JP4952741B2 - MEMORY CONTROLLER, FLASH MEMORY SYSTEM HAVING MEMORY CONTROLLER, AND FLASH MEMORY CONTROL METHOD - Google Patents

Description

  The present invention relates to a memory controller, a flash memory system including the memory controller, and a flash memory control method.

  For example, as disclosed in Patent Document 1, in a memory system using a flash memory, wear leveling (Wear-leveling) is performed so that there is no bias in the number of erasures in a plurality of physical blocks constituting the flash memory. Control is taking place. There are roughly two types of wear leveling control.

  One is a hardware leveling control (dynamic level control) in which a physical block in which data that is not rewritten is stored is left and the number of times of erasure or rewrite of other physical blocks is averaged. This is a control method called a wear-leveling (Dynamic-wear-leveling) control method or a passive wear-leveling control method (hereinafter referred to as dynamic wear leveling). The other is wear leveling control (static wear control) that controls the number of erasures or rewrites of all physical blocks, including physical blocks that store data that will not be rewritten. This is a control method called a static-wear-leveling control method or an active wear-leveling control method (hereinafter referred to as static wear leveling).

  Here, when there are few physical blocks in which data that is not rewritten is stored, the dynamic wear leveling control method is suitable. However, when there are many physical blocks in which data that is not rewritten is stored, the static wear leveling control method is suitable. Therefore, Patent Document 1 discloses dynamic wear according to the frequency of rewriting data stored in the memory system (that is, depending on whether there are many or few physical blocks storing data that is not rewritten). It is described that the wear leveling control is performed among the leveling control method, the static wear leveling control method, or the wear leveling control method combining these.

JP 2007-133683 A

  In Patent Document 1 described above, a user selects a wear leveling control method and sets conditions. However, it is not always easy to select an appropriate wear leveling control method according to the frequency of rewriting data stored in the memory system.

  If the wear leveling control method is not properly selected, the physical wear level control method is not selected, and the physical block with an extremely small number of erasures is left without being subjected to wear leveling control. Or because the static wear leveling control method is selected, unnecessary data transfer may occur between physical blocks. Also, in the static wear leveling control system, if the number of times of erasing each physical block is controlled to increase on average, the data transfer for averaging the number of times of erasure increases, and as a result, the flash The number of erasures as a whole memory increases. Such an increase in the number of erases due to data transfer between physical blocks accelerates the deterioration of the flash memory.

  Therefore, the present invention can reduce the deterioration rate of the flash memory by suppressing an extreme bias in the number of erases in a plurality of physical blocks constituting the flash memory and an increase in the number of erases as the entire flash memory. An object of the present invention is to provide a memory controller, a flash memory system, and a flash memory control method.

A memory controller according to a first aspect of the present invention is a memory controller that controls access to a flash memory to be erased in units of physical blocks based on an access instruction given from a host system,
An erase count management means for managing the erase count of each physical block constituting the flash memory;
Block management means for managing a plurality of sector areas as a logical block by collecting a plurality of sector unit areas to which a logical address given as the access instruction is assigned, and allocating a physical block constituting a flash memory to the logical block; ,
A free block table for searching for a free block that is a physical block in which no valid data is stored is provided for each group of physical blocks defined on the basis of one or more thresholds provided for the number of physical block erasures. A free block search means for searching for the free block belonging to each group using the free block table;
Each group has an earliest block table for searching for the earliest block that is the earliest physical block assigned to the logical block among the physical blocks assigned to the logical block. Earliest block search means for searching for the earliest block belonging to each of the groups using an earliest block table;
Table registration means for registering a physical block assigned to a predetermined logical block in the earliest block table;
Information holding means for holding information for specifying the predetermined logical block;
Data storage means for specifying the logical block to which the area to be accessed belongs based on the access instruction, and storing data given from the host system in the physical block assigned to the specified logical block;
Data transfer means for transferring the data stored in the earliest block detected by the earliest block search means to the empty block detected by the empty block search means;
When a physical block is allocated to the logical block by the block management means in response to the access instruction, the free block search means assigns the group on the lower side, which is the group to which the physical block with a small number of erasures belongs. Search for the free block using the corresponding free block table preferentially,
When a physical block belonging to the group lower than the group to which the free block detected by the free block search means belongs is not registered in the earliest block table, the block management means performs the free block search. Assigning the free block detected by the means to the logical block;
When the physical block belonging to the group lower than the group to which the empty block detected by the empty block searching means belongs is registered in the earliest block table, the earliest block searching means The earliest block belonging to the group lower than the group to which the block belongs is searched, and the data transfer means stores the data stored in the earliest block detected by the earliest block search means as the empty block The block management means assigns the earliest block detected by the earliest block search means to the logical block.

A memory controller according to a second aspect of the present invention is a memory controller that controls access to a flash memory to be erased in units of physical blocks based on an access instruction given from a host system,
Virtual block forming means for forming a plurality of virtual blocks obtained by collecting a plurality of physical blocks constituting the flash memory;
Erasure count management means for managing the erasure count in units of virtual blocks;
A block management means for managing a plurality of sector areas as a logical block by collecting a plurality of sector unit areas to which a logical address given as the access instruction is assigned, and assigning the virtual block to the logical block;
An empty block table for searching for an empty block which is the virtual block in which no valid data is stored is stored in the virtual block defined based on one or more thresholds provided for the number of times of erasing the virtual block. A free block search means for searching for the free blocks belonging to each of the groups using each free block table;
Each group has an earliest block table for searching for the earliest block that is the virtual block assigned to the logical block first among the virtual blocks assigned to the logical block, The earliest block search means for searching for the earliest block belonging to each of the groups using the earliest block table of
Table registration means for registering the virtual block assigned to the predetermined logical block in the earliest block table;
Information holding means for holding information for specifying the predetermined logical block;
Data storage means for specifying the logical block to which the area to be accessed belongs based on the access instruction, and storing data given from a host system in the virtual block assigned to the specified logical block;
Data transfer means for transferring the data stored in the earliest block detected by the earliest block search means to the empty block detected by the empty block search means;
When the virtual block is allocated to the logical block by the block management means in response to the access instruction, the empty block search means is the group on the lower side that is the group to which the virtual block to which the erasure count is small belongs. Search for the free block using the free block table corresponding to the group preferentially,
When the virtual block belonging to the group lower than the group to which the free block detected by the free block searching means belongs is not registered in the earliest block table, the block management means Assigning the free block detected by the search means to the logical block;
When the virtual block belonging to the group lower than the group to which the empty block detected by the empty block searching means belongs is registered in the earliest block table, the earliest block searching means The earliest block belonging to the group lower than the group to which the empty block belongs is searched, and the data transfer means stores the data stored in the earliest block detected by the earliest block search means. Transfer is made to the empty block detected by the empty block search means, and the block management means assigns the earliest block detected by the earliest block search means to the logical block.

  In the memory controller according to the third aspect of the present invention, in the first or second aspect, information for specifying the predetermined logical block is set based on information given from a host system.

  A flash memory system according to a fourth aspect of the present invention includes a memory controller according to any one of the first to third aspects, and a plurality of flash memories controlled by the memory controller.

A method according to a fifth aspect of the present invention is a flash memory control method for controlling access to a flash memory to be erased in units of physical blocks based on an access instruction given from a host system,
An erase count management step for managing the erase count of each physical block constituting the flash memory;
A block management step of managing a plurality of sector areas obtained by collecting a plurality of sector unit areas to which a logical address given as an access instruction is assigned as a logical block, and allocating a physical block constituting a flash memory to the logical block; ,
A free block table for searching for a free block that is a physical block in which no valid data is stored is provided for each group of physical blocks defined on the basis of one or more thresholds provided for the number of physical block erasures. A free block search step of searching for the free block belonging to each of the groups using each of the free block tables;
Each group has an earliest block table for searching for the earliest block that is the earliest physical block assigned to the logical block among the physical blocks assigned to the logical block. An earliest block search step of searching for the earliest block belonging to each of the groups using an earliest block table;
A table registration step of registering a physical block assigned to a predetermined logical block in the earliest block table;
A data storage step of identifying the logical block to which the area to be accessed belongs based on the access instruction, and storing data provided from a host system in a physical block allocated to the identified logical block;
A data transfer step of transferring data stored in the earliest block detected in the earliest block search step to the empty block detected in the empty block search step;
When a physical block is allocated to the logical block by the block management step in response to the access instruction, the free block search step adds the low-order group to the group to which the physical block with a small number of erases belongs. A search for the free block that is preferentially used in the corresponding free block table is performed,
When a physical block belonging to the group lower than the group to which the free block detected by the free block search step belongs is not registered in the earliest block table, the free block search is performed in the block management step. The free block detected by the step is assigned to the logical block;
When a physical block belonging to the group lower than the group to which the empty block detected by the empty block searching step belongs is registered in the earliest block table, the earliest block searching step includes the empty block. The earliest block belonging to the group lower than the group to which the block belongs is searched, and in the data transfer step, the data stored in the earliest block detected by the earliest block search step is free. It is transferred to the empty block detected by the block search step, and the earliest block detected by the earliest block search step is assigned to the logical block in the block management step.

A method according to a sixth aspect of the present invention is a flash memory control method for controlling access to a flash memory that is erased in units of physical blocks based on an access instruction given from a host system,
A virtual block forming step of forming a plurality of virtual blocks obtained by collecting a plurality of physical blocks constituting a flash memory; and an erasure count management step of managing the erase count in units of the virtual blocks;
A block management step of managing a plurality of sector areas obtained by collecting a plurality of sector unit areas to which a logical address given as the access instruction is assigned as a logical block, and assigning the virtual block to the logical block;
An empty block table for searching for an empty block which is the virtual block in which no valid data is stored is stored in the virtual block defined based on one or more thresholds provided for the number of times of erasing the virtual block. A free block search step for searching for the free block belonging to each group using each free block table, which is provided for each group;
Each group has an earliest block table for searching for the earliest block that is the virtual block assigned to the logical block first among the virtual blocks assigned to the logical block, The earliest block search step of searching for the earliest block belonging to each of the groups using the earliest block table of
A table registration step of registering the virtual block assigned to the predetermined logical block in the earliest block table;
A data storage step of identifying the logical block to which the area to be accessed belongs based on the access instruction, and storing data given from a host system in the virtual block allocated to the identified logical block;
A data transfer step of transferring data stored in the earliest block detected in the earliest block search step to the empty block detected in the empty block search step;
When the virtual block is allocated to the logical block by the block management step in response to the access instruction, the lower block that is the group to which the virtual block to which the erasure count is small belongs in the empty block search step belongs to The search for the free block that is preferentially used in the free block table corresponding to the group is performed,
When the virtual block belonging to the group lower than the group to which the free block detected by the free block searching step belongs is not registered in the earliest block table, the free block is used in the block management step. The free block detected by the search step is assigned to the logical block;
When the virtual block belonging to the group lower than the group to which the empty block detected by the empty block searching step belongs is registered in the earliest block table, in the earliest block searching step, The earliest block belonging to the group lower than the group to which the empty block belongs is searched, and the data stored in the earliest block detected by the earliest block search step is the data transfer step. The block is transferred to the empty block detected in the empty block search step, and the earliest block detected in the earliest block search step is assigned to the logical block in the block management step.

  In the method according to the seventh aspect of the present invention, in the fifth or sixth aspect, the predetermined logical block is set based on information given from a host system.

  According to the present invention, one or a plurality of threshold values are provided for the number of times of erasing a physical block, and control is performed such that the number of times of erasing a physical block whose number of erasures exceeds the threshold is suppressed. In this control, data transfer is performed between physical blocks under a predetermined condition. As a result, the extreme number of erase times in a plurality of physical blocks constituting the flash memory is suppressed, and the entire flash memory is controlled. As a result, an increase in the number of erasures is suppressed.

  In this data transfer, a physical block that is a transfer source is limited to a physical block assigned to a specific logical block. In this way, only data corresponding to a specific logical block is targeted for data transfer, and data corresponding to logical blocks other than the specific logical block is excluded from data transfer targets.

1 is a block diagram showing a schematic configuration of a flash memory system according to an embodiment of the present invention. It is a figure which shows the correspondence of a logical block and a physical block. It is explanatory drawing which shows the priority management in a earliest block table. It is explanatory drawing regarding registration to the earliest block table. It is explanatory drawing which shows the structure of an empty block table. It is explanatory drawing regarding the registration to an empty block table. It is explanatory drawing of the table used for the priority management for the management of the frequency | count of erasure | elimination of a physical block, and prior management.

  FIG. 1 is a block diagram schematically showing a flash memory system 1 according to the present embodiment.

  As shown in FIG. 1, the flash memory system 1 includes a flash memory 2 and a memory controller 3 that controls the flash memory 2.

  The flash memory system 1 is connected to the host system 4 via the external bus 13. The host system 4 is composed of a CPU (Central Processing Unit) for controlling the entire operation of the host system 4, a companion chip for transferring information to and from the flash memory system 1, and the like. The host system 4 may be various information processing apparatuses such as a personal computer and a digital still camera that process various types of information such as characters, sounds, and image information.

  As shown in FIG. 1, the memory controller 3 includes a microprocessor 6, a host interface block 7, a work area 8, a buffer 9, a flash memory interface block 10, an ECC (Error Correcting Code) block 11, and a ROM. (Read Only Memory) 12. The memory controller 3 is connected to the flash memory 2 via the internal bus 14. The memory controller 3 constituted by these functional blocks is integrated on one semiconductor chip. Hereinafter, each functional block will be described.

  The host interface block 7 exchanges data, address information, status information, external commands and the like with the host system 4. The external command is a command for the host system 4 to instruct the flash memory system 1 to execute processing. Data or the like supplied from the host system 4 to the flash memory system 1 is taken into the flash memory system 1 (for example, the buffer 9) via the host interface block 7. Data supplied from the flash memory system 1 to the host system 4 is supplied to the host system 4 via the host interface block 7.

  The host interface block 7 includes a command register R1, a sector number register R2, and an LBA register R3. Information given from the host system 4 is written in the command register R1, the sector number register R2, and the LBA register R3. External commands such as a write command and a read command are written in the command register R1. The number of sectors in the access target area is written in the sector number register R2. In the LBA register R3, an LBA (Logical Block Address) (described later) of the head sector of the access target area is written.

  The work area 8 is a work area for temporarily storing data necessary for controlling the flash memory 2, and is composed of a plurality of SRAM (Static Random Access Memory) cells. The work area 8 stores, for example, an address conversion table indicating the correspondence between logical blocks and physical blocks.

  The buffer 9 holds the data read from the flash memory 2 until the host system 4 can receive the data. The buffer 9 holds data to be written to the flash memory 2 until the flash memory 2 is in a writable state.

  The flash memory interface block 10 exchanges data, address information, status information, internal commands, and the like with the flash memory 2 via the internal bus 14. Here, the internal command is a command for the memory controller 3 to instruct the flash memory 2 to execute processing, and the flash memory 2 operates in accordance with the internal command given from the memory controller 3.

  The ECC block 11 generates an error correcting code (ECC) added to data to be written to the flash memory 2 and is included in the read data based on the error correcting code added to the read data. Detect and correct errors.

  The ROM 12 is a non-volatile storage element that stores a program that defines a processing procedure performed by the microprocessor 6. For example, a program that defines a processing procedure such as creation of an address conversion table is stored.

  The microprocessor 6 controls the overall operation of the memory controller 3 in accordance with a program stored in the ROM 12. For example, the microprocessor 6 causes the flash memory interface block 10 to execute processes based on a command set that defines various processes read from the ROM 12.

  The flash memory 2 is a NAND flash memory. The NAND flash memory includes a register and a memory cell array in which a plurality of memory cells are two-dimensionally arranged. The memory cell array includes a plurality of memory cell groups and word lines. Here, the memory cell group is a group in which a plurality of memory cells are connected in series. Each word line is for selecting a specific memory cell in the memory cell group. Data is written from the register to the selected memory cell or data is read from the selected memory cell to the register between the selected memory cell and the register via the word line.

  In the NAND flash memory, a data read operation and a data write operation are performed in units of pages, and a data erase operation is performed in units of blocks (physical blocks). A physical block is composed of a plurality of pages (physical pages). For example, one physical page includes a user area of a predetermined size (for example, 2048 bytes) and a redundant area of a predetermined size (for example, 64 bytes), and one physical block has a predetermined number (for example, 64). It consists of physical pages. The user area is an area for storing data given from the host system 4 and is normally divided into 512-byte storage areas (hereinafter referred to as “logical sector areas”) and managed. The redundant area is an area for storing additional data such as an error correcting code (ECC), logical address information, and a block status (flag).

  The logical address information is information for determining the correspondence between physical blocks and logical blocks. The block status (flag) is a flag indicating pass / fail of the physical block. For the initial defective physical block, a block status (flag) indicating a defective block (physical block in which data cannot be normally written) is written by the manufacturer. Some manufacturers also write a block status (flag) indicating the initial defective physical block in the user area.

  In the writing process of the present embodiment, the host system 4 writes a command code indicating a write command in the command register R1, writes the number of sectors of data to be written in the sector number register R2, and stores in the LBA register R3 The LBA corresponding to the top data to start writing is written. Based on the information written in the sector number register R2 and the LBA register R3, a logical access area that is an access target area is determined and given to the physical block corresponding to the logical block including the logical access area from the host system 4. Data is written.

  When the logical access area determined based on the information written in the sector number register R2 and the LBA register R3 extends over a plurality of logical blocks, the data is obtained by dividing the area for each logical block to which the logical access area belongs. Is written. For example, when the logical access area extends over two logical blocks (the first logical block and the second logical block), the write processing for the physical block corresponding to the first logical block and the second logical block Write processing is performed on the physical block corresponding to the block. Data instructed to be written to the logical access area belonging to the first logical block is written to the physical block corresponding to the first logical block. Data instructed to be written to the logical access area belonging to the second logical block is written to the physical block corresponding to the second logical block.

  The address space on the host system 4 side is managed by an LBA (Logical Block Address) which is a serial number assigned to an area (hereinafter referred to as “logical sector area”) divided in units of sectors (512 bytes). Also, a logical block composed of a plurality of logical sector areas is formed, and one or a plurality of physical blocks are assigned to the logical block.

  The correspondence between logical blocks and physical blocks is often managed on a zone basis. Specifically, one logical zone is composed of a plurality of logical blocks, and address conversion table creation processing and update processing indicating the correspondence between logical blocks and physical blocks are performed in units of logical zones.

  The address conversion table creation process and update process are normally performed on the work area 8. The address conversion table is generally created based on logical address information stored in the redundant area of each physical block at the time of start-up or access, but is stored in the flash memory 2 and flash memory when necessary. 2 may be read out. By doing so, it is possible to obtain the address conversion table in a shorter time than when the logical address information is read from the redundant area of each physical block and the address conversion table is created.

  When the address conversion table is stored in the flash memory 2, it is preferable that the latest address conversion table is always stored in the flash memory 2 in order to cope with a sudden power shutdown. That is, it is preferable that the latest address translation table is stored in the flash memory 2 every time the correspondence between the logical block and the physical block changes (that is, every time the address translation table is updated on the work area 8). . This address conversion table storage process is also performed in units of logical zones, similar to the creation process and update process.

  An example of the correspondence between logical blocks and physical blocks will be described with reference to FIG.

  In the example shown in FIG. 2, 2048000 logical sector areas composed of LBA # 0 to # 20479999 are allocated to storage areas composed of 8192 physical blocks constituting the flash memory 2.

  Each physical block is assigned to a logical block composed of a plurality of logical sector areas. Hereinafter, the serial number assigned to the logical block is referred to as “logical block number (LBN)”. In this example, a logical block is composed of 256 logical sector areas in which LBAs, which are logical addresses in units of sectors, are continuous. That is, the logical block of LBN # 0 is composed of 256 logical sector areas of LBA # 0- # 255, and the logical block of LBN # 1 is composed of 256 logical sector areas of LBA # 256- # 511. Yes. In this way, 8000 logical blocks LBN # 0- # 7999 are configured by 2048000 logical sector areas of LBA # 0- # 20479999.

  Further, a plurality of logical blocks are collected as a logical zone. Hereinafter, the serial number assigned to this logical zone is referred to as “logical zone number (LZN)”. In this example, 8000 logical blocks of LBN # 0- # 7999 are allocated 1000 by 8 to the eight logical zones of LZN # 0- # 7. That is, the logical blocks of LBN # 0- # 999 are in the logical zone of LZN # 0, the logical blocks of LBN # 1000- # 1999 are in the logical zone of LZN # 1, and the logical blocks of LBN # 2000- # 2999 are in LZN #. Similarly, the logical blocks of LBN # 7000- # 7999 are assigned to the logical zone of LZN # 7 in the second logical zone.

  A unique physical block address (PBA) is assigned to 8192 physical blocks constituting the flash memory 2. The physical blocks constituting the flash memory 2 include a congenital defective block that cannot be normally written or erased from the time of shipment, and an acquired block that has become defective due to repeated writing or erasing. There are bad blocks. In general, the number of acquired defective blocks included in the physical block constituting the flash memory 2 is significantly increased when the number of guaranteed erases specified by the manufacturer is exceeded.

  Also, commonly used NAND flash memories are composed of SLC (single level cell) type memory cells and MLC (multi level cell) type memory cells. There is. A flash memory composed of MLC type memory cells generally has a shorter data retention period than a flash memory composed of SLC type memory cells, and the data retention period is significantly shorter when a predetermined number of erasures is exceeded. Become. Therefore, in a flash memory composed of MLC type memory cells, even if a physical block has not reached the guaranteed erase count, the reliability of the data stored in the physical block exceeds the predetermined erase count. descend.

  Data corresponding to each logical block is stored in a physical block assigned to the logical block. The correspondence between the logical block and the physical block is managed by the address conversion table described above. The physical block to be allocated to the logical block is searched using a free block search table that is a table for searching for a physical block in which valid data is not stored.

  When updating the data stored in the flash memory, another physical block that does not contain valid data is newly allocated to the same logical block as the physical block in which the old data is stored. New data is stored in. When all the old data is replaced with new data, the physical block in which the old data is stored becomes a physical block in which valid data is not stored (hereinafter referred to as “invalid block”).

  That is, when all data in a physical block storing data corresponding to any logical block is replaced by data stored in another physical block, the physical block becomes an invalid block. Note that invalid block erasure processing is performed even when all valid data is replaced by data stored in another physical block (that is, when a physical block becomes an invalid block). Or just before the invalid block is assigned to a logical block.

  In the data stored in the flash memory, the physical block of the storage destination usually changes every time it is updated. The transition of the physical block at the storage destination increases for data with a high update frequency, and decreases for data with a low update frequency. Accordingly, the number of times of erasing each physical block is biased due to the difference in the update frequency of the stored data. If this bias in the number of erasures progresses excessively, some physical blocks with a large number of erasures become defective blocks earlier than other physical blocks, or the data retention period becomes significantly shorter than other physical blocks. It is not preferable.

  In order to avoid such a problem, the memory controller 3 according to an embodiment of the present invention sets a threshold value for the number of erase times, and suppresses an increase in the number of erase times of a physical block that has reached the threshold value. Control is in progress. Next, this control method will be described.

  In the present embodiment, two threshold values are provided for the number of times of erasure, and based on these threshold values, a plurality of physical blocks constituting the flash memory 2 are divided into three groups. Note that the number of threshold values to be set is set as appropriate. For example, it may be divided into four groups by three thresholds, or may be divided into five groups by four thresholds (that is, a plurality of physical blocks constituting the flash memory 2). Is divided into (N + 1) groups if there are N thresholds). Further, the threshold value (the number of erasures serving as the threshold) is appropriately set according to the characteristics of the flash memory.

  In the present embodiment, the first threshold value m is set to a value 1000 smaller than the second threshold value n, and the second threshold value n is set so as to coincide with the number of erasures in which the decrease in the data holding period becomes significant. Has been. For example, in the case of a flash memory in which the decrease in the data retention period becomes significant when the number of erases exceeds 5000, the first threshold value m is set to 4000 and the second threshold value n is set to 5000. The memory controller 3 divides a plurality of physical blocks constituting the flash memory 2 into three groups based on the two threshold values. That is, the plurality of physical blocks constituting the flash memory 2 include a first group to which a physical block having an erase count of less than m and a second group to which a physical block having an erase count of m to less than n belongs. And the third group to which the physical block having the erase count of n times or more belongs.

  In this embodiment, the memory controller 3 manages the order in which the physical blocks are assigned to the logical blocks. Hereinafter, this management is referred to as “advanced management”.

  Next, this later management will be described with reference to FIGS. This prior management is performed for each group, and the subsequent management of physical blocks belonging to the first group is performed based on the first earliest block table shown in FIG. The prior management of the physical blocks belonging to the group is performed based on the second earliest block table shown in FIG. 3B, and the prior management of the physical blocks belonging to the third group is shown in FIG. This is performed based on the third earliest block table shown.

  In this prior management, the following priority management, that is, when a physical block is newly assigned to a logical block, the priority of the newly assigned physical block is the highest priority. Management is done. Therefore, the physical block that has been assigned to the logical block has a lower priority. The physical block with the lowest priority corresponds to the physical block assigned to the logical block first among the physical blocks assigned to the logical block. The memory controller 3 can identify the physical block assigned to the logical block first among the physical blocks assigned to the logical block by performing such priority order management.

  In addition, since this prior management is performed for each group, the physical block with the lowest priority in the physical management belonging to the first group is the first physical block belonging to the first group. Corresponds to the physical block assigned to the logical block. Similarly, the physical block with the lowest priority in the prior management of physical blocks belonging to the second group corresponds to the physical block assigned to the logical block first among the physical blocks belonging to the second group. The physical block having the lowest priority in the subsequent management of the physical block belonging to the third group corresponds to the physical block assigned to the logical block first among the physical blocks belonging to the third group.

  For example, in the first earliest block table shown in FIG. 3A, the priority order of the physical blocks belonging to the first group among the physical blocks assigned to the logical block is managed. In the first earliest block table, the physical block priority of PBN # 36 is the highest and the physical block priority of PBN # 857 is the lowest. Therefore, the physical block of PBN # 857 corresponds to the physical block assigned to the logical block first among the physical blocks belonging to the first group. Here, when the physical block of PBN # 523 belonging to the first group is newly assigned to the logical block, the priority of the physical block of PBN # 523 becomes the highest, and the priority of the physical block of PBN # 36 Will be second. Also, when the physical block of PBN # 234 becomes an invalid block, or when the data stored in the physical block of PBN # 234 is erased, the physical block of PBN # 234 is removed from the priority management target. The That is, the priority order of the physical block of PBN # 995, the physical block of PBN # 234, and the physical block of PBN # 645 becomes the order of the physical block of PBN # 995 and the physical block of PBN # 645.

  In the second earliest block table shown in FIG. 3B, the priority order of the physical blocks belonging to the second group among the physical blocks assigned to the logical block is managed. In the second earliest block table, the physical block priority of PBN # 936 is the highest and the physical block priority of PBN # 975 is the lowest. Therefore, the physical block of PBN # 957 corresponds to the physical block assigned to the logical block first among the physical blocks belonging to the second group. In the third earliest block table shown in FIG. 3C, the priorities of the physical blocks belonging to the third group among the physical blocks assigned to the logical block are managed. In the third earliest block table, the physical block priority of PBN # 977 is the highest, and the physical block priority of PBN # 1111 is the lowest. The second earliest block table and the third earliest block table are updated in the same manner as the update of the first earliest block table. That is, when a physical block newly assigned to a logical block has the highest priority and becomes a physical block that does not store valid data, it is removed from the management target of the priority.

  In the present embodiment, since the prior management is performed for each group, the memory controller 3 becomes the registration destination of the physical block based on the number of erases of the physical block newly assigned to the logical block. Determine the earliest block table. That is, as shown in FIG. 4, when the number of erases of the physical block newly assigned to the logical block is less than m times, the memory controller 3 assigns the physical block to the first group corresponding to the first group. When the physical block erase count is m times or more and less than n times, the physical block is registered in the second earliest block table corresponding to the second group, and the physical block is registered. If the number of erases is n or more, the physical block is registered in the third earliest block table corresponding to the third group.

  In this embodiment, an empty block table for searching for a physical block to be allocated to a logical block (that is, a physical block in which valid data is not stored) is also managed for each group. The management of this empty block table will be described with reference to FIGS.

  The first empty block table shown in FIG. 5A, the second empty block table shown in FIG. 5B, and the third empty block table shown in FIG. These bits correspond to any physical block, and the logical value of each bit indicates whether or not the physical block can be assigned to the logical block. Here, a physical block that can be allocated to a logical block is a physical block that is not a bad block, and a physical block in which stored data is erased or a physical block in which valid data is not stored (invalid Block). Hereinafter, a physical block that is not a bad block and has a storage data erased or a physical block in which valid data is not stored (invalid block) is collectively referred to as an “empty block” (that is, the storage data is Including empty blocks that have not been erased.

  In this embodiment, the logical value of a physical block corresponding to an empty block is “1”, and the logical value of a physical block not corresponding to an empty block is “0”. In the first to third empty block tables, the uppermost 8 bits correspond to the physical blocks of PBA # 0 to # 7 in order from the lower bit side (in order from the right bit), and the second from the top. Correspond to the physical block of PBA # 8- # 15 in order from the lower bit side (in order from the right bit), and the third 8 bits from the top in order from the lower bit side (from the right bit) Corresponding to the physical blocks of PBA # 16- # 24, and in the same manner, the physical blocks correspond to the bits in the order of PBA. Therefore, in the example of the first empty block table shown in FIG. 5A, the physical block of PBA # 8, the physical block of PBA # 15, and the like correspond to the empty blocks, and the first block shown in FIG. In the example of the empty block table 2, the physical block of PBA # 0 corresponds to the empty block.

  Further, when the logical value of the bit corresponding to the physical block that has become an empty block is changed to “1”, only the logical value of the bit on any one of the first to third empty block tables is obtained. It is changed to “1”. The table to which the bit to be changed belongs is selected based on the number of times of erasing the physical block that has become an empty block. That is, as shown in FIG. 6, when the number of times of erasing a physical block that has become a free block is less than m times, the memory controller 3 selects the first free block table corresponding to the first group, When the physical block erase count is m times or more and less than n times, the second empty block table corresponding to the second group is selected, and when the physical block erase count is n times or more, the third group is assigned. A corresponding third free block table is selected.

  The search for empty blocks using the first to third empty block tables is usually performed when a physical block is not allocated to a logical block designated as a data write destination given from the host system 4 or when the logical block is specified. This is performed when there is no free area for writing the data given from the host system 4 in the physical block assigned to. The free block detected by the search for the free block is assigned to the logical block designated as the write destination of the data supplied from the host system 4, and the data supplied from the host system 4 is stored.

  In the present embodiment, the first group, the second group, and the third group are grouped in order from the lower group having a smaller number of erasures to the upper group having a larger number of erasures. When searching for an empty block, the table corresponding to the lower group is preferentially used. That is, when a free block is not detected in the search for a free block using the first free block table corresponding to the first group, the free space using the second free block table corresponding to the second group is used. When a block search is performed and an empty block is not detected in the search for an empty block using the second empty block table corresponding to the second group, the third corresponding to the third group is detected. An empty block search using the empty block table is performed. By doing so, physical blocks belonging to the lower group among the empty blocks are preferentially assigned logical blocks. As a result, an increase in the number of erases of physical blocks belonging to the upper group is suppressed.

  However, if the table corresponding to the lower group is preferentially used in this way, an increase in the number of erases of physical blocks belonging to the upper group may not be appropriately suppressed. For example, if data with a low rewrite frequency is stored in a physical block belonging to the lower group, an increase in the number of erases of the physical block belonging to the upper group is not appropriately suppressed. In such a case, if data with low rewrite frequency is stored in a physical block belonging to the upper group, an increase in the number of erasures of the physical block belonging to the upper group is appropriately suppressed. Therefore, in this embodiment, when a free block is detected by searching for a free block, if there is a physical block belonging to a group lower than the group to which the free block belongs, it belongs to the lower group. The data stored in the physical block assigned to the logical block first among the physical blocks is transferred to the empty block detected by the search for the empty block. After this data transfer, the physical block of the transfer source replaces the empty block detected by the empty block search. That is, the physical block of the transfer source is assigned to the logical block designated as the write destination of the data given from the host system 4, and the data given from the host system 4 is stored. By performing such data transfer, data with a low rewrite frequency is stored in the physical block belonging to the upper group, and an increase in the number of erasures of the physical block belonging to the upper group is effectively suppressed.

  In the present embodiment, the grouping is performed in the order of the first group, the second group, and the third group from the lower group to the upper group. Therefore, when an empty block is detected by searching for an empty block using the second empty block table corresponding to the second group or the third empty block table corresponding to the third group, the empty block is There may be a physical block belonging to a lower group than the group to which it belongs. That is, when an empty block is detected by searching for an empty block using the second empty block table, a physical block belonging to the first group becomes a physical block belonging to a group lower than the group to which the empty block belongs. When a free block is detected by searching for a free block using the third free block table, the physical block belonging to the first group and the second group is lower than the group to which the free block belongs. Corresponds to a physical block belonging to a group.

  For example, when an empty block is detected by searching for an empty block using the second empty block table, the search using the first earliest block table (that is, the most physical block belonging to the first group). The search of the physical block previously assigned to the logical block is performed. Then, it is stored in the physical block detected by the search using the first earliest block table (that is, the physical block assigned to the logical block first among the physical blocks belonging to the first group). Existing data is transferred to a free block detected by a search for a free block using the second free block table (that is, a free block belonging to the second group). After this data transfer is completed, the transfer source physical block (the physical block detected by the search using the first earliest block table) is assigned to the logical block, and is given to the transfer source physical block from the host system 4. Stored data is stored. If there is no physical block registered in the first earliest block table, this data transfer is not performed, and an empty block detected by searching for an empty block using the second empty block table is not found. Used as is. That is, an empty block detected by searching for an empty block using the second empty block table (that is, an empty block belonging to the second group) is allocated to the logical block, and is given to the empty block from the host system 4. Data is stored.

  When an empty block is detected by searching for an empty block using the third empty block table, first, a search using the first earliest block table (that is, among the physical blocks belonging to the first group). The search of the physical block assigned to the logical block first) is performed. Then, it is stored in the physical block detected by the search using the first earliest block table (that is, the physical block assigned to the logical block first among the physical blocks belonging to the first group). Existing data is transferred to a free block detected by a free block search using the third free block table (that is, a free block belonging to the third group). After this data transfer is completed, the transfer source physical block (the physical block detected by the search using the first earliest block table) is assigned to the logical block, and is given to the transfer source physical block from the host system 4. Stored data is stored. If there is no physical block registered in the first earliest block table, a search using the second earliest block table (that is, the earliest physical block belonging to the second group). (Retrieval of physical block assigned to logical block) is performed. And it is stored in the physical block detected by the search using the second earliest block table (that is, the physical block assigned to the logical block first among the physical blocks belonging to the second group). Existing data is transferred to a free block detected by a free block search using the third free block table (that is, a free block belonging to the third group). After this data transfer is completed, the transfer source physical block (physical block detected by the search using the second earliest block table) is allocated to the logical block, and given to the transfer source physical block from the host system 4 Stored data is stored. If there is no physical block registered in the second earliest block table, this data transfer is not performed, and an empty block detected by searching for an empty block using the third empty block table. Is used as is. That is, an empty block (that is, an empty block belonging to the third group) detected by searching for an empty block using the third empty block table is assigned to the logical block, and is given to the empty block from the host system 4. Data is stored.

  Next, a management method of the number of times of erasing the physical blocks constituting the flash memory 2 and a priority management method for the prior management will be described with reference to the example shown in FIG. In the table shown in FIG. 7, the priority order link number and the number of erasures of each physical block are described in the order of PBA. In this table, the PBA of the physical block whose priority is one higher than the physical block and the PBA of the physical block whose priority is one lower than the physical block are written as the priority link number. Therefore, the priority link number is deleted for a physical block that is not assigned to a logical block. For the physical block with the highest priority, the PBA of the physical block one priority higher than the physical block is not written, and for the physical block with the lowest priority, the physical block The PBA of the physical block with the next lower priority is not written.

  By referring to this table, for example, in the case of a physical block of PBA # 0, the physical block whose priority is one higher than that physical block is PBA # 118, and whose priority is one lower than that physical block. It can be seen that the physical block is PBA # 572.

  This table also describes the PBA of the physical block with the highest priority and the physical block with the lowest priority of each group. Therefore, when a physical block is newly assigned to a logical block, the physical block with the highest priority changes, so that the description of the PBA indicating the physical block with the highest priority is changed. Also, when the physical block with the lowest priority becomes a physical block that does not store valid data, or all the data stored in the physical block with the lowest priority is transferred to another block. Since the physical block with the lowest priority changes, the description of the PBA indicating the physical block with the lowest priority is changed.

  For example, when the physical block of PBA # 523 belonging to the first group is newly assigned to the logical block, the PBA indicating the highest priority physical block belonging to the first group is changed from # 36 to # 523. Changed to Also, as the priority link number of the physical block with PBA # 36, # 523, which is the PBA indicating the physical block with the next higher priority, is described. As the priority link number of the physical block whose PBA is # 523, # 36 which is the PBA indicating the physical block whose priority is one lower is described.

  The number of times of erasing each physical block is updated every time data stored in each physical block is erased. For example, when the data stored in the physical block with PBA # 0 is deleted, the number of times of deletion described in the table is changed from 1000 to 1001, and the priority link number described in the table is Erased. In addition, since the physical block whose priority is one lower than the physical block of PBA # 118 becomes the physical block of PBA # 572, the priority link number of the physical block of PBA # 118 and the physical block of PBA # 572 is updated. Is done. That is, for the priority link number of the physical block of PBA # 118, the PBA indicating the physical block with the next lower priority is changed from # 0 to # 572. Also, for the priority order link number of the physical block of PBA # 572, the PBA indicating the physical block with the next higher priority is changed from # 0 to # 118.

  In the above-described embodiment, the post-management is performed for all physical blocks assigned to the logical block. However, the physical block to be pre-managed is assigned to the physical block assigned to the specific logical block. It may be limited. For example, the physical block that is the target of the previous management may be limited by the range of the logical block number (LBN) of the logical block assigned to each physical block.

  Further, the setting of the target of the later management may be performed by an instruction from the host system 4. For example, the host system 4 may be provided with an LBA range (hereinafter referred to as a target LBA range) that is a target of prior management or a non-target LBA range (hereinafter referred to as a non-target LBA range). Good.

  When such a setting of the target of the prior management is performed, information regarding the setting is preferably stored in the flash memory 2 and read and held by the memory controller 3 at the time of activation. For example, the memory controller 3 stores information indicating the target LBA range or the non-target LBA range instructed from the host system 4 in a storage area (for example, a work memory 8 or a register in the microprocessor 6) included in the memory controller 3. Hold. Then, the memory controller 3 identifies a logical block (hereinafter referred to as a target logical block) that is subject to prior management based on the information, and performs prior management for only the physical block assigned to the target logical block. The physical block allocated to the logical block that is not subject to the prior management (hereinafter referred to as the non-target logical block) is not subject to the prior management.

  That is, only the physical block assigned to the target logical block is registered in the earliest block table, and the physical block assigned to the non-target logical block is not registered in the earliest block table. Therefore, only the physical block assigned to the target logical block becomes the transfer source physical block in the data transfer from the physical block detected by the search using the earliest block table to the empty block (hereinafter referred to as copy transfer). Therefore, the physical block assigned to the non-target logical block does not become the transfer source physical block in this copy transfer.

  Therefore, the host system 4 can exclude data that is not preferably subject to data transfer between physical blocks by instructing a non-target LBA range. In other words, the physical block that stores data that is not preferably subject to data transfer between physical blocks is set so that it does not become the physical block that is the transfer source in copy transfer, according to the instruction of the non-target LBA range. The

  When the target LBA range is instructed from the host system 4, logical blocks included in the LBA range are subject to prior management, but logical blocks not included in the LBA range are not subject to prior management. When a non-target LBA range is instructed from the host system 4, logical blocks included in the LBA range are not subject to prior management, but logical blocks not included in the LBA range are subject to prior management. It should be noted that whether or not a logical block whose part is included in the target LBA range and the other part is included in the non-target LBA range is subject to the previous management can be appropriately set. In other words, whether or not the logical block corresponding to the LBA range including both the target LBA range and the non-target LBA range is to be subject to prior management can be set as appropriate. For example, depending on which ratio of the target LBA range and the non-target LBA range among the LBA ranges corresponding to the logical block is larger, the memory controller 3 automatically determines whether or not the logical block is the target of the previous management. May be. Specifically, for example, a logical block having a target LBA range larger than the non-target LBA range is a target of prior management, and a logical block having a non-target LBA range larger than the target LBA range is not a target of prior management. .

  Although the embodiment of the present invention has been described above, this is an example for explaining the present invention, and the scope of the present invention is not limited to this embodiment. Of course, various modifications can be made without departing from the scope of the present invention.

For example, in the above-described embodiment, management of the number of physical block erasures, management of the previous block, and management of the empty block table are performed in units of physical blocks. However, a virtual block composed of a plurality of physical blocks for a logical block When assigning blocks, these managements may be performed in units of virtual blocks. That is, when the physical blocks constituting each virtual block are determined by the initial setting or the like, the number of erasures of the physical blocks belonging to the same virtual block is the same. Further, since allocation to logical blocks and erasure of data are performed in units of virtual blocks, post-management and free block table management can also be performed in units of virtual blocks. When the prior management is performed in units of virtual blocks, the above-described priorities may be managed in units of virtual blocks (that is, management is performed using a serial number assigned to the priority link number for the virtual block). When managing the free block table in units of virtual blocks, a virtual block may be assigned to each bit on the table (that is, one virtual block is assigned to one bit).

  In addition, when an empty block is detected by searching for an empty block using the third empty block table, it is registered in both the first earliest block table and the second earliest block table. When the physical block does not exist, a search using the third earliest block table (that is, the search of the physical block assigned to the logical block first among the physical blocks belonging to the third group) is performed. You may do it. Then, the number of erasures of the physical block detected by the search using the third earliest block table is a predetermined number of times than the number of erasures of the empty block detected by the search of the empty block using the third empty block table. If the number is less than the above, the data stored in the physical block detected by the search using the third earliest block table is the empty block detected by the search for the empty block using the third empty block table. You may make it forward to. By doing so, an increase in the difference in the number of erases between physical blocks belonging to the third group is suppressed.

  Note that methods other than the above-described method may be used for the management of the number of physical block erases, the prior management, and the management of the free block table. For example, if the physical block assigned to the logical block first can be specified, the management based on the above-mentioned priority order does not have to be used for the prior management.

DESCRIPTION OF SYMBOLS 1 ... Flash memory system, 2 ... Flash memory, 3 ... Memory controller, 6 ... Microprocessor

Claims (5)

A memory controller that controls access to a flash memory that is erased in units of physical blocks based on an access instruction given from a host system,
An erase count management means for managing the erase count of each physical block constituting the flash memory;
Block management means for managing a plurality of sector areas as a logical block by collecting a plurality of sector unit areas to which a logical address given as the access instruction is assigned, and allocating a physical block constituting a flash memory to the logical block; ,
A free block table for searching for a free block that is a physical block in which no valid data is stored is provided for each group of physical blocks defined on the basis of one or more thresholds provided for the number of physical block erasures. A free block search means for searching for the free block belonging to each group using the free block table;
Each group has an earliest block table for searching for the earliest block that is the earliest physical block assigned to the logical block among the physical blocks assigned to the logical block. Earliest block search means for searching for the earliest block belonging to each of the groups using an earliest block table;
Information holding means for holding information representing a target logical address range or a non-target logical address range specified by the host system;
A physical block allocated to the logical block that is a part of the plurality of logical blocks and that does not belong to the target logical address range or the target logical address range is the first block. Table registration means for registering in the block table;
Data storage means for specifying the logical block to which the area to be accessed belongs based on the access instruction, and storing data given from the host system in the physical block assigned to the specified logical block;
Data transfer means for transferring the data stored in the earliest block detected by the earliest block search means to the empty block detected by the empty block search means;
When a physical block is allocated to the logical block by the block management means in response to the access instruction, the free block search means assigns the group on the lower side, which is the group to which the physical block with a small number of erasures belongs. Search for the free block using the corresponding free block table preferentially,
When a physical block belonging to the group lower than the group to which the free block detected by the free block search means belongs is not registered in the earliest block table, the block management means performs the free block search. Assigning the free block detected by the means to the logical block;
When the physical block belonging to the group lower than the group to which the empty block detected by the empty block searching means belongs is registered in the earliest block table, the earliest block searching means The earliest block belonging to the group lower than the group to which the block belongs is searched, and the data transfer means stores the data stored in the earliest block detected by the earliest block search means as the empty block Transferring to the empty block detected by the block search means, the block management means assigns the earliest block detected by the earliest block search means to the logical block;
Memory controller. A memory controller that controls access to a flash memory that is erased in units of physical blocks based on an access instruction given from a host system,
Virtual block forming means for forming a plurality of virtual blocks obtained by collecting a plurality of physical blocks constituting the flash memory;
Erasure count management means for managing the erasure count in units of virtual blocks;
A block management means for managing a plurality of sector areas as a logical block by collecting a plurality of sector unit areas to which a logical address given as the access instruction is assigned, and assigning the virtual block to the logical block;
An empty block table for searching for an empty block which is the virtual block in which no valid data is stored is stored in the virtual block defined based on one or more thresholds provided for the number of times of erasing the virtual block. A free block search means for searching for the free blocks belonging to each of the groups using each free block table;
Each group has an earliest block table for searching for the earliest block that is the virtual block assigned to the logical block first among the virtual blocks assigned to the logical block, The earliest block search means for searching for the earliest block belonging to each of the groups using the earliest block table of
Information holding means for holding information representing a target logical address range or a non-target logical address range specified by the host system;
The virtual block allocated to the logical block that is part of the plurality of logical blocks and belongs to the logical block that does not belong to the target logical address range or the non-target logical address range. Table registration means for registering in the destination block table;
Data storage means for specifying the logical block to which the area to be accessed belongs based on the access instruction, and storing data given from a host system in the virtual block assigned to the specified logical block;
Data transfer means for transferring the data stored in the earliest block detected by the earliest block search means to the empty block detected by the empty block search means;
When the virtual block is allocated to the logical block by the block management means in response to the access instruction, the empty block search means is the group on the lower side that is the group to which the virtual block to which the erasure count is small belongs. Search for the free block using the free block table corresponding to the group preferentially,
When the virtual block belonging to the group lower than the group to which the free block detected by the free block searching means belongs is not registered in the earliest block table, the block management means Assigning the free block detected by the search means to the logical block;
When the virtual block belonging to the group lower than the group to which the empty block detected by the empty block searching means belongs is registered in the earliest block table, the earliest block searching means The earliest block belonging to the group lower than the group to which the empty block belongs is searched, and the data transfer means stores the data stored in the earliest block detected by the earliest block search means. Transferring to the empty block detected by the empty block search means, the block management means assigns the earliest block detected by the earliest block search means to the logical block;
Memory controller. Flash memory system comprising a memory controller according to claim 1 or 2, wherein a plurality of flash memory controlled by the memory controller. A flash memory control method for controlling access to a flash memory that is erased in units of physical blocks based on an access instruction given from a host system,
An erase count management step for managing the erase count of each physical block constituting the flash memory;
A block management step of managing a plurality of sector areas obtained by collecting a plurality of sector unit areas to which a logical address given as an access instruction is assigned as a logical block, and allocating a physical block constituting a flash memory to the logical block; ,
A free block table for searching for a free block that is a physical block in which no valid data is stored is provided for each group of physical blocks defined on the basis of one or more thresholds provided for the number of physical block erasures. A free block search step of searching for the free block belonging to each of the groups using each of the free block tables;
Each group has an earliest block table for searching for the earliest block that is the earliest physical block assigned to the logical block among the physical blocks assigned to the logical block. An earliest block search step of searching for the earliest block belonging to each of the groups using an earliest block table;
An information holding step for holding information representing a target logical address range or a non-target logical address range specified by the host system;
A physical block allocated to the logical block that is a part of the plurality of logical blocks and that does not belong to the target logical address range or the target logical address range is the first block. A table registration step to register in the block table;
A data storage step of identifying the logical block to which the area to be accessed belongs based on the access instruction, and storing data provided from a host system in a physical block allocated to the identified logical block;
A data transfer step of transferring data stored in the earliest block detected in the earliest block search step to the empty block detected in the empty block search step;
When a physical block is allocated to the logical block by the block management step in response to the access instruction, the free block search step adds the low-order group to the group to which the physical block with a small number of erases belongs. A search for the free block that is preferentially used in the corresponding free block table is performed,
When a physical block belonging to the group lower than the group to which the free block detected by the free block search step belongs is not registered in the earliest block table, the free block search is performed in the block management step. The free block detected by the step is assigned to the logical block;
When a physical block belonging to the group lower than the group to which the empty block detected by the empty block searching step belongs is registered in the earliest block table, the earliest block searching step includes the empty block. The earliest block belonging to the group lower than the group to which the block belongs is searched, and in the data transfer step, the data stored in the earliest block detected by the earliest block search step is free. Transferred to the empty block detected by the block search step, and in the block management step, the earliest block detected by the earliest block search step is assigned to the logical block .
Flash memory control method. A flash memory control method for controlling access to a flash memory that is erased in units of physical blocks based on an access instruction given from a host system,
A virtual block forming step of forming a plurality of virtual blocks obtained by collecting a plurality of physical blocks constituting a flash memory; and an erasure count management step of managing the erase count in units of the virtual blocks;
A block management step of managing a plurality of sector areas obtained by collecting a plurality of sector unit areas to which a logical address given as the access instruction is assigned as a logical block, and assigning the virtual block to the logical block;
An empty block table for searching for an empty block which is the virtual block in which no valid data is stored is stored in the virtual block defined based on one or more thresholds provided for the number of times of erasing the virtual block. A free block search step for searching for the free block belonging to each group using each free block table, which is provided for each group;
Each group has an earliest block table for searching for the earliest block that is the virtual block assigned to the logical block first among the virtual blocks assigned to the logical block, The earliest block search step of searching for the earliest block belonging to each of the groups using the earliest block table of
An information holding step for holding information representing a target logical address range or a non-target logical address range specified by the host system;
The virtual block allocated to the logical block that is part of the plurality of logical blocks and belongs to the logical block that does not belong to the target logical address range or the non-target logical address range. A table registration step to register in the destination block table;
A data storage step of identifying the logical block to which the area to be accessed belongs based on the access instruction, and storing data given from a host system in the virtual block allocated to the identified logical block;
A data transfer step of transferring data stored in the earliest block detected in the earliest block search step to the empty block detected in the empty block search step;
When the virtual block is allocated to the logical block by the block management step in response to the access instruction, the lower block that is the group to which the virtual block to which the erasure count is small belongs in the empty block search step belongs to The search for the free block that is preferentially used in the free block table corresponding to the group is performed,
When the virtual block belonging to the group lower than the group to which the free block detected by the free block searching step belongs is not registered in the earliest block table, the free block is used in the block management step. The free block detected by the search step is assigned to the logical block;
When the virtual block belonging to the group lower than the group to which the empty block detected by the empty block searching step belongs is registered in the earliest block table, in the earliest block searching step, The earliest block belonging to the group lower than the group to which the empty block belongs is searched, and the data stored in the earliest block detected by the earliest block search step is the data transfer step. Transferred to the empty block detected in the empty block search step, and in the block management step, the earliest block detected in the earliest block search step is assigned to the logical block ;
Flash memory control method.

Images