TW201441816A - Data merging method for non-volatile memory and controller and storage apparatus using the same - Google Patents

Abstract

A data merging method for merging valid data of one logical block in a rewritable non-volatile memory module is provided. The method includes assigning a plurality of arrangement physical blocks for the logical block. The method also includes performing a data arrangement operation and a data move operation with a partial synchronization manner to copy the valid data of the logical block into the low physical pages of the arrangement physical blocks from a first data physical block and at least one temporary physical block while programing the valid data of the logical block into a second data physical block from the low physical pages of the arrangement physical blocks in units of each physical page group. The method further includes remapping the logical block to the second physical block. Accordingly, the method can effectively shorten the time of merging valid data and improving the reliability of writing data.

Description

Data merging method, controller and storage device for non-volatile memory

The invention relates to a data merging method for a rewritable non-volatile memory module and a memory controller and a memory storage device using the same.

Digital cameras, mobile phones and MP3s have grown very rapidly in recent years, and the demand for storage media has increased rapidly. Because rewritable non-volatile memory has the characteristics of non-volatile data, power saving, small size, no mechanical structure, fast reading and writing speed, etc., it is most suitable for portable electronic products, such as notebook type. computer. A solid state hard disk is a storage device that uses flash memory as a storage medium. Therefore, in recent years, the flash memory industry has become a very popular part of the electronics industry.

According to the number of bits that each memory cell can store, the (NAND) type flash memory can be divided into single level cell (SLC) NAND type flash memory, multi-level memory cell (Multi Level). Cell, MLC) NAND type fast Flash memory and Trinary Level Cell (TLC) NAND flash memory, in which each memory cell of SLC NAND flash memory can store 1 bit of data (ie, "1" and "0"), each memory cell of the MLC NAND type flash memory can store 2 bits of data and each memory cell of the TLC NAND type flash memory can store 3 bits of data.

The MLC NAND type flash memory has a plurality of physical blocks, and each physical block has a plurality of physical pages.

Specifically, in a NAND type flash memory, a physical page is composed of a plurality of memory cells arranged on the same word line. Since each memory cell of the SLC NAND type flash memory can store one bit of data, in the SLC NAND type flash memory, a plurality of memory cells arranged on the same word line correspond to one entity. page.

Compared with the SLC NAND type flash memory, the floating gate storage layer of each memory cell of the MLC NAND type flash memory can store 2 bits of data, each of which is stored (ie, "11", "10", "01" and "00") include a Least Significant Bit (LSB) and a Most Significant Bit (MSB). For example, the value of the first bit from the left side in the storage state is the LSB, and the value of the second bit from the left side is the MSB. Therefore, a plurality of memory cells arranged on the same character line can form two physical pages, wherein the physical page composed of the LSBs of the memory cells is called a lower physical page, and thus The physical page composed of the MSB of the memory cell is called the upper physical page. In particular, the next page of physical pages The write speed of the face will be faster than the write speed of the physical page of the previous page, and when an error occurs on the stylized page physical page, the data stored on the physical page of the next page may also be lost.

Similarly, in TLC NAND type flash memory, each memory cell can store 3 bits of data, each of which stores state (ie, "111", "110", "101", "100" "011", "010", "001" and "000") include the LSB of the first bit from the left side of each storage state, and the intermediate effective bit of the second bit from the left side. (Center Significant Bit, CSB) and the MSB of the third bit from the left. Therefore, a plurality of memory cells arranged on the same character line can form three physical pages, wherein the physical page composed of the LSBs of the memory cells is called a physical page of the next page, and thus the CSB of the memory cells is composed. The physical page is called the middle page physical page, and the physical page composed of the MSBs of the memory cells is called the upper page physical page. In particular, when storing data in a physical page composed of a plurality of memory cells arranged on the same character line, it is only possible to select only the stylized lower page physical page to store data or to simultaneously use the stylized lower page physical page. The physical page of the middle page and the physical page of the previous page are used to store the data, otherwise the stored data may be lost. For example, if the data is stored only on the page entity page and the middle page entity page, which are composed of a plurality of memory cells arranged on the same word line, the data is read from the next page entity page or the middle page entity page. This read operation will fail.

In addition, when writing data in a physical block, the data must be written in units of physical pages, and the physical page that has been written into the data must be erased before being Used again to write data. In particular, the physical block is the smallest unit of erasure. Therefore, in general, during the writing process of the flash memory module, the physical block is used to write data.

For example, when the data of a certain logical block is stored in a data entity block (hereinafter referred to as the original mapping data entity block) and the host system wants to update the data stored on a certain logical page of a certain logical block. The memory controller of the storage device extracts a physical block from the flash memory module as a temporary physical block corresponding to the logical block, and writes the updated data to the temporary physical block. In the physical page, the time to execute the write command is thus shortened. Then, when the unused physical block in the flash memory module is exhausted, the memory controller performs a data merge (Merge) process on the logical block. For example, in the data merge program, the memory controller extracts an empty physical block as a new data entity block, and all valid data belonging to the logical block from the original mapped physical block and the temporary physical block. Copy to the new data entity block and remap this logical block to this new data entity block.

However, as described above, the reliability of some physical pages in the MLC NAND-type flash memory or the TLC NAND-type flash memory is low, so how to efficiently move data between physical blocks for data merge is The goal of the technical staff in this field.

The invention provides a data merging method, a memory controller and a memory A storage device that improves the efficiency of data merging and the reliability of the data being written.

An exemplary embodiment of the present invention provides a data merging method for merging valid data of a logical block in a rewritable non-volatile memory module, wherein the rewritable non-volatile memory module has multiple physical regions. Block, each physical block has a plurality of physical page groups, each physical page group has at least one lower page physical page and one upper page physical page, and writing data to the next page physical page is faster than writing the data to the previous page The speed of the physical page, the valid data of the logical block is stored in the first data entity block and the at least one temporary physical block, and the valid data of the logical block is to be merged into the second data entity area. Piece. The method for merging data includes: assigning a plurality of collating entity blocks corresponding to the logical block. In addition, the method of merging data also includes performing a data sorting operation and a data moving operation in a partially synchronized manner, wherein the data sorting operation is to use the valid data of the logical block from the first data entity block and the temporary physical block. Sorting to the next page entity page of the physical block, the data moving operation is used to move the valid data of the logical block to the second data entity block from the finishing physical block. Furthermore, the method of merging data further includes re-mapping the logical block to the second data entity block.

In an embodiment of the present invention, the step of performing the data sorting operation and the data moving operation in a partially synchronous manner includes: (a) transferring the logical block from the first data entity block and the temporary storage physical block. The valid data belonging to the plurality of logical pages among the valid data is copied to the next page physical page of the finishing physical block, wherein the number of the logical pages is a predetermined number; (b) from the finishing physical blocks The valid data of the logical pages is copied to the second data entity block, and at the same time Copying, from the first data entity block and the temporary storage entity block, the subsequent valid data belonging to the other logical pages among the valid data of the logical block to the subsequent lower page entity page of the finishing physical block; and (c Repeat steps (a) and (b) until all valid data of the above logical block is copied to the second data entity block.

In an embodiment of the present invention, the step of performing the data sorting operation and the data moving operation in a partially synchronous manner includes: (a) validating the logical block from the first data entity block and the temporary storage physical block. The valid data belonging to one logical page in the data is copied to the next page entity page of the finishing physical block; (b) copying the valid data of the logical page from the finishing physical block to the second physical entity block, and simultaneously Copying valid data belonging to the next logical page from the valid data of the logical block to the subsequent page physical page of the finishing entity block in the first data entity block and the temporary storage entity block; and (c) repeating Steps (a) and (b) are performed until all valid data of the logical block is copied to the second data entity block.

In one embodiment of the invention, the data transfer operation described above is performed by using a copyback instruction.

In an embodiment of the present invention, each of the physical page groups has a medium page physical page, and the data is written to the next page physical page faster than the speed of writing the data to the middle page physical page, and writing The data to the physical page of the middle page is faster than the speed at which the data is written to the physical page of the previous page.

In an embodiment of the present invention, the data merging method further includes: grouping the physical blocks into at least a data area and a temporary storage area, wherein the first data entity block and the second data entity block belong to the data area and Temporary physical block is from temporary storage The area is assigned.

In an embodiment of the present invention, the step of assigning from the temporary storage area as the finishing physical block corresponding to the logical block comprises: extracting three physical blocks from the temporary storage area as corresponding to the logical block. The first finishing physical block, the second finishing physical block, and the third finishing physical block.

An exemplary embodiment of the present invention provides a memory controller for controlling a rewritable non-volatile memory module, wherein the rewritable non-volatile memory module has a plurality of physical blocks, each physical block There are multiple physical page groups, each physical page group has at least one lower page physical page and one last physical page, and the data is written to the next page physical page faster than the data is written to the upper page physical page. The memory controller includes a host interface, a memory interface and a memory management circuit. The host interface is coupled to the host system. The memory interface is coupled to the rewritable non-volatile memory module. The memory management circuit is coupled to the host interface and the memory interface, and is configured to merge the valid data of one logical block into the second physical entity block, wherein the valid data of the logical block is stored in the first data in a dispersed manner. In the physical block and the temporary physical block. Here, the memory management circuit assigns a plurality of collating physical blocks corresponding to the logical block. In addition, the memory management circuit performs the data sorting operation and the data moving operation in a partially synchronized manner, wherein the data sorting operation is used to sort the valid data of the logical block from the first data entity block and the temporary storage physical block to The next page physical page of the physical block is arranged, and the data moving operation is used to move the valid data of the logical block from the finishing physical block to the second physical entity block. Furthermore, the memory management circuit is used to remap this logical block. Shot to the second data entity block.

In an embodiment of the present invention, during the data sorting operation and the data moving operation are performed in a partially synchronous manner, the memory management circuit validates the logical block from the first data entity block and the temporary physical block. The valid data belonging to the plurality of logical pages among the data is copied into the next page physical pages of the finishing physical blocks, wherein the number of the logical pages is a predetermined number. In addition, during the data sorting operation and the data moving operation are performed in a partially synchronous manner, the memory management circuit copies the valid data belonging to the logical pages from the sorted physical blocks to the second data entity block, and simultaneously from the In the data entity block and the temporary storage entity block, the subsequent valid data belonging to the other logical pages among the valid data of the logical block is copied into the subsequent lower page entity pages of the finishing physical blocks.

In an embodiment of the present invention, during the data sorting operation and the data moving operation are performed in a partially synchronous manner, the memory management circuit validates the logical block from the first data entity block and the temporary storage physical block. The valid data belonging to a logical page among the data is copied to the next page entity page of the finishing physical block. In addition, during the data sorting operation and the data moving operation are performed in a partially synchronous manner, the memory management circuit copies the valid data belonging to the logical page from the sorted physical blocks to the second data physical block, and simultaneously from the first In the data entity block and the temporary storage entity block, the valid data belonging to the next logical page among the valid data of the logical block is copied to the subsequent lower page entity page of the finishing physical block.

In an embodiment of the invention, the memory management circuit described above performs a data transfer operation using a copy back instruction.

In an embodiment of the present invention, each physical page group has a medium page physical page, and the data is written to the next page physical page faster than the speed of writing the data to the middle page physical page, and the data is written to The medium page physical page is faster than the data written to the physical page of the previous page.

In an embodiment of the present invention, the memory management circuit groups the physical blocks into at least a data area and a temporary storage area, wherein the first data entity block and the second data entity block belong to the data area and are temporarily stored. The physical block is assigned from the temporary storage area.

In an embodiment of the present invention, the memory management circuit extracts three physical blocks from the temporary storage area as the first finishing physical block, the second finishing physical block, and the third finishing corresponding to the logical block. Physical block.

An exemplary embodiment of the present invention provides a memory storage device including a connector, a rewritable non-volatile memory module, and a memory controller. The connector is coupled to a host system. The rewritable non-volatile memory module has a plurality of physical blocks, each physical block has a plurality of physical page groups, and each physical page group has at least one lower page physical page and one upper page physical page, and is written Entering the data to the next page of the physical page is faster than writing the data to the physical page of the previous page. The memory controller is coupled to the connector and the rewritable non-volatile memory module, and is configured to merge the valid data of one logical block into the second physical entity block, where the effective data of the logical block is dispersed The ground is stored in the first data entity block and the temporary storage entity block. Here, the memory controller assigns a plurality of collating physical blocks corresponding to the logical block. In addition, the memory controller performs data sorting in a partially synchronized manner. And the data moving operation, wherein the data sorting operation is used to sort the valid data of the logical block from the first data entity block and the temporary physical block to the next page physical page of the physical block, and the data is moved. The operation is for moving the valid data of the logical block to the second data entity block from the finishing physical block. Furthermore, the memory controller is further configured to remap the logical block to the second data entity block.

In an embodiment of the present invention, during the data sorting operation and the data moving operation are performed in a partially synchronous manner, the memory controller validates the logical block from the first data entity block and the temporary storage physical block. The valid data belonging to the plurality of logical pages among the data is copied into the next page physical pages of the finishing physical blocks, wherein the number of the logical pages is a predetermined number. In addition, during the data sorting operation and the data moving operation are executed in a partially synchronous manner, the memory controller copies the valid data belonging to the logical pages from the sorted physical blocks to the second data entity block, and simultaneously from the In the data entity block and the temporary storage entity block, the subsequent valid data belonging to the other logical pages among the valid data of the logical block is copied into the subsequent lower page entity pages of the finishing physical blocks.

In an embodiment of the present invention, during the data sorting operation and the data moving operation are performed in a partially synchronous manner, the memory controller validates the logical block from the first data entity block and the temporary storage physical block. The valid data belonging to a logical page among the data is copied to the next page entity page of the finishing physical block. In addition, during the data sorting operation and the data moving operation are performed in a partially synchronous manner, the memory controller copies the valid data belonging to the logical page from the sorted physical blocks to the second data entity block, and simultaneously from the first Data entity block and temporary In the physical block, the valid data belonging to the next logical page among the valid data of the logical block is copied to the subsequent lower page physical page of the finishing physical block.

In an embodiment of the invention, the memory controller described above performs a data transfer operation using a copy back instruction.

In an embodiment of the present invention, each physical page group has a medium page physical page, and the data is written to the next page physical page faster than the speed of writing the data to the middle page physical page, and the data is written to The medium page physical page is faster than the data written to the physical page of the previous page.

In an embodiment of the present invention, the memory controller groups the physical blocks into at least a data area and a temporary storage area, wherein the first data entity block and the second data entity block belong to the data area and are temporarily stored. The physical block is assigned from the temporary storage area.

In an embodiment of the present invention, the memory controller extracts three physical blocks from the temporary storage area as the first finishing physical block, the second finishing physical block, and the third finishing corresponding to the logical block. Physical block.

Based on the above, the data merging method, the memory controller and the memory storage device of the exemplary embodiments of the present invention can effectively improve the reliability of the merged data and shorten the time required to perform data merging.

The above described features and advantages of the present invention will be more apparent from the following description.

1000‧‧‧Host system

1100‧‧‧ computer

1102‧‧‧Microprocessor

1104‧‧‧ Random access memory

1106‧‧‧Input/output devices

1108‧‧‧System Bus

1110‧‧‧Data transmission interface

1202‧‧‧ Mouse

1204‧‧‧ keyboard

1206‧‧‧ display

1208‧‧‧Printer

1212‧‧‧USB flash drive

1214‧‧‧ memory card

1216‧‧‧ Solid State Drive

1310‧‧‧ digital camera

1312‧‧‧SD card

1314‧‧‧MMC card

1316‧‧‧ Memory Stick

1318‧‧‧CF card

1320‧‧‧Embedded storage device

100‧‧‧ memory storage device

102‧‧‧Connector

104‧‧‧ memory controller

106‧‧‧Reusable non-volatile memory module

302‧‧‧Memory Management Circuit

304‧‧‧Host interface

306‧‧‧ memory interface

308‧‧‧ Buffer memory

310‧‧‧Power Management Circuit

312‧‧‧Error checking and correction circuit

502‧‧‧Substitute area

504‧‧‧ temporary storage area

506‧‧‧Information area

410(0)~410(R), 410(R+1)~410(T), 410(T+1)~410(N)‧‧‧ physical blocks

610(0)~610(H)‧‧‧ logical block

Steps of S901, S903, S905, S907‧‧‧ data merge method

S1001, S1003, S1005, S1007‧‧‧ Steps to perform data sorting operations and data movement operations in a partially synchronized manner

FIG. 1A illustrates a host system and a memory storage device according to an exemplary embodiment.

FIG. 1B is a schematic diagram of a computer, an input/output device, and a memory storage device according to an exemplary embodiment of the invention.

FIG. 1C is a schematic diagram of a host system and a memory storage device according to another exemplary embodiment of the invention.

FIG. 2 is a schematic block diagram showing the memory storage device shown in FIG. 1A.

FIG. 3A and FIG. 3B are schematic diagrams showing examples of a memory cell storage architecture and a physical block according to an embodiment of the present exemplary embodiment.

4 is a schematic block diagram of a memory controller according to an exemplary embodiment.

FIG. 5 is a schematic diagram of managing a physical block of a rewritable non-volatile memory module according to an exemplary embodiment.

FIG. 6 is a schematic diagram of writing data according to an example.

FIG. 7 and FIG. 8 are schematic diagrams of a data merge procedure according to an example. FIG. 7 is a schematic diagram showing an example of data sorting operation and FIG. 8 is a schematic diagram showing an example of data shifting operation.

FIG. 9 is a flowchart of a data merging method according to an exemplary embodiment.

FIG. 10 is a detailed flowchart of step S905 according to an exemplary embodiment.

In general, a memory storage device (also referred to as a memory storage system) includes a rewritable non-volatile memory module and controller (also referred to as a control circuit). Typically, the memory storage device is used with a host system to enable the host system to write data to or read data from the memory storage device.

FIG. 1A illustrates a host system and a memory storage device according to an exemplary embodiment.

Referring to FIG. 1A, the host system 1000 generally includes a computer 1100 and an input/output (I/O) device 1106. The computer 1100 includes a microprocessor 1102, a random access memory (RAM) 1104, a system bus 1108, and a data transmission interface 1110. The input/output device 1106 includes a mouse 1202, a keyboard 1204, a display 1206, and a printer 1208 as in FIG. 1B. It must be understood that the device shown in FIG. 1B is not limited to the input/output device 1106, and the input/output device 1106 may further include other devices.

In the embodiment of the present invention, the memory storage device 100 is coupled to other components of the host system 1000 through the data transmission interface 1110. The data can be written to or read from the memory storage device 100 by the operation of the microprocessor 1102, the random access memory 1104, and the input/output device 1106. For example, the memory storage device 100 may be a rewritable non-volatile memory storage device such as a flash drive 1212, a memory card 1214, or a solid state drive (SSD) 1216 as shown in FIG. 1B.

In general, the host system 1000 can be substantially loaded with memory. Set 100 to match any system that stores data. Although in the present exemplary embodiment, the host system 1000 is illustrated by a computer system, in another exemplary embodiment of the present invention, the host system 1000 may be a digital camera, a video camera, a communication device, an audio player, or a video player. And other systems. For example, when the host system is a digital camera (camera) 1310, the rewritable non-volatile memory storage device uses the SD card 1312, the MMC card 1314, the memory stick 1316, the CF card 1318 or Embedded storage device 1320 (shown in Figure 1C). The embedded storage device 1320 includes an embedded multimedia card (Embedded MMC, eMMC). It is worth mentioning that the embedded multimedia card is directly coupled to the substrate of the host system.

FIG. 2 is a schematic block diagram showing the memory storage device shown in FIG. 1A.

Referring to FIG. 2, the memory storage device 100 includes a connector 102, a memory controller 104, and a rewritable non-volatile memory module 106.

In the present exemplary embodiment, connector 102 is compatible with the Secure Digital (SD) interface standard. However, it must be understood that the present invention is not limited thereto, and the connector 102 may also conform to the Institute of Electrical and Electronic Engineers (IEEE) 1394 standard, Parallel Advanced Technology Attachment (PATA) standard. , Peripheral Component Interconnect Express (PCI Express) standard, Universal Serial Bus (USB) standard, Serial Advanced Technology Attachment (SATA) standard, Memory Stick (MS) Interface standard, multimedia memory card (MMC) interface standard, compact flash (Compact Flash, CF) interface standard, Integrated Device Electronics (IDE) standard or other suitable standard.

The memory controller 104 is configured to execute a plurality of logic gates or control commands implemented in a hard type or a firmware type, and perform data in the rewritable non-volatile memory module 106 according to instructions of the host system 1000. Write, read, erase, and merge operations.

The rewritable non-volatile memory module 106 is coupled to the memory controller 104 and has a plurality of physical blocks to store data written by the host system 1000.

In this exemplary embodiment, each physical block has a plurality of physical page groups, and each physical page group includes at least one physical page composed of memory cells located on the same word line, wherein the same physical area belongs to the same physical area. The physical pages of the block must be erased at the same time. In more detail, the physical block is the smallest unit of erasure. That is, each physical block contains one of the smallest number of erased memory cells.

Each physical page typically includes a data bit area and a redundant bit area. The data bit area is used to store the user's data, and the redundant bit area is used to store system data (eg, error checking and correction codes). In this exemplary embodiment, each physical block is composed of 258 physical pages, and the capacity of each physical page is 8 kilobytes (Kilobyte, KB). However, it must be understood that the invention is not limited thereto.

In the exemplary embodiment, the rewritable non-volatile memory module 106 is a third-level storage unit (TLC) NAND-type flash memory module. However, it must be understood that the rewritable non-volatile memory module 106 Not limited to TLC NAND type flash memory modules. In another exemplary embodiment of the present invention, the rewritable non-volatile memory module 106 can also be an MLC NAND type flash memory module or other memory modules having the same characteristics.

FIG. 3A and FIG. 3B are schematic diagrams showing examples of a memory cell storage architecture and a physical block according to an exemplary embodiment of the present invention.

Referring to FIG. 3A, the storage state of each memory cell of the rewritable non-volatile memory module 106 can be identified as "111", "110", "101", "100", "011", "010". "," 001" or "000" (as shown in Figure 3A), where the first bit from the left is the LSB, the second bit from the left is the CSB, and the third from the left The bits are MSB. In addition, a plurality of memory cells arranged on the same character line can form three physical pages, wherein the physical page composed of the LSBs of the memory cells is called a physical page of the next page, and thus the CSB of the memory cells is composed. The physical page is called the middle page physical page, and the physical page composed of the MSBs of the memory cells is called the upper page physical page.

Referring to FIG. 3B, for example, in the exemplary embodiment, a physical block is composed of a plurality of physical page groups (ie, 0th to 85th physical page groups), wherein each physical page group includes an arrangement. The next page entity page, the middle page entity page and the upper page entity page composed of a plurality of memory cells on the same character line. For example, the 0th entity page belonging to the next page entity page, the 1st entity page belonging to the middle page entity page, and the 2nd entity page belonging to the upper page entity page are regarded as one entity page group. Similarly, the 3rd, 4th, and 5th physical pages will be treated as one physical page. Groups, and so on, other entity pages are also differentiated into multiple entity page groups in this way.

4 is a schematic block diagram of a memory controller according to an exemplary embodiment. It should be understood that the memory controller illustrated in FIG. 4 is merely an example, and the present invention is not limited thereto.

Referring to FIG. 4, the memory controller 104 includes a memory management circuit 302, a host interface 304, a memory interface 306, a buffer memory 308, a power management circuit 310, and an error checking and correction circuit 312.

The memory management circuit 302 is used to control the overall operation of the memory controller 104. Specifically, the memory management circuit 302 has a plurality of control commands, and when the memory storage device 100 operates, such control commands are executed to perform operations such as writing, reading, and erasing data.

In the present exemplary embodiment, the control instructions of the memory management circuit 302 are implemented in a firmware version. For example, the memory management circuit 302 has a microprocessor unit (not shown) and a read-only memory (not shown), and such control instructions are programmed into the read-only memory. When the memory storage device 100 is in operation, such control commands are executed by the microprocessor unit to perform operations such as writing, reading, and erasing data.

In another exemplary embodiment of the present invention, the control command of the memory management circuit 302 can also be stored in a specific area of the rewritable non-volatile memory module 106 (for example, the memory module is dedicated to storage). In the system area of the system data). In addition, the memory management circuit 302 has a microprocessor unit (not shown), Read only memory (not shown) and random access memory (not shown). In particular, the read-only memory has a drive code, and when the memory controller 104 is enabled, the microprocessor unit executes the drive code segment to store the rewritable non-volatile memory module 106. The control command is loaded into the random access memory of the memory management circuit 302. After that, the microprocessor unit will run these control commands to perform data writing, reading and erasing operations.

In addition, in another exemplary embodiment of the present invention, the control command of the memory management circuit 302 can also be implemented in a hardware format. For example, the memory management circuit 302 includes a microcontroller, a memory management unit, a memory write unit, a memory read unit, a memory erase unit, and a data processing unit. The memory management unit, the memory writing unit, the memory reading unit, the memory erasing unit and the data processing unit are coupled to the microcontroller. The memory management unit is configured to manage the physical block of the rewritable non-volatile memory module 106; the memory write unit is configured to issue a write command to the rewritable non-volatile memory module 106 to The data is written into the rewritable non-volatile memory module 106; the memory reading unit is configured to issue a read command to the rewritable non-volatile memory module 106 to recover from the rewritable non-volatile memory The module 106 reads the data; the memory erasing unit is configured to issue an erase command to the rewritable non-volatile memory module 106 to erase the data from the rewritable non-volatile memory module 106; The data processing unit is configured to process data to be written to the rewritable non-volatile memory module 106 and data read from the rewritable non-volatile memory module 106.

The host interface 304 is coupled to the memory management circuit 302 and used to connect The instructions and data transmitted by the host system 1000 are received and recognized. In the present exemplary embodiment, host interface 304 is compatible with the SD standard. However, it must be understood that the present invention is not limited thereto, and the host interface 304 may be compatible with the PATA standard, the IEEE 1394 standard, the PCI Express standard, the USB standard, the SATA standard, the MS standard, the MMC standard, the CF standard, the IDE standard, or Other suitable data transmission standards.

The memory interface 306 is coupled to the memory management circuit 302 and is used to access the rewritable non-volatile memory module 106. That is, the data to be written to the rewritable non-volatile memory module 106 is converted to a format acceptable to the rewritable non-volatile memory module 106 via the memory interface 306.

The buffer memory 308 is coupled to the memory management circuit 302 and is used to temporarily store data and instructions from the host system 1000 or data from the rewritable non-volatile memory module 106. For example, the buffer memory 302 can be a static random access memory, a dynamic random access memory, or the like.

The power management circuit 310 is coupled to the memory management circuit 302 and is used to control the power of the memory storage device 100.

The error checking and correction circuit 312 is coupled to the memory management circuit 302 and is used to perform an error correction procedure to ensure the correctness of the data. Specifically, when the host interface 304 receives a host write command from the host system 1000, the error check and correction circuit generates a corresponding error check for the write data (also referred to as update data) corresponding to the host write command. And the error checking and correction code (ECC Code), and the memory management circuit 302 writes the update data and the corresponding error correction code to the rewritable non-volatile memory module. 106. Thereafter, when the memory management circuit 302 reads the data from the rewritable non-volatile memory module 106, the error correction code corresponding to the data is simultaneously read, and the error checking and correction circuit 312 is based on the error correction code. Perform an error correction procedure on the data read.

FIG. 5 is a schematic diagram of managing a physical block of a rewritable non-volatile memory module according to an exemplary embodiment.

Referring to FIG. 5, the rewritable non-volatile memory module 106 has physical blocks 410(0)-410(N), and the memory management circuit 302 of the memory controller 104 will block the physical block 410 (0). The ~410 (N) partition is a replacement area 502, a spare area 504, and a data area 506.

The physical block of the replacement area 502 is used for the bad physical block replacement procedure to replace the damaged physical block. Specifically, if the normal physical block remains in the replacement area 502 and the physical block of the data area 506 or the temporary storage area 504 is damaged, the memory management circuit 302 extracts the normal physical block from the replacement area 502. To replace the damaged physical block.

The physical block of the temporary storage area 504 is used to temporarily store the data written by the host system 1000. The detailed writing method will be described later with the illustration. It is worth mentioning that in the present exemplary embodiment, the memory management circuit 302 is a physical block that operates the temporary storage area 504 using a single page mode. Specifically, in the single page mode, only the next page of the physical page will be used to store the data. That is to say, in the single page mode, the memory management circuit 302 only performs data writing, reading, erasing, etc. on the next page physical page.

The physical block (also referred to as a data entity block) of the data area 506 is used to store data written by the host system 1000. Specifically, the memory management circuit 302 converts the logical access address accessed by the host system 1000 into a corresponding logical block and corresponding logical page and maps the logical page of the logical block to the entity of the data area. The physical page of the block. That is, the physical block of the data area 506 is a physical block that is considered to have been used (eg, data that has been written by the host system). For example, the memory management circuit 302 uses a logical block-physical block mapping table to record the mapping relationship between the logical block and the physical block of the data area 506, where the logical block The logical page in the sequence may correspond to the physical page of the mapped physical block. For example, in the present exemplary embodiment, logical blocks 610(0)-610(H) are configured to map physical blocks of data area 506, where the capacity of one logical block is equal to the capacity of one physical block and the data The number of physical blocks of area 506 must be greater than or equal to the number of logical blocks. That is, the number of physical blocks in the data area 506 will correlate the capacity of the memory storage device 100. In the present exemplary embodiment, the number of logical blocks 610(0)-610(H) is equal to the physical block of the data area 506.

In the present exemplary embodiment, memory management circuit 302 is a physical block that operates data area 506 using a multi-page mode. Specifically, in the multi-page mode, the next page physical page, the middle page physical page, and the upper page physical page of each physical page group of the physical block are used to store data. Furthermore, the physical block operating in the multi-page mode has a shorter lifetime than the physical block operating in the single page mode. Specifically, the number of times each physical block can be written or erased is limited, when one When the number of times a physical block is written exceeds a critical value, the physical block is damaged and cannot be written to the data, and the critical value corresponding to the physical block operating in the multi-page mode is lower than the corresponding value. The threshold of the physical block in which the single page mode is operated.

As described above, the physical block of the temporary storage area 504 and the physical block of the data area 506 are operated by using different modes. Therefore, when a physical block is divided into the temporary storage area 504 or the data area 506, This physical block will only be available for a specific partition. That is, the memory management circuit 302 will independently operate the physical block of the data area 506 and the physical block of the temporary storage area 504 without mixing the physical blocks. For example, when a physical block is divided into the temporary storage area 504, the memory management circuit 302 operates the physical block in the temporary storage area 504 in a single page mode until the physical block is damaged; or when an entity After the block is divided into the data area 506, the memory management circuit 302 operates the physical block in the data area 506 in a multi-page mode until the physical block no longer belongs to the data area 506.

FIG. 6 is a schematic diagram of writing data according to an example.

Referring to FIG. 6, it is assumed that the data entity block 410 (T+1) of the data area 506 has stored data of all logical pages belonging to the logical block 610(0) (ie, the logical block 610(0) is currently mapped. Data entity block 410 (T+1)) and the memory storage device 100 receives a write command from the host system 1000 to store the updated data to the 10th to 100th logical pages of the logical block 610(0), The memory management circuit 302 extracts the physical blocks 410(R+1)~410(R+2) from the temporary storage area 504 as the first and second temporary physical blocks of the corresponding logical block 610(0). The update data belonging to logical block 610(0) is written. Specifically, since the physical block of the temporary storage area 504 can only be a single page The pattern is programmed, so the capacity of the 2 temporary physical blocks is required to be able to store data belonging to 91 logical pages (ie, the 10th to 100th logical pages of logical block 610(0)).

Thereafter, the memory management circuit 302 writes the updated data of the 10th to 95th logical pages to be stored to the logical block 610(0) to the next page entity of the first temporary physical block 410 (R+1). Pages (ie, 0, 3, 6, ... 252, 255 physical pages) and write the updated data of the 96th to 100th logical pages to be stored to the logical block 610(0) to the second temporary physical area The next page of the block 410 (R+2) is the physical page (ie, the 0th, 3rd, 6th, 9th, and 12th physical pages).

In the present exemplary embodiment, after the update data to be stored by the host system 1000 is written to the temporary physical block, the memory management circuit 302 transmits a response (Notation) to the host system 1000 to notify the completed instruction. And, then, when the memory storage device 100 is in an idle state for a period of time (for example, 30 seconds has not received any instructions from the host system 1000) or the number of physical blocks available in the temporary storage area 504 is less than a preset threshold value The memory management circuit 302 will move the valid data belonging to the same logical block to the empty data entity block in the data area 506. For example, the preset threshold will be set to 3. However, it must be understood that the present invention is not limited thereto, and the preset threshold may be other suitable values. Here, the operation of moving the valid data belonging to the same logical block to the empty data entity block of the data area 506 is called data merge operation, and the data merge operation includes data sorting operation and data transfer operation.

7 and FIG. 8 are schematic diagrams of a data merge procedure according to an example FIG. 7 is a schematic diagram showing an example of data sorting operation and FIG. 8 is a schematic diagram showing an example of data moving operation. .

Referring to FIG. 7, it is assumed that the data entity block 410 (T+1), the temporary storage physical block 410 (R+1), and 410 (R+2) respectively store part of the logical page of the logical block 610 (0). When the valid data (as shown in FIG. 6) and the memory management circuit 302 selects to perform the data merge operation on the logical block 610(0), first, the memory management circuit 302 extracts the physical block 410 from the temporary storage area 504 ( R+3), 410(R+4) and 410(R+5) are used as the first to third collating physical blocks of the corresponding logical block 610(0) to store all valid entries belonging to the logical block 610(0). data. Specifically, since the physical blocks of the temporary storage area 504 can only be programmed in a single page mode, the capacity of the three finishing physical blocks can store the data of all logical pages of one logical block.

Thereafter, the memory management circuit 302 will validate the valid data of all logical pages belonging to the logical block 610(0) from the data entity block 410 (T+1) and the temporary storage physical block 410 (R+1) and 410 ( R+2) is sequentially copied into the collating physical blocks 410 (R+3), 410 (R+4), and 410 (R+5).

Specifically, first, the memory management circuit 302 sequentially copies the valid data of the 0th to 9th logical pages belonging to the logical block 610(0) from the data entity block 410 (T+1) to the collation. In the 0th, 3rd, 6th, 24th, and 27th physical pages of the physical block 410 (R+3), and the valid data of the 10th to 85th logical pages belonging to the logical block 610(0) are from the temporary storage entity. The block 410 (R+1) is sequentially copied into the 30th, 33...252, 255 physical pages of the collation physical block 410 (R+3). In more detail, due to the 0th to 9th logical pages belonging to the logical block 610(0) The material management circuit 302 will be from the data entity block 410 (T+1) of the original mapping logic block 610(0) to the 0th to 9th blocks belonging to the logic block 610(0). The data of the logical page is moved to the physical block 410 (R+3). In addition, since the updated data of the 10th to 85th logical pages belonging to the logical block 610(0) has been stored in the temporary physical block 410 (R+1), the memory management circuit 302 will temporarily store the data. The data of the 10th to 85th logical pages belonging to the logical block 610(0) is moved to the physical block 410 (R+3) in the physical block 410 (R+1).

Next, the memory management circuit 302 sequentially stores the valid data of the 86th to 100th logical pages belonging to the logical block 610(0) from the temporary physical blocks 410(R+1) and 410(R+2). Copying to the 0th, 3th, 39th, and 42th physical pages of the finishing physical block 410 (R+4), and validating the valid data of the 101st to 171th logical pages belonging to the logical block 610(0) The physical block 410 (T+1) is sequentially copied into the 45th, 48...252, 255 physical pages of the finishing physical block 410 (R+4). In more detail, since the updated data of the 86th to 95th logical pages belonging to the logical block 610(0) has been stored in the temporary physical block 410 (R+1), the memory management circuit 302 will The data of the 86th to 95th logical pages belonging to the logical block 610(0) is moved from the temporary physical block 410 (R+1) to the physical block 410 (R+4). In addition, since the updated data of the 96th to 100th logical pages belonging to the logical block 610(0) has been stored in the temporary physical block 410 (R+2), the memory management circuit 302 will temporarily store the data. The physical block 410 (R+2) moves the data of the 96th to 100th logical pages belonging to the logical block 610(0) to the physical block 410 (R+4). Furthermore, since the data of the 101st to 171th logical pages belonging to the logical block 610(0) are not updated, the memory is The volume management circuit 302 moves the data of the 101st to 171th logical pages belonging to the logical block 610(0) from the data entity block 410 (T+1) of the original mapping logic block 610(0) to the physical area. Block 410 (R+4).

Then, the memory management circuit 302 sequentially copies the valid data of the 172th to 257th logical pages belonging to the logical block 610(0) from the data entity block 410(T+1) to the sorting entity block 410. (R+5) in the 0th, 3...252, 255 physical pages. In more detail, since the data of the 172th to 257th logical pages belonging to the logical block 610(0) are not updated, the memory management circuit 302 will from the data entity area of the original mapping logical block 610(0). In block 410 (T+1), the data of the 172th to 257th logical pages belonging to the logical block 610(0) are moved to the physical block 410 (R+5).

It is worth mentioning that since all the valid data of logical pages of logical block 610(0) have been copied to the finishing physical blocks 410 (R+3), 410 (R+4) and 410 (R+5) Therefore, in an exemplary embodiment of the present invention, the memory management circuit 302 marks the data entity block 410 (T+1) as a physical block storing invalid data or a data entity block 410 (T+1). ) Perform the erase operation. Similarly, in an exemplary embodiment of the present invention, the memory management circuit 302 marks the temporary physical blocks 410 (R+1) and 410 (R+2) as physical blocks for storing invalid data or for temporary storage. The physical blocks 410 (R+1) and 410 (R+2) perform an erase operation, whereby the physical blocks 410 (R+1) and 410 (R+2) can be used when executing the next write command. To write data.

Referring to FIG. 8, after the valid data of all logical pages of the logical block 610(0) is sorted to the physical blocks 410 (R+3), 410 (R+4), and 410 (R+5), the memory is restored. The volume management circuit 302 extracts the data entity block 410 (T+2) from the data area 506. As a new data entity block corresponding to logical block 610(0). Specifically, the memory management circuit 302 selects an empty physical block from the data area 504 or the stored data is a physical block of invalid data. In particular, if the extracted physical block is a physical block storing invalid data, the memory management circuit 302 first performs an erase operation on the physical block. In other words, invalid data on the physical block must be erased first.

Then, the memory management circuit 302 will validate the valid data of all logical pages of the logical block 610(0) from the collating physical blocks 410 (R+3), 410 (R+4), and 410 (R+5). It is sequentially copied into the extracted data entity block 410 (T+2).

Specifically, the memory management circuit 302 sequentially sorts the valid data of the 0th to 85th logical pages belonging to the logical block 610(0) from the next page physical page of the first collating physical block 410 (R+3). Move to the corresponding page of the physical block 410 (T+2) (for example, the 0th to 85th physical pages of the physical block 410 (T+2)). Next, the memory management circuit 302 sequentially shifts the valid data of the 86th to 171th logical pages belonging to the logical block 610(0) from the next page physical page of the second collation physical block 410 (R+4). To the corresponding page of the physical block 410 (T+2) (for example, the 86th to 171th physical pages of the physical block 410 (T+2)). Then, the memory management circuit 302 sequentially shifts the valid data of the 172th to 257th logical pages belonging to the logical block 610(0) from the next page physical page of the third finishing physical block 410 (R+5). To the corresponding page of the physical block 410 (T+2) (for example, the 172th to 257th physical pages of the physical block 410 (T+2)). That is, since the physical block of the data area 506 is operated in a multi-page mode, all the next page physical pages, the middle page physical pages, and the upper page of the physical block 410 (T+2) Body pages are used to store data. In the present exemplary embodiment, for example, in the data transfer operation, the memory management circuit 302 uses a copyback instruction to move valid data from the collating physical block to the data physical block.

Finally, the memory management circuit 302 will remap the logical block 610(0) to the physical block 410 (T+2) in the logical block-physical block mapping table.

In addition, since the valid data of all logical pages of the logical block 610(0) has been copied into the physical block 410 (T+2), in an exemplary embodiment of the present invention, the memory management circuit 302 The finishing physical blocks 410 (R+3), 410 (R+4), and 410 (R+5) are marked as physical blocks storing invalid data or for the finishing physical blocks 410 (R+3), 410 (R +4) Execute the erase operation with 410 (R+5), whereby the physical blocks 410 (R+3), 410 (R+4) and 410 (R+5) can be executed when the next write command is executed. Used to write data.

It should be noted that in the exemplary embodiment, the memory management circuit 302 performs the above-mentioned data sorting operation (as shown in FIG. 7) and the above data moving operation (as shown in FIG. 8) in a partially synchronized manner. That is, if the memory management circuit 302 wants to store the data in a physical entity block (hereinafter referred to as a first data entity block, for example, the data entity block 410 (T+1) shown in FIG. 7) The physical entity block (for example, the temporary physical blocks 410 (R+1) and 410 (R+2) shown in FIG. 7) are merged into an empty data entity block (hereinafter referred to as a second data entity block). For example, when the data entity block 410 (T+2) shown in FIG. 8 is used, the memory management circuit 302 will simultaneously sort the valid data from the physical block during the data transfer operation (see FIG. 8). The collating physical blocks 410 (R+3), 410 (R+4), and 410 (R+5) shown are copied to the second data. Body block.

7 and FIG. 8 , in the data moving operation, the memory management circuit 302 sequentially sequentially copies the valid data of all the logical pages of the logical block 610 ( 0 ) to the finishing physical block 410 (R). +3), 410 (R+4) and 410 (R+5) on the next page of the entity page. In particular, after the valid data of the 0th logical page of logical block 610(0) has been copied to the finishing physical block 410 (R+3), the memory management circuit 302 continues to issue instructions to the subsequent logical page. The valid data is copied to the collation entity block 410 (R+3), and the release instruction copies the valid data of the 0th logical page of the logical block 610(0) from the collation entity block 410 (R+3) to the data. Physical block 410 (T+2). Then, after the valid data of the first, second, and third logical pages of the logical block 610(0) has been copied to the finishing physical block 410 (R+3), the memory management circuit 302 continues to issue instructions to The valid data of the subsequent logical page is copied to the finishing entity block 410 (R+3), and the release instruction sends the valid data of the first and third logical pages of the logical block 610(0) from the finishing entity block 410 (R+ 3) Copy to data entity block 410 (T+2). Then, after the valid data of the 4th, 5th, and 6th logical pages of the logical block 610(0) has been copied to the finishing physical block 410 (R+3), the memory management circuit 302 continues to issue instructions to The valid data of the subsequent logical page is copied to the collation entity block 410 (R+3), and the release instruction sends the valid data of the 2nd, 4th, and 6th logical pages of the logical block 610(0) from the collation entity block 410 ( R+3) is copied to the data entity block 410 (T+2). Then, after the valid data of the 7th, 8th, and 9th logical pages of the logical block 610(0) has been copied to the finishing physical block 410 (R+3), the memory management circuit 302 continues to issue instructions to Copy the valid data of the subsequent logical page to the finishing entity area At block 410 (R+3), the same instruction copies the valid data of the 5th, 7th, and 9th logical pages of the logical block 610(0) from the collation entity block 410 (R+3) to the data entity block 410. (T+2). By analogy, the memory management circuit 302 synchronously moves the sorted valid data from the collating physical block to the data physical block during the collation of the valid data of the subsequent logical page until all valid data is moved to the data entity. Until the block. That is to say, in this example, the memory management circuit 302 will synchronously arrange portions of the data of a predetermined number of logical pages (for example, three logical pages) while continuing to organize the valid data of the subsequent logical pages. The organized data is moved from the finishing physical block to the data physical block.

It is worth mentioning that the above partial synchronization mode is only an example, and the present invention is not limited thereto. For example, in another exemplary embodiment of the present invention, after the valid data of the 0th logical page of the logical block 610(0) has been copied to the finishing physical block 410 (R+3), the memory management circuit 302 The instruction will continue to be issued to copy the valid data of the subsequent logical page to the finishing entity block 410 (R+3), and the issuing instruction will save the valid data of the 0th logical page of the logical block 610(0) from the finishing physical block. 410 (R+3) is copied to the data entity block 410 (T+2). Then, after the valid data of the second logical page of logical block 610(0) has been copied to the finishing physical block 410 (R+3), the memory management circuit 302 continues to issue instructions to the subsequent logical page. The valid data is copied to the collation entity block 410 (R+3), and the release instruction copies the valid data of the second logical page of the logical block 610(0) from the collation entity block 410 (R+3) to the data entity. Block 410 (T+2). Then, after the valid data of the third logical page of the logical block 610(0) has been copied to the finishing physical block 410 (R+3), the memory management power The path 302 will continue to issue instructions to copy the valid data of the subsequent logical page to the finishing entity block 410 (R+3), while the issuing instruction will validate the valid data of the third logical page of the logical block 610(0) from the finishing entity. Block 410 (R+3) is copied to data entity block 410 (T+2). By analogy, the memory management circuit 302 synchronously moves the valid data from the collating physical block to the data physical block during the collation of the valid data of the subsequent logical page until all valid data is moved to the data physical block. That is to say, in this example, the memory management circuit 302 synchronously reorganizes the collated data from the collating physical block after arranging the materials belonging to one logical page and continuing to organize the valid materials belonging to the next logical page. To the data entity block.

Based on the above, since the data sorting operation and the data moving operation are performed in a partially synchronized manner, the time required to perform the data merge operation is greatly shortened.

FIG. 9 is a flowchart of a data merging method according to an exemplary embodiment.

Referring to FIG. 9, in step S901, the memory management circuit 302 selects a logical block (hereinafter referred to as a target logical block) to perform a data merge operation. Specifically, the memory management circuit 202 determines the logical blocks that need to be executed by the data merge program according to the data temporarily stored in the temporary storage area 504, and selects one logical block from the logical blocks to perform the data merge process. That is to say, the valid data of the target logical block has been distributedly stored in the data entity block (hereinafter referred to as the first data entity block) and at least one temporary physical block.

In step S903, the memory management circuit 302 assigns a corresponding target logic. Multiple defragmentation physical blocks of the block. For example, in the present exemplary embodiment, the memory management circuit 302 extracts three physical blocks from the temporary storage area 504 as the first to third finishing physical blocks of the corresponding target logical block.

In step S905, the memory management circuit 302 performs the data sorting operation and the data moving operation in a partially synchronized manner to sort the valid data of the target logical block from the first data entity block and the at least one temporary physical block. To organize the next page entity page of the physical block, and program the valid data of the target logical block from the finishing physical block to the second data entity block.

In step S907, the memory management circuit 302 remaps the target logical block to the second data entity block, thereby completing the data merge operation.

FIG. 10 is a detailed flowchart of step S905 according to an exemplary embodiment.

Referring to FIG. 10, in step S1001, the memory management circuit 302 will program the valid data of the target logical block from the first data entity block and the at least one temporary physical block to the second data entity. The data of the block is copied to the finishing entity block in units of physical pages.

Next, in step S1003, the memory management circuit 302 determines whether the valid data of all the logical pages of the target logical block have been collated to the finishing physical block.

If the valid data of all logical pages of the target logical block has not been completely organized into the finishing physical block, in step S1005, the memory management circuit 302 will program at least part of the data that has been copied into the finishing physical block to Second data The physical block, at the same time, from the first data entity block and the at least one temporary physical block, copying the data of the valid data of the target logical block to the subsequent entity page of the second data entity block In the physical block. And, after step S1005, step S1003 is executed.

If the valid data of all the logical pages of the target logical block have been organized into the finishing physical block, in step S1007, the memory management circuit 302 will program the data copied into the finishing physical block to the second. Data entity block. And after step S1007, the data sorting operation and the data moving operation are completed.

In summary, the data merging method, the memory controller, and the memory storage device of the exemplary embodiment of the present invention organize the data to be merged into a finishing physical block that operates in a single page mode, and then the data is The stylization to the data entity block is arranged in the physical block, so that the reliability of the stored data can be effectively improved. In addition, the data merging method, the memory controller, and the memory storage device of the exemplary embodiment of the present invention perform the operation of sorting the valid data into the defragmenting physical block and copying the valid data to the data entity block in a partially synchronous manner. This effectively reduces the time required to perform data consolidation.

Although the present invention has been disclosed in the above embodiments, it is not intended to limit the invention, and any one of ordinary skill in the art can make some modifications and refinements without departing from the spirit and scope of the invention. The scope of the invention is defined by the scope of the appended claims.

Steps of S901, S903, S905, S907‧‧‧ data merge method

Claims (34)

A data processing method for a flash memory storage device, wherein the flash memory storage device has a flash memory module, the flash memory module having a plurality of physical blocks, each of the physical regions The block has a plurality of physical pages, and the data processing method includes: storing a first data in a single page mode in a plurality of first physical pages among the plurality of physical pages; and the first data in a multi-page mode Moving from the plurality of first entity pages to the plurality of second entity pages among the plurality of entity pages; moving the first data from the plurality of first entity pages to the plurality of second entity pages During the step, preparing a second material; and writing the second data to the at least one third entity page of the plurality of physical pages in the single page mode. The data processing method of claim 1, wherein the step of moving the first data from the plurality of first entity pages to the plurality of second entity pages in the multi-page mode comprises the plurality of pages A copyback instruction of the mode moves the first data from the plurality of first entity pages to the plurality of second entity pages. The data processing method of claim 1, further comprising: receiving the second data from a host system or reading the second data from the flash memory module; and in a buffer memory Temporarily save the second data. The data processing method according to claim 1, wherein the first data is moved from the plurality of first entity pages to the plurality of second entity pages in the multi-page mode, and the first The step of data includes: moving the first data from the plurality of first entity pages to the plurality of second entity pages in an at least partially synchronized manner and temporarily storing the second data in a buffer memory. The data processing method according to claim 1, wherein the first data is moved from the plurality of first entity pages to the plurality of second entity pages in the multi-page mode, and the first The step of the second data includes: before completely moving the first data from the plurality of physical pages to the plurality of second physical pages in the multi-page mode, first writing the at least one third in the single page mode The second data of the entity page is temporarily stored in a buffer memory. The data processing method of claim 1, wherein a threshold of the number of erasures of a physical block in the plurality of physical blocks storing the data in the multi-page mode is lower than the single page mode. A threshold for the number of erasures of a physical block in which the data is stored. The data processing method of claim 1, wherein a physical block of the plurality of physical blocks storing the data in the multi-page mode is larger than an entity storing the data in the single page mode. A capacity of the block. The data processing method of claim 1, wherein the physical entity page, the middle page physical page, and the upper page entity in a physical block operating in the multi-page mode among the plurality of physical blocks Pages will be used to store data and Only the next page of the physical page in a physical block operating in the single page mode is used to store data. A memory storage device includes: a connector for coupling to a host system; a flash memory module having a plurality of physical blocks, wherein each of the physical blocks has a plurality of physical pages; a memory controller, coupled to the connector and the flash memory module, and configured to store a first data in a single page mode in a plurality of first physical pages among the plurality of physical pages, The memory controller is further configured to move the first data from the plurality of first physical pages to a plurality of second physical pages among the plurality of physical pages in a multi-page mode, wherein the memory controller Further, during the moving of the first data from the plurality of first physical pages to the plurality of second physical pages, preparing a second data, wherein the memory controller is further configured to use the second data to The one-page mode is written to at least one third entity page of the plurality of physical pages. The memory storage device of claim 9, wherein in the operation of moving the first data from the plurality of first physical pages to the plurality of second physical pages in the multi-page mode, The memory controller moves the first data from the plurality of first entity pages to the plurality of second entity pages in a copyback instruction of the multi-page mode. The memory storage device according to claim 9 of the patent application, further comprising a buffer memory, wherein the memory controller receives the second data from the host system or reads the second data from the flash memory module, wherein the memory controller is further used in the buffer memory The second data is temporarily stored in the body. The memory storage device of claim 11, wherein the memory controller moves the first data from the plurality of first physical pages to the plurality of second physical pages in an at least partially synchronized manner and The second data is temporarily stored in the buffer memory. The memory storage device of claim 11, wherein the memory controller moves the first data from the plurality of physical pages to the plurality of second physical pages in the multi-page mode. The second data to be written in the at least one third physical page in the single page mode is temporarily stored in the buffer memory. The memory storage device of claim 9, wherein a threshold of the number of erasures of a physical block in the plurality of physical blocks storing the data in the multi-page mode is lower than the single page mode A threshold value for erasing a physical block of data. The memory storage device of claim 9, wherein a physical block of the plurality of physical blocks storing the data in the multi-page mode is larger than the one storing the data in the single page mode. A capacity of a physical block. The memory storage device of claim 9, wherein a lower of a physical block in the plurality of physical blocks operating in the multi-page mode The physical page, the middle page physical page and the upper page physical page are all used to store data, and only the next page physical page is used to store data in a physical block operated in the single page mode. A data processing method for a flash memory storage device, wherein the flash memory storage device has a flash memory module, the flash memory module has a plurality of physical blocks, and the data processing method The method includes: storing a first data in a single page mode in a first physical block of the plurality of physical blocks; and moving the first data from the first physical block to the plurality in a multi-page mode a second physical block among the physical blocks; temporarily storing a second data before the first data is completely moved from the first physical block to the second physical block; and The second data is written into a third physical block of the plurality of physical blocks in the single page mode. The data processing method of claim 17, wherein the step of moving the first material from the first physical block to the second physical block in the multi-page mode comprises the multi-page mode A copyback instruction moves the first data from the first physical block to the second physical block. The data processing method of claim 17, wherein the second data is temporarily stored before the first physical block is completely moved from the first physical block to the second physical block in the multi-page mode. The step of data includes: selecting a plurality of first entity pages in the first physical block; Moving the first data from the plurality of first entity pages in a multi-page mode by a copyback instruction to a plurality of second entity pages of the second physical block; and intending to use the single page mode The second data written to the third physical block is temporarily stored in a buffer memory. The data processing method of claim 17, wherein the second data is temporarily stored before the first physical block is completely moved from the first physical block to the second physical block in the multi-page mode. The step of data includes: moving the first data from the plurality of physical pages of the first physical block to the plurality of physical pages of the second physical block in an at least partially synchronized manner, and temporarily storing the second data to the first physical data Buffer memory. The data processing method of claim 17, further comprising: receiving the second data from a host system or reading the second data from the flash memory module; and in a buffer memory Temporarily save the second data. The data processing method of claim 17, wherein a threshold of the number of erasures of a physical block in the plurality of physical blocks storing the data in the multi-page mode is lower than the single page mode A threshold for the number of erasures of a physical block in which the data is stored. The data processing method of claim 17, wherein a physical block of the plurality of physical blocks storing the data in the multi-page mode is larger than an entity storing the data in the single page mode; A capacity of the block. The data processing method of claim 17, wherein the physical entity page, the middle page physical page, and the upper page entity in a physical block operating in the multi-page mode among the plurality of physical blocks The pages are used to store the data, and only the next page of the physical page in a physical block operating in the single page mode is used to store the data. A memory storage device includes: a connector coupled to a host system; a flash memory module having a plurality of physical blocks; and a memory controller coupled to the connector and The flash memory module is configured to store a first data in a single page mode in a first physical block of the plurality of physical blocks, wherein the memory controller is used to more than The page mode moves the first data from the first physical block to a second physical block of the plurality of physical blocks; wherein the memory controller is further configured to use the first data by the first Before the physical block is completely moved to the second physical block, a second data is temporarily stored, wherein the memory controller is further configured to write the second data into the plurality of physical blocks in the single page mode. A third physical block in the middle. The memory storage device of claim 25, wherein the memory controller is further configured to move the first data from the first physical block by a copyback instruction in the multi-page mode. To the second physical block. The memory storage device of claim 25, further comprising a buffer memory, The memory controller selects a plurality of first entity pages in the first physical block, and moves the first data from the plurality of first entity pages to a copyback instruction in the multi-page mode to a plurality of second physical pages of the second physical block, wherein the memory controller temporarily stores the second data to be written to the third physical block in the single-page mode to the buffer memory. The memory storage device of claim 25, further comprising a buffer memory, wherein the memory controller is configured to at least partially synchronize the first data from the plurality of physical pages of the first physical block Moving to a plurality of physical pages of the second physical block and temporarily storing the second data in the buffer memory. The memory storage device of claim 25, further comprising a buffer memory, wherein the memory controller receives the second data from the host system or reads the same from the flash memory module The second data, wherein the memory controller is further configured to temporarily store the second data in the buffer memory. The memory storage device of claim 25, wherein a threshold value of an erasure number of a physical block in the plurality of physical blocks storing the data in the multi-page mode is lower than the single page mode A threshold value for erasing a physical block of data. The memory storage device of claim 25, wherein the The capacity of a physical block storing data in the multi-page mode among the plurality of physical blocks is larger than a capacity of a physical block storing the data in the single page mode. The memory storage device of claim 25, wherein the physical page of the lower page, the page of the middle page and the previous page of a physical block operating in the multi-page mode among the plurality of physical blocks The physical page is used to store data, and only the next page of the physical page in a physical block operating in the single page mode is used to store data. A data processing method for a flash memory storage device, wherein the flash memory storage device has a flash memory controller and a flash memory module, the flash memory controller has a buffer memory And the flash memory module has a plurality of physical blocks, and the data processing method includes: storing a first data in a single page mode in a first physical block of the plurality of physical blocks; Transmitting the first data from the first physical block to the second physical block by a copyback instruction in a multi-page mode, wherein the second physical area operating in the multi-page mode The lower page, the middle page and the upper page of the block are used to store data; in the process of moving the first data from the first physical block to the second physical block, Writing a second data of the single-page mode to a third physical block to the buffer memory of the memory controller; and writing the second data from the buffer in the single page mode Entering a third of the plurality of physical blocks Physical block. The data processing method of claim 33, wherein the storage capacity of the physical block in the plurality of physical blocks storing the data in the multi-page mode is a physical area in which the data is stored in the single page mode. Three times the storage capacity of the block.