JP3390740B2 - Method of controlling storage device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [0001] BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an external storage device for a personal computer.
Cannot overwrite flash memory used for
The present invention relates to a storage device control method. [0002] 2. Description of the Related Art Recently, external storage devices such as personal computers have been used.
External storage devices using flash memory
I'm afraid. Flash memory is non-volatile and
No need for backup power supply, electrically rewriting
And it is inexpensive, but has the following problems:
ing.   Write data only after erasing data
Cannot be erased, and the unit of erasure cannot be in bytes.
Yes, in units of sectors, blocks or chips. This
So, like normal memory, data is stored in bytes.
It cannot be rewritten, and
The write or erase speed is slow.   There is a limit on the number of erasures.
It is said that erasure becomes impossible after about one million times. this
Therefore, the number of erasures of a sector or block is averaged.
If not used, the part with the highest number of erases will be defective first
And the usable area decreases. Flash memory
Has the problems described above,
When using memory, use the data save area.
Save and manage data when rewriting data
A table was set up to manage data writing / erasing
In addition, when bad sectors, blocks, etc. occur,
It is necessary to take various measures such as taking remedies.
You. [0003] SUMMARY OF THE INVENTION Flash memory
Advantages such as being non-volatile and electrically rewritable
On the other hand, it has many problems as described above.
Of conventional DRAM, SRAM, etc.
Unlike conductor memory, it must be solved when using it.
There are many issues that must be addressed. The present invention addresses the above problems.
The purpose of the present invention is to
Overwriting cannot be performed as in memory, etc.
To improve the reliability of memory that cannot be erased
Is Rukoto. [0004] FIGS. 1 to 4 show the present invention.
FIG. 1 is a schematic diagram showing the overall configuration of the present invention.
2 to 4 are principle diagrams showing an outline of the present invention.
You. 1 to 4, reference numeral 1 denotes a storage device;
Cannot be written, and
Data can only be erased in certain units, for example
For example, a storage area composed of a flash memory or the like, 1b
Controls writing of data to the storage area 1a,
Control means for erasing the storage area 1a and the like, 1c is a storage area
1a temporarily stores data when writing data
To save the data in the storage area 1a.
The secondary storage media 1d and 1e store the address of the storage area 1a.
Decoding table for coding, 2 is main unit processing unit
It is. In order to solve the above problems, the present invention provides
The invention is readable and writable, as shown in FIGS.
In addition, a storage area where a part of the storage area may be destroyed
1a and the location where data in storage area 1a is stored
Storage device comprising a rewritable decoder 1d
, Two stages of decoders are provided, and a storage area of the storage device 1 is provided.
When part of 1a is destroyed, or when the decorator 1d,
When a part of 1e is destroyed, two-stage decoders 1d and 1e
By rewriting either one or both
And no longer decodes the destroyed part
Things. [0005] In FIG. 1, the main processing unit 2 to the storage device 1
When the write data is sent to the
The data is stored in the primary storage medium 1c. And decode
Decode the address with reference to the table 1d (1e)
The data stored in the primary storage medium 1c is stored in the storage area 1
Write to a new sector or block a. That
The same logical address as the logical address of the data to be written.
If the data of the source already exists in the storage area 1a,
Is the area where the data was erased or stored.
Set the erasable flag on the area. Also stored data
When the erasure flag is set in the area, the control means 1b
Primary storage of non-erasable data in specified units at fixed times
The storage area 1a is deleted by evacuating to the medium 1c, and the primary storage medium is deleted.
Such as writing back the data saved in the body 1c to the storage area 1a.
Processing is performed to create a free area. In addition, control hands
The stage 1b is used when a part of the storage area 1a becomes defective.
Rewrite the decoding table 1d (1e)
Then, the defective part is replaced with a spare area. The storage area 1a shown in FIG.
Lock 1a-1,..., 1a-N and the block
Block of written data consisting of
Storage device to be erased in units, storage area 1 shown in FIG.
Consists of multiple sectors that cannot be written to
Storage device that erases erased data in sector units,
Writing / erasing of data is performed as follows. (1) The storage area 1a includes a plurality of blocks 1a-1,.
-N and a sector that divides the block.
Storage device 1 for erasing written data in block units
In FIG. 2B, as shown in FIG.
1,..., 1a-N, the non-erasable data
When saving to another block, an erasable section
From blocks with a large number of data to m blocks, there is little erasability
N blocks (m and n are arbitrary integers)
Select the selected block and have free sectors at the same time.
Sectors are scattered in this block, and the blocks are uniformly saved.
With this, it is possible to create an empty block
As a result, sectors that are frequently rewritten in each block
Sectors with few replacements are mixed, and blocks
Can be averaged. (2) The storage area 1a includes a plurality of blocks 1 including a save area.
a-1,..., 1a-N and the written data
In a storage device for erasing data in block units, FIG.
Block 1 containing erasable data as shown in (c)
a-1,..., 1a-N, non-erasable data
After moving only the data to the save area,
And erased blocks as new save areas.
Thus, an empty area is created. This allows erasure
Data can be lost and this process must be performed in advance.
Thus, the writing time can be reduced. (3) On the opposite side of the save area with respect to the data write direction
Move only the non-erasable data of the block to the save area.
After that, erase the data in the above block and
Lock as a new evacuation area. This allows the writing area
Space between the write pointer indicating the location and the save area
Regions can be present. Therefore, during processing
Even if there is a disconnection, the data
Can be written, and the processing amount required after the processing is interrupted
Can be reduced. (4) The storage area 1 includes a plurality of blocks 1a including a spare area.
-1,..., 1a-N and written data
In a storage device for erasing data in blocks, FIG.
(A), a write point is reserved as shown in FIG.
Erases blocks that have been set in the storage area and in which data has been written
After moving unusable data to the spare area,
Repeat the process of erasing the data of the last
The number of free sectors in blocks other than the lock is in the spare area.
When it is equal to or larger than the number of sectors,
Blocks other than the block from which the data in the storage area was erased last.
Move to This allows you to create free space
it can. Also, when an empty block becomes unusable,
Recover unusable blocks to create empty blocks
Can be achieved. In addition, the data in the spare area is
Data to a block other than the last erased block.
The write point at the destination,
Search from the block after the deleted block
By erasing data at the end,
All free space between the multiplied block and the write pointer
Simplifies processing after processing is interrupted.
Can be Furthermore, after the processing is interrupted,
When resuming, check if there is any erasable data in the spare area,
Alternatively, depending on whether there is only unerasable data,
Can be determined at the time of
Can be simplified. (5) The storage area 1 has a plurality of blocks 1a-1,..., 1a-
N, and the written data is written in block units.
In the storage device 1 to be erased, the write
By providing a flag F1 indicating presence / absence, as shown in FIG.
Then, the written data is erased with reference to the flag F1.
I do. As a result, unnecessary erasure can be avoided, and
Not only shortening the time to leave, but also
Life can be extended. Also, the erasing process for all storage areas
Flag F2 indicating that the process is being executed is provided.
As a result, during the process of erasing all storage areas,
When interrupted, refer to the running flag and
It can be determined that the erasing process is in progress,
It is possible to respond to interruption of management. (6) Multiple bits corresponding to the same function in storage area 1a
Is recorded, and the flag is determined by the logical product of the flags.
Make settings. As a result, all bits become defective.
Unless it is necessary, the correct judgment value can always be output,
The reliability of the system can be improved. (7) As shown in FIG.
It consists of a number of sectors.
In a storage device that erases in sector units, the same sector
Prepare at least two sectors with the same number, and
If there is data in one of the attached sectors,
At the same time as writing data to the other sector,
Data of the data. This improves writing speed.
Can be up. (8) In the storage device shown in FIG.
Transfer data to the primary storage device 1c
The data in the above sector while
After erasure is completed or after data transfer is completed,
Write data to the data. This improves writing speed
can do. (9) Replace bad sectors in the storage device shown in Fig. 4.
Spare areas 1b-1 and 1b-2 to be written and write sectors
Rewritable decode table 1d for managing
If a bad sector occurs, the decode data
By rewriting the table 1d, the spare area 1b-
1, the write sector is changed to 1b-2, and the spare area 1 is changed.
When b-1 and 1b-2 are gone, decode table
To reconfigure the sector by rewriting the file 1d
Thus, it is possible to relieve when a defective sector occurs. (10) In the storage device shown in FIG.
N + 1 sectors are provided and N + 1
Each sector is shared by N sector numbers, and data is
When writing data to the N + 1 sectors,
Write to existing sector. This improves writing speed.
As well as the extra sectors
The number can be reduced, and the cost can be reduced.
it can. A spare area 1b for replacing a bad sector
-1 and 1b-2, when a bad sector occurs,
The decoding table 1d is rewritten, and the spare area 1b-
Change write block or sector to 1,1b-2
Or reorganize the decoding table 1d
You. As a result, it is possible to relieve when a bad sector occurs.
it can. (11) The storage area 1a is divided into a plurality of
Multiple blocks are provided in the chip,
When writing data to the block, the data is written to a new block.
If the written data already exists in the chip
If the data is erased in blocks,
In the storage device, write the chip to which data is written
Fix it corresponding to the data. Thereby, the control means 1
b needs to manage only the data in each chip, and the entire area
No need to manage. Therefore, in the control means 1b,
The management table can be simplified and the writing speed
Can be improved. Also, the data in each chip
To provide multiple work blocks for writing
When a defect occurs in a work block,
You can rescue lost work blocks. [0007] The invention of claim 1 of the present invention provides the above-described reading and writing system.
And the possibility of destroying part of the storage area
A certain storage area 1a and data in the storage area 1a are stored.
And a rewritable decoder 1d indicating the location where the
In the storage device obtained, as shown in FIG.
Provided that a part of the storage area 1a of the storage device 1 was destroyed.
ComeThe address of the secondary decode unit was rewritten and destroyed
Secondary decoding to prevent decoding to the part
When the unit is destroyed, rewrite the primary decoding unit,Destruction
Since decoding to the part that did
As a coder, such as EEPROM, flash memory, etc.
When using a storage medium that may destroy part of the
The probability of address errors is significantly reduced.
And reliability can be ensured. [0008] Next, embodiments of the present invention will be described. A. 1 illustrates a configuration of a storage device that is a premise of the present invention. FIG. 5 shows a case where a flash memory which is a premise of the present invention is used.
FIG. 2 is a block diagram showing a configuration of a storage device. Smell
20 is a storage device such as a memory card, and 21 is
Controls flash memory data rewriting, erasing, etc.
Controller LSI, 22 is a processor, 23 is various
You can store tables and data when writing data.
SRA used as buffer or save buffer
M and 24 are clock oscillators, 25-1 to 25-5 are flash oscillators.
Flash memory. In the figure, the SRA
The data transferred to M23 is stored in flash memory 25-
When writing to 1 to 25-5, flash as described above.
-Since the memory cannot be overwritten,
In this case, data is written to a new area. Also, the data
In the case of updating, measures such as setting an erasure flag on the old data
Take a place. The old data is flagged for erasure.
, An erasure flag is attached at a predetermined time.
Save only data that is not stored in SRAM 23 in predetermined units
To the area where the flash memory is to be written.
After erasing, rewrite the data to correct the data.
Work. FIG. 6 shows the contents of the SRAM 23 described above.
The SRM 23 has the following data as shown in FIG.
Data is stored. (1) Erase count table Hold the erase count of each block of flash memory
I have. (2) Erasable sector count table Total number of erasable sector flags in each block
Hold the value. (3) Write pointer Chip No. to start writing to flash memory,
Holds block No. and sector address No. (4) WORK-BLOCK-No. Indicates the current WORK-BLOCK-No. And the chip No.
 , Block No. are held. (5) Sort pointer Currently, the chip No., block No.
Data address No. is retained. (6) Rewrite Indicates whether the data to be written is new
You. (7) Write count Manages the number of times the device is written when organizing. (8) Save counter It manages the number of sectors in the evacuation operation during sorting. (9) Number of chips Holds the number of flash-chips mounted on the card. (10) Sector map table Chip No., Block No., and Cell No.
Holder address No. FIGS. 7 and 8 show the flash memory 25-1.
FIG. 25 is a diagram showing the contents of the flash memory.
Each sector has control information and data shown in the following (1) to (5).
Data etc. are stored. In this example,
126 sectors are provided, and each sector has (1)
The management information and data of (5) are stored, and the 126th security
The following management information (6) to (14) is stored after the
ing. (1) Bad flag Indicates the status of this sector.If it cannot be used, write it here.
Make a dent. (2) Erase flag Indicates the data status of this sector and is invalidated by rewriting, etc.
If this happens, write here. (3) Logical address Indicates the logical address of this sector. (4) Data Data written to this sector. (5) Checksum Checksum of written data (6) Bad sector memory Indicates a bad sector in the rearrangement target block. (7) Number of erasures of source Number of erases of the block to be pruned. (8) Evacuation block No. Chip No. and block No. for saving data (9) Erase Start Set when erasing the rearrangement target block is started. (10) Erase End Set when erasing of the rearrangement target block is completed. (11) All erase target Stands when it is determined to be erased during All Erase.
Te (12) Empty block Indicates whether there is data in the block, and if there is data
Is set up. (13) Block status Indicates the status of this block, and stands when it becomes unavailable.
Te FIGS. 9 to 13 show the flash memory 2.
Processor for writing data from 5-1 to 25-5
It is a flowchart which shows the process in 22.
The write operation will be described in detail with reference to FIG. FIG.
In step S1, the current writing of the SRAM 23 is
"Rewrite enabled" indicates whether the data to be updated is new or not.
"No" (see FIG. 6) is set to 0, and in step S2,
It is determined whether or not it is being rearranged. If you are not organizing
Goes to step S5, where the
Set “Number of Writes” to 0
Go to step S7. If you are organizing,
Go to S3, add 1 to the "number of writings", and
At 4, it is determined whether or not the number of writings is 6.
You. If the number of writings is 6, the process proceeds to step S6.
In this case, the number of times of writing is set to 1. In addition, the number of times of writing is 6
If not, go to step S7. After the above processing
As described above, when the number of writings is 2 to 5,
Operation is not performed.
Work to avoid frequent organizing operations.
By setting the number of writes as described above,
The sorting operation is performed once every five times. In step S7
The logical address of the data to be written
It is determined whether or not there is an error.
-Perform processing. When the logical address is not over
Goes to step S8 to determine whether or not there is old data.
That is, whether or not the data to be written is new data
If it is not new data, step S9
In the above, the “rewriting presence / absence” of the SRAM 23 is set to 1
To step S11. Also, for new data
In step S10, the "rewrite"
Is set to 0, and the process proceeds to step S11. Then
In step S11, the data is transferred from the main unit to the SRAM.
Transfer and whether or not there is a transfer error in step S12
Is determined. Error handling if there is a transfer error
Do. If there is no transfer error, step S in FIG.
13 and rewrite or not, that is, with new data
Determine if there is any data, and if it is not new data,
Since the data is updated, the old data is
The erasure bit is written in the sector having the address, and the step S15
go to. In step S15, the SRAM 23
"Number of writes" that manages the number of writes at the time of sorting is 1
Is determined, and if the “write count” is 1,
Is to organize the flash memory after step S16
Do. If the “write count” is not 1,
Go to step S46 of step 13, and after step S46,
Writes data in SRAM23 to flash memory
No. In step S16, when organizing on the SRAM 23,
To manage the number of sectors in the evacuation operation
Is set to 0, and in step S17,
"Arrangement pointer" indicating the sector address No.
(Sector address) is greater than 126 sectors.
Otherwise, if it is not smaller than 126, the sorting is finished
As a result, the procedure goes to step S38 in FIG. Also,
"Arrangement pointer" (sector address) is 126 sectors
If less, go to step S18 and
Determine whether the data pointer is 0, and the rearrangement pointer is not 0
In this case, the procedure goes to step S22. If the rearrangement pointer is 0, step S
In step 19, the number of erasures of each block and the erasable sector
The number is obtained from the table of the SRAM 23, and based on that,
And select the sort target. Then, in step S20
WORK-BLOCK on the flash memory
The chip number to be rearranged is set in the evacuation block number (see Fig. 8).
 And block No. are written. In step S21
To determine if there is a write error,
If there is an error, perform error processing and write error
If there is no, go to step S22. Step S2
In step 2, data to be saved is searched. That is,
Erase flag (see Fig. 7) in rush memory data
If so, go to the next sector. Also,
If there is no logical address, write the erase flag and
To the sector where the logical address is abnormal.
And write the bad flag and go to the next sector. Next, go to step S23 in FIG.
Determine whether there is a write error and write
If there is not, at step S24,
(Sector address) points to the 126th sector
And if the rearrangement pointer is 126,
Indicates that the sorting has been completed, and returns to step S3 in FIG.
Go to 8. If the sort pointer is not 126,
Go to step S25 and save the data to save
From the memory to the SRAM 23, and
And the generated checksum is stored in the flash memory.
Then, it is determined whether or not they match the checksum (see FIG. 7).
If they do not match, check in step S27.
It is determined whether or not the sum is FFh. And Choi
If the checksum is not FFh, the process proceeds to step S28.
And the checksum on the flash memory is FFh
Write the bad flag of the data save source sector
Go to step S29. In step S29, write
It is determined whether or not there is an error, and the process proceeds to step S30.
In step S26, the checksums match.
Or if the checksum is FFh in S27
Goes to step S30, where the write pointer indicates
Search for a writable sector from the
Then, it is determined whether or not there is a writable sector. book
If there is no writable sector, perform error processing,
If there is a writable sector, go to step S32.
To transfer the data on the SRAM 23 to the flash memory.
Moving. In step S33, a write error
To determine if there is a write error and if there is a write error
Returns to step S30. When there is no write error
Go to step S34 of FIG.
Write the sector erase flag. Then, step S3
At 5, it is determined whether or not there is a writing error, and
If there is no writing error, the process proceeds to step S36.
Rewrites the sector map table on the SRAM 23
At the same time, 1 is added to the evacuation counter. Then,
In step S38, the evacuation operation is performed a predetermined number of times.
Number (in this case, 60 times) is determined, and
If the evacuation operation reaches 60 times,
After arranging the memory, the process proceeds to step S38. Also retire
If the evacuation operation has not been performed 60 times, the operation shown in FIG.
Returning to step S22, the above processing is repeated. Whether the retraction counter is 60 or not
In step S17 of FIG.
If it is determined that it is not smaller than 126,
Go to S38, determine whether the evacuation counter is 0
You. When the evacuation counter is 0, that is,
If the logical pointer is not smaller than 126,
On the other hand, if the evacuation operation has not been performed, step S3
9 and in step S39 and subsequent steps,
Processing such as writing to the bad sector memory is performed. Sand
If the save counter is 0, the processing time is short.
Therefore, the information of the defective flag after S39 is stored in the defective section.
Process such as writing to the data memory, and save
If the data is not 0, the process proceeds to step S46 in FIG.
The data on the SRAM 23 to the flash memory.
It performs processing such as moving. In step S39, the evacuated block
Write the information of the defective flag of the disk to the write destination (write destination is WO
RK-BLOCK) bad sector memory (see FIG. 8)
And the number of erasures is displayed in the column of the number of erasures of the rearrangement source (FIG.
See). In step S40, write
Determine if there is an error and if there is no write error
If it is, go to step S41 and delete the arranged block.
Leave. In step S42, is there an erase error?
It is determined whether or not there is no erasure error.
Go to step S43 and write the bad sector memo to the erased block.
Return the defect flag information from the
Then, write the number of erases. Then, to step S44
In the table corresponding to the erased block
Add 1 to the number and set the number of erasable sectors to 0. Steps
In S45, the rearrangement pointer is set to WORK-BLOCK.
And the rearrangement pointer is set to 0. Next, the operation proceeds to step S46, where the write
Search for a writable sector from the sector indicated by the
In step S47, it is determined whether there is a writable sector.
Separate. If there is a writable sector, step
Go to S48 and flash the data on SRAM23.
Move to the memory, and in step S49, write
It is determined whether there is an error. When there is a writing error
In this case, the process returns to step S46, and there is no write error.
In this case, go to step S50. In step S50
Whether or not the rewrite is 1
Determine whether the data to be written is new, and
If the presence or absence is 1, the erase flag is written in step S51.
Erasable sector count table corresponding to the written sector
1 is added to the value of (see FIG. 6), and the procedure goes to step S52.
When the presence or absence of rewriting is 0 in step S50
In step 52, the sector map table
(See FIG. 6), and in step S53, the next
To write data to the sector indicated by the write pointer
Search for a writable sector and finish. B. Write / erase processing. As described above, flash memory has a limited number of erases.
In order to use all the memory effectively,
The number of departures needs to be averaged. Also, the flash menu
Mori cannot overwrite, rewrite data
Write the data to a new sector and
Take appropriate measures, such as setting an erasable flag on the
Erase sector with erasable flag set at time
There is a need. Therefore, the erasable data is erased and
When it is necessary to secure free space for writing data
In both cases, the above-mentioned empty area causes bad blocks, etc.
Take remedies when it becomes unusable due to
You need to create a free space. Below,
Erasure of erasure count (Example 1), erasure of erasable data
Create a free area to secure free space by deleting
2), relief for creating an empty area when a bad block occurs
Implementation of the present invention for area processing such as measures (Example 3)
Here is an example. (1) Embodiment 1 (Average of erase count) As described above, flash memory has a limited number of erases.
In order to use all the memory effectively,
The number of departures needs to be averaged. In this embodiment, each block
The number of erases without being aware of the number of erases
This embodiment shows an example in which the noise can be suppressed.
In addition, additional writing is performed in units of sectors, and
In the memory to be erased, the block is divided into sectors.
As shown below, blocks with many digestible sectors
And blocks with few erasable sectors
To write to a new block,
Averaged. Next, this embodiment will be described. FIG.
4 to 18 show the write processing in this embodiment.
It is a diagram showing, in the present embodiment, a 6-block having 6 sectors.
Write logical sectors A to O to flash memory
An example is shown. In the following description,
The number of empty sectors in the block is N (2 in this embodiment).
M blocks (2 in this embodiment)
), And an erasable sector
M blocks from the largest number (one block in this embodiment)
And n blocks from the smaller number of erasable sectors (this embodiment
In this case, one block is selected and saved at the same time. In addition,
When performing the evacuation operation, as described above, the erasable file
The sector which has no lag is read, and S shown in FIG.
Moved to RAM23, then moved to SRAM23
Write a sector to a new block. In FIG. 14A, the same sector is
There are no sectors A and B in blocks 1 and 2.
C, D, E, F, G, H, I in order from the first sector
Write to. In (b), write sectors C and D
No. In this case, since there are the same sectors C and D,
To the sector where C and D are written
Then, new sectors C and D are written. Next, FIG.
In (c), write sectors J, K, L, M, N, and O.
Get in. In this case, as in the case of FIG.
Since there are no sectors,
Write. In (d), sectors H, I, J, K,
Write M and N. The same sector has been written
To set an erasable flag and to set new sectors H, I, J,
Write K, M, N. In FIGS. 16 (e-1) and (e-2),
Write sectors C and D. At that time, the number of empty blocks
Since there are two, a retreat operation is performed. That is, the figure
In (d) of FIG. 15, the block having the largest number of erasable sectors
Lock 3 and block 4 with the fewest erasable sectors
Select and mix block 3 and block 4 sectors.
And the number of sectors in each block at the save destination becomes uniform.
Thus, the data is written into the empty blocks 5 and 6. As a result, FIG.
As shown in FIG. 6 (e-1), sectors O, I, K, M
Lock 5 and sectors H, J, L, and N block 6
And blocks 3 and 4 become empty blocks.
In this state, data is erased in block 2 in which C and D have already been written.
A flag indicating that the sector C and the sector D can be deleted is set as shown in FIG.
Write to block 5 as shown in FIG. In FIG. 17 (f-1), sectors H,
Write I and J. In this case, the same sector is already
Therefore, the erasable flag is set in blocks 5 and 6.
Then, sectors H and I are written in block 6,
At the time of writing J, there are two empty blocks.
The evacuation operation is performed as described above. That is, as shown in (f-2)
Block 2 having the largest number of erasable sectors,
Select block 6 with the least erasable sector and block
The sector of block 2 and the sector of block 6 are mixed and saved.
Make sure that the number of sectors in each block is
Write to blocks 3 and 4. Next, shown in (f-3)
Is written in sector J. Also in this case, the same sector
The sector of block 4 already written
Set the erasable flag to and write sector J to block 3.
Put in. In FIG. 18 (g), sectors E, F, G
Write. In this case, too, there is the same sector.
An erasable file is written in the written sectors 1 and 3.
Rag, write sectors E, F, and G in blocks 3 and 4.
Put in. FIG. 19 is a flowchart showing the processing of this embodiment.
The processing of this embodiment will be described with reference to FIG. S
In step S1, when data is received from the main unit,
In step S2, the number of blocks with empty sectors is N
It is determined whether or not: Block with empty sectors
If the number is not less than N, the procedure goes to step S7. Free
If the number of blocks with sectors is N or less, step
In S3, m
The block is set as a save target, and erased in step S4.
Save n blocks from blocks with few possible sectors
Elephants. In step S5, the block to be saved
Move data from to the empty block. Write at that time
The method is to change the write block each time data is moved.
Well, after writing to M blocks,
Return to lock. Next, in step S6, the evacuation pair
Erase an elephant block. In step S7, the same
If there is a logical sector, the physical sector that has already been written
Is set to an erasable flag, and in step S8,
Writes the data received from the flash memory
No. By the way, in general, a sector includes a system
Data that is hardly rewritten like a program
Stores stored sectors and data that is constantly rewritten.
There is a sector that has been stored,
In this case, sectors that are hardly rewritten
A large number of blocks has a small number of erases. Therefore, on
As described above, blocks with many erasable sectors
Blocks with many sectors to be replaced) and erasable
Block with few sectors (rewrite is hardly performed)
Block with many sectors).
Writing to each block results in rewriting to each block.
Sectors with a large number of sectors and sectors with a small number of rewrites
And the number of erased blocks can be averaged.
You. In this embodiment, writing is performed based on the above-described principle.
Since the writing process has been performed, the erase
It is possible to reduce the variation in the number of erases without being aware of the number.
And the erasure count is finite like flash memory
It is possible to extend the life of the memory. Also erase
Suppress variations in erase count without counting
Area for counting the number of erases.
It is also possible to manage memory without setting up area
Become. (2) Embodiment 2 (for erasing erasable data)
Creating free space by As mentioned above, flash memory can be overwritten.
Can not be written and write only after erasing
Can not. Therefore, when writing, the same sector is
Set an erasable flag when the
It is necessary to erase the sector where the function flag is set. Book
In the embodiment, as described above, the erasure flag is set.
Area processing to eliminate possible data from storage media
The area processing according to the present embodiment indicates the free time of processing.
Area that can be written to by writing in advance
And the writing time can be shortened. FIGS. 20 to 23 show the writing in this embodiment.
FIG. 9 is a diagram showing write / erase processing, and FIG.
explain. In this embodiment, the following points are assumed.
are doing.   There are six blocks, and six blocks per block.
Read / write is performed in sector units and erased
Is performed in block units.   The write method is a write-once write method, and the same
When writing a logical sector without
An erasable flag is set for the physical sector stored.   Write area 5 blocks + save area for additional writing
One block (block 1 to block 6).   If the write area is exhausted, the erasable
Block with the largest number of physical sectors
Move the data of the available physical sector to the save area, and
Erase block. And evacuate the block
Area, and the existing save area is used as the write area.
(This series of operations is called an evacuation operation).   The address of the sector sent from the main unit is A to P
And   When there is m or more erasable data in a block,
It is a target of the area processing in the present embodiment (see FIG. 20 to FIG.
In the example of 23, m = 1.) Note that FIG.
The arrow from 0 to the right of the block in FIG.
This is the position of the write pointer indicating the location. In FIG. 20A, logical sectors A to G
Write. In this case, the logical sector of the same address
Since there is no
Get in. In (b), logical sectors A and D to G are written.
Put in. In this case, the logical sector (A,
D, E, F, G), so the erasable flag was set.
Later, writing is performed in the same manner as (a). In (c), the logic
Write sectors H to N. In this case, the same address
Since there is no logical sector, the data is written as it is. FIG.
In (d), A, G to L are written. In this case
Are the logical sectors (A, G, H, I, J,
K, L), write after setting the erasable flag.
Put in. In (e), A, H to J are written. This place
In each case, the logical sector of the same address (A, H, I, J,)
Therefore, write after setting the erasable flag.
In (f-1), when writing A, I to L
However, since there is no write area, a save operation is performed. You
That is, the physical sector with the erasable flag set is the highest.
The data of many blocks (block 3) is saved in the save area (block
Block 6), and erases the source block. So
And use that block (block 3) as a save area.
And save the previous save area (block 6) to the write area
And As a result, it becomes like (f-1). Next, as shown in FIG.
A, I to L are written. In this case, the same address
After setting the erasable flag, A,
Write I to L. At this time, the write pointer is saved
It is at the beginning of the area (block 6). Write as above
As a result, the erasable data has been increased.
Region processing is performed. That is, the save area (block
As seen from 3), the block in the direction opposite to the writing direction
With lock (block 2) as the processing target, block 2 and
At 3, the evacuation operation is performed. As a result, as shown in FIG.
Block 2 becomes a save area, and block 3
The non-erasable data of block 2 is moved. Then with the above
Similarly, a retreat operation is performed in blocks 1 and 2. So
As shown in the result (g-2) of FIG.
And the non-erasable data of block 1 is stored in block 2.
Moved. Then, similarly to the above, the blocks 5 and 1 return.
Perform the evacuation operation. As a result, as shown in FIG.
Block 5 becomes the evacuation area, and block 1
The non-erasable data in step 5 is transferred. Block 6
Since there is no erasable data in
Absent. Similarly, the retreat operation is performed in blocks 4 and 5. The result
As a result, as shown in (g-4), the block 4 becomes the save area.
The non-erasable data in block 4 is moved to block 5.
Is done. Then, the evacuation before the processing of the position of the evacuation area is performed
Since the position of the area has been reached, the area processing ends. That's all
By performing such processing, erasable data can be stored on a storage medium.
Disappear from above. Next, as shown in (h), A, H to J
Write. Also in this case, the logical sector of the same address is
Because there is, write after setting the erasable flag. FIG. 24 is a flowchart showing the processing of this embodiment.
The processing of this embodiment will be described with reference to FIG. S
In step S1, the block number of the current save area is
In step S2, the evacuation area is set as a processing target.
I do. In step S3, the processing target is set in the writing direction.
Step forward in the opposite direction. In step S4,
Target and the saved block N saved in step S1
o. are compared, and if they are the same, the process ends. Processing object
If the stored block number is different from the stored block number, go to step S5.
To compare the number of erasable sectors to be processed with a predetermined value m
If the number of erasable sectors is equal to or less than m, step S
3 and the above process is repeated. Number of erasable sectors is m
If it is larger, the process goes to step S6 and the processing target is
In step S7, the write
To the beginning of the save area. In step S8
Move data from the block to be saved. Step
In step S9, the block to be saved is erased, and the
In step S10, the block to be saved is designated as a save area.
Then, the process returns to step S3 to repeat the above processing. In this embodiment, the area is allocated as described above.
Erasable data can be eliminated by performing area processing
It is possible to perform the area processing of this embodiment in advance.
Thus, the writing time can be reduced. Also,
Opposite the writing direction for the block targeted for area processing
As you are proceeding in the direction, the writing indicating the writing location
All free areas exist between the embedded pointer and the save area.
Can be made. As a result, in any case
Even if an interruption occurs, data is not deleted from the position indicated by the write pointer.
Data can be written,
Process can be reduced, and
Even if you do not have, there is no problem. (3) Embodiment 3 (for creating an empty area)
Remedies) As described above, flash memory is used for writing data.
Therefore, it is necessary to secure free space in the memory
However, in the present embodiment, the above-mentioned empty area has a bad block.
If you cannot use it for any reason,
An embodiment is shown in which a new free area is created.
FIG. 25 to FIG. 31 show a process of creating a free area according to the present embodiment.
This embodiment will be described with reference to FIG. The book
The following points are assumed in the embodiment.   There are seven blocks, and six blocks per block.
Read / write is performed in sector units and erased
Is performed in block units.   The write method is a write-once write method, and the same
When writing a logical sector without
An erasable flag is set for the physical sector stored.   One block of spare area (block
0) is secured, and 5 blocks are written for additional writing
+ One save area (block 1 to block 6)
You.   If the write area is exhausted, the erasable
Block with the largest number of physical sectors
Move the data of the available physical sector to the save area, and
Erase block. And evacuate the block
Area, and the existing save area is used as the write area.
(This series of operations is called an evacuation operation).   The address of the sector sent from the main unit is A to P
And In FIG. 25A, sectors A to G are written.
Get in. In this case, there is no same logical sector.
Write from block 1 as is. In (b),
Write the parameters A, D to G. In this case, sector A,
Since there is the same logical sector as DG, the same logical sector
Write after setting the erasable flag to (C) smell
Then, sectors H to N are written. In this case, the same logic
Since there is no sector, write as it is. In FIG. 26 (d)
Then, sectors A and G to L are written. In this case,
Since there is the same logical sector as the data A, GL, the same logical
Writes after setting an erasable flag in the sector. (E)
, Sectors A, H to J are written. In this case,
Since the same logical sector exists as the sectors A and H to J,
Writes after setting an erasable flag in the physical sector. In FIG. 27 (f-1), sectors A and I
To L are written, in this case, there is no write area
Therefore, the write process is performed after the save operation is performed. Sand
The block with the most data with the erasable flag set
Save area for non-erasable data in block (block 3)
Move to (Block 6) and delete the source block
You. Then, the block (block 3) is defined as a save area.
And use the save area up to now as a write area.
Then, in (f-2), sectors A, I to L are written.
The data is written to the block 6 which has become the embedding area. Next, (g)
In the above, A, M, and N are written.
Since there is no write area, the evacuation operation is performed first.
U. That is, the data with the erasable flag set
Non-erasable data in blocks with many (block 5)
Move to the evacuation area (block 3), and
Clear the lock. And that block (block 5)
Used as a save area and write the save area up to now
Area. Here, block 5 becomes unusable due to erasure failure.
If it becomes possible, there will be no evacuation area, so relief measures will be taken.
I do. Therefore, as shown in FIG.
The evacuation operation is performed using block 0 as a temporary evacuation area. sand
In other words, in FIG.
Non-erasable data of the most blocks (block 1)
Data to the temporary save area (block 0)
Erase a block. And the block (block
1) is used as a save area. Then, as shown in (i)
In the spare block (block 0).
U. That is, the spare area (block) used as the temporary save area
Move logical sectors B and C of lock 0) to block 3
You. Now that the spare area and save area have been created,
In (j), logical sectors A, M, and N are stored in block 3.
Write. Next, write sectors A, G, H, O, and P
However, since there is no writing area, FIG.
The evacuation operation is performed as shown in 1). That is, erasable
Of the block with the most data (block 4) can be erased
Data that is not in the backup area (block 1)
Erase the original block. Then, the block
The lock 4) is used as a save area. Then, (k-
Write sectors A, G, H, O, P as shown in 2)
However, the sectors A, G, and H have the same logical address.
Since there is a sector, block 1 after setting the erase flag
And for sectors O and P,
Put in. Next, in (l), sector A is written.
However, since there is no write area, the evacuation operation is performed.
U. As a result, block 2 becomes a save area, and block 2
The sectors D, E, and F that were in 2 move to block 4.
Here, it is assumed that block 2 has become unusable due to a loss failure.
Then, there is no evacuation area, so a rescue measure is taken. sand
In other words, the block (block
Move the data of 3) to the temporary save area (block 0)
To erase the source block. And that block
(Block 3) is used as a save area. The result
As a result, the result is as shown in FIG. This state
The processing was temporarily stopped due to power shortage, etc.
Then, when processing is resumed, the device erases to the spare area
Creates free space because there is no possible sector and there is data
It can be judged as medium. Next, rescue measures to create free space
And sector I and J of block 6 are set in the spare area (block
Move to block 0) and block sectors K and L in block 6.
Move to step 3. As a result, as shown in FIG.
You. Here, the free space excluding the save area sector is used.
Is less than the number of data in the spare area (block 0).
Because it is above, search the write point from the back of the save area
Then, data is transferred from the spare area (block 0) to the vacant area.
Move data. At that time, as shown in FIG.
Move sectors B, C, M, and N in the spare area (block 0)
At this stage, the main unit temporarily stops processing due to power shortage, etc.
Suppose you put on your laptop. In this case, when processing is resumed
On the device side, there is an erasable sector in the spare area,
Judging that data is being moved to the spare area while creating an empty block
Wear. Subsequently, as shown in FIG.
Move logical sectors I and J of (block 0) to block 4
And terminate the processing. FIG. 32 shows a free space creation process in this embodiment.
FIG. 4 is a flowchart showing the processing according to the present embodiment.
Will be described. In step S1, the spare area
It is determined whether or not there is an erasable sector.
Go to step S11. If not, go to step S2.
And set the write pointer to the start of the spare area
In step S3, it is determined whether or not there is data in the spare area.
You. If there is data in the spare area, step S10
If not, go to step S4 and set the save area
This is a spare area. In step S5, the writing area
If there is no erasable sector in each block of
Abort the process because the area cannot be created. Erase
If there is a possible sector, go to step S6 and erase
The block with the largest number of possible sectors is targeted for evacuation, and the
In step S7, data is reserved from the block to be saved.
Move to the area. In step S8, the block to be evacuated is
The block to be evacuated in step S9.
Make the lock a save area. In step S10,
Number of data in the storage area is larger than the number of free sectors other than the save area
It is determined whether the number is large or not.
Return and repeat the above process. If ≤,
Go to step S11, and move the write pointer to the save area.
The search is performed in the writing direction from next, and the process proceeds to step S12.
In this case, data is moved from the spare area to a free sector.
In step S13, the spare area is erased and the processing ends.
You. In this embodiment, an empty area is created as described above.
Unusable blocks occur in the storage medium
Even if remedies can be used to create free space,
To prevent the storage device from becoming unusable.
it can. Also, during the process, the power
Even if the processing is interrupted, it indicates at what stage the processing was interrupted.
Can be easily determined, and
The judgment can be simplified. C. Efficient erasing process Flash memory should be overwritten as described above
Can not be written and can be written only after erasing
Can not. In this embodiment, the flash memory as described above is used.
A flag indicating the presence or absence of writing, and
An in-flight flag indicating that the process is being executed should be provided.
Efficient erasing process when erasing all storage space
And the process was interrupted for some reason during the erase.
Process can be restarted even if
An example is shown. Next, referring to FIG. 33 and FIG.
The processing will be described. In this embodiment, the number of blocks
Are six, and the area for writing the write flag to each block is
Area is provided, and the entire space erasing process is being performed in block 6.
An area for writing an in-execution flag indicating that the operation is in progress is provided.
In FIG. 33 (a), writing of blocks 1, 2, 4
Before writing the write presence / absence flag in blocks 1, 2, 4
Put in. Next, the initialization process to clear the entire space
As shown in FIG.
Write In (b-2), the writing presence / absence flag
Erases block 1 to which data has been written, and writes
Remove the lag. In FIG. 33 (b-3), the presence or absence of writing
Block 2 to which the flag has been written is erased and written.
Turn off the no flag. Here, after erasing block 2,
To the stop sequence due to the cause of
Even if it rises, the running flag is written in block 6.
Immediately, it is determined that initialization is in progress.
Operation can be started. (B-4)
No write is made to the write presence / absence flag in
Therefore, block 4 is erased and the write flag is erased.
You. In (b-5), writing to blocks 1 to 5 is performed.
Block 6 is erased because it is not written in the presence flag
And clear the running flag. FIG. 35 is a flowchart showing the processing of this embodiment.
This embodiment will be described with reference to FIG. Steps
In S1, it is determined whether or not the execution flag is written.
Determine, if yes, go to step S3, if no
Is being executed in the final processing block in step S2
Write the flag. In step S3, the processing target is
Cleared and written to the processing target in step S4
It is determined whether or not there is an existence flag. Write presence flag
If there is no log, go to step S6.
Deletes the processing target in step S5, and
In step S6, the processing target is advanced to the next. Step S7
To determine whether the running flag has been written
If there is, return to step S4 and repeat the above processing.
return. If there is no running flag, the process ends.
In the present embodiment, as described above,
Is provided, unnecessary erasure can be avoided, and
The time required for the media
Can be extended. Also, confirm that initialization is in progress.
The following running flag is provided to cope with unexpected interruption.
can do. D. Improve writing speed and
Allocation of spare area when As mentioned above, flash memory can be overwritten.
Can not be written and write the sector only after erasing.
Can't do it. Therefore, as described above,
At the same time, if there is the same sector,
Sectors with an erasable flag set at the appropriate time
Need to be erased, and it takes time to write. Also,
As described above, flash memory has a limited number of erases.
If you erase more than a specified number of times,
You. Therefore, when a bad sector occurs, a spare
Areas need to be allocated and remedies taken, as shown below.
In this embodiment, it is not possible to overwrite,
In the memory, the time lag at the time of writing
In addition to improving the write speed,
Shows an example of allocating a spare area in the event of occurrence.
You. FIG. 36 is a diagram showing a system configuration of the present embodiment,
An EEPROM 26 is added to the system shown in FIG.
Other configurations are the same as those shown in FIG.
You. Decoding for performing address conversion is performed in the EEPROM 26.
A table is stored in the EEPROM 26
Further address decoding is performed. And flash
When a part of the memory becomes defective, the EEPROM 2
6 is rewritten and remedy is performed. In FIG. 36,
Although the EEPROM 26 is illustrated, the decoding data
Rewritable storage media
Storage media, flash memory, etc.
Can be. (1) Embodiment 1 FIGS. 37 and 38 are diagrams showing the configuration of the present embodiment.
7, 261 is the EEPROM 26 (or FLASH).
Rewritable ROM such as flash memory)
This is a sector conversion unit. 25 is a flash memory
And the flash memory has a first area A, a second area
B and a spare area C are provided, and the first area A, the second area
Area 4 is provided with four sectors 1 to 4 respectively.
Have been. In FIG. 37, data is written to sector 1.
In the case where the first area A and the second area B are
If there is data in the first area A, the second area B
Write data to Also, writing the above data
In the meantime, the sector 1-A in the first area A is erased. That
When a write error occurs, the spare area C is
Assign as an alternate sector. That is, the first area A
If a write error occurs in sector 1-A of
To make the spare area C a substitute for the sector 1-A,
The ROM of the conversion unit 261 is rewritten, and as shown in FIG.
The spare area C is allocated as the sector 1-A. As described above, a write error has occurred.
In this case, a spare area sector is allocated as long as there is a spare area.
In the case of lack of equipment, all data are shown in FIG.
The data is saved in an external area such as the SRAM 23 and the sector conversion unit 2
Reorganized 61 ROMs to provide data space and spare space
Separate. At that time, the sector depends on the system configuration.
Assigned in ascending or descending order
But exclude bad sectors. In this embodiment
Is erasing at the same time as writing, as described above
Therefore, the time lag at the time of writing can be eliminated,
Implanting speed can be improved. Also, the spare area
To be replaced with a spare area when a bad sector occurs.
Remedy when a bad sector occurs.
Can be. (2) Embodiment 2 FIG. 39 and FIG. 40 are views showing the configuration of the present embodiment.
9, 261 is an EEPROM or flash
It consists of rewritable ROM such as flash memory
The sector conversion unit 25 is a flash memory,
The flash memory has a first area A and a spare area C.
In the first area A, four sectors 1 to 4 are provided.
ing. Reference numeral 231 denotes the SRAM 23 shown in FIG.
Is a primary storage medium for the data. In FIG.
When writing data, data is transferred to the primary medium 231.
The sector to be written, and
If there is data, the sector is erased. What
Please note that if the transfer is completed early,
If the transfer is slow, write the data after the transfer is complete.
Put in. If a write error occurs, the section
By rewriting the ROM of the data converter 261, FIG.
0 (when an error occurs in sector 1-A)
Is written in the spare area C.
I do. Unless a normal sector fails,
Writing to the spare area C is not performed. In this embodiment
Transfers the data to the primary storage medium 231 as described above.
While there is data in the sector to be written
In this case, since erasing is performed, writing is performed in the same manner as in the first embodiment.
Time lag at the time of writing
Can be improved. Also, a spare area is provided,
When a good sector occurs, it is replaced with a spare area.
In the event of a bad sector, relief can be performed
You. (3) Embodiment 3 FIGS. 41 and 42 are diagrams showing the configuration of the present embodiment.
1, 261 is an EEPROM or flash
It consists of rewritable ROM such as flash memory
The sector conversion unit 25 is a flash memory,
Flash memory consists of block 1, block 2, and spare block.
Block 1 and spare block 2 are provided.
Lock 1 and block 2 have sectors 1-3A, 1-3
B, ..., 1-3D, 4-6A, 4-6B, ..., 4-6D
Is provided. Then, sectors 1-3A, 1-3
B,..., 1 to 3D each have one physical address.
Where the data of logical addresses 1, 2 or 3 are stored.
Data can be written. That is, the logical address
The area for writing 1, 2, or 3 data has four areas A to D.
When writing data, A
ＤD is written in an empty area. Similarly, sectors
4-6A, 4-6B,..., 4-6D are the same as above.
When writing data at logical addresses 4, 5 or 6.
In this case, the data is written in an empty area of A to D. that's all
As described above, in the present embodiment, n + 1 (in this embodiment,
In this case, n = 3) is divided into sectors, and
One sector space is provided for the logical address to be written.
Have been. Referring to FIG. 41, data is written in sector 1.
, See sectors 1-3A, 1-3B, ..., 1-3D
If sector 1 already exists in these,
Erase sector. Also, it is always empty in the sector
Since there are sectors, empty
Write to existing sector. When writing, write
If an error occurs, a spare block
Assign a lock. For example, sectors 1 to 3A are defective.
42, the spare block 1 is replaced as shown in FIG.
Assign as a replacement block. And the sector conversion unit
261 ROM was rewritten and the block with the error
1 is assigned to the spare block 1. this
Thus, the spare block 1 becomes the block 1. There is a spare
As long as the above processing is done, and there is no spare
Indicates that data is stored in an external device such as the SRAM 23 shown in FIG.
And reorganize the ROM of the sector conversion unit 261
Then, it is divided into data space and spare space. Also,
At times, sectors are ordered in ascending order or depending on system configuration.
Are assigned in descending order in the same way as in the initial state.
Kuta is excluded. Whether the sector conversion unit 261 is a logical sector
When converting from a physical sector to a physical sector, as described above,
Data for each sector 1
So that there are n + 1 writable sectors.
To achieve. And if some sectors are bad
Can rewrite the table of the sector conversion unit 261.
By moving the write block, the (n + 1)
Create a block of sectors. In this embodiment,
As described above, it is divided into n + 1
Data is erased while writing.
Thus, similarly to the first and second embodiments, a time lag at the time of writing is reduced.
Can improve writing speed
You. Also, if a spare area is provided and a bad sector occurs
In the meantime, since a spare area is used, bad sectors
Can be relieved. (4) Embodiment 4 FIGS. 43 and 44 are views showing the configuration of the present embodiment.
In step 3, 261 is an EEPROM or flash
It consists of rewritable ROM such as flash memory
The sector conversion unit 25 is a flash memory,
Flash memory is block 1, block 2, spare 1A ~
1D is provided, and block 1 and block 2
, Sectors 1-3A, 1-3B, ..., 1-3D, 4-6
A, 4 to 6B,..., 4 to 6D are provided. Also,
In the present embodiment, n + 1 (this actual
In the embodiment, n = 3) a block is formed in sector units, and 1
Leave a sector empty. Referring to FIG.
When writing to sector 1, sectors 1-3A, 1-3B,.
Looking at 3D, if sector 1 already exists in these
, The sector is erased. Also, the sector must be empty
Because there is a sector that is
Then, writing is performed to an empty sector. When writing,
If a write error has occurred,
Reserves 1A to 1D are allocated. For example, sectors 1-3A
Becomes defective, as shown in FIG.
A to 1D become block 1, and for sectors 1 to 3A
Allocate spare 1A and write ROM of sector conversion unit 261
In other words, the alternative processing is performed. As a result, the spare sector 1A
Sectors 1 to 3A are provided, and spares 1B to 1D are blocks 1
Become a spare. The above processing is performed as long as there is a spare,
If there is no spare, all data is shown in FIG.
To an external area such as the SRAM 23,
261 ROM, reorganize data space and spare space
Divided into Also, at that time, the sector depends on the system configuration.
Assign ascending or descending order in the same way as the initial state.
However, bad sectors are excluded. Sector conversion unit 26
1 indicates the above when converting from a logical sector to a physical sector.
As shown in FIG.
In the table, there is a table for address conversion based on the decoded data.
N + 1 cells that can be written to sector 1
Are configured to be present. And some sectors
If the data is defective, the data of the sector
By rewriting the table, the
Create a block. In the present embodiment,
Is divided into n + 1 sector units and one sector is left
Beforehand, since erasing is performed at the same time as writing,
Eliminates the time lag at the time of writing as in Examples 1, 2 and 3.
Writing speed can be improved.
Also, a spare area is provided, and when a bad sector occurs,
Since a spare area is used, when a bad sector occurs
Relief can be done. (5) Embodiment 5 FIGS. 45 and 46 are diagrams showing the configuration of the present embodiment.
5 and 46, the same as those shown in FIGS. 37 and 38
Are denoted by the same reference numerals, and 261 is an EEP
Rewriting of ROM or flash memory is possible
Sector conversion unit composed of functional ROM,
Flash memory, and flash memory is a block
1, block 2, spare block 1, spare block 2
Block 1, block 2, sectors 1-3A, 1
... 3B, ..., 1-3D, 4-6A, 4-6B, ..., 4-
6D is provided. In this embodiment, the actual
As in the third embodiment, n + 1 (n = 3 in this embodiment)
It is divided into sectors, and one sector is left. Figure
45, when data is written to sector 1, the sector
Look at 1-3A, 1-3B, ..., 1-3D
If sector 1 already exists, erase that sector
I do. Also, since there are always empty sectors,
At the same time as the erasure described above, writing is performed
U. When writing, if a writing error occurs,
A spare sector is assigned as a substitute sector. For example,
In the case where the first to third members have failed, the spare block 1
And rewrite the ROM of the sector conversion unit 261
And perform an alternative process. Thereby, as shown in FIG.
And the spare block sectors 1, n, and mA are placed in sector 1.
This means that sector 1 exists in the spare block.
Therefore, sectors 2 and 3 remain in the original block. That is, in this embodiment, the block
The structure of the block does not need to be a contiguous sector.
Change the number of configurations. As long as there is a spare,
When there are no more spares, all data is
Is saved in an external area such as the SRAM 23 shown in FIG.
The ROM of the conversion unit 11 is reorganized so that the data space and the spare
Divide into spaces. At that time, the sector is
Ascending or descending order, as in the initial state, depending on the configuration.
However, bad sectors are excluded. In this embodiment,
In addition, as described above, blocks are formed in units of n + 1 sectors.
One sector is left empty and erased simultaneously with writing.
Writing, as in the first, second, third and fourth embodiments.
Time lag can be eliminated and writing speed can be improved.
Can be up. Also, a spare area is provided to
In the event of a stag, it is replaced with a spare area,
Relief when a bad sector occurs can be performed. FIGS. 47 to 49 show processing in the above embodiment.
FIG. 37 is a flowchart showing the system shown in FIG.
Of the present embodiment will be described. In step S1
When there is a write request for sector data from the main unit,
In step S2, the CPU 22 determines whether the controller LS
Check that there is a data write request due to the notification from I21.
And recognize. Note that the controller LSI 21 has a CPU
Notify the data write request by interrupting 22
May be. In step S3, the CPU 22 executes EE
Reads the data in PROM 26 and sets the write location
Recognize and check if sector data is written
, 25-5,.
To access the ECC data or flash
Reference is made to management information and the like stored in the memory. And
In step S5, data is written to the flash memory.
Judge whether or not data is written.
In this case, the procedure goes to step S8 in FIG. Also, the data
If it is written, in step S6, the second
To read information for writing data to the sector,
The CPU 22 accesses the EEPROM 26 and stores the data.
obtain. Then, in step S7, the flash
Erase memory data (error checking to increase efficiency)
Will be performed later), and the process will proceed to step S8 in FIG. In step S8, data is sent from the main unit.
The data is transferred to the SRAM 23, and waits for the end. Steps
At S9, write data to the flash memory
Therefore, the CPU 22 transmits data to the controller LSI 21.
Data is written, and the controller LSI 21
Write data from RAM 23 to flash memory
No. Next, in step S10, the CPU 22
Notification of end of data writing from controller LSI 21
Then, in step S11, a data write error occurs.
It is determined whether or not there was. Data write error
If there is no write, the process ends.
In step S12, there is an alternative sector in step S12.
Judge whether or not there is a replacement sector, and
Go to step S13, and change the sector address.
Change the data by accessing the PROM 26,
It returns to S9. When there is no replacement sector, FIG.
Go to step S14 of step 9 to erase the flash memo being erased.
Flash memory that is being erased
If there is, go to step S15 until the erase is completed.
After waiting, the data writing to the erased sector is completed.
In step S14, the flash memory being erased
If it is determined that there is no
Error to the main unit. On the main unit,
In S17, the data is downloaded and the EEPROM 26 is
Send a command to reorganize. On the storage device side,
In response to these commands, the EEPROM 26 is reorganized,
Two or more sectors for one logical sector sent from the main unit
To be able to choose. This reduces the capacity slightly.
However, it can be used as a storage device. Note that the above processing
Transfer the data to the SRAM 23 and flash
・ Writing data to memory has been explained,
Without providing the SRAM 23, the data is directly stored in the flash memory.
Data can also be written (in this case,
Steps S8 and S9 of the chart are one process). E. Reliability of decoding part for address translation
Improvement Hereinafter, an embodiment of the first aspect of the present invention will be described.
You. Translation to translate logical address to physical address
ROM may be used as a table. Also,
Flash EEPROM, EEPROM instead of ROM
Although it is possible to use OM or the like,
Because it is not necessary to make the part variable,
No. On the other hand, flash memory, due to the nature of its medium,
There is a limit to the number of erases.
The dress cannot be used. Therefore, the above-mentioned D.I. Fruit
As shown in Examples (1) to (5), the logical address is
As a conversion table for converting to addresses,
Rewriteable EEPROM, EEPROM, etc.
If a bad sector occurs using the decoding unit,
Rewrite the code part and make it look like there is no bad address
Is used. However, as a conversion table
Using flash EEPROM, EEPROM, etc.
If these flash EEPROM, EEPROM
M also has a limit on the number of erasures, and erases more than a predetermined
When doing so, a defect occurs. This embodiment, as described above,
Flash EEPROM, EEPROM as decoding unit
When a rewritable storage medium such as M is used
, The decoding section is made into two stages,
Extend the life of the code section, and improve the
The example which improves reliability is shown. FIG. 50 shows the structure of the decoding unit of this embodiment.
FIG. 5A shows a normal case, and FIG.
(C), the decoding unit becomes defective
301, a primary decoding unit, 302
Is a secondary decoding unit, and 303 is a flash memory section.
Data. In the figure, the flash memory
If the sector is normal, as shown in FIG.
The code part 301 is used as the decoded value of the address of the sector 0.
And outputs the address "0000h" to the secondary decoding unit 3.
02 is the physical address as the decoded value of “0000h”
"5555h" is output and the sector of the flash memory is output.
0 physical address is specified. Where the flash
・ The number of erasures of the memory exceeds the limit and the sector becomes defective.
The physical address of sector 0 is reserved for address 8888h.
When switching to a ready sector, as shown in FIG.
As described above, the secondary decoding unit 302 is rewritten to “0000”
h ”from“ 5555h ”to“ 8888h ”
Switch to As a result, the physical address is set as sector 0.
"8888h" is assigned. As above
And how many times the secondary decoding unit (or primary decoding unit)
Overwriting limit value by rewriting,
Next decoding unitButWhen it becomes defective, as shown in FIG.
The primary decoding section 301 is rewritten and its decoded value
Stagger. That is,Secondary decoding section 302 decodes
"2222h" which outputs "8888h" as the value is
Output as code valuePrimary decoding section 301To
rewrite. As a result, the secondary decoding unit 302 is defective.
The physical address "8888h"
Can be output as As described above, the decoding unit is divided into two stages,
If one of the decoding units becomes defective, the other decoding unit
Rewrite the section to square the life of the decoding section
can do. For example, until a sector becomes bad
Number of times L (rewrite limit value), secondary decoding unit is defective
The number of times until it becomes M (rewrite limit value), primary deco
The number of times until the card part becomes defective is N times (rewrite limit
Value), the number of times until the address becomes an error is N
× M × L times. Basically, the rewrite limit is flash
100,000 to 1,000,000 times in EEPROM, and in EEPROM
It is said to be about 10,000 times, so the decoding part is actually
Just by two stages, sufficient reliability can be secured
You. F. Write time estimation process In flash memory, writing data
When saving, it is necessary to perform evacuation processing, etc.
It takes longer to write than in the case of. In this embodiment,
The write time to the flash memory and estimate the power consumption.
An embodiment is shown in which force estimation, abnormality detection, and the like are performed. Figure
51 shows a schematic configuration of the entire system of the present embodiment. 1 is
The storage device 1a cannot be overwritten and
Data can only be erased in predetermined units, such as
Cannot be configured, such as a flash memory
Storage area, 1b is a controller, 2 is a main processor
It is. Next, this embodiment will be described. Now storage
Is in the state shown in FIG.
Let us consider the case of writing data in the data. First, the book in FIG.
The number of sectors to be written is sent to the processor 22 from the body. Writing
When the number of sectors to be included is sent to the processor 22, the processor
Then, the write time is obtained from the number of sectors to be written. This
Here, when there is no block to be written as shown in FIG.
Since the save operation is performed, the write time t is
I will be asked. t = [time required for writing one sector] × 3 + [data
Evacuation time] (seconds) Processor 22 calculates the write time as described above
And return it to the main unit. The main unit uses the writing time
The power consumption W1 at the time of the write operation is obtained by the following equation. W1 = t ÷ 3600 × [power consumption of write operation]
(W) The main body reads the remaining power W2 from the power supply.
Compared with the obtained write power, and if W2> W1,
Write data. On the other hand, the body is
The obtained writing time t is compared with the actual writing time.
When the writing is not completed even after the writing time t
Determines that the storage device is abnormal, and notifies the user, for example.
Or take measures such as stopping the writing process. FIG. 53 shows the processing of the main body in this embodiment.
This flowchart is used to explain the present embodiment.
You. In step S1, the data to be written
Send the size to storage and get time to write
You. In step S2, the time required for writing,
Power consumption per hour by writing power
obtain. In step S3, the remaining power and
Compare the power used, and if the remaining power is less,
Perform error processing. If the remaining power is more,
Go to step S4 and write data, go to step S5
The actual writing time is longer than expected
It is determined whether or not it is, and if there is more, an abnormal process is performed. Also, calligraphy
If the recording time is within the expected time, step S
6 to determine whether or not the writing is completed,
If not, the process returns to step S5, and when writing is completed.
If so, the process ends. As described above, in this embodiment, the writing
The power required for writing
Since it is estimated, whether writing can be done with the remaining power
Can be determined. For this reason, battery drive
Power is lost during writing
Can be prevented from being interrupted on the way.
System reliability can be improved. Also, the actual book
By comparing the writing time with the estimated writing time,
Can detect abnormalities in the storage device,
By notifying the user by displaying the
This allows the user to determine the abnormality of the storage device.
You. G. Improvement of reliability in flag judgment processing Normally, a flag is determined by assigning one bit to one function.
It is often specified. However, the flash menu
A fatal accident that the memory cannot be erased due to over-erasure
It has a good cause structurally.
It falls into a state where it does not return to the low level while staying at the low level.
Therefore, the flag on the flash memory is set to 1 bit 1 bit.
If the function was supported, the bit became defective.
Sometimes, the validity of the flag is lost. This embodiment
Is on the flash memory as shown in FIG.
Flags for erase flags, bad flags, parity, etc.
Multiple flags and determine the flag by logical product.
Address the structural disadvantages of flash memory described above
And shows an embodiment in which the reliability of flag determination is improved.
You. (1) Embodiment 1 FIG. 54 is a diagram showing the first embodiment, and FIG.
Register, (b) is an AND circuit for determination, (c) is
4 shows a truth table. In this embodiment, FIG.
As shown in (a), the flag b0,
Prepare b4 and store it in the flag register. And
When performing the lag determination, as shown in FIG.
By determining the logical product of the flags of
Get fruit. Since the determination is made as described above, FIG.
Bit b0 is fixed at a high level as in row number 2 of
However, as a result of the logical product, a correct result can be obtained.
Can be. Similarly, as in row number 3, bit b4 is
Even if it is fixed at a high level,
Can get the correct result. (2) Embodiment 2 FIG. 55 shows a second embodiment, and FIG. 55 (a) shows the first embodiment.
A flag register, (b) is a second flag register,
(C) shows an AND circuit for performing flag determination. Book
Embodiments use different flag registers for flags for the same function.
The data is stored in the data as shown in FIG.
The flag is stored in bit b0 of the first flag register,
Bit b0 of the second flag register shown in FIG.
The bit stored in the first flag register in bit b2 is
The flags corresponding to the same function are stored. And
When the flag is determined, the logical multiplication shown in FIG.
The logical product of the flags is obtained by the path. In this embodiment
However, as in the first embodiment, some of the flags are fixed at a high level.
Even if it is determined, a correct determination result can be obtained. More than
As described above, in the present embodiment, the determination of the flag is performed by a plurality of bits.
All bits are defective.
Unless it is necessary, the correct judgment value can always be output,
The reliability of the flag determination can be improved. H. Management table savings In this embodiment, the storage device is composed of a plurality of chips, and each chip
Flash memory with multiple blocks inside
In the system, the management table is simplified and
Example where each block can be used on average
Is shown. In the embodiment described above, data is
There are no restrictions on the chips and blocks to be written, and data is
Because the lock sector could be written, the entire area was
Although it was necessary to provide a table to manage,
The data to be written in each chip is fixed,
Move the block to be written in the
Data writing without having to manage the entire area.
Flash memory management
It simplifies and averages the use of each block.
You. Also, in the present embodiment, the
Data is smaller than the total capacity of the chip by a certain amount.
Quantitative blocks when work blocks and bad blocks occur
Relief block. And the data in the block
The sequence of data is fixed, and when rewriting data,
If there is valid data in the old block in
The data is saved in the SRAM 23 shown in FIG.
Write to the empty area along with the new data and write
Erase old data block if successful
are doing. Next, this embodiment will be described. FIG. 56 shows the present embodiment.
FIG. 7 is a diagram showing a configuration in a chip, a block, and a sector in an example.
Yes, as shown in the figure, the chip is 2 Mbytes,
No. of 4Kbyte per block. 000-No. 511
Is divided into 512 blocks, and in each block,
No. of 528 bytes in one sector. 00-No. 07 8
Is provided. The block No. in FIG. You
And sector No. Indicates a physical address. Further
In the sector, as shown in the figure, the actual data of 256 bytes
Data area, ECC area for Long instruction, defective data file
Lag, bad block flag, block address, 256byt
e is provided with an actual data area ECC area. Bro
Block address stored in the SRAM 23.
Indicates the address value of the block pointer and the block address.
Sectors that have been determined to have no actual data to be written.
And In this way, hot-line insertion and removal
When the number of sector addresses is large, it is determined that the data is normal
be able to. If the number of sector addresses is equal,
Judgment is made based on the correctness of ECC and checksum. Bad block
The flag indicates the condition of the block.
Fh: normal block, $ FFh: bad block
You. Bad data flag indicates data condition
Normal data at FFh, bad data at FFh
And ECC area for Long instruction is up to 4 bytes
It is. The ECC area indicates the condition of the actual data.
Data correction and error
Perform detection. FIGS. 57 to 61 show the writing in this embodiment.
FIG. 9 is a diagram showing only processing, and the present embodiment will be described with reference to FIG.
You. In this embodiment, the chip is No. 00-No.
04, 5 pieces each with 4 work blocks
The writing and management method in the case is shown. FIG.
This shows the state where all blocks have been erased.
Thus, each chip has a work block of work 01 to 0
4 are provided, and the data is the above work block
It is written to work01-04. Write is sector
This is possible in units, but after writing sector 005,
After writing the upper sector, such as writing data 004
Cannot write to the lower sector. Also, insert the tip
Data does not move beyond
The order of the data is fixed. Further, for example, logical blocks
Write only sectors 000-005 to address 00
Write data does not fill all eight sectors
Write data in sectors 000-005
Then, only the block address is written in sectors 006 and 007.
Get in. In this case, one block is composed of eight sectors.
Also, since there are five chips, the logical address n
The data is obtained by dividing the quotient of (logical address n ÷ 8) by 5.
Is written to the remaining chips obtained when example
For example, when n = 121, 121 ÷ 8 is 15 and the remainder is 1
15 ÷ 5 is 5 and the remainder is 0. 00 of
Written to the chip. The data write processing in this embodiment is
It is performed as follows. First, all the blocks shown in FIG.
From the erased state, the sector No. 000-039 (Logical
FIG. 58 shows the result of continuous writing of 40 sectors
Data is written to work01 of each chip as shown
It is. That is, the sector No. to be written first. 00
Since the logical block address of 00 to 007 is 00, the main body
From the SRAM 23 to the SRAM 23. 000-007 days
Transfer the chip No. 00 write pointer
Data is written from the indicated work01. Note that the logical block
Lock addresses are managed in chip units. And this
In the case of, the logical block address 00 does not exist in the past.
Therefore, chip No. Work01 of 00 is work0
Move to the free area next to 4. 8 sector write
Upon completion, the write pointer is set to the chip No. 01 w
ork01, and sector No. 008
Write the data of .about.015 to work01. As described above, writing is performed on chip N
o. 04, the write pointer is set to chip N
o. It moves to work02 of 00. Next, as shown in FIG.
From the state, as shown in FIG. 120-19
Continuous writing of 72 sectors of 1 (logical address)
You. In the same manner as described above, the sector number is stored in the SRAM 23 from the main unit.
The data of the sector Nos. 12
Since 0 to 127 have a logical block address of 03,
No. Is work02 pointed to by the 00 write pointer?
Write the data. Then, as described above, the chip N
o. 00 work02 and work03 work01
Move to the next free space. Similarly, the sector No.
The data of 128 to 191 is stored in the chip No. 00-No. 0
Write to 4. Here, from the state of FIG.
o. When two sectors 002 to 003 are continuously written,
As shown in FIG. First, the sector is stored in the SRAM 23.
No. 002 to 003 are transferred. in this case,
If the write data is a sector No. Since it is 002-003,
Logical block address 00 is specified, which is
Therefore, it is determined that the old data needs to be saved. Ma
If the write data is a sector No. 002-003,
Not written from the beginning of logical block address 00
The data of the old logical block address 00 is stored in the SRAM2
Evacuate to 3. Note that the write data is
When two sectors from the beginning of the address
Data of the old logical block address after processing the data first
Is performed. Next, the old data 0 saved in the SRAM 23
00, 001, 004, 005, 006, 007 and new data
Data 002, 003 and the chip N indicated by the write pointer
o. Write to work04 of 00. At that time, write
Is 000,001,002,003,004,005,
Write in the order of 006 and 007. Next, the logical block
Since address 00 already exists, the old logical block
Erases the block address. In addition, old work04
Move to logical block address 00. As mentioned above,
Power is turned off after continuous writing
When turned ON again, as shown in FIG.
The block is specified after the used block. Also,
Specify the work block as described above,
Unable to specify all work blocks even when searching to the end
If it does, return to the beginning of the chip and
Is specified. In the above process, a block erase error
When the error occurs, 00h is set to the defect flag and
Block cannot be written or read. sand
In other words, by coping with one work block,
When there are no more work blocks, the system will write
Not allowed. Also, an error occurred during data backup
In this case, write the data after setting the bad data flag.
Get in. In the read operation, an error occurs in that block.
Is detected, the block is regarded as a bad block.
Do not treat. If a data write error occurs,
The block is regarded as a bad block. Setting a bad flag
Save the data that has been written once,
Set the flag only and write again. Bad block hula
In the same way as normal writing, write
Stand up against. If an error is detected during data deletion
In that case, the block is regarded as a bad block. Bad bro
How to set the check flag, set only the bad block flag.
Instead, write ALL [00] including actual data.
Get in. Defect processing is performed on all sectors of the block.
Do. As described above, in this embodiment, each channel
The data to be written in the chip is fixed,
Since the block to be written is moved and written,
There is no need to manage the area. For this reason, the data is
Only need to be managed within the server, saving management tables
And the speed of the write operation can be increased.
You. Also, write data to the work block,
Since the block is configured to move within the chip,
Each block can be used on average. [0072] As described above, according to the present invention,
Has two stages of decoders, and a part of the storage area of the storage device is
When destroyed or when part of the decorator was destroyed
Two-stage decodingDaDestroyed by rewriting
Since decoding to parts is not performed, decoding
One such as EEPROM, flash memory, etc.
When using a storage medium that may cause
Even reduce the probability of address errors
And reliability can be ensured.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a principle diagram showing an overall schematic configuration of the present invention. FIG. 2 is a principle diagram (1) showing an outline of the present invention. FIG. 3 is a principle diagram (2) showing an outline of the present invention. FIG. 4 is a principle view (3) showing an outline of the present invention. FIG. 5 is a block diagram showing a configuration of a storage device as a premise of the present invention. FIG. 6 is a diagram showing the contents of an SRAM in the storage device. FIG. 7 is a diagram showing the contents of a flash memory in a storage device. FIG. 8 is a diagram (continued) showing the contents of the flash memory in the storage device. FIG. 9 is a flowchart illustrating a writing process in the storage device. FIG. 10 is a flowchart (continued) showing a write process in the storage device. FIG. 11 is a flowchart (continued) showing processing at the time of writing in the storage device. FIG. 12 is a flowchart (continued) showing processing at the time of writing in the storage device. FIG. 13 is a flowchart (continued) showing processing at the time of writing in the storage device. FIG. 14 is a diagram showing an embodiment for suppressing variation in the number of times of erasing. FIG. 15 is a diagram (continued) showing an embodiment for suppressing variation in the number of erasures. FIG. 16 is a diagram (continued) showing an embodiment for suppressing variation in the number of erasures. FIG. 17 is a diagram (continued) showing an embodiment for suppressing variations in the number of erasures. FIG. 18 is a diagram (continued) showing an embodiment for suppressing variation in the number of erasures. FIG. 19 is a flowchart illustrating a process according to an embodiment for suppressing variation in the number of erasures. FIG. 20 is a diagram showing an embodiment of a process for eliminating erasable data. FIG. 21 is a diagram (continued) showing an embodiment of a process for eliminating erasable data. FIG. 22 is a diagram (continued) illustrating an embodiment of a process for eliminating erasable data. FIG. 23 is a diagram (continued) showing an embodiment of a process for eliminating erasable data. FIG. 24 is a flowchart showing an embodiment of a process for eliminating erasable data. FIG. 25 is a diagram showing an embodiment of a free area creation process. FIG. 26 is a diagram (continued) illustrating an embodiment of a process of creating a free area; FIG. 27 is a diagram (continued) illustrating an embodiment of a process of creating a free area; FIG. 28 is a diagram (continued) illustrating an embodiment of a process of creating a free area; FIG. 29 is a diagram (continued) illustrating an embodiment of a process of creating a free area; FIG. 30 is a diagram (continued) illustrating an embodiment of a process of creating a free area; FIG. 31 is a diagram (continued) illustrating an embodiment of a process of creating a free area; FIG. 32 is a flowchart of an embodiment of a process of creating a free area. FIG. 33 is a diagram illustrating an example of an erasing process. FIG. 34 is a diagram (continued) illustrating an example of an erasing process. FIG. 35 is a flowchart of an embodiment of an erasing process. FIG. 36 is a diagram showing a system configuration of an embodiment for improving a writing speed. FIG. 37 is a diagram showing a first embodiment for improving the writing speed. FIG. 38 is a diagram (continued) showing the first embodiment for improving the writing speed. FIG. 39 is a diagram showing a second embodiment for improving the writing speed. FIG. 40 is a diagram (continued) showing a second embodiment for improving the writing speed. FIG. 41 is a diagram showing a third embodiment for improving the writing speed. FIG. 42 is a diagram (continued) showing a third embodiment for improving the writing speed. FIG. 43 is a diagram showing a fourth embodiment for improving the writing speed. FIG. 44 is a diagram (continued) showing a fourth embodiment for improving the writing speed. FIG. 45 is a diagram showing a fifth embodiment for improving the writing speed. FIG. 46 is a diagram (continued) showing a fifth embodiment for improving the writing speed. FIG. 47 is a flowchart of an embodiment for improving the writing speed. FIG. 48 is a flowchart (continued) of the embodiment for improving the writing speed. FIG. 49 is a flowchart (continued) of the embodiment for improving the writing speed. FIG. 50 is a diagram showing an embodiment in which the address conversion decoding unit has two stages. FIG. 51 is a diagram showing a schematic configuration of an entire system for performing a writing time estimation process. FIG. 52 is a diagram illustrating a state of a storage area in the embodiment of the writing time estimation processing. FIG. 53 is a flowchart of an example of a writing time estimation process. FIG. 54 is a diagram illustrating a first example of a flag determination process. FIG. 55 is a diagram illustrating a second example of the flag determination process. FIG. 56 is a diagram showing a configuration of a storage area in an embodiment for saving a management table. FIG. 57 is a diagram showing an embodiment for saving the management table. FIG. 58 is a diagram (continued) showing an embodiment for saving the management table; FIG. 59 is a diagram (continued) showing an embodiment for saving the management table; FIG. 60 is a diagram (continued) showing an embodiment for saving the management table. FIG. 61 is a diagram (continued) showing an embodiment for saving the management table. [Description of Signs] 1,20 Storage device 1a Storage area 1b Primary storage medium 1c Control means 1d, 1e Decoding table 2 Main body processing device 21 Controller LSI 22 Processor 23 SRAM 24 Clock oscillator 25-1 to 25-5 Flash memory 26 EEPROM 261 Sector converter 301 Primary decoder 302 Secondary decoder

──────────────────────────────────────────────────続 き Continued on the front page (72) Inventor Tomohiro Hayashi 4-36, Honcho, Naka-ku, Yokohama-shi, Kanagawa Prefecture Within Fujitsu Computer Technology Co., Ltd. (72) Inventor Shogo Shibasaki 4-36, Honcho, Naka-ku, Yokohama-shi, Kanagawa Prefecture Shares (72) Inventor Hiroyuki Ito 4-36 Honcho, Naka-ku, Yokohama-shi, Kanagawa Prefecture Fujitsu Computer Technology Co., Ltd. Masaru Takehara 4-36, Honcho, Naka-ku, Yokohama-shi, Kanagawa Prefecture Fujitsu Computer Ltd. (56) References JP-A-5-233470 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) G06F 12/16 G11C 16/06 G11C 29/00

Claims (1)

(57) [Claims 1] A storage area that is readable and writable and that may partially destroy the storage area, and a place where data in the storage area is stored a method of controlling a memory device having a rewritable decoder, as the decoder, to output the address of the storage area
Output the secondary decoding unit and the address of the secondary decoding unit
A primary decoding unit is provided that performs the above-mentioned secondary decoding when a part of the storage area of the storage device is destroyed.
Rewrite the address in the code section to
Code is not executed, and when the secondary decoding unit is destroyed, the primary decoding unit is
A method for controlling a storage device, wherein rewriting is performed so that decoding of a destroyed portion is not performed.