JP2749752B2 - Memory write control device - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method of controlling the number of times of writing to an electrically rewritable memory.
The present invention relates to a control device . 2. Description of the Related Art Conventional EEPROMs (Electrical Erasa
ble and Programmable ROM) has a small capacity,
There were many external circuits necessary for writing. Further, it has only a mode for erasing all data in the chip. Recently, as the capacity has increased, it has become possible to connect to the address bus and data bus of the CPU with almost no need for external circuits, and it has become possible to erase only 1-byte data in the EEPROM. With the above improvements, it has become possible to replace the function configured by the conventional random access memory (RAM) depending on the purpose of use. For example, there is a storage medium called a memory card for holding a program, a sentence, an external character, and the like created by a conventional small personal computer and a Japanese word processor. this is,
The RAM and battery were installed in the memory card so that the program and text were stored when inserted into the main body of a personal computer, Japanese word processor, etc. when necessary, and the data was stored even when the memory card was pulled out. . Therefore, it was considered that the battery could be eliminated by configuring the memory card with an EEPROM. [0004] However, EEPRO
M has a restriction that it cannot be freely rewritten many times unlike a conventional RAM. That is, by writing data to a memory card exceeding a preset number of times of writing, data to be stored can be stored. There were problems such as disappearing. [0005] Of the data written in the EEPROM, there are data that is frequently rewritten and data that is rewritten less frequently, and when the number of rewrites of frequently rewritten data exceeds a predetermined value, rewriting to the EEPROM is stopped. There has been a problem that rewriting is impossible even though it is possible. SUMMARY OF THE INVENTION The present invention has been made to solve the above-described problem, and determines the number of rewrites for each area of a memory divided into a plurality of areas. Use less space first
It is another object of the present invention to provide a memory write control device capable of greatly extending the write life of a memory having a limited number of write operations by controlling the rewriting of the memory. A memory write control device according to the present invention is a memory write control device capable of electrically rewriting information written in a memory, wherein the information is divided into a plurality of areas. Storage means for storing the number of rewrites for each area of the memory, determination means for determining the number of rewrites for each area stored in the storage means, and the number of rewrites for each area determined by the determination means. based, as the difference between the number of times of rewriting each region is reduced, writing
So that the area with the least number of replacements is preferentially used
Correction means for controlling rewriting of the memory . According to the present invention, the correction means determines the number of rewrites for each area, which is determined by the determination means for determining the number of rewrites for each area, stored in the storage means .
So that the area with the least number of replacements is preferentially used
By controlling the rewriting of the memory, the frequency of writing to each area is dispersed. FIG. 1 is a block diagram of a memory according to an embodiment of the present invention.
And FIG. 11 is a block diagram illustrating a configuration of an information processing device to which the embedded control device can be applied .
8 having a RAM 11b and stored in the ROM 11a,
Each unit is controlled according to a program according to the flowchart shown in FIG. Reference numeral 12 denotes input means for instructing data writing and data erasing in the EEPROM 1 as programmable read only memory means set in the data writing device 13. The CPU 11 has accumulators ACC and BCC for transferring data. [0010] The write control of the memory configured as described above.
The control device includes a memory (EEPRO) divided into a plurality of areas.
M1) Number of rewrites for each area (number of rewrites described later)
WCNT) (“0” of the EEPROM 1).
-1 "), and stored in the storage means.
Judging means for judging the number of rewrites for each area (CPU 1
1 is a program stored in the ROM 11a and the RAM 11b.
Is determined by executing the program).
Of the number of rewrites based on the
Write to the memory to use less space
Correction means for correcting (when the CPU 11 has the ROM 11a
11b by executing the program stored therein) and
And the CPU 11 performs the rewriting number W of the EEPROM 1
Based on the number of rewrites for each area stored in the CNT
EEP so that an area with a small number of rewrites is used.
Correct writing to ROM1 and write to each area
Disperse frequency. [0011] FIG. 2, port of EEPROM1 shown in FIG. 1
It is a schematic diagram explaining the memory map of an inter block . EEPROM1 shown in this figure, for example, writing capacity 327 6 8 bytes × 8 bits, the write count is 1
It is assumed that it has been set to 10,000 times. In addition, EEPROM1
Part of the storage area of, composed of the pointer blocks 1a and spare pointer block SPB1～SPB50. Pointer block 1a has 5 addresses (1 byte each)
The number of rewrites WCNT, for example, “1388 ₁₆ ” is stored in the address storage bytes “0 to 1”. One byte at the address “2” of the pointer block 1a is
A directory DB, for example, “01 ₁₆ ” is stored.
Further, one byte at address "3" of the pointer block 1a is an unused start block number OSB, for example, "33".
₁₆ "is memorized. Also, 1-byte "4" address pointer block 1a stores unused end block number 0EB, for example, "7 A ₁₆ '. [0012] Figure 3 is a schematic diagram showing a storage area of the E EPROM1. In this figure, reference numeral 21 denotes a block number, for example, 127 blocks BLOCK1 to BLOCK
CK127. Each block is, for example, 2
It is composed of 56 bytes, and the first two bytes store the number of times the block has been updated, that is, the number of updates described later. Next, the following 253 bytes are stored data D
ATA is stored, the last byte has a continuation block area CB indicating whether the storage data DATA stays in this block or extends to another block. If the storage data DATA extends to another block, In the continuation block area CB, a block number to be continued is described, and when the storage data DATA does not reach other blocks, “FF ₁₆ ” is stored in the continuation block area CB. FIG. 4 is a schematic diagram for explaining the structure of each directory block shown in FIG. In this figure,
Reference numeral 30 denotes a directory block specified in the directory DB, and reference numeral 31 denotes an update counter of the directory block 30, which is composed of, for example, 2 bytes. Reference numeral 32 denotes a file area in which each file name is stored in 12 bytes.
A start block number area (SB) 33 is composed of, for example, 1 byte and stores a start block number of a file. Reference numeral 34 denotes an end block number area (EB), which is composed of, for example, 1 byte and stores the end block number of the file. Reference numeral 35 denotes a chain block area (CB) which stores the presence or absence of a directory block continuing from the directory block 30. For example, the chain block area 35 is “FF ₁₆ ”. The directory block 30 is composed of, for example, 18 file areas 32. Next, referring to FIG. 2 and FIG.
The structure of the PROM 1 will be described. As shown in FIG. 2 , for example, if “1388 ₁₆ ” is stored in the number of rewrites WCNT of the pointer block 1a, 500
Indicates that 0 updates have been made, and directory D
Since “01 ₁₆ ” is stored in B, the block number of the directory block 30 specified in the directory DB is “1”, and “142F ₁₆ ” is stored in the update counter 31 of the directory block 30. ing.
This means that this directory block 30 is "5167"
In the file (File) 1 (file name) of the file area 32, the start block number area 33 is “02 ₁₆ ” and the end block number area 34 is “0516”.
It starts with CK2 and ends with block BLOCK5. The file 2 file area 32, a start Proc number area 33 "0A _16", since the end Proc number area 34 is "0Al6" begins Proc BLOCK10, Proc B
It ends with LOCK15. Further, since the start block number area 33 of the file 3 (file name) of the file area 32 is “15 ₁₆ ” and the end block number area 34 is “18 ₁₆ ”, the block BLO
It starts with CK21 and ends with block BLOCK24. Since “FF ₁₆ ” is written after file 3 in file area 32, this file area 3
2 ends with file 3. FIG. 5 is a schematic diagram for explaining the state of the unused EEPROM 1, and the same components as those in FIGS. 1 and 4 are denoted by the same reference numerals. As shown in this figure, the unused EE
Number of rewrites WCN of pointer block 1a of PROM1
T is “0001 ₁₆ ”, directory DB is “01 ₁₆ ”,
The unused start block number OSB “02 ₁₆ ” and the unused end block number OEB “7F ₁₆ ” are stored at addresses 0 to 4 of the pointer block 1a, respectively. Thereby, referring to the block BLOCK1 specified in the directory DB, “0001 ₁₆ ” is written in the update counter 31, and
“FF ₁₆ ” is written in the file 1 of the file area 32, and “F ₁₆ ” is further written in the chain block area 35.
F ₁₆ ”has been written, indicating that the EEPROM 1 is in an unused state. Further, “02 ₁₆ ” and “7F ₁₆ ” are written in the start block number OSB and the end block number OEB of the pointer block 1a, respectively. That is, the block BLOCK2
２７127, “0001 ₁₆ ” is written in the first two bytes, and the block BLOCK is stored in the chain block area 35 indicating the continuation of each succeeding block in the last one byte.
“03 to 7F ₁₆ ” is written to the blocks 2 to 126, and the chain block area 35 of the block BLOCK 127 is written.
Is written with "FF". Thus, each of the blocks BLOCK2 to BLOCK127 has a single chain structure. FIG. 6 is a schematic diagram for explaining a write operation to the EEPROM 1 shown in FIG. 1, and the same components as those in FIGS. 2 and 4 are denoted by the same reference numerals. It is assumed that the state immediately before the writing is as shown in FIG. First, the start of the file area 32 of each block BLOCK is "00".
₁₆ ". In the case of FIG. 4, file 2
There is "00 ₁₆ " between the file 3 and the file 3, and the name of the file 4 is written in 12 bytes there, and the start block number OS of the unused block of the pointer block 1a is written.
B indicates the block BLOCK, that is, “ 3
The 2 bytes information at the head of " 3 ₁₆ ", that is, the update counter 31 is incremented by "1", and if the added value exceeds, for example, 10,000 times, the block indicated by the chain block area 35 of that block. The same operation is performed on BLOCK, and the update counter 31 finds a block BLOCK of 10,000 or less, and checks the block BL.
The number of the OCK is written in the start block number OSB of the pointer block 1a. If the number is 10,000 or less, the data of the file 4 is written in the block BLOCK 51 (253).
Byte), and if the block BLOCK 51 overflows, the update counter 31 of the block BLOCK indicated by the chain block area 35 of the block BLOCK 51 is incremented by “1”, and the added value exceeds 10,000 times, for example. It is checked whether the update counter 31 of the designated block BLOCK exceeds 10,000 times, and the block BLOC whose update count is 10,000 times or less is checked.
Write to the K chain block area 35. In this way, data is written, and blocks whose update count exceeds 10,000 are eliminated. The same operation is performed until there is no more write data, and the chain block area 35 of the last written block BLOCK is written.
Is rewritten to a new unused start block number OSB, and the number of rewrites WCNT of the pointer block 1a is incremented by “1” to “1389”.
₁₆ ", and the block BLO to which data was last written
The CK chain block area 35 is set to “FF ₁₆ ”. Then, when the number of the block BLOCK in which the last data has been written is written in the end block number area 34 storing the last block number of the directory block 30 and the update counter 31 is incremented by "1", the result is shown in FIG. So, the update counter 31
Becomes “1430 ₁₆ ”, the start block number area 33 of the file 4 becomes “33 ₁₆ ”, and the end block number area 34 becomes “37 ₁₆ ”. Next, referring to FIG.
The operation of deleting the file 1 written in the file 1 will be described. Block BLO that becomes directory block 30
File 1 is searched from CK1, and the first two bytes of the file area 32 are set to “00 ₁₆ ”. Then, the update counter 31 of the directory block 30 is incremented by "1", and the data of the start block number area 33 and the end block number area 34 of the file 1 are referred to, and the data of the block indicated by the end block number OEB of the pointer block 1a is referred to. The contents of the chain block area 35 (which was "FF ₁₆ " until immediately before deletion) is changed to the contents of the start block number area 33, and the update counter 31 of this block is incremented by "1". That is, the file 4 just deleted is connected to the end of the unused block. In this way, the update counter 31 approaches 10,000 times while updating and deleting the file many times while moving the update counter 31 forward. Next, an access process when the directory block update counter 31 reaches 10,000 times will be described. First, the block BLOCK indicated by the contents of the start block number OSB of the pointer block 1a
The content of the chain block area 35 is a new start block number OSB. Then transfers the contents other than the information of the update counter 31 de <br/> I recto Li block 30 to the block. Then, a new directory block number is written to the directory DB of the pointer block 1, and the pointer block 1a rewrite count WCNT and the update counter 31 are incremented by "1". On the other hand, when the number of rewrites WCNT of the pointer block 1a exceeds 10,000, data other than the information of the number of rewrites WCNT is transferred to the nearest spare pointer block among the spare pointer blocks SPB1 to SPB50. Number of rewrites WCNT of new pointer block
(0000 ₁₆ ) is incremented by “1” to “0001”
₁₆ ". In this case, the number of rewrites WCNT of the discarded pointer block 1a becomes 10,000 or more, and the number of rewrites WCNT of the true pointer block 1a becomes 10,000 or less. Thus, the directory block 30
And the writing / deletion of the pointer block 1a. Also, the blocks used by the deleted file are turned to the end of unused blocks. This is to average the number of times unused blocks are used. However, if the used file remains unchanged without being updated, the block in which the file is used remains unchanged in the update count. For example, if the initially created file remains registered as it is, the imbalance occurs such that the other block has been updated 5000 times or more, but this file alone is 2 times. Therefore, a correction process for averaging the usage state of the EEPROM 1 is performed. The condition for starting the correction process in this embodiment is the time when the following conditions (a) and (b) are satisfied. (A) When the value of the number of rewrites WCNT of the pointer block 1a becomes an integral multiple of 256. (B) Update counter 31 of a block constituting a file
At the time point when the difference between the lowest value of the average value and the average value of the update counter 31 of the unused block exceeds 256. That is, a new file is created,
After deletion, the number of rewrites WCNT of the pointer block 1a is looked at, and it is just an integer multiple of 256, and 16 bytes of 2 bytes.
When the lower digit of the radix becomes “00 ₁₆ ”, the files are inspected in the order registered in the directory block 30. Then, an update average value is calculated by adding the update counter 31 of the blocks constituting the file and dividing by the number of blocks constituting the file. Next, the same update average value is issued for the next file and compared, and the lower one is left as a comparison control, and compared with the update average value of the next file. Find the file that Then, the average of the update counts of the unused blocks is calculated, and the difference from the file having the lowest average update count is calculated. Next, a correction processing operation in the information processing apparatus according to the present invention will be described with reference to FIG.
FIG. 7 is a schematic diagram illustrating a correction processing operation in the information processing apparatus according to the present invention. In these figures, a file 41 is composed of blocks BLOCK5 to BLOCK7, and has the lowest average number of updates. An unused block group 42 is indicated by an unused start block number OSB of the pointer block 1a. The unused block group 42 includes blocks BLOCK50, 10, 11, 1
8, 55, 80, and 81 have one chain structure. First, in order to connect the start block of the file 41 to the subsequent stage of the block BLOCK 81 of the unused block group 42, as shown in FIG.
The contents of the chain block area 35 of the LOCK 81 are changed from “FF ₁₆ ” to the contents of the chain block area 35 of “05 ₁₆ ”, that is, the block BLOCK 5 of the file 41 as shown in FIG. Blocks BLOCK 5 to 7 (until the contents of the chain block area 35 of the file 41 become “FF ₁₆ ”) are converted to blocks BLOCK 81 of the unused block group 42.
The start pointer of the file 41 of the directory block 30 is changed to the block BLOCK50, and the end pointer is changed to the block BLOCK11, as shown in FIG. Next, the unused start block number OSB of the unused block group 42
To “12 ₁₆ ” and the unused end block number OEB to “0B ₁₆ ”. As a result, it becomes possible to use again the blocks constituting the registered file, and the EE
The entire block of the PROM 1 is used and updated on average. [0024] FIG 8 is a flowchart illustrating an example of a data rewriting control procedure to EEPROM1 in the information processing apparatus according to the present invention. Note that (1) to (1)
8) shows each step. First, an empty area of the directory block 30 is searched, and a new file name is written (1). Next, the unused start block number OS
B is stored in the accumulator ACC of the CPU 11 (2). The number of times of rewriting WCNT of the block designated by the accumulator ACC is updated by "+1" (3). here,
It is determined whether the number of rewrites WCNT has exceeded "10000" (4). If YES, the continuation block area CB of the block indicated by the accumulator ACC is stored in the accumulator ACC (5), and the process returns to step (3).
If NO, the contents of the accumulator ACC are written into the start block number area (SB) 33 of the directory block 30 (6). Next, data is written to the data area of the block designated by the accumulator ACC (7). Here, it is determined whether the write data exceeds 2 35 bytes (8), YES if the accumulator ACC is to store the continuation block area CB of the block that instructs the accumulator BCC (9). Next, the number of times W of rewriting the block designated by the accumulator BCC is W
The CNT is updated by "+1" (10). Next, it is determined whether or not the number of rewrites WCNT has exceeded "10000" (11). If YES, the continuation block area CB of the block designated by the accumulator BCC is stored in the accumulator BCC (12), and the process returns to step (10). ,
If NO, the contents of the accumulator BCC are written into the continuation block area CB of the block designated by the accumulator ACC (13), and the process returns to step (7). On the other hand, if the determination in step (8) is NO, the continuation block area CB specified by the accumulator ACC is written to the unused start block number OSB (14). Next, the number of times W of rewriting of the pointer block 1a is W
The CNT is updated by "+1" (15). Next, the continuation block area CB of the block designated by the accumulator ACC
"FF ₁₆ " is written to (16). Then, the contents of the accumulator ACC are written to the end block number area 34 at the new file position of the directory block 30 (17).
Next, the number of rewrites WCN of the directory block 30
T is updated (18). FIG. 9 is a flowchart showing an example of a correction control processing procedure in the information processing apparatus according to the present invention.
Note that (1) to (7) indicate each step. First, it is determined whether or not the number of rewrites WCNT of the pointer block 1a is an integral multiple of "256" (1).
If S, the average value of the update counter 31 of the blocks constituting each file registered in the directory block 30 is calculated, and the file with the least number of updates is found.
(2). Next, the average value of the update count of the unused block is calculated (3). Next, the minimum value of the average value of the update counters of the blocks constituting the file is subtracted from the average value of the update counters of the unused blocks, and “25” is obtained from the subtraction value.
It is determined whether the value obtained by subtracting "6" is positive (4). If NO, the process returns. If YES, the head of the file corresponding to the end of the unused block is connected (5). Next, the contents of the connected file are transferred to an unused block (6), and the start pointer and end pointer of the connected file in the directory block 30 are changed (7).
And return. As described above, according to the present invention,
Correction means based on the number of rewrites of each area determined by the judgment means for judging the number of rewrites for each area stored in the storage means for storing the number of rewrites for each area of the memory divided into a plurality of areas But the number of rewrites is small
Rewrite the memory so that the
, The frequency of writing to each area can be dispersed, and the writing life of the memory can be greatly extended. Therefore, there is an effect that data can be prevented from being altered or lost due to data rewriting on a memory means such as a programmable read only memory.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a write control of a memory showing one embodiment of the present invention.
1 is a block diagram illustrating a configuration of an information processing device to which the device can be applied . FIG. 2 is a schematic diagram illustrating a memory map of the EEPROM shown in FIG. FIG. 3 is a schematic diagram showing a storage area of the EEPROM shown in FIG. FIG. 4 is a schematic diagram illustrating the structure of each directory block shown in FIG. FIG. 5 is a schematic diagram illustrating a state of an unused EEPROM. FIG. 6 is a schematic diagram illustrating a write operation to the EEPROM shown in FIG. 1; FIG. 7 is a schematic diagram illustrating a correction processing operation in the information processing apparatus according to the invention. FIG. 8 shows an EEPRO in the information processing apparatus according to the present invention.
It is a flowchart which shows an example of the data rewriting control processing procedure to M. FIG. 9 is a flowchart illustrating an example of a correction control processing procedure in the information processing apparatus according to the present invention. [Description of Signs] 1 EEPROM 1a Pointer block 21 Block number 30 Directory block 31 Update counter 32 File area 33 Start block number area 34 End block number area 35 Chain block area 41 File 42 Unused block group

Claims (1)

(57) [Claims] A memory write control device capable of electrically rewriting information written in a memory, comprising: storage means for storing the number of rewrites for each area of a memory divided into a plurality of areas; Determining means for determining the number of times of rewriting for each area; and, based on the number of times of rewriting for each area determined by the determining means , giving priority to an area having a small number of times of rewriting so as to reduce the difference in the number of times of rewriting for each area. To use
And a correction means for controlling rewriting of the memory.