US20100023680A1 - Method for Controlling Non-Volatile Semiconductor Memory System - Google Patents

Method for Controlling Non-Volatile Semiconductor Memory System Download PDF

Info

Publication number
US20100023680A1
US20100023680A1 US12/564,770 US56477009A US2010023680A1 US 20100023680 A1 US20100023680 A1 US 20100023680A1 US 56477009 A US56477009 A US 56477009A US 2010023680 A1 US2010023680 A1 US 2010023680A1
Authority
US
United States
Prior art keywords
block
logical
data
blocks
physical
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Abandoned
Application number
US12/564,770
Inventor
Yoshiyuki Tanaka
Makoto Yatabe
Takeaki Sato
Kazuya Kawamoto
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Toshiba Corp
Original Assignee
Toshiba Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from JP9214561A external-priority patent/JPH10124384A/en
Application filed by Toshiba Corp filed Critical Toshiba Corp
Priority to US12/564,770 priority Critical patent/US20100023680A1/en
Publication of US20100023680A1 publication Critical patent/US20100023680A1/en
Abandoned legal-status Critical Current

Links

Images

Classifications

    • GPHYSICS
    • G11INFORMATION STORAGE
    • G11CSTATIC STORES
    • G11C16/00Erasable programmable read-only memories
    • GPHYSICS
    • G11INFORMATION STORAGE
    • G11CSTATIC STORES
    • G11C16/00Erasable programmable read-only memories
    • G11C16/02Erasable programmable read-only memories electrically programmable
    • G11C16/06Auxiliary circuits, e.g. for writing into memory
    • G11C16/10Programming or data input circuits
    • G11C16/102External programming circuits, e.g. EPROM programmers; In-circuit programming or reprogramming; EPROM emulators
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F12/00Accessing, addressing or allocating within memory systems or architectures
    • G06F12/02Addressing or allocation; Relocation
    • G06F12/0223User address space allocation, e.g. contiguous or non contiguous base addressing
    • G06F12/023Free address space management
    • G06F12/0238Memory management in non-volatile memory, e.g. resistive RAM or ferroelectric memory
    • G06F12/0246Memory management in non-volatile memory, e.g. resistive RAM or ferroelectric memory in block erasable memory, e.g. flash memory
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601Interfaces specially adapted for storage systems
    • G06F3/0602Interfaces specially adapted for storage systems specifically adapted to achieve a particular effect
    • G06F3/061Improving I/O performance
    • G06F3/0613Improving I/O performance in relation to throughput
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601Interfaces specially adapted for storage systems
    • G06F3/0628Interfaces specially adapted for storage systems making use of a particular technique
    • G06F3/0638Organizing or formatting or addressing of data
    • G06F3/064Management of blocks
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601Interfaces specially adapted for storage systems
    • G06F3/0668Interfaces specially adapted for storage systems adopting a particular infrastructure
    • G06F3/0671In-line storage system
    • G06F3/0673Single storage device
    • G06F3/0679Non-volatile semiconductor memory device, e.g. flash memory, one time programmable memory [OTP]
    • GPHYSICS
    • G11INFORMATION STORAGE
    • G11CSTATIC STORES
    • G11C16/00Erasable programmable read-only memories
    • G11C16/02Erasable programmable read-only memories electrically programmable
    • G11C16/06Auxiliary circuits, e.g. for writing into memory
    • G11C16/22Safety or protection circuits preventing unauthorised or accidental access to memory cells
    • G11C16/225Preventing erasure, programming or reading when power supply voltages are outside the required ranges
    • GPHYSICS
    • G11INFORMATION STORAGE
    • G11CSTATIC STORES
    • G11C5/00Details of stores covered by group G11C11/00
    • G11C5/14Power supply arrangements, e.g. power down, chip selection or deselection, layout of wirings or power grids, or multiple supply levels
    • G11C5/143Detection of memory cassette insertion or removal; Continuity checks of supply or ground lines; Detection of supply variations, interruptions or levels ; Switching between alternative supplies
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601Interfaces specially adapted for storage systems
    • G06F3/0628Interfaces specially adapted for storage systems making use of a particular technique
    • G06F3/0662Virtualisation aspects
    • G06F3/0664Virtualisation aspects at device level, e.g. emulation of a storage device or system
    • GPHYSICS
    • G11INFORMATION STORAGE
    • G11CSTATIC STORES
    • G11C29/00Checking stores for correct operation ; Subsequent repair; Testing stores during standby or offline operation
    • G11C29/70Masking faults in memories by using spares or by reconfiguring
    • G11C29/76Masking faults in memories by using spares or by reconfiguring using address translation or modifications

Definitions

  • the present invention relates generally to a method for controlling a memory system. More specifically, the invention relates to a method for controlling a non-volatile semiconductor memory system, such as a non-volatile semiconductor memory card.
  • a flash memory card like that shown in FIG. 1 has widely used as a storage medium for a portable information apparatus, such as a digital still camera or a PDA.
  • NAND-type flash memories have been used and sold.
  • a flash memory card of this type has a thin plastic package 1 having a slightly recessed portion, in which a flash memory 2 having a flat electrode with 22 pins is mounted.
  • This flash memory card is electrically connected to a host system via a dedicated connector to input and output data.
  • FIG. 2 shows, as an example of a flash memory, a 16-Mbit NAND-type flash memory divided into 512 physical memory blocks. Each of these blocks is a minimum unit for an erase operation. One block is divided into 16 pages. One page is a basic unit of writing and read-out. One page comprises 264 bytes. Among the 264 bytes, 256 bytes are a user data area (data division), and the remaining 8 bytes (redundant division) are used for storing an error correcting code and a management information service.
  • FIG. 3 Five hundred logical blocks (identified by logical block address (LBA)), are set in FIG. 3 .
  • One logical block corresponds to 8 continuous sectors. That is, logical block 0 means logical sectors 0 to 7 .
  • FIG. 4 shows an exemplary data storing method.
  • Data Area- 1 stores the first half 0 to 255 bytes of the data of 512 bytes.
  • Data Area- 2 stores the second half 256 to 511 bytes of the data of 512 bytes.
  • the User Data Area is open to store user data, so that the use thereof is entrusted to the user.
  • a Data Status Area indicates whether data are presumed reliable. Although an “FFh” is usually set, a “00h” is set when defective data are written (e.g., when the desired threshold voltage for the memory cell cannot be achieved or another error condition is associated with the cell).
  • a Block Status Area indicates whether a block is good or defective.
  • a Block Address Area- 1 indicates a logical address information of a block. Furthermore, since 8 sectors forming one logical block correspond to one of 512 physical blocks, the same values for all the data are written in the same block. Similarly, in a Block Address Area- 2 , the same contents as the data of the Block Address Area- 1 are written.
  • An ECC Area- 1 is a 3-byte ECC code of even page data (256 bytes).
  • An ECC Area- 2 is a 3-byte ECC code of odd page data (256 bytes).
  • FIG. 5 shows, as another example of a flash memory, a 64-Mbit NAND-type flash memory that is divided into 1024 physical memory blocks. Each of these blocks is a minimum unit of erase.
  • One block is divided into 16 pages.
  • One page is a basic unit of writing and read-out.
  • One page comprises 528 bytes. Among the 528 bytes, 512 bytes are used for a user data area (data division), and the remaining 16 bytes (redundant division) are used for storing an error correcting code and a management information service.
  • One thousand logical blocks are set in FIG. 6 .
  • One logical block corresponds to 16 continuous sectors. That is, logical block 0 includes logical sectors 0 to 15 .
  • FIG. 7 shows a method for storing data in the 64-Mbit NAND-type flash memory.
  • the control of such a memory card adopts an additional writing system for writing updated data in a previously erased area when data are updated and subsequently erasing an area in which the original data were stored. Therefore, a physical block, in which data corresponding to a certain logical block exist, is not fixed and is always moving in the memory. Also, as shown in FIG. 8 , the redundant division of the physical block stores therein a logical block address information indicating which logical block corresponds to the data held in the physical block.
  • the Block Address Area- 1 and the Block Address Area- 2 in FIGS. 4 and 7 correspond to the corresponding logical addresses.
  • the memory system searches areas for storing the logical block address information of all the physical blocks to prepare a translation table between logical blocks and physical blocks on a system RAM, usually when a power supply is turned on. After the tables are prepared once, it is possible to immediately determine the physical block corresponding to the logical block by referring to the tables. The searching operation for all the blocks is carried out once when the power supply is turned on. Naturally, if the position of the corresponding physical block changes after the data are updated, the memory system updates a logical address/physical address translation table to get ready for the next access.
  • FIG. 9 shows a logical address/physical address translation table of a conventional 16-Mbit NAND-type flash memory.
  • the data of one logical block i.e., the data of 8 continuous sectors, exist in any one of 512 physical blocks in the flash memory. In order to select one block from the 512 physical blocks, 9 bits are required. If the table is formed so that an offset logical block address directly indicates a physical block for convenience of a software, 2 bytes are required for one logical block, so that a RAM area of 1 KB in total is required.
  • the address of a physical block, in which the information of logical block 5 is stored is an address offset by 5 words (10 bytes) from the top of the table.
  • FIG. 10 shows a logical address/physical address translation table of a conventional 64-Mbit NAND-type flash memory.
  • the data of one logical block i.e., the data of 16 continuous sectors, exist in any one of 1024 physical blocks in the flash memory.
  • the selection of one physical block from 1024 physical blocks needs 10 bits, so that a RAM area of 2 KB in total is required. For that reason, a very large RAM area is required similar to the 16-bit NAND-type flash memory.
  • the number of blocks is 8192 for the 1-Gbit capacity memories, so that a RAM capacity of 16 KB is required.
  • the Block Address Area of the redundant division of a physical block of the 16-Mbit NAND-type flash memory shown in FIG. 7 stores therein a logical block address information indicating which logical block corresponds to the data held in the physical block.
  • FIG. 11 shows the structure of the Block Address Area of the redundant division of each physical block.
  • the writing and erase for a flash memory will be briefly described below.
  • the writing in a flash memory is carried out page by page.
  • one page has 528 bytes.
  • erase is carried out in units of a block.
  • one block is formed by 16 pages.
  • Parts (a) and (b) of FIG. 12 show conventional DOS format parameters, wherein part (a) shows the parameters in the case of a cluster size of 4 KB and part (b) shows the parameters in the case of a cluster size of 8 KB.
  • cluster means a basic minimum unit of file management on the DOS. Even when a small file is stored, the file occupies one cluster. When the file size is large, the file is managed as a chain of a plurality of clusters, and its management information service is stored in a FAT (file allocation table).
  • the size of the cluster, the management method for the FAT and so forth are managed in a sector called a boot sector.
  • a boot sector When one device is managed as a plurality of drives, its information is stored in a master boot sector.
  • the OS issues a write command for each cluster.
  • Part (a) of FIG. 12 shows the case of a cluster size of 4 KB.
  • a master boot sector is arranged in logical sector 0
  • a boot sector is arranged in logical sector 16 .
  • FATs are arranged in logical sectors 17 through 22
  • the copies of the FATs are arranged in logical sectors 23 through 28 .
  • directories are arranged in logical sectors 29 through 44
  • file data areas are arranged in and after logical sector 45 .
  • Part (b) of FIG. 12 shows the case of cluster size of 8 KB.
  • a master boot sector is arranged in logical sector 0
  • a boot sector is arranged in logical sector 16 .
  • FATs are arranged in logical sectors 17 through 19
  • the copies of the FATs are arranged in logical sectors 20 through 22 .
  • directories are arranged in logical sectors 23 through 38
  • file data areas are arranged in and after logical sector 39 .
  • the number of the write and erase operations actually executed in the device is far greater than the number of updated sectors, so that there is the second problem in that the rewriting speed viewed from the outside decreases.
  • FIG. 16 shows the relationship between a management area and a data area when an erase command is executed.
  • FIG. 16 for example, when erase commands for File- 1 and File- 4 are executed, the File- 1 and File- 4 are open, and a del.mark is stored. At this time, the File- 1 and the File- 4 are not erased from the data area.
  • the ECC Area- 1 shown in FIG. 4 is a 3-byte ECC code of an even page data (256 bytes).
  • the ECC Area- 2 is a 3-byte ECC code of an odd page data (256 bytes).
  • the ECC error correcting code
  • the system utilizes this error correcting code to determine whether a read data has an error. When an error exists, the system can correct the error.
  • the required error correcting capability depends on the reliability of the flash memory itself, e.g., the cell structure of the memory. Flash memories have a plurality of data storing methods. When these flash memories are used for a system, such as a digital still camera and a PDA, error correction will be considered.
  • a first flash memory card holds binary values “0” and “1” corresponding to the different threshold values of a memory cell shown in FIG. 17 .
  • the memory card can therefore use a 1-bit error correcting code to provide ECC for one page (256 bytes).
  • a second flash memory card holds four-valued values “00”, “01”, “10” and “11” (2 bits) corresponding to the multiple values of the threshold of a memory cell shown in FIG. 18 .
  • the memory card that stores multiple bits per cell, illustrated in FIG. 18 therefore uses a 2-bit error correcting code to provide ECC for one page (256 bytes) since there is a possibility that 2-bit data may be erroneous when one memory cell is defective.
  • the algorithm for generating a code and detecting and correcting an error for 1-bit error correction is different from that for 2-bit error correction.
  • a method for controlling a memory system which uses the terminals of a connector sequentially connected to the terminals of a memory card (a storage medium), which is disclosed in Japanese Patent Laid-Open Nos. 8-90969 and 8-202509.
  • a method for controlling a memory system using a storage medium which is inserted into an electronic apparatus via a connector to add a memory function thereto, said storage medium having ground terminals, power supply terminals, control terminals and data input/output terminals, said connector having a function of being sequentially connected to each of said terminals, wherein when said storage medium is inserted into said connector, said ground terminals and control terminals of said storage medium are connected to corresponding terminals of said connector before said power supply terminals and data input/output terminals of said storage medium are connected to corresponding terminals of said connector.
  • a connector to add a memory-function thereto, said storage medium having ground terminals, power supply terminals, control terminals and data input/output terminals, said connector having a function of being sequentially disconnected from each of said terminals wherein when said storage medium is ejected from said connector, said power supply terminals and data input/output terminals of said storage medium are disconnected from corresponding terminals of said connector before said ground terminals and control terminals of said storage medium are disconnected from corresponding terminals of said connector.
  • a method for controlling a memory system using a storage medium which is inserted into an electronic apparatus via a connector to add a memory function thereto, said storage medium having ground terminals, power supply terminals, a first control terminal, a second control terminal and data input/output terminals, said connector having a function of being sequentially connected to each of said terminals, wherein when said storage medium is inserted into said connector, said ground terminals and first control terminals of said storage medium are connected to corresponding terminals of said connector before said power supply terminals and data input/output terminal of said storage medium are connected to corresponding terminals of said connector, and said second control terminal being connected to a corresponding terminal of said connector before said data input/output terminals are connected to a corresponding terminal of said connector.
  • a method for controlling a memory system using a storage medium which is inserted into an electronic apparatus via a connector to add a memory function thereto, said storage medium having ground terminals, a power supply terminals, a first control terminal, a second control terminal and data input/output terminals, said connector having a function of being sequentially connected to each of said terminals, wherein when said storage medium is ejected from said connector, said data input/output terminals are disconnected from corresponding terminals of said connector before said second control terminal is disconnected from a corresponding terminal of said connector, and said power supply terminals and data input/output terminals of said storage medium being disconnected from corresponding terminals of said connector before said ground terminals and first control terminal of said storage medium are disconnected from corresponding terminals of said connector.
  • a method for controlling a memory system having a storage medium which is inserted into an electronic apparatus via a connector to add a memory function thereto, said storage medium having ground terminals, power supply terminals, first control terminal, a second control terminal, data input/output terminals and an insertion/ejection detecting terminal, said connector having a function of being sequentially connected to each of said terminals, wherein when said storage medium is inserted into said connector, said ground terminals and first control terminals of said storage medium are connected to corresponding terminals of said connector before said power supply terminals and data input/output terminals of said storage medium are connected to corresponding terminals of said connector, said second control terminal of said storage medium being connected to a corresponding terminal of said connector before said data input/output terminals of said storage medium are connected to corresponding terminals of said connector, and said insertion/ejection detecting terminal being connected to a corresponding terminal of said connector after all of said terminals of said storage medium are inserted.
  • a method for controlling a memory system having a storage medium which is inserted into an electronic apparatus via a connector to add a memory function thereto, said storage medium having ground terminals, power supply terminals, a first control terminal, a second control terminal, data input/output terminals and an insertion/ejection detecting terminal, said connector having a function of being sequentially connected to each of said terminals, wherein when said storage medium is ejected from said connector, said insertion/ejection detecting terminal is disconnected from a corresponding terminal of said connector before said data input/output terminals of said storage medium are disconnected from corresponding terminals of said connector, said data input/output terminals of said storage medium being disconnected from corresponding terminals of said connector before said power supply terminals and second control terminal of said storage medium are disconnected from corresponding terminals of said connector, and said power supply terminals of said storage medium being disconnected from corresponding terminals of said connector before said ground terminals and first control terminal of said storage medium are disconnected from corresponding terminal
  • a method for controlling a memory system which includes:
  • logical blocks managed by the system physical blocks for storing therein data corresponding to said logical blocks, said physical blocks comprising a plurality of memory cells; redundant divisions included in a corresponding said physical blocks for storing therein addresses of corresponding said logical blocks; and physical block areas formed by at least two of said physical blocks, wherein a logical address/physical address translation table is prepared for managing corresponding relationships between said logical blocks and said physical blocks.
  • a method for controlling a memory system which is a semiconductor memory system for storing a file managed according to predetermined units, in a storage area divided into second predetermined units.
  • a method for controlling a memory system which includes:
  • a method for controlling a memory system which includes:
  • a method for controlling a non-volatile semiconductor memory system which comprises:
  • dividing a cell array into a plurality of physical blocks storing each information corresponding to relationship between said physical block and logical block which is managed by said system, in each said physical block; and in order to form a table for managing corresponding relationships between said logical blocks and said physical blocks, in a random access memory in said system, sequentially preparing required corresponding relationships of corresponding relationships between said logical blocks and said physical blocks, in said random access memory in said system in accordance with accesses from a host.
  • a method for controlling a non-volatile semiconductor memory system which comprises:
  • a method for controlling a non-volatile semiconductor memory system which comprises:
  • said method further comprising:
  • a method for controlling a non-volatile semiconductor memory system which further comprises:
  • a method for controlling a non-volatile semiconductor memory system which allows any one of various memory units to be detachably mounted in a body of said memory system, said method comprising selecting a corresponding one of various error correcting means of said body in accordance with said one of various memory units mounted in said body, to carry out error correction.
  • FIG. 1 is an outside drawing of a memory card
  • FIG. 2 is a table showing a physical block construction of a flash memory
  • FIG. 3 is a table showing data in a personal computer
  • FIG. 4 is a table showing a method for storing data in a flash memory
  • FIG. 5 is a table showing a physical block construction of a flash memory
  • FIG. 6 is a table showing a physical block construction of a flash memory
  • FIG. 7 is a table showing a method for storing data in a flash memory
  • FIG. 8 is a conventional logical/physical block translation table
  • FIG. 9 is a logical address/physical address translation table in a conventional memory system
  • FIG. 10 is a logical address/physical address translation table in a 64-Mbit NAND-type flash memory system
  • FIG. 11 is a table showing a data configuration of a block address of a flash memory
  • FIG. 12 is a table showing parameters in a conventional DOS format
  • FIG. 13 is a chart showing a conventional rewrite sequence
  • FIG. 14 is a chart showing a conventional rewrite sequence
  • FIG. 15 is a chart showing a conventional rewrite sequence
  • FIG. 16 is a table showing a conventional relationship between a management area and a data area when am erase command is executed
  • FIG. 17 shows an example of how single-bit per cell data are stored in a flash memory
  • FIG. 18 shows an example of how multi-bit per cell data are stored in a flash memory
  • FIG. 19 is a table showing a conventional system
  • FIG. 20 is an outside drawing of a memory card for use in a memory system according to the present invention.
  • FIG. 21 is a table showing signal names of the respective terminals of the card of FIG. 20 ;
  • FIG. 22 is a main flow chart of a memory system using a method for controlling a memory system according to the present invention.
  • FIGS. 23( a ) and 23 ( b ) are a schematic view showing the relationship between a power supply voltage and an appearance of a memory card
  • FIGS. 24( a ) and 24 ( b ) are a schematic view showing a method for electrically detecting a power supply voltage of a memory card
  • FIG. 25 is an outside drawing of a 5 V dedicated connector
  • FIG. 26 is an outside drawing of a 3.3 V dedicated connector
  • FIG. 27 is an outside drawing of a 5 V/3.3 V dedicated connector
  • FIG. 28 is a schematic view showing a method for detecting the insertion/ejection of a memory card
  • FIG. 29 is a schematic view showing a method for detecting the insertion/ejection of a memory card when the memory card corresponds to a PC card adapter;
  • FIG. 30 is a table showing an ECC data construction in a memory system according to the present invention.
  • FIG. 31 is a view showing conditions for calculating an ECC code in a memory system according to the present invention.
  • FIG. 32 is a control flow chart of a memory system according to the present invention when a power supply is turned on;
  • FIG. 33 is a flow chart for preparing a logical address/physical address translation table in a memory system according to the present invention.
  • FIG. 34 is a logical address/physical address translation table in a memory system according to the present invention.
  • FIG. 35 is a flow chart of a memory system according to the present invention when read-out is carried out
  • FIG. 36 is a flow chart of a memory system according to the present invention writing is carried out
  • FIG. 37 is a view showing parameters of a DOS format according to the present invention.
  • FIG. 38 is a chart showing a rewrite sequence according to the present invention.
  • FIG. 39 is a chart showing a rewrite sequence according to the present invention.
  • FIG. 40 is a chart showing a rewrite sequence according to the present invention.
  • FIG. 41 is a table showing the relationship between a management area and a data area according to the present invention when an erase command is executed;
  • FIGS. 42( a ) and 42 ( b ) are a table of an example of a logical block/physical block translation table in a preferred embodiment of the present invention.
  • FIG. 43 is a flow chart of an example of a method for preparing a logic block/physical block translation table in this preferred embodiment
  • FIG. 44 is a logical/physical block translation table in the preferred embodiment.
  • FIG. 45 is a table showing a method for expressing physical block addresses in the preferred embodiment.
  • FIG. 46 is a table for explaining a conventional embodiment when a defective block is replaced with a redundant block
  • FIG. 47 is a table for explaining a conventional example of a replacement of a defective block with a redundant block
  • FIG. 48 is a table for explaining a preferred embodiment when a defective block is replaced with a redundant block in view of a zone division control as a premise;
  • FIG. 49 is a schematic view of a PC card adapter according to the present invention.
  • FIG. 50 is a flow chart when a flash memory card is inserted into a PC card adapter
  • FIG. 51 is a table showing parameters of another preferred embodiment of a DOS format of a flash memory according to the present invention.
  • FIG. 52 is a table showing a method for using an ECC code area.
  • FIG. 20 shows a flash memory card for use in a memory system according to the present invention
  • FIG. 21 shows signal names of the respective pins of a flat electrode of the memory card.
  • FIG. 22 shows a main control flow for use in the flash memory card. The operations of the respective steps in this flow will be described below.
  • the start of processing of a memory card assumes that the memory card is normally inserted into a connector. There is a possibility that physical or data destruction may be caused if the system accesses the memory card when the memory card is incompletely inserted. Therefore, in the control flow for the memory system according to the present invention, there is provided means for detecting whether a memory card is normally inserted. This means may be structural or electrical.
  • a detecting switch may be provided in the connector at a position (e.g., the innermost part of the connector) so that the detecting switch can transmit a detection signal to the system after the memory card is held at a normal contact position.
  • a system mechanism for ensuring that the memory card is held at a normal position, not for detecting insertion may be provided.
  • a lid may be provided in an inlet for the memory card so that the lid pushes the memory card to a home position when the lid is closed.
  • means for electrically detecting the insertion of the memory cart may be provided.
  • a power supply voltage detecting step B will be described in detail below.
  • the memory cards include a product having a power supply voltage of 5 V and a product having a power supply voltage of 3.3 V. If a voltage of 5 V is applied to a memory card operated by a power supply voltage of 3.3 V, there is a possibility that voltage destruction and so forth may be caused. In order to avoid this, the system is designed to detect a power supply voltage for a memory card. Methods for detecting the power supply voltage may include a method for determining the power supply voltage based on the appearance of the memory card, and a method for electrically determining the power supply voltage.
  • FIGS. 23( a ) and 23 ( b ) show an example of a method for determining a power supply voltage for a memory card based on the appearance of the memory card.
  • the 5 V product has a cutout on the left side of the upper end thereof as shown in FIG. 23( a ), and the 3.3 V product has a cutout on the right side of the upper end thereof as shown in FIG. 23( b ).
  • the power supply voltage for the memory card it is possible to easily visually determine the power supply voltage for the memory card.
  • FIGS. 24( a ) and 24 ( b ) show an example of a method for electrically detecting a power supply voltage for a memory card.
  • a number 17 pin is used for detecting the power supply voltage.
  • the number 17 pin is electrically on the floating condition as shown in FIG. 24( a ), and in the 3.3 V product, the number 17 pin is short-circuited to a VCC line on the surface of the memory card.
  • the number 17 pin is pulled down to a VSS via a resistor.
  • a voltage of 3.3 V is supplied to a number 12 pin and a number 22 pin as a VCC.
  • a voltage of 5 V is not applied to a power supply voltage terminal.
  • the system monitors the level of the number 17 pin. When the level of the number 17 pin is a “L” level, the system determines that the memory card is the 5 V product, and when the level of the number 17 pin is a “H” level, the system determines that the memory card is the 3.3 V product. In accordance with the determined results, a normal power supply voltage is applied to the power supply voltage terminal.
  • FIG. 25 shows an example of a dedicated system for the 5 V product.
  • the 5 V dedicated system may use a connector with a cutout detecting mechanism so as to prevent the 3.3 V product from being inserted.
  • the 3.3 V product can be inserted upside down, it is possible to prevent the 3.3 V product from electrically contacting the system.
  • FIG. 25 is simplified in order to facilitate better understanding thereof, the following consideration is given.
  • a misinsertion preventing mechanism utilizing the cutout must function before the terminals of the memory card contact the terminals of the connector.
  • the misinsertion preventing mechanism may be positioned in the vicinity of the inlet of the connector so as to prevent the memory card from electrically contacting the system when misinsertion is carried out.
  • FIG. 26 shows an example of a 3.3 V dedicated system.
  • the 3.3 V dedicated system uses a connector with a cutout detecting mechanism so as to prevent the 5 V product from being inserted. In this case, it is also possible to prevent the 3.3 V product from being inserted upside down. Although the 5 V product can be inserted upside down, it is possible to prevent the 5 V product from electrically contacting the system.
  • FIG. 27 shows an example of a 5 V/3.3 V dual purpose system.
  • this system is designed to prevent a power supply of 5 V from being applied to the 3.3 V product. That is, a power supply voltage of 5 V may not be outputted when a 3.3 V memory card is inserted. The supply voltage of 5 V is not applied to the device until the power supply voltage is electrically detected to verify that the inserted memory card is the 5 V product.
  • a capacity detecting step C will be described below.
  • Flash memories having different memory capacities and interface specifications.
  • a maker code, device code or the like of the device is determined to inhibit new access when an unsupported code is determined.
  • a normal power supply voltage is inputted.
  • a physical format checking step D will be described below.
  • the memory card has a physical format for storing data.
  • the physical format thereof is checked. If the memory card has an unsupported format, it is rejected without altering data. If the system executes a physical format again when an unknown physical format product is inserted, there is a risk that the processing for innate and acquired defective blocks of the flash memory may be incomplete, so care should be taken.
  • a 16-Mbit NAND-type flash memory is divided into 512 physical blocks. In the top block thereof a physical format, an attribute information of the card and so forth are written. Since the remaining blocks are used as data areas, it may be determined whether a supportable physical format is provided by examining data of the top block.
  • a logical format step E is carried out to allow the memory card to be accessed as a device on the DOS. It is also required to check whether the logical format of the memory card is supported. If the cluster delimiter of the DOS is coincident with the delimiter of the physical block of the NAND flash memory, it is possible to carry out a higher speed operation.
  • the respective operations indicate the memory access operations of the system, and include read-out, write and erase operations.
  • ejection is detected in a manner similar to the above described insertion detection.
  • the operation of the memory system ends.
  • a number 1 pin terminal and number 10 pin terminal of the connector are connected to a VSS, and a number 11 pin terminal of the connector is pulled up to a VCC level.
  • the potential level of the connector terminal is held at a “H” level by means of a pull-up resistor.
  • the number 11 pin of the memory card contacts the connector terminal, the number 11 pin is connected to VSS via the number 1 pin and number 10 pin, so that the potential level is changed to a “L” level.
  • hot-line insertion/ejection means the insertion/ejection of a memory card while a power supply voltage is applied to a connector. In order to prevent malfunction and electrical destruction of the memory, the following care should be taken.
  • control signals (/CE, /WE, CLE, etc.) and an I/O pin are undefined while a power supply voltage is supplied, there is a possibility that an unintended write or erase command or the like may be received.
  • the device condition is a data output condition.
  • the system side condition is also a data output condition, there is a possibility that collision occurs on a data bus, having an undesired influence on the device.
  • a potential higher than the power supply voltage is applied to an I/O pin, there is a possibility that current flows into the device to cause the latch up and so forth.
  • the contact and removal sequences for connector terminals are provided, and the contact and removal are sequentially carried out, so that it is possible to achieve the hot-line insertion/ejection.
  • the undermentioned four-stage sequence is suitable for use in a PC card adapter or the like since there is a little limit to the system side.
  • the contact sequence for the connector and the limitations on the system side are as follows.
  • the contact with a VCC terminal is made to establish the ground level of the chip, and the level of a CLE terminal is fixed to a “L” level (an inactive state) so as to avoid the problem that an erroneous command is input with noise and so forth in a subsequent sequence that erroneously erases the device.
  • the /CE terminal is fixed to the “H” level (an inactive state) so that the state of the output pin of the device is a high impedance state, thereby preventing the collision of data on the bus regardless of the bus control state on the system side.
  • the problem that current flows into the output pin from the bus line is avoided since the VCC reaches a predetermined level at the second stage.
  • the number 11 pin finally contacts the connector terminal. As described above, the number 11 pin serves as an insertion/ejection detecting terminal.
  • a hot-line insertion/ejection in a two-stage sequence may be used.
  • the contact sequence for the connector and the limitations on the system side are as follows.
  • the contact with the VCC terminal is made to establish the ground level of a chip, and the level of the CLE terminal is fixed to the “L” level (the inactive state) so as to avoid the problem that an erroneous command is input with noise and so forth in a subsequent sequence to erroneously erase the device.
  • the VCC, /CE, I/O pins and so forth do not completely simultaneously contact the connector terminals, and any one of unintended contacts may occur at the contact timing.
  • the I/O bus is held in the high impedance state on the system side, it is possible to avoid the collision of data and the flow of current into the pin on the bus.
  • the connector for the four-stage contact sequence may also be applied to a PC card type adapter.
  • a typical PC card has two pins, ⁇ CD 1 and ⁇ CD 2 pins, which are defined as card detecting pins, and the insertion/ejection of the PC card is typically detected on the system side while both pins are pulled up.
  • ⁇ CD 1 and ⁇ CD 2 pins As a signal generating method for the ⁇ CD 1 and ⁇ CD 2 pins in the PC card adapter, an example of a method using the 11 pin of the memory card is shown in FIG. 29 . In this case, a standard sequence for a personal computer will be described below. When only a PC card adapter is inserted into a personal computer, the personal computer determines that no card is inserted.
  • the VSS When a memory card is inserted into the personal computer, the VSS ( 1 , 10 pins) first contacts a four-stage contact sequence connector. At this stage, GND level is supplied from the personal computer. After sequential connection proceeds and all the pins are connected, the 11 pin is finally connected. At this stage, the levels of the ⁇ CD 1 and ⁇ CD 2 are reduced to the “L” level, and the personal computer side recognizes that the card is inserted, so that the initializing processing is started and a power supply is turned on the personal computer side.
  • the memory card When a memory card is ejected, the memory card is removed in the reverse sequence to the above described contact sequence. First, the ejection of the memory card is detected by means of the 11 pin. Then, a write enable signal /WP is enabled to reset the erase or write operation. The time required for reset is 1.5 seconds or less at the maximum in an erase mode. In order to normally carry out the reset operation, the power supply voltage must be supplied to the memory card, and the logic in the memory card must be normally operated. When the four-stage removal sequence is used, it is ideal that the reset operation is completed before the VCC pin is removed and that the reset operation is completed before the I/O pin and so forth are removed at the third stage.
  • the second-stage removal sequence can be achieved by removing the pins in the reverse sequence to the above described two-stage insertion sequence.
  • a three-stage contact/removal sequence for simultaneously carrying out the third and fourth stages of the four-stage contact/removal sequence is effective in data protection during the hot-line insertion/ejection.
  • the memory system control method of the present invention it is possible to improve the stability of operation during the insertion/ejection of the memory card when the insertion/ejection of the memory card is detected, and it is possible to protect data during the hot-line insertion/ejection.
  • a flash memory card which has a reduced capacity of a RAM area necessary for a logical address/physical address translation table.
  • the structure of a physical block and the data construction in the physical block in a memory system according to the present invention are the same as those described in the description of the prior art, and shown in FIGS. 2 and 4 .
  • FIG. 4 the meanings of the respective bytes in a physical block in a memory system according to the present invention will be described.
  • Data Area- 1 stores the data of the first half 0 to 255 bytes of the data of 512.
  • Data Area- 2 stores the data of the second half 256 to 511 bytes of the data of 512 bytes.
  • the data of a User Data Area are open to a user, and the use thereof is entrusted to the user.
  • a Data Status Area indicates whether data are normal. Although an “FFh” is usually set, a “00h” is set when abnormal data are written.
  • a Block Status Area indicates whether a block is good or defective.
  • a Block Address Area- 1 indicates a logical address information of a block. Furthermore, since 8 sectors forming one logical block correspond to one of 512 physical blocks, the same value for all the data is written in the same block. Similarly, in a Block Address- 2 , the same contents as the data of the Block Address Area- 1 are written.
  • An ECC Area- 1 is a 3-byte ECC code of even page data (256 bytes).
  • An ECC Area- 2 is a 3-byte ECC code of odd page data (256 bytes).
  • ECC Error Correction Code
  • bit 0 of the input at the first byte is the first bit (address: 00000000 000) of 2048 bits
  • bit 7 of the input at the 255th byte is the 2048th bit (address: 11111111 111) of 2048 bits.
  • the ECC code (line parity (LP) and column parity (CP)) is calculated as an odd parity of 1024 bits meeting the conditions shown in FIG. 31 .
  • FIG. 32 is a control flow chart for the second preferred embodiment of a flash memory card according to the present invention when a power supply is turned on. This control flow will be described below.
  • Step S 1 A power supply voltage is detected in response to the insertion of the memory card into the connector.
  • Step S 2 An ID code of the memory card is read out, and the storage capacity is read.
  • Step S 3 If an ID code that is not supported by the system is read out, it is rejected.
  • Step S 4 The physical format is checked. The information on the top block of the physical address is read.
  • Step S 5 If a physical format that is not supported by the system is designated, it is rejected.
  • Step S 6 A logical address/physical address translation table is prepared, and an erased area for write in the next write operation is selected.
  • FIG. 33 shows a flow chart for preparing a logical address/physical address translation table, which is shown at the aforementioned step S 6 and which is prepared when a power supply is turned on.
  • Step S 1 A RAM area for storing therein a logical address/physical address translation table is reset.
  • Step S 2 A table area for storing therein an erased block used for the next data writing is reset.
  • Step S 3 Search is started from physical block 1 .
  • Step S 4 A redundant division of the block is read out.
  • Step S 5 On the basis of data in a predetermined area, it is determined whether the block is a normal block. When the block is a defective block, the subsequent processing is not required, and the search for the next block is carried out.
  • Step S 6 It is determined whether the area is an erased area.
  • Step S 7 If it is an erased block, it is stored in the table as a proposed block used when the next write is carried out.
  • Step S 8 If it is not the erased area, a logical address information area is extracted. At this time, a parity check is carried out to check validity.
  • Step S 9 On the basis of the above contents, a logical address/physical address translation table is prepared.
  • Step S 10 Physical block number is counted up.
  • Step S 11 After 512 blocks are searched, the routine ends.
  • FIG. 34 shows a logical address/physical address translation table prepared according to the aforementioned flow.
  • the term “physical block area” shown in FIG. 34 means an aggregate of two continuous physical blocks.
  • physical block area 0 indicates physical block 0 and physical block 1 .
  • one physical block area is allocated to one logical block.
  • physical block area 5 is allocated to logical block 0
  • actual data of logical block 0 are stored in physical block 10 or 11 . Therefore, when access is actually carried out, it is required to search for a data area indicating the relationship with the logical addresses of the redundant divisions of the physical blocks 10 and 11 to determine which actually stores data of the logical block 0 .
  • it is enough to read only a very limited area there is little influence on the performance of memory access.
  • 256 (512/2) physical block areas exist as a whole and can be described by 8-bit address data. If the table is formed so that an address offset directly indicates a physical block for convenience of the software, it is required to provide 1 byte for one block, i.e., a RAM area of 0.5 KB in total. For example, the address of a physical block area storing therein the information on physical block 5 is stored in the fifth byte offset from the top of the table.
  • the 0.5 KB RAM capacity is a half of 1 KB RAM capacity conventionally required.
  • a general purpose CPU has a RAM area of about 1 KB. According to this preferred embodiment, it is possible to reduce the RAM area by 0.5 KB to obtain a great advantage.
  • the present invention should not be limited to the above described preferred embodiment, but four physical blocks may be defined as a physical block area, or a larger number of blocks may be supposed.
  • FIG. 35 is an operation flow chart of a memory card in this preferred embodiment when read-out is carried out. This flow will be described below.
  • Step S 1 The top sector address for carrying out read-out and the number of transfer sectors are received from the host.
  • Step S 2 It is verified whether the read-out range is a valid range.
  • Step S 3 The sector is converted to a logical block. Since one block comprises 8 sectors in the case of a 16-Mbit product, one block is divided by 8.
  • Step S 4 A physical block area, in which the corresponding logical block exists, is obtained by referring to a logical address/physical address translation table.
  • Step S 5 The logical address information areas of the physical blocks in the block area are examined, and it is determined which physical block stores therein data of the designated logical block.
  • Step S 6 The data of one sector is read out of the identified physical block. For example, when the sector number is 0, the data of the top 2 pages of the physical block are read, and when the sector number is 7, the data of the end 2 pages of the physical block are read. In one physical block, the data of 8 sectors are arranged in sequence.
  • Step S 7 Error check for the read-out data is carried out, and the presence of an error is checked.
  • Step S 8 When an error is detected, it is determined whether the data can be corrected.
  • Step S 9 When an error is detected and when the data can be corrected, the data is corrected.
  • Step S 10 After the number of sectors required by the host are read out, the routine ends.
  • Step S 11 It is determined whether the next read sector exceeds the boundary between physical blocks. For example, when the sector is transferred from sector 7 to sector 8 , data exist in different physical blocks, so that the system refers to the logical address/physical address translation table again.
  • Step S 12 When the read-out is continued in the same block, pages to be read out are counted up.
  • Step S 13 When the block to be read is moved to another block, the logical block is counted up, and the count of pages is also reset.
  • the write operation is roughly divided into the following three processes.
  • the case where, e.g., logical sector 3 , is rewritten will be described.
  • sector 3 is updated, the processing for one block is required since data of 8 sectors, sectors 0 to 7 , exist in the same block.
  • the operations of copy/updated data writing/copy for one block are basically carried out.
  • a write in sectors 0 through 7 is carried out, all the data of one block are updated, so that the copying operation is clearly unnecessary.
  • the branch of the undermentioned flow chart proceeds mainly while determining whether a sector to be written is a data update or copying operation.
  • FIG. 36 is an operation flow chart of a flash memory card in this preferred embodiment when write is carried out. This flow will be described below.
  • Step S 1 The top sector address for carrying out read-out and the number of transfer sectors are received from the host.
  • Step S 2 This information is converted to a logical block number to refer to the logical address/physical address translation table. An actual physical block is selected from two blocks of a logical block area similar to the read-out operation. Data to be copied are read out from the selected block.
  • Step S 3 The processing is started from the top of the physical block.
  • Step S 4 It is determined whether the copying operation in the first half of the block or the update of data is carried out.
  • Step S 5 When the copying operation is carried out, data are read out of the original block, and write is carried out in a new block.
  • Step S 6 The processing for the next sector is carried out.
  • Step S 7 When it is determined at step S 4 that an update of data is carried out the write is carried out on the basis of the updated data received from the host.
  • Step S 8 The processing for the next sector is carried out.
  • Step S 9 It is checked whether write has been carried out for the number of sectors required by the host.
  • Step S 10 If it is determined at step S 9 that write has been carried out for the required number of sectors, it is determined whether the process has reached the boundary between blocks. If unwritten areas remain, the copying operation on the second half of the block is carried out. If the process has reached the boundary between blocks, it is not required to carry out the copying operation.
  • Step S 11 Data are read from the original block to be written in a new block.
  • Step S 12 The processing for the next sector is carried out.
  • Step S 13 When it is determined at step S 9 that the number of sectors required to be written by the host is not completed, it is required to carry out more writing. However, when the process reaches the boundary between blocks, the processing for the next physical block is carried out.
  • Step S 14 Before the processing for the next block is carried out or before the processing is completed, the logical address/physical address translation table is updated on the basis of the written results, and the physical block, in which the original data exist, is erased. In addition, a proposed area is registered as a new write area for the next processing.
  • Step S 15 The processing for the next block is carried out.
  • the memory system control method in this preferred embodiment it is possible to considerably reduce the RAM area.
  • the RAM capacity required for the logical address/physical address translation table is large, and the system cannot be constructed only by the integrated RAM of the general purpose CPU, so that the external RAM must be provided.
  • the memory card system of the present invention uses a flash memory control method, which can reduce the RAM capacity required for the logical address/physical address translation table and which can control only the integrated RAM of the general purpose CPU.
  • the external RAM which has been conventionally required, is not required, so that it is possible to considerably reduce costs.
  • a flash memory card which can carry out the write operation of data at a higher speed when it is used in the DOS format.
  • Parts (a) and (b) of FIG. 37 show DOS format parameters in this preferred embodiment, wherein part (a) shows the case of a cluster size of 4 KB, and part (b) shows the case of a cluster size of 8 KB.
  • the cluster size is 4 KB
  • a master boot sector is arranged in logical master 0
  • a boot sector is arranged in logical sector 19 .
  • FATs are arranged in logical sectors 20 through 25
  • the copies of the FATs are arranged in logical sectors 26 through 31 .
  • directories are arranged in logical sectors 32 through 47
  • file data areas are arranged in and after logical sector 48 .
  • a master boot sector is arranged in logical sector 0
  • a boot sector is arranged in logical sector 25
  • FATs are arranged in logical sectors 26 through 28
  • the copies of the FATs are arranged in logical sectors 29 through 31
  • directories are arranged in logical sectors 32 through 47
  • file data areas are arranged in and after logical sector 49 .
  • the rewriting speed of the conventional memory system shown in FIGS. 12 through 15 can be compared with the present strategy for the case of a cluster size of 4 KB.
  • the write operation for 48 sectors and the erase operation for three blocks are carried out to update 8 KB data.
  • the write operation for 32 sectors and the erase operation for two blocks are carried out to achieve the same result, so that the rewriting time can be reduced to two thirds.
  • the conventional memory system performs a write operation for 32 sectors and an erase operation for two blocks to update 8 KB data
  • the present memory performs a write operation for 16 sectors and an erase operation for one block, so that the rewriting time can be reduced to half according to the present invention.
  • the cluster delimiter serving as a unit of file management on the DOS does not straddle the boundary between physical blocks of the flash memory, so that the rewriting speed can be increased.
  • the write operation for 32 sectors and the erase operation for two blocks are carried out in order to update 8 KB data
  • the write operation for 16 sectors and the erase operation for one block are carried out in order to update 8 KB data, so that the rewriting time is reduced to half. That is, if the size of a cluster is the same size as that of a physical block, a high-speed write can be achieved. Also, if the size of a cluster is an integer times as large as the size of a physical block, the same advantage can be obtained.
  • a flash memory card which improves the processing speed when a write command is executed after erase.
  • a mark indicating that a corresponding file is invalid is put on a directory, and a memory area having been occupied by the corresponding file is not only open on a FAT (file allocation table), but the data division of the file body is erased on the flash memory. That is, when the erase of the file is commanded, the erase operation from the open cluster area is carried out.
  • FIG. 41 shows the relationship between a management area and a data area when an erase command is executed.
  • the File- 1 and File- 4 in the management area are open, and a del.mark is stored.
  • the areas having stored File- 1 and File- 4 in the data area are erased.
  • the advantage of this preferred embodiment is most remarkable when the cluster size is the same as the block size of the flash memory.
  • the cluster size is smaller than the block size of the flash memory, a part of a block is erased.
  • the processing is not only complicated, but there are also some cases where it is not possible to erase only a part of a block on the specification.
  • the cluster size is the same as the block size, the cluster can be open by simply erasing the block. Also if the cluster size is an integer times as large as the physical block size, the same advantage can be obtained.
  • the present invention can be embodied in various ways without departing from the principle of the invention.
  • the execution timing for erase operation should not be limited thereto.
  • the erase operation from all the clusters may be executed when the format operation is executed.
  • the memory card may be shipped after erasing the cluster in the data area. When a memory card is shipped, a shipping test for the memory card is generally carried out.
  • directories and FATs are rewritten when this test is completed, it is not only possible to obtain no file state, but it is also possible to achieve a high-speed file writing without the need of end user's correction when the memory card is delivered to the end user if the erase operation from the data area is executed according to this preferred embodiment.
  • the memory card is used for a digital still camera or the like
  • operations including the erase of a file in the camera and the erase of an image file, such as a reformat are carried out, if a cluster area storing the file body is erased simultaneously when the FAT and so forth are rewritten, it is possible to achieve a high-speed writing in the subsequent image writing sequence, so that it is possible to continuously take pictures and to capture moving pictures.
  • the cluster size is an integer times as large as the block size and if the cluster delimiter is coincident with the delimiter of block size, the file body part can be easily erased.
  • unnecessary file parts may be automatically erased when a power supply is turned on in a digital camera or the like.
  • the cluster delimiter serving as a unit of file management on the DOS does not straddle the boundary between physical blocks, so that it is possible to reduce the numbers of erase and write operations occurring when data are rewritten, thereby increasing the rewriting speed. Moreover, when a flash memory having a limited number of rewrite operations is used, it is possible to increase the life of the memory.
  • a flash memory card which reduces the capacity of a RAM area required for a logical address/physical address translation table.
  • case 1 is a state accessing the first half 250 logical blocks of 500 logical blocks and that case 2 is a state accessing the second half 250 logical blocks of 500 logical blocks.
  • case 1 a table of the first half 250 blocks is held in the logical address/physical address translation table.
  • case 2 a table of the second half 250 blocks is held in the logical address/physical address translation table.
  • FIG. 42( a ) shows the state of the table in the case 1
  • FIG. 42( b ) shows the state of the table in the case 2.
  • the table of the second half 250 logical blocks exists on the table in a certain moment, if the range accessed from the host is a range of the second half 250 through 499 logical blocks, it is possible to search the correspondence between the logical blocks and the physical blocks by using the existing table.
  • the table of the first half 250 logical blocks exists on the table in a certain moment, if the range accessed from the host is a range of the second half 250 through 499 logical blocks, it is not possible to search the correspondence between the logical blocks and the physical blocks by using the existing table. Therefore, in this case, the logical address/physical address translation table corresponding to the second half 250 logical blocks is remade. This needs to refer to all the areas of the flash memory again.
  • the table of the second half 250 logical blocks exists on the table in a certain moment, if the range accessed from the host is a range of the first half 0 through 249 logical blocks, it is not possible to search the correspondence between the logical blocks and the physical blocks by using the existing table. Therefore, in this case, the logical address/physical address translation table corresponding to the first half 250 logical blocks is remade. This needs to refer to all the areas of the flash memory again.
  • the required logical address/physical address translation table is remade by referring to all the areas of the flash memory again.
  • FIG. 43 shows a flow chart for preparing the logical address/physical address translation table in this case.
  • Step S 0 The presence of a required logical address/physical address translation table is checked at the beginning of access, and if it is required, the routine goes to a table preparing routine.
  • Step S 1 A RAM area for storing therein a logical address/physical address translation table is reset.
  • Step S 2 Search is started from the top of a physical block.
  • Step S 3 The redundant division of the block is read out.
  • Step S 4 On the basis of data in a predetermined area, it is determined whether the block is a normal block.
  • the subsequent processes are not required, and the search for the next block is carried out.
  • Step S 5 It is determined whether the block is an erased area.
  • Step S 6 If the block is an erased area, it is stored on the table as a proposed block used for the next write.
  • Step S 7 If the block is not an erased area, a logical address information area is extracted.
  • Step S 8 A logical address/physical address translation table is prepared on the basis of the above described contents.
  • Step S 9 The physical block number is counted up.
  • the logical address/physical address translation table is prepared if necessary.
  • the flash memory has been divided into two parts, the first and second halves, the present invention should not be limited thereto, but the flash memory may be divided into an optional number of parts.
  • the number and size of divided parts may be optionally determined.
  • the flash memory card serves as a device of the DOS to store therein an image file or the like.
  • file management areas i.e., a master boot sector, a partition boot sector, a file allocation table (FAT) and a directory area.
  • FAT file allocation table
  • directory area i.e., a directory area.
  • file management areas are frequently accessed each time the update or access of the file is carried out.
  • the area corresponding to the file management area is one area, and each of other file data storing areas are divided into two parts.
  • the logical address/physical address translation table corresponding to the file management area may always be held.
  • a flash memory card which reduces the capacity of a RAM area required for a logical address/physical address translation table.
  • a physical block address of the flash memory is divided into a plurality of logical areas (which will be hereinafter referred to as “zones”).
  • the number of physical blocks allocated to each zone i.e., the capacity of each zone, may be uniform or non-uniform.
  • the number of zones may be one or plural.
  • the number of zones may be even or odd.
  • FIG. 44 shows the structure of a physical block and the relationship between logical addresses and physical addresses when a 16-Mbit NAND-type flash memory is divided into two zones having the same capacity.
  • the number of required logical blocks is defined to be 500, and the values of logical block addresses are 0 through 499.
  • the redundant division of a physical block stores therein a logical block address information indicating which logical block the data held in the physical block correspond to.
  • logical block addresses are given from 0 in series for each zone. Therefore, logical block addresses 0 through 249 are allocated to zone 1 , and each of logical blocks corresponds to any one of 256 physical blocks of physical block addresses 0 through 255 .
  • any one of logical block addresses 0 through 249 is stored in the redundant division of a physical block.
  • logical block addresses 250 through 499 are allocated to zone 2 , and each of logical blocks corresponds to any one of 256 physical blocks of physical block addresses 256 through 511 .
  • any one of logical block addresses 0 through 249 is stored in the redundant division of a physical block.
  • An additional write system during data update is executed in a zone including a logical block address to be updated, and a physical block having data corresponding to a certain logical block is not fixed and is always moving in each zone.
  • the logical block address information stored in the redundant divisions of all the physical blocks in the zone is searched, and a translation table between logical blocks and physical blocks is prepared on a system RAM.
  • the zone prepared on the RAM may be any one of the available zones.
  • the FATs and directory areas serving as management information services are arranged in the first zone, so that it is efficient to prepare the table of the top zone.
  • the table prepared on the RAM should not be limited to one zone, but it may be prepared for a plurality of zones so far as the RAM capacity permits.
  • zone 1 An area including physical block addresses 0 through 255 is called zone 1 .
  • zone 2 an area including physical block addresses 256 through 511 is called zone 2 .
  • the host remembers which zone belongs to the translation table currently prepared on the RAM.
  • the translation table on the RAM is zone 1 , no table is prepared.
  • the translation table is updated to get ready for the next access.
  • the translation table in the RAM is for zone 1 , the logical block address information of the redundant divisions of physical block addresses 256 through 511 included in zone 2 is searched, and the translation table of zone 2 is prepared on the RAM.
  • the translation table is updated to get ready for the next access.
  • the translation table is prepared in accordance with the accessed logical block, it is possible to easily reduce the RAM area in comparison with the conventional RAM area. Also, when the table of zone 2 has been prepared on the system RAM, it is possible to easily access a target address by the similar process.
  • the translation table between logical blocks and physical blocks is prepared on the RAM.
  • the values of physical block addresses on the translation table may be 0 through 255.
  • the physical block is actually accessed, it is possible to easily obtain a new physical block address to be inputted to the flash memory by adding 256 x zone number to the physical address on the translation table as OFFSET.
  • a physical block address may be expressed by 8 bits, i.e., 1 byte. Therefore, the RAM capacity, which has conventionally required 1 Kbyte, can be reduced to half. If a logical block address increases, the capacity of the logical address/physical address translation table increases. Therefore, the advantages of this preferred embodiment increases as the capacity of the flash memory increases.
  • a large-capacity flash memory can be controlled by a block address, which can be stored in the Block Address Area of the redundant division of the physical block shown in FIG. 11 . That is, if one zone is divided so as to be formed by a physical block of a block address value, which can be stored in the Block Address Area of the redundant division, it is possible to correspond to a large-capacity logical address.
  • FIG. 46 shows the structure of a physical block when a 16-Mbit NAND-type flash memory is divided into four equal capacity zones.
  • the number of required logical blocks is defined to be 500, so that the values of logical block addresses are 0 through 499.
  • the redundant division of the physical block stores therein a logical block address information indicating which logical block corresponds to the data held in the physical block.
  • the logical block addresses are given in series from 0 for each zone. Physical block addresses corresponding to these logical block addresses are as follows. The logical block addresses 0 through 124 are allocated to zone 1 , and each of the logical blocks corresponds to any one of 128 physical blocks having physical block addresses 0 through 127 .
  • the logical block addresses 125 through 249 are allocated to zone 2 , and each of the logical blocks corresponds to any one of 128 physical blocks having physical block addresses 128 through 255 .
  • the logical block addresses 126 through 374 are allocated to zone 3 , and each of the logical blocks corresponds to any one of 128 physical blocks having physical block addresses 256 through 383 .
  • the logical block addresses 384 through 499 are allocated to zone 4 , and each of the logical blocks corresponds to any one of 128 physical blocks having physical block addresses 384 through 511 .
  • the additional write system during data update is executed only in a zone including a logical block to be written, and a physical block storing therein data corresponding to a certain logical block address is not fixed and is always moving in the zone.
  • the redundant division of each physical block address stores therein a logical block address information indicating which logical block address the data held in the physical block belong to.
  • 125 logical blocks are allocated to each zone, and 128 physical blocks are allocated to each zone. If a physical block is divided into four zones as described above, it is possible to obtain the same advantages as those in the case where a physical block is divided into two zones.
  • the redundancy operation of a 16-Mbit NAND-type flash memory as an example of a conventional flash memory, which does not carry out the division of a physical block into zones, will be described.
  • the 16-Mbit NAND-type flash memory has 512 physical blocks, and as shown in FIG. 3 , the number of logical blocks viewed from the host is defined to be 500.
  • one block must be provided for storing therein information for easily constructing a PC card ATA interface, and one block must be provided for adopting the additional write system. Therefore, 502 blocks must be provided for controlling this flash memory. Accordingly, the 16-Mbit NAND-type flash memory permits the existence of up to 10 defective blocks.
  • the flash memory has the upper limit of the number of rewrite operations.
  • a defective block may be produced therein. Therefore, in order to ensure a sufficient storage region, the flash memory must have a writable and erasable effective block as a block replaced when the defective block is produced.
  • a flash memory having 10 or more defective blocks can not be shipped since it does not have a sufficient capacity when it is shipped, so that the yield of products is decreased.
  • the 16-Mbit NAND-type flash memory is provided with a plurality of redundant blocks in addition to 512 body blocks, in order to ensure a large number of effective blocks to be prepared for the occurrence of defective blocks in the flash memory and in order to improve the yield of products when the flash memory is shipped.
  • These redundant blocks replace defective blocks, which have been produced in the 512 body blocks, by means of a redundant circuit as a hardware when the flash memory is shipped.
  • the redundant circuit When the redundant circuit is used, the redundant blocks are allocated to addresses of defective blocks. After the replacement in hardware, if the address of a defective block is selected, the replacement redundant block is selected. Since the number of the redundant blocks is not infinite, a defective block, which has not been replaced when the flash memory is shipped, is treated as an innate defective block. In addition, a defective block, which is produced when the flash memory is used by the user, is treated as an acquired defective block. These innate and acquired defective blocks are relieved by the effective blocks.
  • the replacement of the body blocks and redundant blocks is usually carried out by means of the redundant circuit as a hardware. As shown in FIG. 47 , the replacement is carried out sequentially from a defective block having a small block address or a large block address for convenience of replacement operation. Therefore, after the replacement of blocks, there are blocks, which can be written or erased sequentially from a small block address or a large block address.
  • the conventional flash memory control method requires 500 logical blocks, one block for storing therein information for easily constructing a PC card ATA interface, and one block as an empty area to use the additional write system, with respect to 512 physical blocks.
  • the conventional flash memory can be controlled if it has 502 rewritable blocks. Therefore, the conventional control method permits 10 defective blocks with respect to 512 physical blocks.
  • zone 1 one block for storing therein information for easily constructing a PC card ATA interface is required, and one block as an empty area is required since the additional write system is adopted. Therefore, with respect to 125 logical blocks, 127 rewritable physical blocks are required, and the number of defective blocks, which are permitted in the zone and which can not be rewritten, is up to 1. With respect to zones 2 through 4 , since additional one block is required as an additionally writing empty block, 126 rewritable physical blocks are required, and the number of defective block, which are permitted in the zone and which can not be rewritten, is up to 2. In order to simplify the explanation, it is assumed that the number of defective blocks permitted with respect to zones 1 through 4 is 1. Therefore, the specification of the number of permitted defective blocks is very severe in comparison with the specification of the number of defective blocks permitted by the conventional control method.
  • the flash memory has four redundant blocks.
  • the defective blocks after replacement are three blocks having physical block addresses 132 , 385 and 389 .
  • These blocks are regarded as innate blocks, and innate defective block marks are put on the redundant divisions of the blocks. Since up to 10 innate defective blocks are permitted in the conventional control method, there is particularly no problem to control the flash memory. However, when it is premised that the flash memory is divided into zones to be controlled, two defective blocks having physical block addresses 385 and 389 exist in zone 4 , so that it is not possible to control the flash memory and it is not possible to ship this product.
  • the defective blocks existing in each zone are replaced with redundant blocks so that all the zones meet the number of effective blocks.
  • FIG. 48 shows the state of a physical block when defective blocks existing in each zone are replaced with redundant blocks so that the number of defective blocks exceeds the number permitted in each zone. Similar to the flash memory shown in FIG. 46 , the flash memory shown in FIG. 48 has seven defective blocks having block addresses 2 , 5 , 129 , 131 , 132 , 385 and 389 and four redundant blocks. It can be seen that it is possible to ship products, which can not be shipped by the conventional replacement method, as good product. An example of replacement procedure will be described below.
  • Various methods for replacing body blocks with redundant blocks may be considered. While the flash memory has been divided into four zones in this preferred embodiment, it may be divided into two zones or odd zones. In addition, the capacities (the number of blocks) of divided zones may be different. In either case, in this preferred embodiment, body blocks are replaced with redundant blocks while monitoring the number of defective blocks existing in each zone so that the number of defective blocks existing in each zone after replacement operation exceeds the number of defective blocks permitted in each zone, and the replacement procedure may be modified without departing from the principle of the invention.
  • a table is prepared every zone serving as an object to be accessed unlike conventional methods using a table wherein logical blocks have a one to one correspondence with physical blocks of a flash memory, it is possible to reduce a RAM area required for the table, and it is possible to control a memory only by means of an integrated RAM of a general purpose CPU although this can not be achieved by the conventional method, so that it is possible to considerably reduce costs in comparison with conventional methods using an external RAM.
  • these advantages are remarkable when a non-volatile semiconductor memory having a large number of physical blocks is controlled.
  • a flash memory for use in a system using a control method for allocating a plurality of physical blocks to a plurality of logical zones to prepare a translation table between logical blocks and the physical blocks every zone to carry out memory access
  • defective blocks of the body are replaced with redundant blocks while monitoring the number of defective blocks existing in each zone so that the number of defective blocks existing in each zone after replacement exceeds the number of defective blocks permitted for each zone, although the replacement has been carried out sequentially from a small (or large) address in conventional methods. Therefore, it is possible to reduce the number of products, which can not be shipped since the products have unusable zones, so that it is possible to improve the yield of products.
  • a flash memory card which can support any types of binary/multi-valued flash memories.
  • FIG. 49 is a schematic view of a PC card adapter, to which this preferred embodiment is applied.
  • a flash memory card 101 A is a binary flash memory card shown in FIG. 17
  • a flash memory card 101 B is a four-valued flash memory card shown in FIG. 18 .
  • a PC card adapter 102 is provided for transferring the data of the flash memory cards 101 A and 101 B.
  • the PC card adapter 102 has a 68-pin connector for a PC card slot, and a connector for a flash memory card.
  • a controller 103 for controlling the flash memory card and for electrically interfacing with the PC card slot, an oscillator 104 for a CPU provided in the controller, a RAM 105 for buffer and so forth.
  • the controller 103 includes circuits for two types of error correction, an ECC circuit 1 and an ECC circuit 2 , which are directly related to the present invention.
  • FIG. 50 shows an example of control when a power supply is turned on.
  • Step S 1 Detection of Power Supply Voltage
  • the memory cards include a product having a power supply voltage of 5 V and a product having a power supply voltage of 3.3 V.
  • a power supply voltage of 5 V is applied to a memory card operated by a power supply voltage of 3.3 V, there is a possibility that voltage destruction and so forth may be caused. In order to avoid this, the system detects the power supply voltage.
  • flash memory cards having different storage capacities or interface specifications.
  • the system determines a maker code, a device code or the like of the device.
  • the determined code is an unsupported code, a new access is not carried out.
  • a normal power supply voltage is inputted.
  • a first method is a method for determining a device code similar to the above described steps S 2 and S 3 .
  • the device code may be changed by the memory construction.
  • FIG. 51 shows the constructions of data and redundant divisions of a 16-Mbit flash memory in this preferred embodiment.
  • the different point from the conventional memory shown in FIG. 4 is that a User Data Area of the redundant division is allocated to each of an ECC Flag Area and an ECC Area- 3 .
  • three areas of ECC Area- 1 through ECC Area- 3 are defined as areas for storing ECC codes.
  • FIG. 52 a method for using the three areas will be described.
  • Information on the ECC system is stored in an ECC Flag Area byte. For example, in the case of ECC system 1 , “AAh” is defined as Flag data, and in the case of ECC system 2 , “55h” is defined as Flag data.
  • a 3-byte ECC code of an even page data (256 bytes) enters the ECC Area- 1
  • a 3-byte ECC code of an odd page data (256 bytes) enters the ECC Area- 2 .
  • the ECC Area- 3 is null. It is possible to correct a 2-bit error of each of 256 bytes in the ECC Area- 1 and ECC Area- 2 .
  • ECC codes with respect to 512 bytes are dispersed throughout the ECC Area- 1 , ECC Area- 2 and ECC Area- 3 . In this case, it is possible to correct a 2-bit error in 512-byte data.
  • the controller 103 in the PC card adapter 102 of FIG. 49 reads the ECC Flag Area out to determine the ECC system when the flash memory card is inserted into the adapter.
  • the code generation and an error detection circuit which correspond to each of the ECC system 1 and ECC system 2 , are selected.
  • the CPU in the controller controls the flow of data between the host and the RAM for buffer and the flash memory.
  • the ECC circuit need not always be hardware.
  • the generation of the ECC codes and so forth may be carried out by software.
  • the selection of the ECC system includes the selection that no ECC is used.
  • the ECC is not essential. In this case, it may be defined that the ECC is not used if the content of the above described ECC Flag Area is “FFh”.
  • the ECC system has been defined for every flash memory in the above described preferred embodiment, the present invention should not be limited.
  • the ECC system may be switched every optional unit, such as every sector or block.
  • the ECC system may be switched each time access is carried out for each of the above described units without simply determining the ECC system when a power supply is turned on.
  • the operation for converting data read out by the ECC system 1 to the ECC system 2 to restore the converted data may be supposed.
  • the present invention should not be limited thereto.
  • the present invention may be applied to various apparatuses, such as a digital still camera, a PDA, a word processor and a voice recorder.
  • a flash memory card having a very wide use range may be treated by one system, the flexibility can be remarkably improved.
  • the contact and breakaway of various pins are carried out in sequence, so that it is possible to improve the stability of operation and to surely protect data.
  • a plurality of physical blocks are allocated to one logical block, so that it is possible to reduce a RAM area required for the table.
  • the cluster delimiter does not straddle a block serving as an erase unit, so that it is possible to decrease the numbers of erase and write operations occurring when data are rewritten.
  • the management area for physical blocks is open, and simultaneously, the data area is also erased, so that it is possible to improve the processing speed when a subsequent write command is executed.
  • defective blocks are replaced with redundant blocks so that the number of defective blocks existing in each logical zone does not exceed a predetermined value after replacement, so that it is possible to improve the yield of products.
  • the error correction algorithm is selected in accordance with the kind of storage medium used, so that it is possible to use various storage media to improve the flexibility.

Abstract

In a memory system using a storage medium, which is inserted into an electronic apparatus via a connector to add a memory function thereto, the storage medium has a GROUND terminal, a power supply terminal, a control terminal and a data input/output terminal, and the connector has a function of being sequentially connected to each of the terminals. When the storage medium is inserted into the connector, the GROUND terminal and control terminal of the storage medium are connected to corresponding terminals of the connector before the power supply terminal and data input/output terminal of the storage medium are connected to corresponding terminals of the connector. Thus, it is possible to improve the stability when a memory card is inserted into or ejected from the memory system.

Description

    RELATED APPLICATION
  • This is a continuation of continuation application Ser. No. 11/931,101, filed Oct. 31, 2007, which is a continuation of divisional application Ser. No. 10/913,865, filed Aug. 6, 2004, which is a divisional of application Ser. No. 09/130,818, filed Aug. 7, 1998, which application is hereby incorporated in its entirety. Parent application Ser. No. 09/130,818 claimed priority under 35 U.S.C. §119 from Japanese patent application Serial No. 214561/1997 filed Aug. 8, 1997 and Japanese patent application Serial No. 119099/1998 filed Apr. 28, 1998. The present divisional application claims priority from these Japanese applications under 35 U.S.C. §§119-120.
  • BACKGROUND OF THE INVENTION
  • 1. Field of the Invention
  • The present invention relates generally to a method for controlling a memory system. More specifically, the invention relates to a method for controlling a non-volatile semiconductor memory system, such as a non-volatile semiconductor memory card.
  • 2. Related Background Art
  • In recent years, a flash memory card like that shown in FIG. 1 has widely used as a storage medium for a portable information apparatus, such as a digital still camera or a PDA. NAND-type flash memories have been used and sold.
  • As shown in FIG. 1, a flash memory card of this type has a thin plastic package 1 having a slightly recessed portion, in which a flash memory 2 having a flat electrode with 22 pins is mounted. This flash memory card is electrically connected to a host system via a dedicated connector to input and output data.
  • FIG. 2 shows, as an example of a flash memory, a 16-Mbit NAND-type flash memory divided into 512 physical memory blocks. Each of these blocks is a minimum unit for an erase operation. One block is divided into 16 pages. One page is a basic unit of writing and read-out. One page comprises 264 bytes. Among the 264 bytes, 256 bytes are a user data area (data division), and the remaining 8 bytes (redundant division) are used for storing an error correcting code and a management information service.
  • On the other hand, personal computers manage data as logical blocks shown in FIG. 3. Five hundred logical blocks (identified by logical block address (LBA)), are set in FIG. 3. One logical block corresponds to 8 continuous sectors. That is, logical block 0 means logical sectors 0 to 7.
  • In ordinary personal computers, data are managed every sector (512 bytes). Therefore, the memory card stores therein data of one sector of a logical block using 2 pages of a physical block as a pair to carry out data management in units of 512 bytes. FIG. 4 shows an exemplary data storing method.
  • Unused normal blocks of both of the data division and redundant division are set to be “FFh”. The meanings of the respective bytes will be described below. Data Area-1 stores the first half 0 to 255 bytes of the data of 512 bytes. Data Area-2 stores the second half 256 to 511 bytes of the data of 512 bytes. The User Data Area is open to store user data, so that the use thereof is entrusted to the user. A Data Status Area indicates whether data are presumed reliable. Although an “FFh” is usually set, a “00h” is set when defective data are written (e.g., when the desired threshold voltage for the memory cell cannot be achieved or another error condition is associated with the cell). A Block Status Area indicates whether a block is good or defective. Although the “FFh” is usually set, the “00h” (initial defective block) or an “F0h” (acquired defective block) is set in the case of a defective block. Furthermore, the same value is written in the Block Status Area for all data in one block. A Block Address Area-1 indicates a logical address information of a block. Furthermore, since 8 sectors forming one logical block correspond to one of 512 physical blocks, the same values for all the data are written in the same block. Similarly, in a Block Address Area-2, the same contents as the data of the Block Address Area-1 are written. An ECC Area-1 is a 3-byte ECC code of even page data (256 bytes). An ECC Area-2 is a 3-byte ECC code of odd page data (256 bytes).
  • FIG. 5 shows, as another example of a flash memory, a 64-Mbit NAND-type flash memory that is divided into 1024 physical memory blocks. Each of these blocks is a minimum unit of erase. One block is divided into 16 pages. One page is a basic unit of writing and read-out. One page comprises 528 bytes. Among the 528 bytes, 512 bytes are used for a user data area (data division), and the remaining 16 bytes (redundant division) are used for storing an error correcting code and a management information service. One thousand logical blocks are set in FIG. 6. One logical block corresponds to 16 continuous sectors. That is, logical block 0 includes logical sectors 0 to 15. FIG. 7 shows a method for storing data in the 64-Mbit NAND-type flash memory.
  • The control of such a memory card adopts an additional writing system for writing updated data in a previously erased area when data are updated and subsequently erasing an area in which the original data were stored. Therefore, a physical block, in which data corresponding to a certain logical block exist, is not fixed and is always moving in the memory. Also, as shown in FIG. 8, the redundant division of the physical block stores therein a logical block address information indicating which logical block corresponds to the data held in the physical block. The Block Address Area-1 and the Block Address Area-2 in FIGS. 4 and 7 correspond to the corresponding logical addresses.
  • Therefore, the memory system searches areas for storing the logical block address information of all the physical blocks to prepare a translation table between logical blocks and physical blocks on a system RAM, usually when a power supply is turned on. After the tables are prepared once, it is possible to immediately determine the physical block corresponding to the logical block by referring to the tables. The searching operation for all the blocks is carried out once when the power supply is turned on. Naturally, if the position of the corresponding physical block changes after the data are updated, the memory system updates a logical address/physical address translation table to get ready for the next access.
  • However, in conventional memory systems, there is a first problem in that the RAM area required for the logical address/physical address translation table is large. The contents thereof will be described in detail below.
  • FIG. 9 shows a logical address/physical address translation table of a conventional 16-Mbit NAND-type flash memory. As described above, the data of one logical block, i.e., the data of 8 continuous sectors, exist in any one of 512 physical blocks in the flash memory. In order to select one block from the 512 physical blocks, 9 bits are required. If the table is formed so that an offset logical block address directly indicates a physical block for convenience of a software, 2 bytes are required for one logical block, so that a RAM area of 1 KB in total is required. For example, the address of a physical block, in which the information of logical block 5 is stored, is an address offset by 5 words (10 bytes) from the top of the table.
  • Thus, in the conventional method, there is a problem in that the RAM area required for the logical address/physical address translation table is very large. In general, a general purpose CPU has a RAM of about 1 KB on board as an integrated RAM. Therefore, conventionally, the logical address/physical address translation table must use 1 KB, and system configuration requires more than the integrated RAM, so that an external RAM must be provided for the system. This is a great factor in an increase in costs.
  • FIG. 10 shows a logical address/physical address translation table of a conventional 64-Mbit NAND-type flash memory. In this case, the data of one logical block, i.e., the data of 16 continuous sectors, exist in any one of 1024 physical blocks in the flash memory. The selection of one physical block from 1024 physical blocks needs 10 bits, so that a RAM area of 2 KB in total is required. For that reason, a very large RAM area is required similar to the 16-bit NAND-type flash memory.
  • This problem is more serious as the capacity of the flash memory increases. For example, the number of blocks is 8192 for the 1-Gbit capacity memories, so that a RAM capacity of 16 KB is required.
  • If the memory capacity increases more, there is a second problem in that the logical address can not be stored in the redundant division of the physical block of the flash memory. The Block Address Area of the redundant division of a physical block of the 16-Mbit NAND-type flash memory shown in FIG. 7 stores therein a logical block address information indicating which logical block corresponds to the data held in the physical block. FIG. 11 shows the structure of the Block Address Area of the redundant division of each physical block. In FIG. 11, four bits of D4 through D7 of a number 262 byte of an even page and a number 259 byte of an odd page are “0”, “0”, “0” and “1”, and one bit of D0 of a number 263 byte of an even page and a number 260 byte of an odd page has a fixed value “1”. Therefore, the maximum value of a block address capable of being stored is 2047 expressed by BA0 through BA10. Since 4096 physical blocks exist in a 512-Mbit NAND-type flash memory, it is not possible to store the address unless the description method for the Block Address Area is changed. Since the method for controlling the redundant division is different from those for conventional flash memories, it is not possible to control a high-capacity flash memory unless the host prepares two kinds of programs. This is a problem in that there is insufficient capacity in the program storing area of the host system for the necessary programs.
  • The writing and erase for a flash memory will be briefly described below. The writing in a flash memory is carried out page by page. In the case of a 64-Mbit NAND-type EEPROM, one page has 528 bytes. In addition, erase is carried out in units of a block. In the case of the 64-Mbit NAND-type EEPROM, one block is formed by 16 pages. Thus, in the NAND EEPROM, the unit of writing is different from the unit of erase. Therefore, it is not possible to erase only a certain page to update data.
  • When a flash memory card is used for a personal computer, it is generally treated as a drive under the control of the DOS. Parts (a) and (b) of FIG. 12 show conventional DOS format parameters, wherein part (a) shows the parameters in the case of a cluster size of 4 KB and part (b) shows the parameters in the case of a cluster size of 8 KB. The term “cluster” means a basic minimum unit of file management on the DOS. Even when a small file is stored, the file occupies one cluster. When the file size is large, the file is managed as a chain of a plurality of clusters, and its management information service is stored in a FAT (file allocation table). The size of the cluster, the management method for the FAT and so forth are managed in a sector called a boot sector. When one device is managed as a plurality of drives, its information is stored in a master boot sector. In order to carry out the writing in a file, the OS issues a write command for each cluster.
  • Part (a) of FIG. 12 shows the case of a cluster size of 4 KB. A master boot sector is arranged in logical sector 0, and a boot sector is arranged in logical sector 16. In addition, FATs are arranged in logical sectors 17 through 22, and the copies of the FATs are arranged in logical sectors 23 through 28. Moreover, directories are arranged in logical sectors 29 through 44, and file data areas are arranged in and after logical sector 45.
  • Part (b) of FIG. 12 shows the case of cluster size of 8 KB. A master boot sector is arranged in logical sector 0, and a boot sector is arranged in logical sector 16. In addition, FATs are arranged in logical sectors 17 through 19, and the copies of the FATs are arranged in logical sectors 20 through 22. Moreover, directories are arranged in logical sectors 23 through 38, and file data areas are arranged in and after logical sector 39.
  • Referring to FIG. 13, a conventional rewrite sequence in the case of a cluster size of 4 KB will be described. Since the cluster size is 4 KB, a write command for continuous 8 sectors is issued from the OS. At this time, the writing (data update) in logical sectors 45 through 52 (cluster A) is carried out.
  • (1) An erased new area is searched, and logical sectors 32 through 44 are copied on the new area NAND Block C from the original block.
    (2) The new data of logical sectors 45 through 47 are written in the new area NAND Block C.
    (3) The original block NAND Block A is erased.
    (4) The logical address/physical address translation table is updated.
    (5) An erased new area is searched, and the new data of logical sectors 48 through 52 are written in the new area NAND Block D.
    (6) The data of logical sectors 53 through 63 of the original block NAND Block B are copied on the new area NAND Block D.
    (7) The original block NAND Block B is erased.
    (8) The logical address/physical address translation table is updated.
  • Therefore, when 8 sectors are rewritten, as viewed from the outside, logical sectors 32 through 63, i.e., 32 sectors in total (32 pages) are written, and the NAND Block A and the NAND Block B, i.e., 2 blocks in total, are erased within the actual device.
  • Referring to FIG. 14, a writing sequence in cluster B will be described below. In this case, the writing (data update) in logical sectors 53 through 60 (cluster B) is carried out.
  • (1) An erased new area is searched, and logical sectors 48 through 52 are copied on the new area NAND Block C from the original block NAND Block B.
    (2) The new data of logical sectors 53 through 60 are written in the new area NAND Block C.
    (3) Logical sectors 61 through 63 are copied on the new area NAND Block C from the original block NAND Block B.
    (4) The original block NAND Block B is erased.
    (5) The logical address/physical address translation table is updated.
    Therefore, when 8 sectors are rewritten, as viewed from the outside, logical sectors 48 through 63, i.e., 16 sectors in total (16 pages), are written and the NAND Block A, i.e., one block, is erased within the actual device.
  • Referring to FIG. 15, a conventional rewriting sequence in the case of a cluster size of 8 KB will be described. Since the cluster size is 8 KB, a write command for 16 continuous sectors is issued from the OS. At this time, the writing (data update) in logical sectors 39 through 54 (cluster A) is carried out.
  • (1) An erased new area is searched, and logical sectors 32 through 38 are copied on the new area NAND Block C from the original block NAND Block A.
    (2) The new data of logical sectors 39 through 47 are written in the new area NAND Block C.
    (3) The original block NAND Block A is erased.
    (4) The logical address/physical address translation table is updated.
    (5) An erased new area is searched, and the new data of logical sectors 48 through 54 are written in the new area NAND Block D.
    (6) The data of logical sectors 55 through 63 of the original block NAND Block B are copied on the new area NAND Block D.
    (7) The original block NAND block B is erased.
    (8) The logical address/physical address translation table is updated.
    Therefore, when 16 sectors are rewritten if viewed from the outside, the write operation in logical sectors 32 through 63, i.e., 32 sectors in total (32 pages), and the erase operation from the NAND Block A and the NAND Block B, i.e., 2 blocks in total, are carried out in an actual device.
  • Comparing the cluster of 4 KB with the cluster of 8 KB when the same 8 KB data are written, in the case of the cluster size of 4 KB, the processing is divided into two write operations, so that the write operation consists of 48 sectors in total and an erase operation for three blocks is carried out. On the other hand, comparing with the cluster size of 8 KB in the case of the cluster size of 8 KB, the processing is concentrated on one writing, so that the write operation consists of 32 sectors in total and an erase operation for two blocks is carried out.
  • Thus, in the conventional memory system, when viewed from the outside, the number of the write and erase operations actually executed in the device is far greater than the number of updated sectors, so that there is the second problem in that the rewriting speed viewed from the outside decreases.
  • The operation of the conventional memory system when executing a file erase command will be described below. In an ordinary DOS file system, when the file erase command is executed, a mark indicating that a corresponding file is invalid is put on a directory, and a memory area having been occupied by the corresponding file is identified as open on a FAT (file allocation table). Therefore, the data division of the file body remains on the flash memory without being erased. FIG. 16 shows the relationship between a management area and a data area when an erase command is executed. In FIG. 16, for example, when erase commands for File-1 and File-4 are executed, the File-1 and File-4 are open, and a del.mark is stored. At this time, the File-1 and the File-4 are not erased from the data area.
  • For that reason, when a subsequent write command is executed, it is first required to carry out the erase operation of the flash memory when a data division of a new file is written in the open area. Consequently, the erase operation of the flash memory must be always carried out when the file writing is carried out, so that there is a third problem in that the file writing speed deteriorates.
  • The ECC Area-1 shown in FIG. 4 is a 3-byte ECC code of an even page data (256 bytes). The ECC Area-2 is a 3-byte ECC code of an odd page data (256 bytes).
  • The ECC (error correcting code) means a code for correcting an error. The system utilizes this error correcting code to determine whether a read data has an error. When an error exists, the system can correct the error. The required error correcting capability depends on the reliability of the flash memory itself, e.g., the cell structure of the memory. Flash memories have a plurality of data storing methods. When these flash memories are used for a system, such as a digital still camera and a PDA, error correction will be considered.
  • For example, a first flash memory card holds binary values “0” and “1” corresponding to the different threshold values of a memory cell shown in FIG. 17. The memory card can therefore use a 1-bit error correcting code to provide ECC for one page (256 bytes). A second flash memory card holds four-valued values “00”, “01”, “10” and “11” (2 bits) corresponding to the multiple values of the threshold of a memory cell shown in FIG. 18. The memory card that stores multiple bits per cell, illustrated in FIG. 18, therefore uses a 2-bit error correcting code to provide ECC for one page (256 bytes) since there is a possibility that 2-bit data may be erroneous when one memory cell is defective. The algorithm for generating a code and detecting and correcting an error for 1-bit error correction is different from that for 2-bit error correction.
  • Conventional systems (e.g., digital still cameras, PDAs) have only one kind of error correcting algorithm on board. For that reason, there is a fourth problem in that it is possible to read only one of the above described first (single bit) and second (multi-bit storage per cell) memory cards. This is an obstacle to the enhancement of flexibility of flash memories on the market.
  • SUMMARY OF THE INVENTION
  • It is therefore an object of the present invention to eliminate the aforementioned problems and to provide a method for controlling a memory system, which improve the stability of operation when the insertion/ejection of a memory card is carried out.
  • It is another object of the present invention to provide a method for reducing the capacity of a RAM required for a logical address/physical address translation table to control a flash memory by means of only an integrated RAM of a general purpose CPU, thereby dispensing with an external RAM, which has been conventionally required, to considerably reduce costs.
  • It is another object of the present invention to provide a method for controlling a memory system so as to prevent the delimiter of a cluster serving as a basic unit of file management on the DOS from straddling a block serving as a unit of erase, thereby providing a high-speed data writing.
  • It is a further object of the present invention to open a management area for a physical block while erasing its data area when an erase command is executed, thereby improving a processing speed when a subsequent write command is executed.
  • It is a still further object of the present invention to provide a system capable of supporting any one of two kinds of flash memories, such as binary/multi-valued flash memories, or a plurality of flash memories.
  • In order to accomplish the aforementioned and other objects, according to the present invention, there is provided a method for controlling a memory system, which uses the terminals of a connector sequentially connected to the terminals of a memory card (a storage medium), which is disclosed in Japanese Patent Laid-Open Nos. 8-90969 and 8-202509.
  • According to a first aspect of the present invention, there is provided a method for controlling a memory system using a storage medium, which is inserted into an electronic apparatus via a connector to add a memory function thereto, said storage medium having ground terminals, power supply terminals, control terminals and data input/output terminals, said connector having a function of being sequentially connected to each of said terminals, wherein when said storage medium is inserted into said connector, said ground terminals and control terminals of said storage medium are connected to corresponding terminals of said connector before said power supply terminals and data input/output terminals of said storage medium are connected to corresponding terminals of said connector.
  • According to a second aspect of the present invention, there is which is inserted into an electronic apparatus via a connector to add a memory-function thereto, said storage medium having ground terminals, power supply terminals, control terminals and data input/output terminals, said connector having a function of being sequentially disconnected from each of said terminals wherein when said storage medium is ejected from said connector, said power supply terminals and data input/output terminals of said storage medium are disconnected from corresponding terminals of said connector before said ground terminals and control terminals of said storage medium are disconnected from corresponding terminals of said connector.
  • According to a third aspect of the present invention, there is provided a method for controlling a memory system using a storage medium, which is inserted into an electronic apparatus via a connector to add a memory function thereto, said storage medium having ground terminals, power supply terminals, a first control terminal, a second control terminal and data input/output terminals, said connector having a function of being sequentially connected to each of said terminals, wherein when said storage medium is inserted into said connector, said ground terminals and first control terminals of said storage medium are connected to corresponding terminals of said connector before said power supply terminals and data input/output terminal of said storage medium are connected to corresponding terminals of said connector, and said second control terminal being connected to a corresponding terminal of said connector before said data input/output terminals are connected to a corresponding terminal of said connector.
  • According to a fourth aspect of the present invention, there is provided a method for controlling a memory system using a storage medium, which is inserted into an electronic apparatus via a connector to add a memory function thereto, said storage medium having ground terminals, a power supply terminals, a first control terminal, a second control terminal and data input/output terminals, said connector having a function of being sequentially connected to each of said terminals, wherein when said storage medium is ejected from said connector, said data input/output terminals are disconnected from corresponding terminals of said connector before said second control terminal is disconnected from a corresponding terminal of said connector, and said power supply terminals and data input/output terminals of said storage medium being disconnected from corresponding terminals of said connector before said ground terminals and first control terminal of said storage medium are disconnected from corresponding terminals of said connector.
  • According to a fifth aspect of the present invention, there is provided a method for controlling a memory system having a storage medium, which is inserted into an electronic apparatus via a connector to add a memory function thereto, said storage medium having ground terminals, power supply terminals, first control terminal, a second control terminal, data input/output terminals and an insertion/ejection detecting terminal, said connector having a function of being sequentially connected to each of said terminals, wherein when said storage medium is inserted into said connector, said ground terminals and first control terminals of said storage medium are connected to corresponding terminals of said connector before said power supply terminals and data input/output terminals of said storage medium are connected to corresponding terminals of said connector, said second control terminal of said storage medium being connected to a corresponding terminal of said connector before said data input/output terminals of said storage medium are connected to corresponding terminals of said connector, and said insertion/ejection detecting terminal being connected to a corresponding terminal of said connector after all of said terminals of said storage medium are inserted.
  • According to a sixth aspect of the present invention, there is provided a method for controlling a memory system having a storage medium which is inserted into an electronic apparatus via a connector to add a memory function thereto, said storage medium having ground terminals, power supply terminals, a first control terminal, a second control terminal, data input/output terminals and an insertion/ejection detecting terminal, said connector having a function of being sequentially connected to each of said terminals, wherein when said storage medium is ejected from said connector, said insertion/ejection detecting terminal is disconnected from a corresponding terminal of said connector before said data input/output terminals of said storage medium are disconnected from corresponding terminals of said connector, said data input/output terminals of said storage medium being disconnected from corresponding terminals of said connector before said power supply terminals and second control terminal of said storage medium are disconnected from corresponding terminals of said connector, and said power supply terminals of said storage medium being disconnected from corresponding terminals of said connector before said ground terminals and first control terminal of said storage medium are disconnected from corresponding terminals of said connector.
  • According to a seventh aspect of the present invention, there is provided a method for controlling a memory system which includes:
  • logical blocks managed by the system; physical blocks for storing therein data corresponding to said logical blocks, said physical blocks comprising a plurality of memory cells; redundant divisions included in a corresponding said physical blocks for storing therein addresses of corresponding said logical blocks; and physical block areas formed by at least two of said physical blocks, wherein a logical address/physical address translation table is prepared for managing corresponding relationships between said logical blocks and said physical blocks.
  • According to an eighth aspect of the present invention, there is provided a method for controlling a memory system, which is a semiconductor memory system for storing a file managed according to predetermined units, in a storage area divided into second predetermined units.
  • According to a ninth aspect of the present invention, there is provided a method for controlling a memory system which includes:
  • files managed by the system; data areas for storing therein the contents of said files; and management areas for storing therein a corresponding relationships between said files and said data areas, wherein when said file is erased, it is marked that said data areas corresponding to said management areas are empty area, to erase said corresponding data areas.
  • According to a tenth aspect of the present invention, there is provided a method for controlling a memory system which includes:
  • files managed by the system; data areas for storing therein the contents of said files; and management areas for storing therein a corresponding relationships between said files and said data areas, wherein when said file is erased, it is marked that said data areas corresponding to said management areas are empty area, and the contents of said management areas are detected, on the basis of signals inputted to said memory system, to erase said data areas.
  • According to an eleventh aspect of the present invention, there is provided a method for controlling a non-volatile semiconductor memory system, which comprises:
  • dividing a cell array into a plurality of physical blocks; storing each information corresponding to relationship between said physical block and logical block which is managed by said system, in each said physical block; and in order to form a table for managing corresponding relationships between said logical blocks and said physical blocks, in a random access memory in said system, sequentially preparing required corresponding relationships of corresponding relationships between said logical blocks and said physical blocks, in said random access memory in said system in accordance with accesses from a host.
  • According to a twelfth aspect of the present invention, there is provided a method for controlling a non-volatile semiconductor memory system which comprises:
  • dividing a cell array of non-volatile semiconductor memory cells into a plurality of physical blocks; storing each information corresponding to relationship between said physical block and logical block address in each said physical block, which logical blocks are managed by said system, in order to form an address translation table for managing corresponding relationships between said physical blocks and said logical block addresses, in a random access memory in said memory system, forming a plurality of areas, each area being formed by an aggregate of at least one of said plurality of physical blocks, controlling said system so that data in predetermined address of said logical block are stored in said predetermined area, forming an address translation table corresponding to said predetermined area in which data in said predetermined address of said logical block are stored, if necessary, when said non-volatile semiconductor memory is accessed.
  • According to a thirteenth aspect of the present invention, there is provided a method for controlling a non-volatile semiconductor memory system which comprises:
  • dividing a cell array of non-volatile memory cells into a plurality of physical blocks; storing each information corresponding to relationship between said physical block and logical block in a storage region of each of said physical blocks, which logical blocks are managed by said system; and forming a table for managing a corresponding relationship between said logical blocks and said physical blocks of a flash memory. In a random access memory of said memory system, said method further comprising:
  • ensuring an area formed by one or a plurality of physical blocks, on a cell array of said flash memory; in every memory access time, searching said object area of physical blocks, forming said table for managing said corresponding relationship between said logical blocks and said physical blocks, on said random access memory of said system, allowing to select physical blocks corresponding to said logical blocks by using said table.
  • According to a further aspect of the present invention, there is provided a method for controlling a non-volatile semiconductor memory system, which further comprises:
  • providing a function of selectively replacing defective physical blocks including defective cells with redundant physical blocks; and
  • managing said functions, for said each area, so that the number of defective physical blocks is less than or equal to a predetermined number.
  • According to a fifteenth further aspect of the present invention, there is provided a method for controlling a non-volatile semiconductor memory system, which allows any one of various memory units to be detachably mounted in a body of said memory system, said method comprising selecting a corresponding one of various error correcting means of said body in accordance with said one of various memory units mounted in said body, to carry out error correction.
  • BRIEF DESCRIPTION OF THE DRAWINGS
  • The present invention will be understood more fully from the detailed description given herebelow and from the accompanying drawings of the preferred embodiments of the invention. However, the drawings are not intended to imply limitation of the invention to a specific embodiment, but are for explanation and understanding only.
  • In the drawings:
  • FIG. 1 is an outside drawing of a memory card;
  • FIG. 2 is a table showing a physical block construction of a flash memory;
  • FIG. 3 is a table showing data in a personal computer;
  • FIG. 4 is a table showing a method for storing data in a flash memory;
  • FIG. 5 is a table showing a physical block construction of a flash memory;
  • FIG. 6 is a table showing a physical block construction of a flash memory;
  • FIG. 7 is a table showing a method for storing data in a flash memory;
  • FIG. 8 is a conventional logical/physical block translation table;
  • FIG. 9 is a logical address/physical address translation table in a conventional memory system;
  • FIG. 10 is a logical address/physical address translation table in a 64-Mbit NAND-type flash memory system;
  • FIG. 11 is a table showing a data configuration of a block address of a flash memory;
  • FIG. 12 is a table showing parameters in a conventional DOS format;
  • FIG. 13 is a chart showing a conventional rewrite sequence;
  • FIG. 14 is a chart showing a conventional rewrite sequence;
  • FIG. 15 is a chart showing a conventional rewrite sequence;
  • FIG. 16 is a table showing a conventional relationship between a management area and a data area when am erase command is executed;
  • FIG. 17 shows an example of how single-bit per cell data are stored in a flash memory;
  • FIG. 18 shows an example of how multi-bit per cell data are stored in a flash memory;
  • FIG. 19 is a table showing a conventional system;
  • FIG. 20 is an outside drawing of a memory card for use in a memory system according to the present invention;
  • FIG. 21 is a table showing signal names of the respective terminals of the card of FIG. 20;
  • FIG. 22 is a main flow chart of a memory system using a method for controlling a memory system according to the present invention;
  • FIGS. 23( a) and 23(b) are a schematic view showing the relationship between a power supply voltage and an appearance of a memory card;
  • FIGS. 24( a) and 24(b) are a schematic view showing a method for electrically detecting a power supply voltage of a memory card;
  • FIG. 25 is an outside drawing of a 5 V dedicated connector;
  • FIG. 26 is an outside drawing of a 3.3 V dedicated connector;
  • FIG. 27 is an outside drawing of a 5 V/3.3 V dedicated connector;
  • FIG. 28 is a schematic view showing a method for detecting the insertion/ejection of a memory card;
  • FIG. 29 is a schematic view showing a method for detecting the insertion/ejection of a memory card when the memory card corresponds to a PC card adapter;
  • FIG. 30 is a table showing an ECC data construction in a memory system according to the present invention;
  • FIG. 31 is a view showing conditions for calculating an ECC code in a memory system according to the present invention;
  • FIG. 32 is a control flow chart of a memory system according to the present invention when a power supply is turned on;
  • FIG. 33 is a flow chart for preparing a logical address/physical address translation table in a memory system according to the present invention;
  • FIG. 34 is a logical address/physical address translation table in a memory system according to the present invention;
  • FIG. 35 is a flow chart of a memory system according to the present invention when read-out is carried out;
  • FIG. 36 is a flow chart of a memory system according to the present invention writing is carried out;
  • FIG. 37 is a view showing parameters of a DOS format according to the present invention;
  • FIG. 38 is a chart showing a rewrite sequence according to the present invention;
  • FIG. 39 is a chart showing a rewrite sequence according to the present invention;
  • FIG. 40 is a chart showing a rewrite sequence according to the present invention;
  • FIG. 41 is a table showing the relationship between a management area and a data area according to the present invention when an erase command is executed;
  • FIGS. 42( a) and 42(b) are a table of an example of a logical block/physical block translation table in a preferred embodiment of the present invention;
  • FIG. 43 is a flow chart of an example of a method for preparing a logic block/physical block translation table in this preferred embodiment;
  • FIG. 44 is a logical/physical block translation table in the preferred embodiment;
  • FIG. 45 is a table showing a method for expressing physical block addresses in the preferred embodiment;
  • FIG. 46 is a table for explaining a conventional embodiment when a defective block is replaced with a redundant block;
  • FIG. 47 is a table for explaining a conventional example of a replacement of a defective block with a redundant block;
  • FIG. 48 is a table for explaining a preferred embodiment when a defective block is replaced with a redundant block in view of a zone division control as a premise;
  • FIG. 49 is a schematic view of a PC card adapter according to the present invention;
  • FIG. 50 is a flow chart when a flash memory card is inserted into a PC card adapter;
  • FIG. 51 is a table showing parameters of another preferred embodiment of a DOS format of a flash memory according to the present invention; and
  • FIG. 52 is a table showing a method for using an ECC code area.
  • DESCRIPTION OF THE PREFERRED EMBODIMENTS
  • Referring now to the accompanying drawings, the preferred embodiments of a memory system control method according to the present invention will be described below.
  • FIG. 20 shows a flash memory card for use in a memory system according to the present invention, and FIG. 21 shows signal names of the respective pins of a flat electrode of the memory card.
  • FIG. 22 shows a main control flow for use in the flash memory card. The operations of the respective steps in this flow will be described below.
  • First, an insertion detecting step A will be described below.
  • As described in the description of the prior art, the start of processing of a memory card assumes that the memory card is normally inserted into a connector. There is a possibility that physical or data destruction may be caused if the system accesses the memory card when the memory card is incompletely inserted. Therefore, in the control flow for the memory system according to the present invention, there is provided means for detecting whether a memory card is normally inserted. This means may be structural or electrical.
  • For example, a detecting switch may be provided in the connector at a position (e.g., the innermost part of the connector) so that the detecting switch can transmit a detection signal to the system after the memory card is held at a normal contact position.
  • Alternatively, there may be provided a system mechanism for ensuring that the memory card is held at a normal position, not for detecting insertion. For example, a lid may be provided in an inlet for the memory card so that the lid pushes the memory card to a home position when the lid is closed.
  • In addition, means for electrically detecting the insertion of the memory cart may be provided.
  • A power supply voltage detecting step B will be described in detail below.
  • The memory cards include a product having a power supply voltage of 5 V and a product having a power supply voltage of 3.3 V. If a voltage of 5 V is applied to a memory card operated by a power supply voltage of 3.3 V, there is a possibility that voltage destruction and so forth may be caused. In order to avoid this, the system is designed to detect a power supply voltage for a memory card. Methods for detecting the power supply voltage may include a method for determining the power supply voltage based on the appearance of the memory card, and a method for electrically determining the power supply voltage.
  • FIGS. 23( a) and 23(b) show an example of a method for determining a power supply voltage for a memory card based on the appearance of the memory card. The 5 V product has a cutout on the left side of the upper end thereof as shown in FIG. 23( a), and the 3.3 V product has a cutout on the right side of the upper end thereof as shown in FIG. 23( b). Thus, it is possible to easily visually determine the power supply voltage for the memory card.
  • FIGS. 24( a) and 24(b) show an example of a method for electrically detecting a power supply voltage for a memory card. In this case, a number 17 pin is used for detecting the power supply voltage. For example, in the 5 V product, the number 17 pin is electrically on the floating condition as shown in FIG. 24( a), and in the 3.3 V product, the number 17 pin is short-circuited to a VCC line on the surface of the memory card.
  • On the system side, the number 17 pin is pulled down to a VSS via a resistor. On the system side, a voltage of 3.3 V is supplied to a number 12 pin and a number 22 pin as a VCC. At this stage, a voltage of 5 V is not applied to a power supply voltage terminal. The system monitors the level of the number 17 pin. When the level of the number 17 pin is a “L” level, the system determines that the memory card is the 5 V product, and when the level of the number 17 pin is a “H” level, the system determines that the memory card is the 3.3 V product. In accordance with the determined results, a normal power supply voltage is applied to the power supply voltage terminal.
  • The electrical detection may be omitted by the improvement of a connector. FIG. 25 shows an example of a dedicated system for the 5 V product. The 5 V dedicated system may use a connector with a cutout detecting mechanism so as to prevent the 3.3 V product from being inserted. In this case, it is also possible to prevent the memory card serving as the 5 V product from being inserted upside down. Although the 3.3 V product can be inserted upside down, it is possible to prevent the 3.3 V product from electrically contacting the system. Although FIG. 25 is simplified in order to facilitate better understanding thereof, the following consideration is given. In the case of a system wherein a supply voltage of 5 V is always supplied to a connector, a misinsertion preventing mechanism utilizing the cutout must function before the terminals of the memory card contact the terminals of the connector. For example, the misinsertion preventing mechanism may be positioned in the vicinity of the inlet of the connector so as to prevent the memory card from electrically contacting the system when misinsertion is carried out.
  • FIG. 26 shows an example of a 3.3 V dedicated system. The 3.3 V dedicated system uses a connector with a cutout detecting mechanism so as to prevent the 5 V product from being inserted. In this case, it is also possible to prevent the 3.3 V product from being inserted upside down. Although the 5 V product can be inserted upside down, it is possible to prevent the 5 V product from electrically contacting the system.
  • FIG. 27 shows an example of a 5 V/3.3 V dual purpose system. In the case of the dual purpose system, either of the 5 V and 3.3 V products may be inserted into the system. Therefore, this system is designed to prevent a power supply of 5 V from being applied to the 3.3 V product. That is, a power supply voltage of 5 V may not be outputted when a 3.3 V memory card is inserted. The supply voltage of 5 V is not applied to the device until the power supply voltage is electrically detected to verify that the inserted memory card is the 5 V product.
  • A capacity detecting step C will be described below.
  • There are various flash memories having different memory capacities and interface specifications. When a memory card is inserted into the system, a maker code, device code or the like of the device is determined to inhibit new access when an unsupported code is determined. In order to read the maker code, device code or the like, a normal power supply voltage is inputted.
  • A physical format checking step D will be described below.
  • The memory card has a physical format for storing data. When a memory card is inserted into the system, the physical format thereof is checked. If the memory card has an unsupported format, it is rejected without altering data. If the system executes a physical format again when an unknown physical format product is inserted, there is a risk that the processing for innate and acquired defective blocks of the flash memory may be incomplete, so care should be taken. For example, a 16-Mbit NAND-type flash memory is divided into 512 physical blocks. In the top block thereof a physical format, an attribute information of the card and so forth are written. Since the remaining blocks are used as data areas, it may be determined whether a supportable physical format is provided by examining data of the top block.
  • A logical format step E is carried out to allow the memory card to be accessed as a device on the DOS. It is also required to check whether the logical format of the memory card is supported. If the cluster delimiter of the DOS is coincident with the delimiter of the physical block of the NAND flash memory, it is possible to carry out a higher speed operation.
  • At step F, the respective operations indicate the memory access operations of the system, and include read-out, write and erase operations.
  • At an ejection detecting step G, ejection is detected in a manner similar to the above described insertion detection. When ejection is detected, the operation of the memory system ends.
  • For the above described control flow for the memory card, methods for electrically detecting the insertion/ejection of a memory card are disclosed in Japanese Patent Laid-Open Nos. 8-90969 and 8-202509. A method for controlling a memory system using connector terminals sequentially connected to the terminals of a memory card (a storage medium) will be described below.
  • Referring to FIG. 28, a method for detecting the insertion/ejection of a memory card in the memory system using the connector will be described below.
  • A number 1 pin terminal and number 10 pin terminal of the connector are connected to a VSS, and a number 11 pin terminal of the connector is pulled up to a VCC level. When the number 11 pin terminal of the memory card does not contact a connector terminal, the potential level of the connector terminal is held at a “H” level by means of a pull-up resistor. When the number 11 pin of the memory card contacts the connector terminal, the number 11 pin is connected to VSS via the number 1 pin and number 10 pin, so that the potential level is changed to a “L” level. Therefore, it is possible to electrically detect the insertion of the memory card on the basis of the transition of the potential of the connector terminal of the number 11 pin to the “L” level, and the ejection thereof on the basis of the transition from the “L” level to the “H” level. In order to electrically detect the insertion by means of the number 11 pin, it is required to allow terminals other than the number 11 pin to contact the terminals of the connector if the number 11 pin contacts the connector terminal. The value of the pull-up resistor may be adjusted so as to meet the required specifications of the system.
  • In the above described memory system using the connector, a control method during the insertion/ejection of the memory card will be described in detail below. In this preferred embodiment, as a practical example, a hot-line insertion/ejection will be described. The term “hot-line insertion/ejection” means the insertion/ejection of a memory card while a power supply voltage is applied to a connector. In order to prevent malfunction and electrical destruction of the memory, the following care should be taken.
  • If control signals (/CE, /WE, CLE, etc.) and an I/O pin are undefined while a power supply voltage is supplied, there is a possibility that an unintended write or erase command or the like may be received. In addition, if /CE=“L” and /RE=“L” while a power supply voltage is supplied, the device condition is a data output condition. At this time, if the system side condition is also a data output condition, there is a possibility that collision occurs on a data bus, having an undesired influence on the device. Generally, in a CMOS device, if a potential higher than the power supply voltage is applied to an I/O pin, there is a possibility that current flows into the device to cause the latch up and so forth. An example of a method for avoiding the above described matters is described below. For example, the contact and removal sequences for connector terminals are provided, and the contact and removal are sequentially carried out, so that it is possible to achieve the hot-line insertion/ejection. For example, the undermentioned four-stage sequence is suitable for use in a PC card adapter or the like since there is a little limit to the system side. The contact sequence for the connector and the limitations on the system side are as follows.
  • TABLE 1
    Connector Contact Sequence
    Contact Contact System Side
    Sequence Pin Setting Condition
    First Stage VSS(1, 10), CLE, ALE, /WP When Not Accessing
    Second Stage VCC, /CE, /RE, /WE Memory, Each Control
    Third Stage Pin at First, Second, Third Signal (/CE, /WE, /RE,
    Stages or More CLE, ALE, etc.)
    Fourth Stage 11 Pin Should Be Held On
    Inactive Stage
  • In this case, at the first stage, the contact with a VCC terminal is made to establish the ground level of the chip, and the level of a CLE terminal is fixed to a “L” level (an inactive state) so as to avoid the problem that an erroneous command is input with noise and so forth in a subsequent sequence that erroneously erases the device. At the second stage, the /CE terminal is fixed to the “H” level (an inactive state) so that the state of the output pin of the device is a high impedance state, thereby preventing the collision of data on the bus regardless of the bus control state on the system side. At the third stage, the problem that current flows into the output pin from the bus line is avoided since the VCC reaches a predetermined level at the second stage. At the fourth stage, the number 11 pin finally contacts the connector terminal. As described above, the number 11 pin serves as an insertion/ejection detecting terminal.
  • In addition to the hot-line insertion/ejection in the four-stage sequence, a hot-line insertion/ejection in a two-stage sequence may be used. In this case, the contact sequence for the connector and the limitations on the system side are as follows.
  • TABLE 2
    Connector Contact Sequence
    Contact Contact System Side
    Sequence Pin Setting Condition
    First Stage VSS(1, 10), CLE, ALE, /WP When Not Accessing
    Second Stage Pin other than Pin Memory, Each Control
    Contacting At First Signal (/CE, WE, RE,
    Stage CLE, ALE, etc.) Should
    Be Held On Inactive
    Stage I/O Pin Should Be
    Held in Hi-z State
  • In this case, at the first stage, the contact with the VCC terminal is made to establish the ground level of a chip, and the level of the CLE terminal is fixed to the “L” level (the inactive state) so as to avoid the problem that an erroneous command is input with noise and so forth in a subsequent sequence to erroneously erase the device. At the second stage, the VCC, /CE, I/O pins and so forth do not completely simultaneously contact the connector terminals, and any one of unintended contacts may occur at the contact timing. However, if the I/O bus is held in the high impedance state on the system side, it is possible to avoid the collision of data and the flow of current into the pin on the bus.
  • The connector for the four-stage contact sequence may also be applied to a PC card type adapter. A typical PC card has two pins, −CD 1 and −CD2 pins, which are defined as card detecting pins, and the insertion/ejection of the PC card is typically detected on the system side while both pins are pulled up. As a signal generating method for the −CD 1 and −CD2 pins in the PC card adapter, an example of a method using the 11 pin of the memory card is shown in FIG. 29. In this case, a standard sequence for a personal computer will be described below. When only a PC card adapter is inserted into a personal computer, the personal computer determines that no card is inserted. When a memory card is inserted into the personal computer, the VSS (1, 10 pins) first contacts a four-stage contact sequence connector. At this stage, GND level is supplied from the personal computer. After sequential connection proceeds and all the pins are connected, the 11 pin is finally connected. At this stage, the levels of the −CD 1 and −CD2 are reduced to the “L” level, and the personal computer side recognizes that the card is inserted, so that the initializing processing is started and a power supply is turned on the personal computer side.
  • The processing performed when a memory card is ejected while accessing the memory card will be described. Since this action should be basically an inhibit action, it is naturally conceived that means, such as an access lamp, for informing the user that the personal computer is accessing the memory card is provided. However, if the memory card is erroneously ejected during erase or write operation, there is a possibility that data other than those of an accessed block may be corrupted. For that reason, the following sequence is carried out to reduce the possibility that data other than those of a selected block may be corrupted. An example of a connector of a four-stage insertion/ejection sequence will be described below.
  • When a memory card is ejected, the memory card is removed in the reverse sequence to the above described contact sequence. First, the ejection of the memory card is detected by means of the 11 pin. Then, a write enable signal /WP is enabled to reset the erase or write operation. The time required for reset is 1.5 seconds or less at the maximum in an erase mode. In order to normally carry out the reset operation, the power supply voltage must be supplied to the memory card, and the logic in the memory card must be normally operated. When the four-stage removal sequence is used, it is ideal that the reset operation is completed before the VCC pin is removed and that the reset operation is completed before the I/O pin and so forth are removed at the third stage.
  • In addition, the second-stage removal sequence can be achieved by removing the pins in the reverse sequence to the above described two-stage insertion sequence.
  • Moreover, a three-stage contact/removal sequence for simultaneously carrying out the third and fourth stages of the four-stage contact/removal sequence is effective in data protection during the hot-line insertion/ejection.
  • As described above, according to the memory system control method of the present invention, it is possible to improve the stability of operation during the insertion/ejection of the memory card when the insertion/ejection of the memory card is detected, and it is possible to protect data during the hot-line insertion/ejection.
  • The second preferred embodiment of the present invention will be described in detail below.
  • In this preferred embodiment, there is provided a flash memory card, which has a reduced capacity of a RAM area necessary for a logical address/physical address translation table.
  • First, the structure of a physical block and the relationship with a logical block in a memory system according to the present invention will be described below.
  • The structure of a physical block and the data construction in the physical block in a memory system according to the present invention are the same as those described in the description of the prior art, and shown in FIGS. 2 and 4. Referring to FIG. 4, the meanings of the respective bytes in a physical block in a memory system according to the present invention will be described.
  • Data Area-1 stores the data of the first half 0 to 255 bytes of the data of 512. Data Area-2 stores the data of the second half 256 to 511 bytes of the data of 512 bytes. The data of a User Data Area are open to a user, and the use thereof is entrusted to the user. A Data Status Area indicates whether data are normal. Although an “FFh” is usually set, a “00h” is set when abnormal data are written. A Block Status Area indicates whether a block is good or defective.
  • Although the “FFh” is usually set, the “00h” (an initial defective block) or an “F0h” (an acquired defective block) is set in the case of a defective block. When a block has “0”s of two bits or more, it is determined that the block is a defective block. Furthermore, the same value for all the data is written in the same block. A Block Address Area-1 indicates a logical address information of a block. Furthermore, since 8 sectors forming one logical block correspond to one of 512 physical blocks, the same value for all the data is written in the same block. Similarly, in a Block Address-2, the same contents as the data of the Block Address Area-1 are written. An ECC Area-1 is a 3-byte ECC code of even page data (256 bytes). An ECC Area-2 is a 3-byte ECC code of odd page data (256 bytes).
  • The ECC will be described below. In this preferred embodiment, an ECC code is generated for every data page of 256 bytes (2048 bits). ECC data of 22 bits are used for 256 bytes so as to have a correction function of 1 bit. FIG. 30 shows the arrangement of 256 bytes of data.
  • That is, bit 0 of the input at the first byte is the first bit (address: 00000000 000) of 2048 bits, and bit 7 of the input at the 255th byte is the 2048th bit (address: 11111111 111) of 2048 bits.
  • The ECC code (line parity (LP) and column parity (CP)) is calculated as an odd parity of 1024 bits meeting the conditions shown in FIG. 31.
  • Column parities CP0 through CP5 are updated each time data of 1 byte (8 bits) are inputted. When the ECC code is generated by software, there may be provided a method, which previously has the calculated results of column parities for an input of 1 byte (256 ways) on an ROM in the system. By this method, it is not required to carry out calculation every bit, so that it is possible to considerably reduce the time required for calculation. Alternatively, there may be provided a method for deriving the calculated results of column parities for an input of 1 byte (256 ways) at one time when a power supply is turned on, to hold the calculated results on a RAM. In comparison with the former method, it is required to provide the RAM area although it is not required to provide a ROM.
  • FIG. 32 is a control flow chart for the second preferred embodiment of a flash memory card according to the present invention when a power supply is turned on. This control flow will be described below.
  • (Step S1) A power supply voltage is detected in response to the insertion of the memory card into the connector.
  • (Step S2) An ID code of the memory card is read out, and the storage capacity is read.
  • (Step S3) If an ID code that is not supported by the system is read out, it is rejected.
  • (Step S4) The physical format is checked. The information on the top block of the physical address is read.
  • (Step S5) If a physical format that is not supported by the system is designated, it is rejected.
  • (Step S6) A logical address/physical address translation table is prepared, and an erased area for write in the next write operation is selected.
  • FIG. 33 shows a flow chart for preparing a logical address/physical address translation table, which is shown at the aforementioned step S6 and which is prepared when a power supply is turned on.
  • (Step S1) A RAM area for storing therein a logical address/physical address translation table is reset.
  • (Step S2) A table area for storing therein an erased block used for the next data writing is reset.
  • (Step S3) Search is started from physical block 1.
  • (Step S4) A redundant division of the block is read out.
  • (Step S5) On the basis of data in a predetermined area, it is determined whether the block is a normal block. When the block is a defective block, the subsequent processing is not required, and the search for the next block is carried out.
  • (Step S6) It is determined whether the area is an erased area.
  • (Step S7) If it is an erased block, it is stored in the table as a proposed block used when the next write is carried out.
  • (Step S8) If it is not the erased area, a logical address information area is extracted. At this time, a parity check is carried out to check validity.
  • (Step S9) On the basis of the above contents, a logical address/physical address translation table is prepared.
  • (Step S10) Physical block number is counted up.
  • (Step S11) After 512 blocks are searched, the routine ends.
  • FIG. 34 shows a logical address/physical address translation table prepared according to the aforementioned flow. The term “physical block area” shown in FIG. 34 means an aggregate of two continuous physical blocks. For example, physical block area 0 indicates physical block 0 and physical block 1. In this table, one physical block area is allocated to one logical block. For example, when physical block area 5 is allocated to logical block 0, actual data of logical block 0 are stored in physical block 10 or 11. Therefore, when access is actually carried out, it is required to search for a data area indicating the relationship with the logical addresses of the redundant divisions of the physical blocks 10 and 11 to determine which actually stores data of the logical block 0. However, since it is enough to read only a very limited area, there is little influence on the performance of memory access.
  • At this time, 256 (512/2) physical block areas exist as a whole and can be described by 8-bit address data. If the table is formed so that an address offset directly indicates a physical block for convenience of the software, it is required to provide 1 byte for one block, i.e., a RAM area of 0.5 KB in total. For example, the address of a physical block area storing therein the information on physical block 5 is stored in the fifth byte offset from the top of the table.
  • The 0.5 KB RAM capacity is a half of 1 KB RAM capacity conventionally required. Usually, a general purpose CPU has a RAM area of about 1 KB. According to this preferred embodiment, it is possible to reduce the RAM area by 0.5 KB to obtain a great advantage.
  • That is, it is possible to form the system by using an empty area of 0.5 KB without the need of an external RAM increasing costs, so that it is possible to reduce costs.
  • The present invention should not be limited to the above described preferred embodiment, but four physical blocks may be defined as a physical block area, or a larger number of blocks may be supposed.
  • FIG. 35 is an operation flow chart of a memory card in this preferred embodiment when read-out is carried out. This flow will be described below.
  • (Step S1) The top sector address for carrying out read-out and the number of transfer sectors are received from the host.
  • (Step S2) It is verified whether the read-out range is a valid range.
  • (Step S3) The sector is converted to a logical block. Since one block comprises 8 sectors in the case of a 16-Mbit product, one block is divided by 8.
  • (Step S4) A physical block area, in which the corresponding logical block exists, is obtained by referring to a logical address/physical address translation table.
  • (Step S5) The logical address information areas of the physical blocks in the block area are examined, and it is determined which physical block stores therein data of the designated logical block.
  • (Step S6) The data of one sector is read out of the identified physical block. For example, when the sector number is 0, the data of the top 2 pages of the physical block are read, and when the sector number is 7, the data of the end 2 pages of the physical block are read. In one physical block, the data of 8 sectors are arranged in sequence.
  • (Step S7) Error check for the read-out data is carried out, and the presence of an error is checked.
  • (Step S8) When an error is detected, it is determined whether the data can be corrected.
  • (Step S9) When an error is detected and when the data can be corrected, the data is corrected.
  • (Step S10) After the number of sectors required by the host are read out, the routine ends.
  • (Step S11) It is determined whether the next read sector exceeds the boundary between physical blocks. For example, when the sector is transferred from sector 7 to sector 8, data exist in different physical blocks, so that the system refers to the logical address/physical address translation table again.
  • (Step S12) When the read-out is continued in the same block, pages to be read out are counted up.
  • (Step S13) When the block to be read is moved to another block, the logical block is counted up, and the count of pages is also reset.
  • The write operation in this preferred embodiment will be described below.
  • Basically, the write operation is roughly divided into the following three processes. The case where, e.g., logical sector 3, is rewritten will be described. Although only sector 3 is updated, the processing for one block is required since data of 8 sectors, sectors 0 to 7, exist in the same block.
  • First, since the data of logical sectors 0, 1 and 2 are not updated, the data of logical sectors 0, 1 and 2 are copied on a physical block to be newly written from the originally stored physical block.
  • Secondly, since logical sector 3 is updated, it is not required to copy the original data, and the data supplied from the host are written in a block to be newly written.
  • Thirdly, since the data of logical sectors 4 through 7 are not updated, the data of logical sectors 4 through 7 are copied on a physical block to be newly written from the originally stored physical block.
  • As described above, the operations of copy/updated data writing/copy for one block are basically carried out. When a write in sectors 0 through 7 is carried out, all the data of one block are updated, so that the copying operation is clearly unnecessary. The branch of the undermentioned flow chart proceeds mainly while determining whether a sector to be written is a data update or copying operation.
  • FIG. 36 is an operation flow chart of a flash memory card in this preferred embodiment when write is carried out. This flow will be described below.
  • (Step S1) The top sector address for carrying out read-out and the number of transfer sectors are received from the host.
  • (Step S2) This information is converted to a logical block number to refer to the logical address/physical address translation table. An actual physical block is selected from two blocks of a logical block area similar to the read-out operation. Data to be copied are read out from the selected block.
  • (Step S3) The processing is started from the top of the physical block.
  • (Step S4) It is determined whether the copying operation in the first half of the block or the update of data is carried out.
  • (Step S5) When the copying operation is carried out, data are read out of the original block, and write is carried out in a new block.
  • (Step S6) The processing for the next sector is carried out.
  • (Step S7) When it is determined at step S4 that an update of data is carried out the write is carried out on the basis of the updated data received from the host.
  • (Step S8) The processing for the next sector is carried out.
  • (Step S9) It is checked whether write has been carried out for the number of sectors required by the host.
  • (Step S10) If it is determined at step S9 that write has been carried out for the required number of sectors, it is determined whether the process has reached the boundary between blocks. If unwritten areas remain, the copying operation on the second half of the block is carried out. If the process has reached the boundary between blocks, it is not required to carry out the copying operation.
  • (Step S11) Data are read from the original block to be written in a new block.
  • (Step S12) The processing for the next sector is carried out.
  • (Step S13) When it is determined at step S9 that the number of sectors required to be written by the host is not completed, it is required to carry out more writing. However, when the process reaches the boundary between blocks, the processing for the next physical block is carried out.
  • (Step S14) Before the processing for the next block is carried out or before the processing is completed, the logical address/physical address translation table is updated on the basis of the written results, and the physical block, in which the original data exist, is erased. In addition, a proposed area is registered as a new write area for the next processing.
  • (Step S15) The processing for the next block is carried out.
  • As described above, according to the memory system control method in this preferred embodiment, it is possible to considerably reduce the RAM area. In the conventional memory card system, the RAM capacity required for the logical address/physical address translation table is large, and the system cannot be constructed only by the integrated RAM of the general purpose CPU, so that the external RAM must be provided. On the other hand, the memory card system of the present invention uses a flash memory control method, which can reduce the RAM capacity required for the logical address/physical address translation table and which can control only the integrated RAM of the general purpose CPU. Thus, the external RAM, which has been conventionally required, is not required, so that it is possible to considerably reduce costs.
  • The third preferred embodiment of the present invention will be described in detail below.
  • In this preferred embodiment, there is provided a flash memory card, which can carry out the write operation of data at a higher speed when it is used in the DOS format.
  • Parts (a) and (b) of FIG. 37 show DOS format parameters in this preferred embodiment, wherein part (a) shows the case of a cluster size of 4 KB, and part (b) shows the case of a cluster size of 8 KB. When the cluster size is 4 KB, a master boot sector is arranged in logical master 0, and a boot sector is arranged in logical sector 19. In addition, FATs are arranged in logical sectors 20 through 25, and the copies of the FATs are arranged in logical sectors 26 through 31. Moreover, directories are arranged in logical sectors 32 through 47, and file data areas are arranged in and after logical sector 48. When the cluster size is 8 KB, a master boot sector is arranged in logical sector 0, and a boot sector is arranged in logical sector 25. In addition, FATs are arranged in logical sectors 26 through 28, and the copies of the FATs are arranged in logical sectors 29 through 31. Moreover, directories are arranged in logical sectors 32 through 47, and file data areas are arranged in and after logical sector 49. Thus, in either case of cluster sizes of 4 KB and 8 KB, parameters are set so that the cluster delimiter does not straddle the boundary of physical block. This is achieved by adjusting a place, in which a boot sector is arranged, of the DOS format parameters.
  • First, referring to FIG. 38, an example of a write sequence in the case of a cluster size of 4 KB will be described. Since the cluster size is 4 KB, a write command for 8 continuous sectors is issued from the OS. At this time, write (data update) in logical sectors 48 through 55 (cluster A) is carried out.
  • (1) An erased new area is searched, and new data of logical sectors 48 through 55 are written in a new area NAND Block C.
    (2) The original data of logical sector 56 through 63 are copied on the new area NAND Block C.
    (3) The original block NAND Block B is erased.
    (4) The logical address/physical address translation table is updated.
  • Therefore, if the rewrite is carried out for 8 sectors when viewed from the outside, logical sectors 48 through 63, i.e., 16 sectors in total (16 pages), are written and the NAND Block B, i.e., one block, is erased within the actual device.
  • Referring to FIG. 39, the write sequence in cluster B will be described below. In this case, the write (data update) in logical sectors 56 through 63 (cluster B) is carried out.
  • (1) An erased new area is searched, and the original data of logical sectors 48 through 55 are copied on the new area NAND Block C.
    (2) New data of logical sectors 56 through 63 are written in the new area NAND Block C.
    (3) The original block NAND Block B is erased.
    (4) The logical address/physical address translation table is updated.
  • Therefore, if rewrite is carried out for 8 sectors when viewed from the outside, logical sectors 48 through 63, i.e., 16 sectors in total (16 pages), are written and the NAND Block B, i.e., one block, is erased within the actual device.
  • Referring to FIG. 40, an example of a write sequence in the case of a cluster size of 8 KB will be described below. Since the cluster size is 8 KB, a write command for 16 continuous sectors is issued from the OS. At this time, write (data update) in logical sectors 48 through 55 (cluster A) is carried out.
  • (1) An erased new area is searched, and new data of logical sectors 48 through 63 are written in a new area NAND Block C.
    (2) The original block NAND Block B is erased.
    (3) The logical address/physical address translation table is updated.
  • Therefore, if rewrite is carried out for 16 sectors when viewed from the outside, logical sectors 48 through 63, i.e., 16 sectors in total (16 pages), are written and the NAND Block B, i.e., one block, is erased within the actual device.
  • Comparing the cluster size of 4 KB with the cluster size of 8 KB when the same 8 KB data are written, in the case of the cluster size of 4 KB, the processing is divided into two write operations, so that the write operation in 32 sectors in total and the erase operation from two blocks are carried out. On the other hand, in the case of the cluster size of 8 KB, the processing is concentrated on one writing, so that the write operation in 16 sectors in total and the erase operation from one block are carried out.
  • The rewriting speed of the conventional memory system shown in FIGS. 12 through 15 can be compared with the present strategy for the case of a cluster size of 4 KB. In the conventional memory system, the write operation for 48 sectors and the erase operation for three blocks are carried out to update 8 KB data. In the present memory system, the write operation for 32 sectors and the erase operation for two blocks are carried out to achieve the same result, so that the rewriting time can be reduced to two thirds. In the case of the cluster size of 8 KB, the conventional memory system performs a write operation for 32 sectors and an erase operation for two blocks to update 8 KB data, whereas the present memory performs a write operation for 16 sectors and an erase operation for one block, so that the rewriting time can be reduced to half according to the present invention.
  • Thus, the cluster delimiter serving as a unit of file management on the DOS does not straddle the boundary between physical blocks of the flash memory, so that the rewriting speed can be increased.
  • In addition, in the case of a cluster size of 4 KB, the write operation for 32 sectors and the erase operation for two blocks are carried out in order to update 8 KB data, whereas in the case of a cluster size of 8 KB, the write operation for 16 sectors and the erase operation for one block are carried out in order to update 8 KB data, so that the rewriting time is reduced to half. That is, if the size of a cluster is the same size as that of a physical block, a high-speed write can be achieved. Also, if the size of a cluster is an integer times as large as the size of a physical block, the same advantage can be obtained.
  • The fourth preferred embodiment of the present invention will be described below.
  • In this preferred embodiment, there is provided a flash memory card, which improves the processing speed when a write command is executed after erase.
  • In the flash memory card system in this preferred embodiment, unlike the file erase in the ordinary DOS, a mark indicating that a corresponding file is invalid is put on a directory, and a memory area having been occupied by the corresponding file is not only open on a FAT (file allocation table), but the data division of the file body is erased on the flash memory. That is, when the erase of the file is commanded, the erase operation from the open cluster area is carried out.
  • FIG. 41 shows the relationship between a management area and a data area when an erase command is executed. In FIG. 41, for example, when erase commands for File-1 and File 4 are executed, the File-1 and File-4 in the management area are open, and a del.mark is stored. In addition, the areas having stored File-1 and File-4 in the data area are erased.
  • Therefore, since the cluster selected when the next new file write command occurs has been erased, write can be immediately carried out, so that the file writing speed can be improved. Since the erasing time is generally longer than the writing time in the flash memory, it is possible to remarkably improve the file writing speed according to the present invention.
  • As can be seen from the above described third preferred embodiment, the advantage of this preferred embodiment is most remarkable when the cluster size is the same as the block size of the flash memory. When the cluster size is smaller than the block size of the flash memory, a part of a block is erased. In this case, the processing is not only complicated, but there are also some cases where it is not possible to erase only a part of a block on the specification. If the cluster size is the same as the block size, the cluster can be open by simply erasing the block. Also if the cluster size is an integer times as large as the physical block size, the same advantage can be obtained.
  • The present invention can be embodied in various ways without departing from the principle of the invention. For example, in this preferred embodiment, while the erase operation from the corresponding cluster in the data area has been executed when the file is erased, the execution timing for erase operation should not be limited thereto. For example, the erase operation from all the clusters may be executed when the format operation is executed. The memory card may be shipped after erasing the cluster in the data area. When a memory card is shipped, a shipping test for the memory card is generally carried out. If directories and FATs are rewritten when this test is completed, it is not only possible to obtain no file state, but it is also possible to achieve a high-speed file writing without the need of end user's correction when the memory card is delivered to the end user if the erase operation from the data area is executed according to this preferred embodiment.
  • In a case where the memory card is used for a digital still camera or the like, when operations including the erase of a file in the camera and the erase of an image file, such as a reformat, are carried out, if a cluster area storing the file body is erased simultaneously when the FAT and so forth are rewritten, it is possible to achieve a high-speed writing in the subsequent image writing sequence, so that it is possible to continuously take pictures and to capture moving pictures. In this case, if the cluster size is an integer times as large as the block size and if the cluster delimiter is coincident with the delimiter of block size, the file body part can be easily erased. In addition, unnecessary file parts may be automatically erased when a power supply is turned on in a digital camera or the like. Thus, it is also possible to increase the writing speed without giving the user trouble, with respect to a memory card wherein the file erase has been carried out by simply updating the FATs and so forth in a personal computer. This timing should not be limited to the time when the power supply is turned on, but it may be any time. The writing speed can also be increased when the memory card of the present invention is used for a computer system or the like.
  • In addition, according to this preferred embodiment, since it is possible to reduce the numbers of erase and write operations occurring when data are rewritten, it is also possible to increase the life of a flash memory when a flash memory having a limited number of rewrite operations is used.
  • As described above, according to the memory system control method in this preferred embodiment, the cluster delimiter serving as a unit of file management on the DOS does not straddle the boundary between physical blocks, so that it is possible to reduce the numbers of erase and write operations occurring when data are rewritten, thereby increasing the rewriting speed. Moreover, when a flash memory having a limited number of rewrite operations is used, it is possible to increase the life of the memory.
  • The fifth preferred embodiment of the present invention will be described in detail below.
  • In this preferred embodiment, there is provided a flash memory card, which reduces the capacity of a RAM area required for a logical address/physical address translation table.
  • In this preferred embodiment, when a logical address/physical address translation table is prepared, access from a host is divided into two kinds. For example, it is assumed that case 1 is a state accessing the first half 250 logical blocks of 500 logical blocks and that case 2 is a state accessing the second half 250 logical blocks of 500 logical blocks. In case 1, a table of the first half 250 blocks is held in the logical address/physical address translation table. In case 2, a table of the second half 250 blocks is held in the logical address/physical address translation table. FIG. 42( a) shows the state of the table in the case 1, and FIG. 42( b) shows the state of the table in the case 2.
  • Assuming that the table of the first half 250 logical blocks exists on the table in a certain moment, if the range accessed from the host is a range of the first half 0 through 249 logical blocks, it is possible to search the correspondence between the logical blocks and the physical blocks by using the existing table.
  • Similarly, assuming that the table of the second half 250 logical blocks exists on the table in a certain moment, if the range accessed from the host is a range of the second half 250 through 499 logical blocks, it is possible to search the correspondence between the logical blocks and the physical blocks by using the existing table.
  • Then, assuming that the table of the first half 250 logical blocks exists on the table in a certain moment, if the range accessed from the host is a range of the second half 250 through 499 logical blocks, it is not possible to search the correspondence between the logical blocks and the physical blocks by using the existing table. Therefore, in this case, the logical address/physical address translation table corresponding to the second half 250 logical blocks is remade. This needs to refer to all the areas of the flash memory again.
  • Similarly, assuming that the table of the second half 250 logical blocks exists on the table in a certain moment, if the range accessed from the host is a range of the first half 0 through 249 logical blocks, it is not possible to search the correspondence between the logical blocks and the physical blocks by using the existing table. Therefore, in this case, the logical address/physical address translation table corresponding to the first half 250 logical blocks is remade. This needs to refer to all the areas of the flash memory again.
  • Thus, if the logical address/physical address translation table corresponding to the area accessed from the host does not exist on the RAM, the required logical address/physical address translation table is remade by referring to all the areas of the flash memory again.
  • FIG. 43 shows a flow chart for preparing the logical address/physical address translation table in this case.
  • (Step S0) The presence of a required logical address/physical address translation table is checked at the beginning of access, and if it is required, the routine goes to a table preparing routine.
  • (Step S1) A RAM area for storing therein a logical address/physical address translation table is reset.
  • (Step S2) Search is started from the top of a physical block.
  • (Step S3) The redundant division of the block is read out.
  • (Step S4) On the basis of data in a predetermined area, it is determined whether the block is a normal block.
  • When the block is a defective block, the subsequent processes are not required, and the search for the next block is carried out.
  • (Step S5) It is determined whether the block is an erased area.
  • (Step S6) If the block is an erased area, it is stored on the table as a proposed block used for the next write.
  • (Step S7) If the block is not an erased area, a logical address information area is extracted.
  • (Step S8) A logical address/physical address translation table is prepared on the basis of the above described contents.
  • (Step S9) The physical block number is counted up.
  • After the search for all the blocks is completed, the routine ends.
  • In accordance with the above described operations, the logical address/physical address translation table is prepared if necessary.
  • The present invention should not be limited to the above described preferred embodiment, but the invention can be embodied in various ways without departing from the principle of the invention.
  • For example, in the above described preferred embodiment, while the flash memory has been divided into two parts, the first and second halves, the present invention should not be limited thereto, but the flash memory may be divided into an optional number of parts.
  • In addition, in the above described preferred embodiment, while the flash memory has been divided into the first and second halves having the same size, the present invention should not be limited thereto.
  • Thus, the number and size of divided parts may be optionally determined.
  • It is not always required to hold only one table when the flash memory is divided into three or more areas. For example, it is assumed that the flash memory card serves as a device of the DOS to store therein an image file or the like. In the top of an ordinary device, there are file management areas, i.e., a master boot sector, a partition boot sector, a file allocation table (FAT) and a directory area. These file management areas are frequently accessed each time the update or access of the file is carried out. In this case, the area corresponding to the file management area is one area, and each of other file data storing areas are divided into two parts. The logical address/physical address translation table corresponding to the file management area may always be held. In this case, when a file is written in the second half of the logical address, it is not required to alternately frequently remake the file management area and the second half of the logical address, so that it is possible to prevent the deterioration of performance. While the file management area has been described, the same advantage can be obtained if the table is always held for areas other than the file management area.
  • As described above, according to this preferred embodiment, all of the correspondence relationships between logical blocks and physical blocks must not be always held on the RAM, and the correspondence relationship of only a required area is prepared on the RAM in the system one by one in accordance with the access from the host. Therefore, in comparison with the case where the correspondence relationships in all the areas are always held on the RAM, it is possible to reduce the minimum required RAM area, and it is possible to control the memory only by the integrated RAM of the general purpose CPU although this has not been able to be achieved. Thus, it is possible to greatly reduce costs in comparison with a conventional system having an external RAM.
  • The sixth preferred embodiment of the present invention will be described in detail below.
  • In this preferred embodiment, there is provided a flash memory card, which reduces the capacity of a RAM area required for a logical address/physical address translation table.
  • In this preferred embodiment, a physical block address of the flash memory is divided into a plurality of logical areas (which will be hereinafter referred to as “zones”). In this case, the number of physical blocks allocated to each zone, i.e., the capacity of each zone, may be uniform or non-uniform. In addition, the number of zones may be one or plural. Moreover, the number of zones may be even or odd.
  • FIG. 44 shows the structure of a physical block and the relationship between logical addresses and physical addresses when a 16-Mbit NAND-type flash memory is divided into two zones having the same capacity.
  • When the host controls the flash memory, the number of required logical blocks is defined to be 500, and the values of logical block addresses are 0 through 499. The redundant division of a physical block stores therein a logical block address information indicating which logical block the data held in the physical block correspond to. In this preferred embodiment, logical block addresses are given from 0 in series for each zone. Therefore, logical block addresses 0 through 249 are allocated to zone 1, and each of logical blocks corresponds to any one of 256 physical blocks of physical block addresses 0 through 255. In addition, any one of logical block addresses 0 through 249 is stored in the redundant division of a physical block. Moreover, logical block addresses 250 through 499 are allocated to zone 2, and each of logical blocks corresponds to any one of 256 physical blocks of physical block addresses 256 through 511. In addition, any one of logical block addresses 0 through 249 is stored in the redundant division of a physical block.
  • An additional write system during data update is executed in a zone including a logical block address to be updated, and a physical block having data corresponding to a certain logical block is not fixed and is always moving in each zone.
  • In this preferred embodiment, usually when a power supply is turned on, the logical block address information stored in the redundant divisions of all the physical blocks in the zone is searched, and a translation table between logical blocks and physical blocks is prepared on a system RAM. The zone prepared on the RAM may be any one of the available zones. Usually, when the flash memory is used under the control of the DOS, the FATs and directory areas serving as management information services are arranged in the first zone, so that it is efficient to prepare the table of the top zone. In addition, the table prepared on the RAM should not be limited to one zone, but it may be prepared for a plurality of zones so far as the RAM capacity permits.
  • A process for preparing a translation table for an access demand from a host will be described below. An area including physical block addresses 0 through 255 is called zone 1. In addition, an area including physical block addresses 256 through 511 is called zone 2. The host remembers which zone belongs to the translation table currently prepared on the RAM.
  • When a translation table of zone 1 is prepared on the RAM and when an access demand for logical block address 128 is made, the procedure will be described below.
  • (1) If n is derived so that logical block address 128-250 x(n−1)<250, n=1, so that it can be seen that the translation table of zone 1 is required.
    (2) Since the translation table on the RAM is zone 1, no table is prepared.
    (3) Since the values of addresses on the translation table are 0 through 249, an address to be accessed on the translation table is derived from logical block address 128 demanded by the host. From 128-250×(1−1)=128, a physical block address corresponding to address 128 on the translation table may be accessed.
    (4) When rewrite occurs, the translation table is updated to get ready for the next access.
  • Then, when a translation table of zone 1 has been prepared on the RAM and when an access demand for logical block address 324 is made, the procedure will be described below.
  • (1) If n is derived so that logical block address 324-250 x(n−1)<250, n=2, so that it can be seen that a translation table of zone 2 is required.
    (2) Since the translation table in the RAM is for zone 1, the logical block address information of the redundant divisions of physical block addresses 256 through 511 included in zone 2 is searched, and the translation table of zone 2 is prepared on the RAM.
    (3) Since the values of addresses on the translation table are 0 through 249, an address to be referred on the translation table is derived from logical block address 324 demanded by the host. From 324-250×(2−1)=74, a physical block address corresponding to address 74 on the translation table may be accessed.
    (4) When rewrite occurs, the translation table is updated to get ready for the next access.
  • As described above, if the translation table is prepared in accordance with the accessed logical block, it is possible to easily reduce the RAM area in comparison with the conventional RAM area. Also, when the table of zone 2 has been prepared on the system RAM, it is possible to easily access a target address by the similar process.
  • When the logical address/physical address translation table is prepared, the translation table between logical blocks and physical blocks is prepared on the RAM. At this time, since a zone number to be prepared is known, the values of physical block addresses on the translation table may be 0 through 255. When the physical block is actually accessed, it is possible to easily obtain a new physical block address to be inputted to the flash memory by adding 256 x zone number to the physical address on the translation table as OFFSET.
  • When a 16-Mbit NAND-type flash memory is used, the conventional control method requires 9 bits to express a physical block address, and uses 2 bytes for convenience of a software. In this preferred embodiment, as shown in FIG. 45, a physical block address may be expressed by 8 bits, i.e., 1 byte. Therefore, the RAM capacity, which has conventionally required 1 Kbyte, can be reduced to half. If a logical block address increases, the capacity of the logical address/physical address translation table increases. Therefore, the advantages of this preferred embodiment increases as the capacity of the flash memory increases.
  • In addition, according to this preferred embodiment, a large-capacity flash memory can be controlled by a block address, which can be stored in the Block Address Area of the redundant division of the physical block shown in FIG. 11. That is, if one zone is divided so as to be formed by a physical block of a block address value, which can be stored in the Block Address Area of the redundant division, it is possible to correspond to a large-capacity logical address.
  • FIG. 46 shows the structure of a physical block when a 16-Mbit NAND-type flash memory is divided into four equal capacity zones.
  • When the host controls the flash memory, the number of required logical blocks is defined to be 500, so that the values of logical block addresses are 0 through 499. The redundant division of the physical block stores therein a logical block address information indicating which logical block corresponds to the data held in the physical block. In this preferred embodiment, the logical block addresses are given in series from 0 for each zone. Physical block addresses corresponding to these logical block addresses are as follows. The logical block addresses 0 through 124 are allocated to zone 1, and each of the logical blocks corresponds to any one of 128 physical blocks having physical block addresses 0 through 127. The logical block addresses 125 through 249 are allocated to zone 2, and each of the logical blocks corresponds to any one of 128 physical blocks having physical block addresses 128 through 255. The logical block addresses 126 through 374 are allocated to zone 3, and each of the logical blocks corresponds to any one of 128 physical blocks having physical block addresses 256 through 383. The logical block addresses 384 through 499 are allocated to zone 4, and each of the logical blocks corresponds to any one of 128 physical blocks having physical block addresses 384 through 511.
  • The additional write system during data update is executed only in a zone including a logical block to be written, and a physical block storing therein data corresponding to a certain logical block address is not fixed and is always moving in the zone. The redundant division of each physical block address stores therein a logical block address information indicating which logical block address the data held in the physical block belong to.
  • As described above, 125 logical blocks are allocated to each zone, and 128 physical blocks are allocated to each zone. If a physical block is divided into four zones as described above, it is possible to obtain the same advantages as those in the case where a physical block is divided into two zones.
  • An example of redundancy operation in this preferred embodiment in the case of the physical block divided into four zones will be described below.
  • Before describing the redundancy operation in this preferred embodiment, the redundancy operation of a 16-Mbit NAND-type flash memory as an example of a conventional flash memory, which does not carry out the division of a physical block into zones, will be described. As shown in FIG. 2, the 16-Mbit NAND-type flash memory has 512 physical blocks, and as shown in FIG. 3, the number of logical blocks viewed from the host is defined to be 500. In addition, one block must be provided for storing therein information for easily constructing a PC card ATA interface, and one block must be provided for adopting the additional write system. Therefore, 502 blocks must be provided for controlling this flash memory. Accordingly, the 16-Mbit NAND-type flash memory permits the existence of up to 10 defective blocks.
  • However, the flash memory has the upper limit of the number of rewrite operations. When the flash memory is used, a defective block may be produced therein. Therefore, in order to ensure a sufficient storage region, the flash memory must have a writable and erasable effective block as a block replaced when the defective block is produced. In addition, there is a problem in that a flash memory having 10 or more defective blocks can not be shipped since it does not have a sufficient capacity when it is shipped, so that the yield of products is decreased.
  • For that reason, the 16-Mbit NAND-type flash memory is provided with a plurality of redundant blocks in addition to 512 body blocks, in order to ensure a large number of effective blocks to be prepared for the occurrence of defective blocks in the flash memory and in order to improve the yield of products when the flash memory is shipped. These redundant blocks replace defective blocks, which have been produced in the 512 body blocks, by means of a redundant circuit as a hardware when the flash memory is shipped. When the redundant circuit is used, the redundant blocks are allocated to addresses of defective blocks. After the replacement in hardware, if the address of a defective block is selected, the replacement redundant block is selected. Since the number of the redundant blocks is not infinite, a defective block, which has not been replaced when the flash memory is shipped, is treated as an innate defective block. In addition, a defective block, which is produced when the flash memory is used by the user, is treated as an acquired defective block. These innate and acquired defective blocks are relieved by the effective blocks.
  • The replacement of the body blocks and redundant blocks is usually carried out by means of the redundant circuit as a hardware. As shown in FIG. 47, the replacement is carried out sequentially from a defective block having a small block address or a large block address for convenience of replacement operation. Therefore, after the replacement of blocks, there are blocks, which can be written or erased sequentially from a small block address or a large block address.
  • Thus, as described above, the conventional flash memory control method requires 500 logical blocks, one block for storing therein information for easily constructing a PC card ATA interface, and one block as an empty area to use the additional write system, with respect to 512 physical blocks. Thus, the conventional flash memory can be controlled if it has 502 rewritable blocks. Therefore, the conventional control method permits 10 defective blocks with respect to 512 physical blocks.
  • The redundancy operation in this preferred embodiment will be described below. With respect to zone 1, one block for storing therein information for easily constructing a PC card ATA interface is required, and one block as an empty area is required since the additional write system is adopted. Therefore, with respect to 125 logical blocks, 127 rewritable physical blocks are required, and the number of defective blocks, which are permitted in the zone and which can not be rewritten, is up to 1. With respect to zones 2 through 4, since additional one block is required as an additionally writing empty block, 126 rewritable physical blocks are required, and the number of defective block, which are permitted in the zone and which can not be rewritten, is up to 2. In order to simplify the explanation, it is assumed that the number of defective blocks permitted with respect to zones 1 through 4 is 1. Therefore, the specification of the number of permitted defective blocks is very severe in comparison with the specification of the number of defective blocks permitted by the conventional control method.
  • In the 16-Mbit NAND-type flash memory shown in FIG. 46, it is assumed that, for example, 7 blocks having block addresses 2, 5, 129, 131, 132, 385 and 389 are defective blocks, and that the flash memory has four redundant blocks. As shown in FIG. 46, when the redundant circuit is used and when defective blocks are replaced with redundant blocks in sequence from a defective block having a small block address similar to the conventional method, the defective blocks after replacement are three blocks having physical block addresses 132, 385 and 389. These blocks are regarded as innate blocks, and innate defective block marks are put on the redundant divisions of the blocks. Since up to 10 innate defective blocks are permitted in the conventional control method, there is particularly no problem to control the flash memory. However, when it is premised that the flash memory is divided into zones to be controlled, two defective blocks having physical block addresses 385 and 389 exist in zone 4, so that it is not possible to control the flash memory and it is not possible to ship this product.
  • Thus, when defective blocks exist in the flash memory in this preferred embodiment, if the replacement is carried out sequentially from a defective block having a small block address or a large block address for simple convenience of replacement operation similar to the conventional flash memory, it is not possible to ensure required good blocks, and the possibility of occurrence of unusable zones is increased.
  • Thus, in this preferred embodiment, when the flash memory is divided into zones to be controlled, the defective blocks existing in each zone are replaced with redundant blocks so that all the zones meet the number of effective blocks.
  • FIG. 48 shows the state of a physical block when defective blocks existing in each zone are replaced with redundant blocks so that the number of defective blocks exceeds the number permitted in each zone. Similar to the flash memory shown in FIG. 46, the flash memory shown in FIG. 48 has seven defective blocks having block addresses 2, 5, 129, 131, 132, 385 and 389 and four redundant blocks. It can be seen that it is possible to ship products, which can not be shipped by the conventional replacement method, as good product. An example of replacement procedure will be described below.
  • (1) Zones 1 through 4 are searched, and the following variables are derived. Defective physical block addresses are extracted. Z(n) BA(m) (n=1˜4, m=1˜the number of defective blocks in each zone)
  • The number of defective blocks for each zone is derived. Z(n) BN (n=1˜4)
  • (2) Even if all of redundant blocks are used with respect to one of n=1˜4 of Z(n) BN, if at least one does not meet the number of effective blocks, the replacement is not carried out and the operation ends.
    (3) Among n=1˜4 of Z(n) BN, the maximum n is extracted.
    (4) With respect to n extracted in (3), a block corresponding to a physical block stored in Z(n) BA(m) with respect to the minimum or maximum m for storing therein a block address information among Z(n) BA(m) (m=1˜the number of defective blocks in the zone) is replaced with a redundant block.
  • (5) Z(n) BN=Z(n) BN−1
  • (6) With respect to m selected in (3), the block address information of Z(n) BA(m) is deleted.
    (7) The processes (3) through (6) are repeated. After all the redundant blocks are used, the process (9) is carried out.
    (8) If Z(n) BN (n=1˜4) are equal to each other, (3) through (7) are repeated with respect to the zone of the minimum or maximum n.
    (9) Z(n) BN (n=1˜4) is checked. When n exceeding the specification exists, the product is regarded as a defective.
  • (10) End
  • Various methods for replacing body blocks with redundant blocks may be considered. While the flash memory has been divided into four zones in this preferred embodiment, it may be divided into two zones or odd zones. In addition, the capacities (the number of blocks) of divided zones may be different. In either case, in this preferred embodiment, body blocks are replaced with redundant blocks while monitoring the number of defective blocks existing in each zone so that the number of defective blocks existing in each zone after replacement operation exceeds the number of defective blocks permitted in each zone, and the replacement procedure may be modified without departing from the principle of the invention.
  • As described above, according to this preferred embodiment, since a table is prepared every zone serving as an object to be accessed unlike conventional methods using a table wherein logical blocks have a one to one correspondence with physical blocks of a flash memory, it is possible to reduce a RAM area required for the table, and it is possible to control a memory only by means of an integrated RAM of a general purpose CPU although this can not be achieved by the conventional method, so that it is possible to considerably reduce costs in comparison with conventional methods using an external RAM. In particular, these advantages are remarkable when a non-volatile semiconductor memory having a large number of physical blocks is controlled. In addition, it is possible to cope with a large-capacity logical address by a bit number determined by a Block Address Area of a redundant division of a physical block.
  • In addition, according to this preferred embodiment, in a flash memory for use in a system using a control method for allocating a plurality of physical blocks to a plurality of logical zones to prepare a translation table between logical blocks and the physical blocks every zone to carry out memory access, defective blocks of the body are replaced with redundant blocks while monitoring the number of defective blocks existing in each zone so that the number of defective blocks existing in each zone after replacement exceeds the number of defective blocks permitted for each zone, although the replacement has been carried out sequentially from a small (or large) address in conventional methods. Therefore, it is possible to reduce the number of products, which can not be shipped since the products have unusable zones, so that it is possible to improve the yield of products.
  • The seventh preferred embodiment of the present invention will be described below.
  • In this preferred embodiment, there is provided a flash memory card, which can support any types of binary/multi-valued flash memories.
  • FIG. 49 is a schematic view of a PC card adapter, to which this preferred embodiment is applied. A flash memory card 101A is a binary flash memory card shown in FIG. 17, and a flash memory card 101B is a four-valued flash memory card shown in FIG. 18. A PC card adapter 102 is provided for transferring the data of the flash memory cards 101A and 101B. The PC card adapter 102 has a 68-pin connector for a PC card slot, and a connector for a flash memory card. In the PC card adapter 102, there is provided a controller 103 for controlling the flash memory card and for electrically interfacing with the PC card slot, an oscillator 104 for a CPU provided in the controller, a RAM 105 for buffer and so forth. The controller 103 includes circuits for two types of error correction, an ECC circuit 1 and an ECC circuit 2, which are directly related to the present invention.
  • FIG. 50 shows an example of control when a power supply is turned on.
  • (Step S1) Detection of Power Supply Voltage
  • The memory cards include a product having a power supply voltage of 5 V and a product having a power supply voltage of 3.3 V. When a power supply voltage of 5 V is applied to a memory card operated by a power supply voltage of 3.3 V, there is a possibility that voltage destruction and so forth may be caused. In order to avoid this, the system detects the power supply voltage.
  • (Steps S2 and S3) ID Check
  • There are various kinds of flash memory cards having different storage capacities or interface specifications. When a memory card is inserted into the system, the system determines a maker code, a device code or the like of the device. When the determined code is an unsupported code, a new access is not carried out. In order to read the maker code, the device code or the like, a normal power supply voltage is inputted.
  • (Steps S4 through S8) ECC Check
  • The system of ECC is checked herein.
  • Examples of methods for recognizing the ECC system will be described below. A first method is a method for determining a device code similar to the above described steps S2 and S3. For example, if flash memories have the same capacity, the device code may be changed by the memory construction.
  • FIG. 51 shows the constructions of data and redundant divisions of a 16-Mbit flash memory in this preferred embodiment. The different point from the conventional memory shown in FIG. 4 is that a User Data Area of the redundant division is allocated to each of an ECC Flag Area and an ECC Area-3. As shown in FIG. 51, three areas of ECC Area-1 through ECC Area-3 are defined as areas for storing ECC codes. Referring to FIG. 52, a method for using the three areas will be described. Information on the ECC system is stored in an ECC Flag Area byte. For example, in the case of ECC system 1, “AAh” is defined as Flag data, and in the case of ECC system 2, “55h” is defined as Flag data.
  • In the case of a flash memory card based on the EGC system 1, a 3-byte ECC code of an even page data (256 bytes) enters the ECC Area-1, and a 3-byte ECC code of an odd page data (256 bytes) enters the ECC Area-2. The ECC Area-3 is null. It is possible to correct a 2-bit error of each of 256 bytes in the ECC Area-1 and ECC Area-2.
  • In the case of a flash memory card based on the ECC system 2, ECC codes with respect to 512 bytes are dispersed throughout the ECC Area-1, ECC Area-2 and ECC Area-3. In this case, it is possible to correct a 2-bit error in 512-byte data.
  • The controller 103 in the PC card adapter 102 of FIG. 49 reads the ECC Flag Area out to determine the ECC system when the flash memory card is inserted into the adapter.
  • In FIG. 49, the code generation and an error detection circuit, which correspond to each of the ECC system 1 and ECC system 2, are selected. The CPU in the controller controls the flow of data between the host and the RAM for buffer and the flash memory. In FIG. 49, the ECC circuit need not always be hardware. The generation of the ECC codes and so forth may be carried out by software.
  • The present invention should not be limited to the above described preferred embodiment. The present invention may be embodied in various ways without departing from the principle of the invention.
  • While the existence of two kinds of ECC systems has been assumed in the above described preferred embodiment, the present invention should not be limited thereto, but three or more kinds of ECC systems may be set.
  • The selection of the ECC system includes the selection that no ECC is used. In the fields of a flash memory card having very high reliability, and in the field of data, which does not require particularly high reliability, such as data in voice field or the like, the ECC is not essential. In this case, it may be defined that the ECC is not used if the content of the above described ECC Flag Area is “FFh”.
  • While the ECC system has been defined for every flash memory in the above described preferred embodiment, the present invention should not be limited. For example, the ECC system may be switched every optional unit, such as every sector or block. In this case, the ECC system may be switched each time access is carried out for each of the above described units without simply determining the ECC system when a power supply is turned on. In addition, for example, the operation for converting data read out by the ECC system 1 to the ECC system 2 to restore the converted data may be supposed.
  • While the PC card adapter has been used in the above described preferred embodiment, the present invention should not be limited thereto. The present invention may be applied to various apparatuses, such as a digital still camera, a PDA, a word processor and a voice recorder. Thus, according to this preferred embodiment, since a flash memory card having a very wide use range may be treated by one system, the flexibility can be remarkably improved.
  • According to the present invention, when a storage medium is inserted into or ejected from a connector in an electronic apparatus, the contact and breakaway of various pins are carried out in sequence, so that it is possible to improve the stability of operation and to surely protect data.
  • In addition, according to the present invention, when the translation between logical addresses and physical addresses is carried out by means of a translation table, a plurality of physical blocks are allocated to one logical block, so that it is possible to reduce a RAM area required for the table.
  • Moreover, according to the present invention, the cluster delimiter does not straddle a block serving as an erase unit, so that it is possible to decrease the numbers of erase and write operations occurring when data are rewritten.
  • In addition, according to the present invention, when an erase command is executed, the management area for physical blocks is open, and simultaneously, the data area is also erased, so that it is possible to improve the processing speed when a subsequent write command is executed.
  • In addition, according to the present invention, only the correspondence relationship in a required area of the correspondence relationships between logical blocks and physical blocks is prepared one by one, so that it is possible to reduce the RAM area required at the minimum.
  • In addition, according to the present invention, defective blocks are replaced with redundant blocks so that the number of defective blocks existing in each logical zone does not exceed a predetermined value after replacement, so that it is possible to improve the yield of products.
  • Moreover, according to the present invention, the error correction algorithm is selected in accordance with the kind of storage medium used, so that it is possible to use various storage media to improve the flexibility.
  • While the present invention has been disclosed in terms of the preferred embodiment in order to facilitate better understanding thereof it should be appreciated that the invention can be embodied in various ways without departing from the principle of the invention. Therefore, the invention should be understood to include all possible embodiments and modification to the shown embodiments which can be embodied without departing from the principle of the invention as set forth in the appended claims.

Claims (15)

1-8. (canceled)
9. A method for controlling a memory system which includes a non-volatile semiconductor memory having a plurality of first predetermined units, comprising:
receiving file data, the file data being organized into second predetermined units; and
storing the file data so that a boundary of the second predetermined units is to be aligned with a boundary of the first predetermined units,
wherein the first predetermined unit is a minimum erase unit of the non-volatile semiconductor memory.
10. A method for controlling a memory system as set forth in claim 9, wherein the second predetermined unit is a cluster and each one of the clusters consists of the same number of sectors.
11. A method for controlling a memory system as set forth in claim 10, wherein the non-volatile semiconductor memory is a NAND-type flash memory including a plurality of blocks and the block is the first predetermined unit.
12. A method for controlling a memory system as set forth in claim 11, wherein two or more entire clusters are capable of being stored in the block.
13. A method for controlling a memory system as set forth in claim 12, further comprising:
receiving updated file data which supersedes at least one of the clusters in an original block, the original block storing original file data;
writing the updated file data in an empty block which is different from the original block, the empty block is prepared before writing the updated file data;
copying the clusters which are not superseded by the updated file data to the empty block; and
erasing the original file data stored in the original block.
14. A method for controlling a memory system as set forth in claim 11, wherein a size of the block is n (n is integer) times larger than a size of the cluster.
15. A method for controlling a memory system which includes a non-volatile semiconductor memory having a plurality of first predetermined units, comprising:
receiving file data, the file data being organized into second predetermined units; and
storing the file data so that the second predetermined units do not straddle a boundary of the first predetermined units,
wherein the first predetermined unit is a minimum erase unit of the non-volatile semiconductor memory.
16. A method for controlling a memory system as set forth in claim 15, wherein the second predetermined unit is a cluster and each one of the clusters consists of the same number of sectors.
17. A method for controlling a memory system as set forth in claim 16, wherein the non-volatile semiconductor memory is a NAND-type flash memory including a plurality of blocks and the block is the first predetermined unit.
18. A method for controlling a memory system as set forth in claim 17, wherein two or more entire clusters are capable of being stored in the block.
19. A method for controlling a memory system as set forth in claim 18, further comprising:
receiving updated file data which supersedes at least one of the clusters in an original block, the original block storing original file data;
writing the updated file data in an empty block which is different from the original block, the empty block is prepared before writing the updated file data;
copying the clusters which are not superseded by the updated file data to the empty block; and
erasing the original file data stored in the original block.
20. A method for controlling a memory system as set forth in claim 17, wherein a size of the block is n (n is integer) times larger than a size of the cluster.
21. A method for controlling a memory system as set forth in claim 11,
wherein the file system is a FAT file system.
22. A method for controlling a memory system as set forth in claim 17,
wherein the file system is a FAT file system.
US12/564,770 1997-08-08 2009-09-22 Method for Controlling Non-Volatile Semiconductor Memory System Abandoned US20100023680A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US12/564,770 US20100023680A1 (en) 1997-08-08 2009-09-22 Method for Controlling Non-Volatile Semiconductor Memory System

Applications Claiming Priority (8)

Application Number Priority Date Filing Date Title
JP09-214561 1997-08-08
JP9214561A JPH10124384A (en) 1996-08-28 1997-08-08 Method for controlling non-volatile semiconductor memory
JP11909998A JP4079506B2 (en) 1997-08-08 1998-04-28 Method for controlling nonvolatile semiconductor memory system
JP10-119099 1998-04-28
US09/130,818 US6845438B1 (en) 1997-08-08 1998-08-07 Method for controlling non-volatile semiconductor memory system by using look up table
US10/913,865 US8756401B2 (en) 1997-08-08 2004-08-06 Method for controlling non-volatile semiconductor memory system
US11/931,101 US8230156B2 (en) 1997-08-08 2007-10-31 Method for controlling non-volatile semiconductor memory system
US12/564,770 US20100023680A1 (en) 1997-08-08 2009-09-22 Method for Controlling Non-Volatile Semiconductor Memory System

Related Parent Applications (1)

Application Number Title Priority Date Filing Date
US11/931,101 Continuation US8230156B2 (en) 1997-08-08 2007-10-31 Method for controlling non-volatile semiconductor memory system

Publications (1)

Publication Number Publication Date
US20100023680A1 true US20100023680A1 (en) 2010-01-28

Family

ID=26456898

Family Applications (6)

Application Number Title Priority Date Filing Date
US09/130,818 Expired - Fee Related US6845438B1 (en) 1997-08-08 1998-08-07 Method for controlling non-volatile semiconductor memory system by using look up table
US10/786,960 Expired - Fee Related US7054991B2 (en) 1997-08-08 2004-02-25 Method for controlling non-volatile semiconductor memory system
US10/913,865 Expired - Fee Related US8756401B2 (en) 1997-08-08 2004-08-06 Method for controlling non-volatile semiconductor memory system
US11/931,101 Expired - Fee Related US8230156B2 (en) 1997-08-08 2007-10-31 Method for controlling non-volatile semiconductor memory system
US12/564,770 Abandoned US20100023680A1 (en) 1997-08-08 2009-09-22 Method for Controlling Non-Volatile Semiconductor Memory System
US12/702,134 Expired - Fee Related US9009387B2 (en) 1997-08-08 2010-02-08 Method for controlling non-volatile semiconductor memory system

Family Applications Before (4)

Application Number Title Priority Date Filing Date
US09/130,818 Expired - Fee Related US6845438B1 (en) 1997-08-08 1998-08-07 Method for controlling non-volatile semiconductor memory system by using look up table
US10/786,960 Expired - Fee Related US7054991B2 (en) 1997-08-08 2004-02-25 Method for controlling non-volatile semiconductor memory system
US10/913,865 Expired - Fee Related US8756401B2 (en) 1997-08-08 2004-08-06 Method for controlling non-volatile semiconductor memory system
US11/931,101 Expired - Fee Related US8230156B2 (en) 1997-08-08 2007-10-31 Method for controlling non-volatile semiconductor memory system

Family Applications After (1)

Application Number Title Priority Date Filing Date
US12/702,134 Expired - Fee Related US9009387B2 (en) 1997-08-08 2010-02-08 Method for controlling non-volatile semiconductor memory system

Country Status (6)

Country Link
US (6) US6845438B1 (en)
EP (2) EP1533703A3 (en)
JP (1) JP4079506B2 (en)
KR (2) KR100365385B1 (en)
CA (1) CA2245271C (en)
TW (1) TW392173B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9021184B2 (en) 2011-05-20 2015-04-28 Kabushiki Kaisha Tokai Rika Denki Seisakusho Flash memory device

Families Citing this family (89)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP4079506B2 (en) 1997-08-08 2008-04-23 株式会社東芝 Method for controlling nonvolatile semiconductor memory system
JP3544476B2 (en) 1998-09-11 2004-07-21 富士通株式会社 Creating a memory management table
JP3389186B2 (en) 1999-04-27 2003-03-24 松下電器産業株式会社 Semiconductor memory card and reading device
JP3249959B2 (en) * 1999-10-12 2002-01-28 株式会社ソニー・コンピュータエンタテインメント Portable storage device and memory card
RU2257609C2 (en) * 1999-10-21 2005-07-27 Мацусита Электрик Индастриал Ко., Лтд. Device for accessing semiconductor memory card, computer- readable data carrier, initialization method and semiconductor memory card
US6377500B1 (en) 1999-11-11 2002-04-23 Kabushiki Kaisha Toshiba Memory system with a non-volatile memory, having address translating function
US8266367B2 (en) * 2003-12-02 2012-09-11 Super Talent Electronics, Inc. Multi-level striping and truncation channel-equalization for flash-memory system
US8037234B2 (en) * 2003-12-02 2011-10-11 Super Talent Electronics, Inc. Command queuing smart storage transfer manager for striping data to raw-NAND flash modules
US8341332B2 (en) * 2003-12-02 2012-12-25 Super Talent Electronics, Inc. Multi-level controller with smart storage transfer manager for interleaving multiple single-chip flash memory devices
US6426893B1 (en) 2000-02-17 2002-07-30 Sandisk Corporation Flash eeprom system with simultaneous multiple data sector programming and storage of physical block characteristics in other designated blocks
CZ296706B6 (en) * 2000-02-21 2006-05-17 Trek 2000 International Ltd Portable data storage device
JP3942807B2 (en) 2000-06-06 2007-07-11 株式会社ルネサステクノロジ Semiconductor memory device with block alignment function
US7295443B2 (en) 2000-07-06 2007-11-13 Onspec Electronic, Inc. Smartconnect universal flash media card adapters
US6438638B1 (en) 2000-07-06 2002-08-20 Onspec Electronic, Inc. Flashtoaster for reading several types of flash-memory cards with or without a PC
JP2002091806A (en) * 2000-09-20 2002-03-29 Olympus Optical Co Ltd Filing system
JP2002132456A (en) * 2000-10-20 2002-05-10 Sony Corp Information recording device, information reproducing device, information recording method, information reproducing method, information recording medium and program providing medium
US6763424B2 (en) * 2001-01-19 2004-07-13 Sandisk Corporation Partial block data programming and reading operations in a non-volatile memory
JP4812192B2 (en) * 2001-07-27 2011-11-09 パナソニック株式会社 Flash memory device and method for merging data stored therein
TW539946B (en) * 2001-08-07 2003-07-01 Solid State System Company Ltd Window-based flash memory storage system, and the management method and the access method thereof
US6948026B2 (en) * 2001-08-24 2005-09-20 Micron Technology, Inc. Erase block management
KR100437609B1 (en) * 2001-09-20 2004-06-30 주식회사 하이닉스반도체 Method for address translation of semiconductor memory device and apparatus the same
KR100449708B1 (en) * 2001-11-16 2004-09-22 삼성전자주식회사 Flash memory management method
JP2004062328A (en) * 2002-07-25 2004-02-26 Renesas Technology Corp Flash storage medium loading nand-type flash memory
US7627229B2 (en) * 2002-07-26 2009-12-01 Fujifilm Corporation Moving image recording apparatus and method of recording moving image
US7234036B1 (en) * 2002-10-28 2007-06-19 Sandisk Corporation Method and apparatus for resolving physical blocks associated with a common logical block
DE60316171T2 (en) * 2002-10-28 2008-05-29 SanDisk Corp., Milpitas AUTOMATIC WEARING COMPENSATION IN A NON-VOLATILE STORAGE SYSTEM
US6901498B2 (en) * 2002-12-09 2005-05-31 Sandisk Corporation Zone boundary adjustment for defects in non-volatile memories
JP2005108304A (en) 2003-09-29 2005-04-21 Toshiba Corp Semiconductor memory and its control method
US8176238B2 (en) * 2003-12-02 2012-05-08 Super Talent Electronics, Inc. Command queuing smart storage transfer manager for striping data to raw-NAND flash modules
US20090193184A1 (en) * 2003-12-02 2009-07-30 Super Talent Electronics Inc. Hybrid 2-Level Mapping Tables for Hybrid Block- and Page-Mode Flash-Memory System
US20090204872A1 (en) * 2003-12-02 2009-08-13 Super Talent Electronics Inc. Command Queuing Smart Storage Transfer Manager for Striping Data to Raw-NAND Flash Modules
JP3892851B2 (en) * 2004-02-04 2007-03-14 株式会社東芝 Memory card and semiconductor device
WO2005083573A1 (en) * 2004-02-27 2005-09-09 Matsushita Electric Industrial Co., Ltd. Semiconductor memory device
JP4791356B2 (en) * 2004-05-06 2011-10-12 パナソニック株式会社 Semiconductor memory device
JP4710274B2 (en) * 2004-08-02 2011-06-29 ソニー株式会社 MEMORY DEVICE, MEMORY DEVICE CONTROL METHOD, AND DATA PROCESSING SYSTEM
US20110029723A1 (en) * 2004-08-06 2011-02-03 Super Talent Electronics, Inc. Non-Volatile Memory Based Computer Systems
KR100876084B1 (en) * 2007-02-13 2008-12-26 삼성전자주식회사 Computing system capable of delivering deletion information to flash storage
US20060184718A1 (en) * 2005-02-16 2006-08-17 Sinclair Alan W Direct file data programming and deletion in flash memories
JP4643315B2 (en) 2005-03-11 2011-03-02 株式会社東芝 Semiconductor integrated circuit device
US7644092B2 (en) * 2005-05-03 2010-01-05 Kabushiki Kaisha Toshiba System for managing data on memory device using file system
TWI275101B (en) * 2005-05-24 2007-03-01 Prolific Technology Inc Flash memory storage system
US8429326B2 (en) * 2005-09-12 2013-04-23 Mediatek Inc. Method and system for NAND-flash identification without reading device ID table
US7702821B2 (en) 2005-09-15 2010-04-20 Eye-Fi, Inc. Content-aware digital media storage device and methods of using the same
US7681109B2 (en) * 2005-10-13 2010-03-16 Ramot At Tel Aviv University Ltd. Method of error correction in MBC flash memory
US7644251B2 (en) * 2005-12-19 2010-01-05 Sigmatel, Inc. Non-volatile solid-state memory controller
TW200734875A (en) * 2006-03-08 2007-09-16 Quanta Comp Inc System and method for detecting peripheral device
JP4840975B2 (en) * 2006-03-31 2011-12-21 株式会社リコー Fade-in memory, information recording method and information reproducing method
JP2008009527A (en) * 2006-06-27 2008-01-17 Toshiba Corp Memory system
JP4182993B2 (en) * 2006-06-30 2008-11-19 Tdk株式会社 MEMORY CONTROLLER, FLASH MEMORY SYSTEM HAVING MEMORY CONTROLLER, AND FLASH MEMORY CONTROL METHOD
JP4956068B2 (en) * 2006-06-30 2012-06-20 株式会社東芝 Semiconductor memory device and control method thereof
US20080148132A1 (en) * 2006-10-26 2008-06-19 Mavila Rajith K Error detection and correction scheme for multi-level cell NAND flash
JP5171840B2 (en) * 2006-11-24 2013-03-27 サンドフォース インコーポレイテッド Method and storage system for delaying operation to shorten memory life
US7818701B1 (en) * 2006-12-22 2010-10-19 Cypress Semiconductor Corporation Memory controller with variable zone size
JP4978224B2 (en) * 2007-02-08 2012-07-18 カシオ計算機株式会社 Photoelectric conversion device and display panel having the same
KR100817087B1 (en) 2007-02-13 2008-03-27 삼성전자주식회사 Method for operating buffer cache of storage device including flash memory
KR100882740B1 (en) * 2007-02-22 2009-02-09 삼성전자주식회사 Method and storage device of mapping a nonvolatile memory based on map history
TW200839620A (en) * 2007-03-16 2008-10-01 Realtek Semiconductor Corp Card reader controller and its method of detecting interferences
US7984173B2 (en) * 2007-05-30 2011-07-19 Red Hat, Inc. Channel bundling
JP4794530B2 (en) * 2007-10-26 2011-10-19 ルネサスエレクトロニクス株式会社 Semiconductor device and mobile phone
KR101449524B1 (en) 2008-03-12 2014-10-14 삼성전자주식회사 Storage device and computing system
US7637061B2 (en) * 2008-03-13 2009-12-29 Plinths And Caissons, Llc Grave marker grid support system
US8843691B2 (en) * 2008-06-25 2014-09-23 Stec, Inc. Prioritized erasure of data blocks in a flash storage device
US7996736B2 (en) * 2008-10-26 2011-08-09 Sandisk 3D Llc Bad page marking strategy for fast readout in memory
US7899967B2 (en) * 2008-11-24 2011-03-01 Mediatek Inc. Systems for accessing memory card and methods for accessing memory card by a control unit
US8386736B2 (en) * 2008-12-18 2013-02-26 Spansion Llc Rapid memory buffer write storage system and method
US8392687B2 (en) 2009-01-21 2013-03-05 Micron Technology, Inc. Solid state memory formatting
US20100286962A1 (en) * 2009-05-05 2010-11-11 Cheng Uei Precision Industry Co., Ltd. System and method of recording measurement data
SG183887A1 (en) * 2010-03-05 2012-10-30 Eye Fi Inc Endless memory
JP5077385B2 (en) 2010-04-15 2012-11-21 株式会社デンソー Vehicle navigation device
US8429391B2 (en) * 2010-04-16 2013-04-23 Micron Technology, Inc. Boot partitions in memory devices and systems
KR101936311B1 (en) * 2010-12-03 2019-01-09 삼성전자주식회사 Method of processing data
JP5221699B2 (en) * 2011-03-23 2013-06-26 株式会社東芝 Semiconductor memory device
JP5382084B2 (en) * 2011-10-07 2014-01-08 株式会社デンソー Vehicle equipment
WO2013098463A1 (en) * 2011-12-29 2013-07-04 Nokia Corporation Method for erasing data entity in memory module
US8924636B2 (en) 2012-02-23 2014-12-30 Kabushiki Kaisha Toshiba Management information generating method, logical block constructing method, and semiconductor memory device
JP2013200584A (en) * 2012-03-23 2013-10-03 Sony Corp Electronic device
TWI486767B (en) * 2012-06-22 2015-06-01 Phison Electronics Corp Data storing method, and memory controller and memory storage apparatus using the same
US8914670B2 (en) 2012-11-07 2014-12-16 Apple Inc. Redundancy schemes for non-volatile memory using parity zones having new and old parity blocks
JP6300202B2 (en) * 2014-03-03 2018-03-28 パナソニックIpマネジメント株式会社 Memory card and memory card control device
US9710199B2 (en) * 2014-11-07 2017-07-18 International Business Machines Corporation Non-volatile memory data storage with low read amplification
US10162700B2 (en) 2014-12-23 2018-12-25 International Business Machines Corporation Workload-adaptive data packing algorithm
US9712190B2 (en) 2015-09-24 2017-07-18 International Business Machines Corporation Data packing for compression-enabled storage systems
US9870285B2 (en) 2015-11-18 2018-01-16 International Business Machines Corporation Selectively de-straddling data pages in non-volatile memory
USD786878S1 (en) 2015-12-04 2017-05-16 Capital One Services, Llc Payment card chip
CN109901956A (en) * 2017-12-08 2019-06-18 英业达科技有限公司 The system and method for memory integrated testability
US11593258B2 (en) * 2019-11-26 2023-02-28 Micron Technology, Inc. Enhanced filesystem support for zone namespace memory
US20220382487A1 (en) * 2019-12-31 2022-12-01 Micron Technology, Inc. Mobile storage random read performance estimation enhancements
CN112486854B (en) * 2020-12-28 2022-08-23 潍柴动力股份有限公司 NAND Flash Nand Flash storage management method and device
US11733927B2 (en) * 2021-11-30 2023-08-22 Microsoft Technology Licensing, Llc Hybrid solid-state drive

Citations (30)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US3432795A (en) * 1967-02-23 1969-03-11 Sylvania Electric Prod Electrical connector having facile engagement means
US4849944A (en) * 1986-08-18 1989-07-18 Tokyo Electric Company, Ltd. Connecting structure for connecting a memory unit to a memory unit controller
US4889495A (en) * 1988-09-08 1989-12-26 Mitsubishi Denki Kabushiki Kaisha Connecting mechanism for memory card
US5210855A (en) * 1989-06-09 1993-05-11 International Business Machines Corporation System for computer peripheral bus for allowing hot extraction on insertion without disrupting adjacent devices
US5329491A (en) * 1993-06-30 1994-07-12 Intel Corporation Nonvolatile memory card with automatic power supply configuration
US5438573A (en) * 1991-09-13 1995-08-01 Sundisk Corporation Flash EEPROM array data and header file structure
US5450365A (en) * 1993-05-20 1995-09-12 Fuji Photo Film Co., Ltd. Memory card control device
US5457658A (en) * 1993-02-24 1995-10-10 International Business Machines Corporation Nonvolatile memory with cluster-erase flash capability and solid state file apparatus using the same
US5463772A (en) * 1993-04-23 1995-10-31 Hewlett-Packard Company Transparent peripheral file systems with on-board compression, decompression, and space management
US5471604A (en) * 1992-10-30 1995-11-28 Intel Corporation Method for locating sector data in a memory disk by examining a plurality of headers near an initial pointer
US5537592A (en) * 1989-05-08 1996-07-16 Alphatronix System and method for reading and writing disks formatted for an operating system foreign to the host computer
US5544356A (en) * 1990-12-31 1996-08-06 Intel Corporation Block-erasable non-volatile semiconductor memory which tracks and stores the total number of write/erase cycles for each block
US5555391A (en) * 1993-12-23 1996-09-10 Unisys Corporation System and method for storing partial blocks of file data in a file cache system by merging partial updated blocks with file block to be written
US5579502A (en) * 1991-08-09 1996-11-26 Kabushiki Kaisha Toshiba Memory card apparatus using EEPROMS for storing data and an interface buffer for buffering data transfer between the EEPROMS and an external device
US5584030A (en) * 1991-07-31 1996-12-10 3Com Corporation Method and apparatus for live insertion and removal of electronic sub-assemblies
US5630093A (en) * 1990-12-31 1997-05-13 Intel Corporation Disk emulation for a non-volatile semiconductor memory utilizing a mapping table
US5636347A (en) * 1994-09-27 1997-06-03 Intel Corporation Computer card insertion detection circuit
US5717951A (en) * 1995-08-07 1998-02-10 Yabumoto; Kan W. Method for storing and retrieving information on a magnetic storage medium via data blocks of variable sizes
US5828892A (en) * 1996-02-27 1998-10-27 Mitsubishi Denki Kabushiki Kaisha Memory cards capable of operating with more than one power supply voltage
US5832525A (en) * 1996-06-24 1998-11-03 Sun Microsystems, Inc. Disk fragmentation reduction using file allocation tables
US5905993A (en) * 1994-11-09 1999-05-18 Mitsubishi Denki Kabushiki Kaisha Flash memory card with block memory address arrangement
US5937423A (en) * 1996-12-26 1999-08-10 Intel Corporation Register interface for flash EEPROM memory arrays
US6081878A (en) * 1997-03-31 2000-06-27 Lexar Media, Inc. Increasing the memory performance of flash memory devices by writing sectors simultaneously to multiple flash memory devices
US6189081B1 (en) * 1996-05-24 2001-02-13 Nec Corporation Non-volatile semiconductor storage with memory requirement and availability comparison means and method
US6278678B1 (en) * 1999-02-12 2001-08-21 Sony Corporation Editing apparatus, editing method, and recording medium
US20020069314A1 (en) * 1996-01-08 2002-06-06 Shigenori Miyauchi Semiconductor storage device
US6611907B1 (en) * 1999-10-21 2003-08-26 Matsushita Electric Industrial Co., Ltd. Semiconductor memory card access apparatus, a computer-readable recording medium, an initialization method, and a semiconductor memory card
US20050144363A1 (en) * 2003-12-30 2005-06-30 Sinclair Alan W. Data boundary management
US20050193161A1 (en) * 2004-02-26 2005-09-01 Lee Charles C. System and method for controlling flash memory
US20090180209A1 (en) * 2005-09-22 2009-07-16 Matsushita Electric Industrial Co., Ltd. Information recording medium access device, and data recording method

Family Cites Families (38)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US3828327A (en) * 1973-04-30 1974-08-06 Ibm Simplified storage protection and address translation under system mode control in a data processing system
FR2553541B1 (en) * 1983-10-17 1992-02-28 Inst Nat Rech Inf Automat DEVICE AND METHOD FOR FAST AND STABLE STORAGE OF INFORMATION
JPS60160491A (en) * 1984-01-31 1985-08-22 Toshiba Corp Ic card
US4794524A (en) * 1984-07-03 1988-12-27 Zilog, Inc. Pipelined single chip microprocessor having on-chip cache and on-chip memory management unit
JPS61148551A (en) * 1984-12-24 1986-07-07 Hitachi Ltd Address converting system
JP2511950B2 (en) 1987-03-27 1996-07-03 三菱電機株式会社 Interface circuit
US4980816A (en) * 1987-12-18 1990-12-25 Nec Corporation Translation look-aside buffer control system with multiple prioritized buffers
JPH01296482A (en) 1988-05-24 1989-11-29 Mitsubishi Electric Corp Memory card
US5404476A (en) * 1989-02-10 1995-04-04 Nec Corporation Multiprocessing system having a single translation lookaside buffer with reduced processor overhead
JPH0484216A (en) 1990-07-26 1992-03-17 Toshiba Corp Data erasing method for semiconductor disk device
US5430857A (en) * 1993-01-04 1995-07-04 Intel Corporation Method and apparatus for translating logical addresses into physical addresses using odd/even translation tables
DE4339179A1 (en) 1993-04-06 1994-10-13 Hewlett Packard Co Method and apparatus for preventing signal corruption on a common computer bus
WO1994027222A1 (en) * 1993-05-10 1994-11-24 Jochen Liedtke Process for converting a virtual store address with a first length into a real address with a second length
US5479628A (en) * 1993-10-12 1995-12-26 Wang Laboratories, Inc. Virtual address translation hardware assist circuit and method
US5386402A (en) * 1993-11-29 1995-01-31 Kabushiki Kaisha Toshiba Access control apparatus and address translation method for disk storage device
JP2556285B2 (en) 1994-02-15 1996-11-20 ダイソー株式会社 Method for producing chlorinated polyethylene cross-linked composition
JP2609431B2 (en) 1994-09-22 1997-05-14 山一電機株式会社 Memory card
JPH08211973A (en) 1995-01-26 1996-08-20 Mitsubishi Electric Corp Pc card
JP3706167B2 (en) 1995-02-16 2005-10-12 株式会社ルネサステクノロジ Semiconductor disk device
US5644731A (en) 1995-07-07 1997-07-01 Sun Microsystems, Inc. Method and apparatus for hot plugging/unplugging a sub-system to an electrically powered system
JP3604466B2 (en) 1995-09-13 2004-12-22 株式会社ルネサステクノロジ Flash disk card
JPH0981459A (en) * 1995-09-19 1997-03-28 Hitachi Ltd Address conversion buffer device
JP3727982B2 (en) 1995-09-28 2005-12-21 キヤノン株式会社 Flash ROM management method and apparatus
JPH09198201A (en) 1996-01-16 1997-07-31 Toshiba Corp Semiconductor disk device and method for controlling number of rewriting thereof
JP3197815B2 (en) 1996-04-15 2001-08-13 インターナショナル・ビジネス・マシーンズ・コーポレ−ション Semiconductor memory device and control method thereof
JP2843306B2 (en) * 1996-08-02 1999-01-06 山一電機株式会社 Protection device for IC memory card
JPH1049257A (en) 1996-08-05 1998-02-20 Yokogawa Electric Corp Information processor
JP3588231B2 (en) 1997-08-04 2004-11-10 東京エレクトロンデバイス株式会社 Data processing system and block erase type storage medium
JP4079506B2 (en) 1997-08-08 2008-04-23 株式会社東芝 Method for controlling nonvolatile semiconductor memory system
JP3524428B2 (en) * 1999-04-20 2004-05-10 東京エレクトロンデバイス株式会社 Storage device, storage system, memory management method, and recording medium
US6301646B1 (en) * 1999-07-30 2001-10-09 Curl Corporation Pointer verification system and method
US6377500B1 (en) * 1999-11-11 2002-04-23 Kabushiki Kaisha Toshiba Memory system with a non-volatile memory, having address translating function
US7003771B1 (en) * 2000-06-08 2006-02-21 International Business Machines Corporation Logically partitioned processing system having hypervisor for creating a new translation table in response to OS request to directly access the non-assignable resource
US6684289B1 (en) * 2000-11-22 2004-01-27 Sandisk Corporation Techniques for operating non-volatile memory systems with data sectors having different sizes than the sizes of the pages and/or blocks of the memory
JP4037605B2 (en) * 2000-12-04 2008-01-23 株式会社東芝 Nonvolatile memory unit controller, memory system having the controller, and nonvolatile memory unit control method
JP3579389B2 (en) * 2001-11-30 2004-10-20 株式会社東芝 Disk array device and data recovery method in the device
JP4206688B2 (en) * 2002-04-15 2009-01-14 ソニー株式会社 Data processing apparatus and data processing method
US7076634B2 (en) * 2003-04-24 2006-07-11 International Business Machines Corporation Address translation manager and method for a logically partitioned computer system

Patent Citations (31)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US3432795A (en) * 1967-02-23 1969-03-11 Sylvania Electric Prod Electrical connector having facile engagement means
US4849944A (en) * 1986-08-18 1989-07-18 Tokyo Electric Company, Ltd. Connecting structure for connecting a memory unit to a memory unit controller
US4889495A (en) * 1988-09-08 1989-12-26 Mitsubishi Denki Kabushiki Kaisha Connecting mechanism for memory card
US5537592A (en) * 1989-05-08 1996-07-16 Alphatronix System and method for reading and writing disks formatted for an operating system foreign to the host computer
US5210855A (en) * 1989-06-09 1993-05-11 International Business Machines Corporation System for computer peripheral bus for allowing hot extraction on insertion without disrupting adjacent devices
US5544356A (en) * 1990-12-31 1996-08-06 Intel Corporation Block-erasable non-volatile semiconductor memory which tracks and stores the total number of write/erase cycles for each block
US5630093A (en) * 1990-12-31 1997-05-13 Intel Corporation Disk emulation for a non-volatile semiconductor memory utilizing a mapping table
US5584030A (en) * 1991-07-31 1996-12-10 3Com Corporation Method and apparatus for live insertion and removal of electronic sub-assemblies
US5579502A (en) * 1991-08-09 1996-11-26 Kabushiki Kaisha Toshiba Memory card apparatus using EEPROMS for storing data and an interface buffer for buffering data transfer between the EEPROMS and an external device
US5438573A (en) * 1991-09-13 1995-08-01 Sundisk Corporation Flash EEPROM array data and header file structure
US5471604A (en) * 1992-10-30 1995-11-28 Intel Corporation Method for locating sector data in a memory disk by examining a plurality of headers near an initial pointer
US5457658A (en) * 1993-02-24 1995-10-10 International Business Machines Corporation Nonvolatile memory with cluster-erase flash capability and solid state file apparatus using the same
US5463772A (en) * 1993-04-23 1995-10-31 Hewlett-Packard Company Transparent peripheral file systems with on-board compression, decompression, and space management
US5450365A (en) * 1993-05-20 1995-09-12 Fuji Photo Film Co., Ltd. Memory card control device
US5329491A (en) * 1993-06-30 1994-07-12 Intel Corporation Nonvolatile memory card with automatic power supply configuration
US5555391A (en) * 1993-12-23 1996-09-10 Unisys Corporation System and method for storing partial blocks of file data in a file cache system by merging partial updated blocks with file block to be written
US5636347A (en) * 1994-09-27 1997-06-03 Intel Corporation Computer card insertion detection circuit
US5905993A (en) * 1994-11-09 1999-05-18 Mitsubishi Denki Kabushiki Kaisha Flash memory card with block memory address arrangement
US5717951A (en) * 1995-08-07 1998-02-10 Yabumoto; Kan W. Method for storing and retrieving information on a magnetic storage medium via data blocks of variable sizes
US20020069314A1 (en) * 1996-01-08 2002-06-06 Shigenori Miyauchi Semiconductor storage device
US5828892A (en) * 1996-02-27 1998-10-27 Mitsubishi Denki Kabushiki Kaisha Memory cards capable of operating with more than one power supply voltage
US6189081B1 (en) * 1996-05-24 2001-02-13 Nec Corporation Non-volatile semiconductor storage with memory requirement and availability comparison means and method
US5832525A (en) * 1996-06-24 1998-11-03 Sun Microsystems, Inc. Disk fragmentation reduction using file allocation tables
US5937423A (en) * 1996-12-26 1999-08-10 Intel Corporation Register interface for flash EEPROM memory arrays
US6081878A (en) * 1997-03-31 2000-06-27 Lexar Media, Inc. Increasing the memory performance of flash memory devices by writing sectors simultaneously to multiple flash memory devices
US6278678B1 (en) * 1999-02-12 2001-08-21 Sony Corporation Editing apparatus, editing method, and recording medium
US6611907B1 (en) * 1999-10-21 2003-08-26 Matsushita Electric Industrial Co., Ltd. Semiconductor memory card access apparatus, a computer-readable recording medium, an initialization method, and a semiconductor memory card
US20030200411A1 (en) * 1999-10-21 2003-10-23 Takuji Maeda Semiconductor memory card access apparatus, a computer-readable recording medium, an initialization method, and a semiconductor memory card
US20050144363A1 (en) * 2003-12-30 2005-06-30 Sinclair Alan W. Data boundary management
US20050193161A1 (en) * 2004-02-26 2005-09-01 Lee Charles C. System and method for controlling flash memory
US20090180209A1 (en) * 2005-09-22 2009-07-16 Matsushita Electric Industrial Co., Ltd. Information recording medium access device, and data recording method

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9021184B2 (en) 2011-05-20 2015-04-28 Kabushiki Kaisha Tokai Rika Denki Seisakusho Flash memory device

Also Published As

Publication number Publication date
KR100438012B1 (en) 2004-07-02
JPH11110283A (en) 1999-04-23
EP0896280A2 (en) 1999-02-10
EP1533703A3 (en) 2007-10-24
US7054991B2 (en) 2006-05-30
US8230156B2 (en) 2012-07-24
US20050007865A1 (en) 2005-01-13
US6845438B1 (en) 2005-01-18
US20080059695A1 (en) 2008-03-06
CA2245271C (en) 2002-01-01
KR20020075310A (en) 2002-10-04
EP0896280A3 (en) 2004-05-12
US20050005059A1 (en) 2005-01-06
US8756401B2 (en) 2014-06-17
KR19990023469A (en) 1999-03-25
KR100365385B1 (en) 2003-04-08
US20100138606A1 (en) 2010-06-03
JP4079506B2 (en) 2008-04-23
TW392173B (en) 2000-06-01
US9009387B2 (en) 2015-04-14
CA2245271A1 (en) 1999-02-08
EP1533703A2 (en) 2005-05-25

Similar Documents

Publication Publication Date Title
US8230156B2 (en) Method for controlling non-volatile semiconductor memory system
US7116578B2 (en) Non-volatile memory device and data storing method
US7616485B2 (en) Semiconductor memory device having faulty cells
US7102943B2 (en) Non-volatile semiconductor memory device
US8352672B2 (en) Memory system with nonvolatile memory
US9383929B2 (en) Data storing method and memory controller and memory storage device using the same
CN113360427A (en) Data storage device and operation method thereof
CN103678162B (en) System data storage method, memory controller and memory storage device
US20220318157A1 (en) Control method of flash memory controller and associated flash memory controller and storage device
JP2007293917A (en) Method for controlling memory system
US8276033B2 (en) Data writing method for a flash memory, and flash memory controller and flash memory storage apparatus using the same
CN113138720B (en) Data storage method, memory control circuit unit and memory storage device
CN106326131B (en) Memory management method, memory control circuit unit and memory storage device
CN116225993A (en) Memory operation control method, memory device and memory control circuit unit
CN113094307A (en) Mapping information management method, memory storage device and memory controller
CN114627941A (en) Memory management method, memory storage device and memory control circuit unit
CN111858389A (en) Data writing method, memory control circuit unit and memory storage device
CN113220502A (en) Memory management method, memory storage device and memory controller
CN114840444A (en) Memory system and operating method thereof

Legal Events

Date Code Title Description
STCB Information on status: application discontinuation

Free format text: ABANDONED -- FAILURE TO RESPOND TO AN OFFICE ACTION