the life of a flash memory having a limitation on the number of erasures is managed by determining whether or not the number of erasures has reached. 4. A flash memory erase number management method according to claim 3, further comprising a cluster information area indicating a state of the cluster, wherein the cluster information area includes erase information of each sector in the cluster and information on the entire cluster. A method of managing the number of times of erasure of a flash memory, wherein information in a cluster is obtained only by reading a cluster information area by recording the information. 5. A method according to claim 1, wherein said specific data pattern is rewritten by writing a specific data pattern to a specific address of a management sector. Detecting the fact that the management sector has been initialized by detecting that the management sector has been initialized. 6. An MPU, a main storage means and an erasure count are limited, and erasure is performed for each sector which is a specific unit byte.
By erasing, the data in the sector all have a value of “1” in bit units, and the write operation is performed with respect to the bit value from “1” to “0”, “1” to “1”, and “0”. In a data processing device having a flash memory which is a non-volatile memory where “0” is permitted, the flash memory includes:
Having a management sector that is a single sector assigned;
This management sector is divided into areas respectively corresponding to the plurality of sectors to be managed, and each time the management target sector is erased, a continuous byte in a plurality of bytes of the area corresponding to the management target sector in the management sector is erased. A data processing apparatus for managing the life of a flash memory having a limited number of erases by sequentially writing "0" to bits and counting the number of erases. 7. An MPU, a main storage means, and the number of erasures are limited, and erasing is performed for each sector which is a specific unit byte.
By erasing, the data in the sector becomes all "0" in bit units, and the write operation is performed from "0" to "1", "1" to "1", and "0" to the bit value. In a data processing device having a flash memory which is a non-volatile memory where “0” is permitted, the flash memory includes:
Having a management sector that is a single sector assigned;
This management sector is divided into areas respectively corresponding to the plurality of sectors to be managed, and each time the management target sector is erased, a continuous byte in a plurality of bytes of the area corresponding to the management target sector in the management sector is erased. A data processing apparatus for managing the life of a flash memory having a limited number of erases by sequentially writing "0" to bits and counting the number of erases. 8. The data processing device according to claim 6, wherein the flash memory allocates one management sector to a cluster which is a group of a plurality of sectors, and stores the number of erasures in the management sector. It has erase counters for the number of sectors to be managed, and periodically reads a specific address of the erase counter, compares the data at that address with predetermined specific data, and determines whether the erase count has reached the specified erase count. A data processing device for managing the life of a flash memory having a limited number of erasures by judging whether or not the data is erased. 9. The data processing apparatus according to claim 8, wherein
The flash memory has a cluster information area indicating the state of the cluster, and erase information of each sector in the cluster and information on the entire cluster are recorded in the cluster information area, so that the read operation of the cluster information area is performed. A data processing device characterized in that information in a cluster is obtained only by using the information. 10. A data processing device according to claim 6, wherein a specific data pattern is written in a specific address of a management sector to detect that the specific data pattern has been rewritten. By doing
A data processing device for detecting that the management sector has been initialized.