WO2022142544A1

WO2022142544A1 - Method for preventing data loss from flash memory, solid state drive controller, solid state drive

Info

Publication number: WO2022142544A1
Application number: PCT/CN2021/121393
Authority: WO
Inventors: 张吉兴; 印中举; 于洲
Original assignee: 深圳大普微电子科技有限公司
Priority date: 2020-12-29
Filing date: 2021-09-28
Publication date: 2022-07-07
Also published as: CN112632643A

Abstract

A method for preventing data loss from a flash memory, a solid state drive controller, and a solid state drive. The method comprises: when it is detected that a power-down behavior occurs in a solid state drive, and the page written when the power-down behavior occurs is an upper page/extra page, obtaining the number of a target segment where a lower page corresponding to the upper page/extra page is located; obtaining backup data of the target segment and first lower page data according to the number; and restoring second lower page data according to the backup data and the first lower page data. The present invention can solve the problem of sharing a physical page, improve the stability of flash memory data, and do not require additional costs.

Description

Method for preventing data loss from flash memory, solid state drive controller, solid state drive

CROSS-REFERENCE TO RELATED APPLICATIONS

This application claims the priority of the Chinese patent application filed on December 29, 2020 with the application number 202011595224.3 and the invention titled "Method for Preventing Data Loss in Flash Memory, Solid State Disk Controller, Solid State Disk", the entire contents of which are Incorporated herein by reference.

technical field

The present application relates to a solid-state hard disk control technology, and in particular, to a method for preventing data loss in flash memory, a solid-state hard disk controller, and a solid-state hard disk.

Background technique

Solid State Drives (SSD) are hard drives made of solid-state electronic memory chip arrays, which have been widely used in various occasions. Due to their excellent indicators in performance, power consumption, and environmental adaptability, they are gradually becoming Replace traditional hard drives. At present, whether it is the SSD consumer market or the enterprise market, TLC flash memory has become the current mainstream storage medium.

A cell (cell) of TLC flash memory can represent 3 bits, 128K cells constitute a word line, the first bit of all cells in the word line constitutes a slow page (Lower Page), and the second and third bits constitute a slow page respectively. Fast page (Upper Page) and super page (Extra Page). We call 3 physical pages in a word line shared physical pages. TLC flash memory physical word lines are written in two ways, one pass and two pass. The so-called two-pass writing method means that the first step master writes the slow page into the flash word line, the second step master writes the fast page and the super page together into the flash word line, and then all cells in the entire word line are written. will remain in a stable state. However, if an abnormal power failure occurs during the above-mentioned second step of writing, not only the data currently written in the word line will be lost, but also the data previously successfully written in the slow page will be lost. For the host, the data loss of the slow page is unacceptable, because the SSD master will inform the host that the data of the slow page has been successfully written to the flash memory before the second step above.

In order to solve the above problems, the related art configures a backup power supply for the main power supply, and when the main power supply fails abnormally, the backup power supply intervenes to work. However, the backup power capacity of the capacitor will weaken with the use of time. In the later life of the SSD, the actual backup power capacity of the capacitor cannot meet the requirements of the initial design. At this time, there is a high probability that the second step will still occur in writing data. In case of abnormal power failure. To this end, the related art also chooses to use a capacitor with super power backup capability, but the capacitor is also relatively expensive. Therefore, it is of great significance to successfully solve the problem of sharing physical pages caused by abnormal power failure of the SSD without additional cost.

SUMMARY OF THE INVENTION

The technical problem mainly solved by the embodiments of the present application is how to successfully solve the problem of sharing physical pages caused by abnormal power failure of the SSD without additional cost.

In order to solve the above-mentioned technical problems, a technical solution adopted by the embodiments of the present application is to provide a method for preventing data loss of flash memory, including:

When it is detected that the solid-state drive has a power-down behavior, and the page written when the power-down behavior occurs is a fast page superpage, obtain the number of the target segment where the slow page corresponding to the fast page superpage is located;

Acquire the backup data and the first slow page data of the target segment according to the number;

Calculate and restore the second slow page data according to the backup data and the first slow page data.

Optionally, the obtaining the serial number of the target segment where the slow page corresponding to the fast page super page is located includes:

Obtain the physical page number of the slow page corresponding to the fast page superpage, and a fixed step size, where the fixed step size is used to represent the serial number interval length between the slow page and the fast page superpage of a shared physical page;

The number of the target segment where the slow page corresponding to the fast page superpage is located is calculated according to the physical page number and the fixed step size.

Optionally, the acquiring the physical page number of the slow page corresponding to the fast page superpage, and the fixed step size specifically include:

Obtain the physical page number in the TLC physical block of the slow page corresponding to the fast page superpage, and the fixed step size corresponding to the TLC physical block.

Optionally, obtaining the backup data and the first slow page data of the target segment according to the serial number includes:

Find the number of the target segment from the SLC backup physical block, and read the backup data corresponding to the number of the target segment; wherein, the backup data is based on the first slow The page data and the second slow page data are calculated and stored in the SLC backup physical block;

The unmissed slow page data in the physical page corresponding to the number of the target segment is read from the TLC physical block, and the unmissed slow page data constitutes the first slow page data.

Optionally, the backup data is XOR data, and the XOR data is obtained by performing an XOR operation on the non-lost slow page data and the second slow page data.

Optionally, the method further includes: backing up the backup data in a physical block;

Wherein, backing up the backup data in the physical block includes:

Erase the SLC backup physical block;

Divide the TLC physical block into the target segment according to the characteristics of the flash memory;

Write physical pages into the divided target segments in sequence;

When the physical page of one target segment is written, read all the slow page data in the one target segment, and calculate all the slow page data to obtain the corresponding data of all the slow pages in the one target segment the calculation result;

The calculation result is saved in the physical page of the SLC backup physical block, and the page number of the physical page of the SLC backup physical block is the number of the target segment corresponding to the calculation result.

Optionally, before starting to write data to the current target segment, the step of backing up the backup data in the physical block is performed on the previous target segment of the current target segment.

Optionally, the calculating and restoring the second slow page data according to the backup data and the first slow page data includes:

Performing an XOR operation on the backup data and the first slow page data, and the result of the XOR operation is the second slow page data.

In order to solve the above technical problem, another technical solution adopted by the embodiments of the present application is to provide a solid-state hard disk controller, including: at least one processor; and a memory, where a program is stored in the memory; When the at least one processor is executed, the program causes the solid-state disk controller to execute the method described above.

In order to solve the above technical problem, another technical solution adopted by the embodiments of the present application is to provide a solid-state hard disk, including: the above-mentioned solid-state hard disk controller, and at least one flash memory medium communicatively connected to the solid-state hard disk controller .

In order to solve the above technical problem, another technical solution adopted by the embodiments of the present application is to provide a computer program product, wherein the computer program product includes a computer program stored on a solid-state hard disk, and the computer program includes program instructions. When the program instructions are executed by the solid-state hard disk controller, the solid-state hard disk controller is made to execute the above-mentioned method.

Different from the situation in the related art, the embodiment of the present application obtains the corresponding superpage of the fast page by detecting that the solid-state drive has a power-down behavior, and the page written when the power-down behavior occurs is a fast page superpage. The number of the target segment where the slow page is located; obtain the backup data and the first slow page data of the target segment according to the number; calculate and restore the second slow page data according to the backup data and the first slow page data. By dividing the physical block into target segments, the second slow page data for recovering lost data can be calculated according to the backup data of the target segment and the first slow page data. A method for preventing data loss in flash memory, a solid state disk controller, and a solid state disk can solve the problem of shared physical pages, improve the stability of flash memory data, and do not require additional costs.

Description of drawings

One or more embodiments are exemplified by the accompanying drawings, which are not intended to limit the embodiments, and elements with the same reference numerals in the drawings represent similar elements, unless otherwise specified. It is stated that the figures in the accompanying drawings do not constitute a scale limitation.

FIG. 1 is a flowchart of a method for preventing data loss of flash memory provided by an embodiment of the present application;

2 is a schematic diagram of the composition of a plane (Plane) in the TLC flash memory provided by an embodiment of the present application;

3 is a schematic diagram of a writing process flow of a physical page in the target segment provided by an embodiment of the present application;

4 is a schematic diagram of a writing process flow of data in a backup physical block provided by an embodiment of the present application;

5 is a schematic diagram of backing up and restoring the backup data of the target segment for TLC flash memory provided by an embodiment of the present application;

FIG. 6 is a schematic diagram of a hardware structure of a solid-state hard disk controller 20 according to an embodiment of the present application.

Detailed ways

In order to make the purposes, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments It is a part of the embodiments of the present application, but not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

In addition, the technical features involved in the various embodiments of the present application described below can be combined with each other as long as there is no conflict with each other.

The method for preventing flash data loss, the solid-state disk controller, and the solid-state disk provided by the embodiments of the present application divide a physical page into multiple target segments according to the characteristics of the shared physical page problem, and a target segment includes slow pages, fast pages and Superpage, the fast page and superpage corresponding to the slow page in each target segment are in the next target segment. In view of the fact that the power failure occurs when writing the fast page and the super page, the data of the slow page corresponding to the fast page and the super page will be lost, and the slow page is located in the previous target segment. The SSD will notify the host that the slow page data is successfully written to the flash memory. Therefore, if the data of the current slow page is lost, it is unacceptable for the host. Based on this, the embodiment of the present application backs up the backup data by acquiring the backup data corresponding to all the slow page data in the target segment, and when the data of a slow page in the target segment is lost, according to the The data of the lost slow page is obtained by calculating the backup data and the data of other slow pages that are not lost in the target segment, so as to restore the data of the slow page. Thus, the stability of the data of the flash memory is guaranteed, and other costs do not need to be increased.

Specifically, please refer to FIG. 1 , which is a flowchart of a method for preventing data loss in a flash memory provided by an embodiment of the present application. The method includes:

S101. When it is detected that the solid-state hard disk has a power-down behavior, and the page written when the power-down behavior occurs is a fast page superpage, obtain the number of the target segment where the slow page corresponding to the fast page superpage is located;

Wherein, the obtaining the number of the target segment where the slow page corresponding to the fast page super page is located includes:

Wherein, the acquiring the physical page number of the slow page corresponding to the fast page super page, and the fixed step size specifically include:

S102. Acquire the backup data and the first slow page data of the target segment according to the serial number;

The acquiring the backup data and the first slow page data of the target segment according to the number includes: searching for the number of the target segment from the SLC backup physical block, and reading the backup data corresponding to the number of the target segment ; wherein, the backup data is calculated according to the first slow page data and the second slow page data when the solid state drive is not powered off, and is stored in the SLC backup physical block; The unmissed slow page data in the physical page corresponding to the number of the target segment is read from the TLC physical block, and the unmissed slow page data constitutes the first slow page data.

The backup data is XOR data, and the XOR data is obtained by performing an XOR operation on the non-lost slow page data and the second slow page data.

S103. Calculate and restore the second slow page data according to the backup data and the first slow page data.

The first slow page data is data that is not lost in a target segment, and the second slow page data is data that is lost in the target segment. The backup data is calculated based on the first slow page data and the second slow page data. When the second slow page data needs to be restored, the first slow page data and the backup data can be obtained. The data calculation restores the second slow page data, for example, performing an exclusive OR operation on the first slow page data and the backup data. In addition, it should be noted that the calculation of the backup data is performed before the data of the second slow page is not lost.

The above-mentioned method for preventing flash data loss will be specifically described below based on TLC flash memory. The process mainly includes dividing target segments, calculating backup data, backing up backup data, and restoring lost data.

The embodiment of the present application mainly solves the problem of sharing physical pages. The so-called shared physical page problem is that a physical cell (cell) is shared by two pages (pages), and two bits in one cell are mapped to two In different logical pages, when a write operation is performed on one of the bits (whether it is MSB or LSB), if the power is suddenly turned off, the data in this unit is bound to be lost, that is, the data in the two pages will be lost. Data will be wrong.

It can be known that a cell of the current mainstream storage medium TLC flash memory can represent 3 bits. The first bit of all cells in the word line constitutes a slow page, and the second and third bits constitute a fast page and a super page, respectively. Three physical pages in a word line are shared physical pages. Among them, slow pages, fast pages and super pages are associated with each other.

In this embodiment, when the solid state drive is powered down, if the object currently being written is the fast page superpage, it is determined that there is a risk of data loss in the slow page associated with the fast page superpage. Therefore, , the method for preventing data loss of flash memory in this embodiment is mainly aimed at preventing data loss of the slow page.

Please refer to FIG. 2. FIG. 2 is a schematic diagram of the structure of a plane (Plane) in the TLC flash memory, and the Plane includes a page of registers and several physical pages. When writing data, the SSD master first transfers 16KB of data to the page register, and then the flash memory writes the data in the page register to the physical page of the flash memory. Figure 2 shows the positional relationship of all shared physical pages in a physical block in Plane. The serial numbers of the physical pages in Figure 2 (such as 36, 37, 38, 39, etc.) are the writing order, and each dotted line links a The group shares physical pages. For example, in Figure 2, physical page 62 (slow page), physical page 132 (fast page), and physical page 133 (super page) together constitute the shared physical page, that is, these three physical pages are physically for the same word line. As can be seen from FIG. 2 , there is a fixed step size between the serial numbers of the slow page of the physical page and the superpage of the fast page, for example, the fixed step size is 70 in FIG. 2 . The fixed step size is determined by the 3D stacking characteristics of the TLC flash memory.

Based on the characteristics of the above-mentioned TLC flash memory, in this embodiment, all physical pages in one plane are divided into multiple target segments. Wherein, the method for dividing the target segment includes: obtaining the serial number of the physical page and the fixed step size; dividing all the physical pages into multiple targets equally according to the serial number of the physical page and the fixed step size Each target segment contains the same number of physical pages, and includes slow pages, fast pages and super pages, and the fast pages and super pages corresponding to any slow page in each target segment are under the in a target segment. For example, in FIG. 2, the fast page and super page corresponding to physical page 65 (slow page) are in the target segment x+1, namely physical page 135 (fast page) and physical page 136 (super page). 2, the physical page corresponding to target segment x includes physical page 60 to physical page 131; the physical page corresponding to target segment x+1 includes physical page 132 to physical page 203;

Please refer to FIG. 3 . FIG. 3 is a schematic diagram of a writing process of a physical page in the target segment. As shown in Figure 3, first the master writes the next physical page k; judges whether the physical page k is a slow page, if it is not a slow page, then continues to write the next physical page, if it is a slow page, judges the Whether physical page k is the last slow page; if the physical page k is the last slow page, calculate all the slow page data in the target segment to obtain backup data, write the backup data into the backup physical block, and then Then write to the next physical page to back up the data corresponding to the next target segment; wherein, if the physical page k is not the last slow page, the master continues to write the next physical page.

It can be known that the writing sequence of TLC physical pages in the plane Plane is carried out in the manner of slow page, fast page superpage, slow page, fast page superpage... . If an abnormal power failure occurs and the slow page is being written, the data of the slow page will be lost, and the SSD will not send a notification to the host that the data writing is complete, that is, the data lost in the slow page will be lost. It is allowed, and the fast page and super page associated with the slow page are written later, which has not had time to write data, so it will not affect the fast page and super page. If an abnormal power failure occurs and the fast page superpage is being written, the data of the fast page superpage will be lost, and the data of the slow page associated with the fast page superpage will be lost. The write operation to the associated slow page has been completed before the write operation, and the solid-state drive has reported to the host that the part of the data has been written, but in fact the current data of the slow page has been lost.

Based on the above characteristics, when a fast page superpage is written and an abnormal power failure occurs, the data of the associated slow page will be lost, and at most only one slow page will be lost, that is, the fast page will be lost. The data of the slow page associated with the page superpage is lost, and the data of other slow pages in the target segment is intact. Therefore, when all the slow page data of a target segment is written, calculate all the slow page data in the target segment to obtain backup data, and back up the backup data; or, start writing to the current target segment Before entering data, the slow page data in the previous target segment of the current target segment is calculated to obtain backup data, and the backup data is backed up.

Wherein, the backup data is used to convert the backup data and the corresponding The unmissed slow page data in the target segment is calculated to obtain the lost slow page data.

For example, suppose the data of a slow page consists of 4 (actually hundreds of thousands) bits, and there are 4 slow pages in a target segment, i.e.,

slow page a = [0,1,1,1];

slowpage b = [1,0,0,1];

slowpage c=[0,0,0,1];

slowpage d=[1,0,0,0];

XOR the slow page a, slow page b, slow page c and slow page d to obtain backup data, that is:

After the SSD is powered on, it is found that the data of the slow page d is lost, then the master reads out the data slow page a, slow page b, slow page c, XOR, and performs XOR calculation on them. The XOR result is equal to the slow page d. data, i.e. slow pages

It should be noted that the above is only an example of how to recover lost data, in which the way of XOR calculation is used to restore lost data, in addition to the way of XOR calculation, other ways can also be used, not limited to XOR .

The following describes how to back up the backup data. A backup of backed up data is performed before the data is restored.

Please refer to FIG. 4 . FIG. 4 is a schematic diagram of a writing process of data in a backup physical block. As shown in Figure 4, first erase the SLC backup physical block; divide the TLC physical block into a target segment according to the characteristics of the flash memory; write the physical pages into the target segment in sequence; judge whether the data in the target segment has been written, If it is not finished, continue to write physical pages into the target segment in sequence. If it has been written, read out all the slow page data in the target segment and perform calculation (for example, XOR calculation), The calculation result is obtained, and the calculation result is saved in the SLC backup physical block; then, the above-mentioned writing process is continued to be performed on the next SLC backup physical block.

Please refer to FIG. 5 , which is a schematic diagram of backing up and restoring the backup data of the target segment with respect to the TLC flash memory. Among them, the master configures a physical block in SLC mode for each plane of the flash memory as a backup module. The physical block in SLC mode has the characteristics of high performance, long life, and low capacity, and the writing method of the physical block in SLC mode is one pass, that is, there is naturally no problem of sharing physical pages. Therefore, this embodiment adopts the SLC backup physical block as the backup block.

The calculation result of all slow page data in a target segment is stored in the physical page of the SLC backup physical block, and the page number of the physical page of the SLC backup physical block is the page number of the target segment corresponding to the calculation result. Numbering.

When data needs to be restored, the backup data is obtained from the SLC backup physical block according to the number of the target segment where the slow page where the data is lost is located, and then combined with other slow pages without data loss in the target segment where the slow page is located. The data is calculated together to obtain the lost data of the slow page.

The above is to back up the calculation result of all slow page data in a target segment, instead of backing up all slow page data in a target segment. Therefore, the problem of shared physical pages is solved and the efficiency of data backup is also improved.

For example, there are 72/3=24 slow pages in each target segment, so the time spent backing up all slow page data in a target segment = time spent reading 24 SLC physical pages + time spent writing 1 SLC physical page = 24 *30us+300us=1020us, so we only need 1020us*segment number=1020us*(720/72)=10200us time to back up the slow page data in the entire TLC physical block.

And if it is to back up all the slow page data in a target segment, that is, after each slow page data is written, the master immediately sends a Copy Back command (the function of the flash memory chip can copy a certain physical device in the same Plane. Copy the page data to another physical page) Write the data in the slow page to the SLC physical page of the backup physical block. But doing this is equivalent to increasing the writing time of all slow pages by T_copyback (Copy Back time = time spent reading SLC physical page + time spent writing SLC physical page ~ = 30us+300us=330us) time, assuming a physical There are 720 physical pages in the block (that is, there are 240 SLC physical pages), then the total time spent in backing up all slow page data in a TLC physical block=330*240=79200us.

In comparison, the data backup method of the present application significantly improves the data backup efficiency, reduces the time consumption, and can solve the problem of sharing physical pages with a minimal performance cost.

In this embodiment, the TLC physical page is divided into several target segments, and the calculation results of all the slow page data in each target segment are backed up to the SLC backup physical block respectively. When a certain fast page superpage in the TLC physical page is When an abnormal power failure occurs during the write operation, obtain the number of the last target segment of the target segment where the fast page superpage is located, and obtain the backup from the SLC backup physical block according to the number of the last target segment. data, and calculate the backup data with the slow page data without the lost data, so as to obtain the lost slow page data. The method provided by this embodiment can solve the problem of sharing physical pages, improve the stability of flash memory data, and does not require additional cost.

It should be noted that the above is for TLC-type physical blocks, and for other types of physical blocks, such as MLC-type physical blocks, the problem of sharing physical pages can also be solved based on the same application concept above. MLC-type physical blocks There is also the problem of shared physical pages, and the backed up data can also be stored in SLC-type physical blocks.

Please refer to FIG. 6. FIG. 6 is a schematic diagram of a hardware structure of a solid-state disk controller 20 provided by an embodiment of the present application. The solid-state disk controller 20 can be used to execute the above-mentioned method for preventing data loss in a flash memory. The processor 20 includes: one or more processors 21 and a memory 22, and one processor 21 is taken as an example in FIG. 6 .

The processor 21 and the memory 22 may be connected by a bus or in other ways, and the connection by a bus is taken as an example in FIG. 6 .

As a non-volatile computer-readable storage medium, the memory 22 can be used to store non-volatile software programs, non-volatile computer-executable programs and modules, such as corresponding to the method for preventing flash data loss in the embodiments of the present application. program instructions/modules. The processor 21 executes various functional applications and data processing of the SSD controller by running the non-volatile software programs, instructions and modules stored in the memory 22, that is, implementing the method for preventing flash data loss in the above method embodiments.

The memory 22 may include a stored program area and a stored data area, wherein the stored program area may store an operating system, an application program required for at least one function. Additionally, memory 22 may include high-speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid state storage device. In some embodiments, memory 22 may optionally include memory located remotely from processor 21 .

The one or more modules are stored in the memory 22, and when executed by the one or more processors 21, execute the method for preventing flash data loss in any of the above method embodiments, for example, execute the above-described method. The method steps in FIG. 1 , FIG. 2 , and FIG. 3 .

The above product can execute the method provided by the embodiments of the present application, and has functional modules corresponding to the execution method. For technical details not described in detail in this embodiment, reference may be made to the methods provided in the embodiments of this application.

In the solid-state disk controller provided in this embodiment, when it is detected that the solid-state disk has a power-down behavior, and the page written when the power-down behavior occurs is a fast page superpage, it obtains the corresponding superpage of the fast page. The number of the target segment where the slow page is located; the backup data and the first slow page data of the target segment are obtained according to the number; the second slow page data is calculated according to the backup data and the first slow page data, and the first slow page data is calculated. The second slow page data is used to restore the lost data of the slow page. By dividing the physical block into target segments, the second slow page data for recovering lost data can be calculated according to the backup data of the target segment and the first slow page data. The SSD controller can solve the problem of shared physical pages and improve the stability of flash data without additional cost.

Embodiments of the present application further provide a solid-state hard disk, where the solid-state hard disk includes the solid-state hard disk controller 20 described in the foregoing embodiments, and at least one flash memory medium. The solid state disk controller 20 is communicatively connected to the at least one flash memory medium.

The solid-state disk provided by this embodiment acquires the slow page corresponding to the fast-page superpage by detecting that the solid-state disk has a power-down behavior, and the page written when the power-down behavior occurs is a fast page superpage The number of the target segment; obtain the backup data and the first slow page data of the target segment according to the number; calculate the second slow page data according to the backup data and the first slow page data, the second slow page The page data is used to recover the lost data of the slow page. By dividing the physical block into target segments, the second slow page data for recovering lost data can be calculated according to the backup data of the target segment and the first slow page data. Solid-state drives can solve the problem of shared physical pages, improve the stability of flash data, and do not require additional costs.

An embodiment of the present application provides a computer program product, where the computer program product includes a computer program stored on a solid-state disk, and the computer program includes program instructions, when the program instructions are executed by the solid-state disk controller, The SSD controller is enabled to execute the method for preventing flash data loss in any of the above method embodiments, for example, execute the above-described method steps S101 to S103 in FIG. 1 , and execute the processes in FIGS. 3 and 4 .

The device embodiments described above are only illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in One place, or it can be distributed over multiple network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution in this embodiment.

From the description of the above embodiments, those of ordinary skill in the art can clearly understand that each embodiment can be implemented by means of software plus a general hardware platform, and certainly can also be implemented by hardware. Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium, and the program is During execution, it may include the processes of the embodiments of the above-mentioned methods. Wherein, the storage medium may be a magnetic disk, an optical disk, a read-only memory (Read-Only Memory, ROM) or a random access memory (Random Access Memory, RAM) or the like.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, but not to limit them; under the thinking of the present application, the technical features in the above embodiments or different embodiments can also be combined, The steps may be carried out in any order, and there are many other variations of the different aspects of the present application as described above, which are not provided in detail for the sake of brevity; although the present application has been The skilled person should understand that it is still possible to modify the technical solutions recorded in the foregoing embodiments, or to perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the implementation of the application. scope of technical solutions.

Claims

A method for preventing data loss in flash memory, comprising:

When it is detected that the solid-state drive has a power-down behavior, and the page written when the power-down behavior occurs is a fast page superpage, obtain the number of the target segment where the slow page corresponding to the fast page superpage is located;

Acquire the backup data and the first slow page data of the target segment according to the number;

Calculate and restore the second slow page data according to the backup data and the first slow page data.
The method according to claim 1, wherein the acquiring the number of the target segment where the slow page corresponding to the fast page super page is located comprises:

Obtain the physical page number of the slow page corresponding to the fast page superpage, and a fixed step size, where the fixed step size is used to represent the serial number interval length between the slow page and the fast page superpage of a shared physical page;

The number of the target segment where the slow page corresponding to the fast page superpage is located is calculated according to the physical page number and the fixed step size.
The method according to claim 2, wherein the acquiring the physical page number of the slow page corresponding to the fast page superpage, and the fixed step size specifically comprises:

Obtain the physical page number in the TLC physical block of the slow page corresponding to the fast page superpage, and the fixed step size corresponding to the TLC physical block.
The method according to claim 3, wherein the acquiring the backup data and the first slow page data of the target segment according to the serial number comprises:

Find the number of the target segment from the SLC backup physical block, and read the backup data corresponding to the number of the target segment; wherein, the backup data is based on the first slow The page data and the second slow page data are calculated and stored in the SLC backup physical block;

The unmissed slow page data in the physical page corresponding to the number of the target segment is read from the TLC physical block, and the unmissed slow page data constitutes the first slow page data.
The method according to claim 4, wherein the backup data is XOR data, and the XOR data is obtained by performing an XOR operation on the non-lost slow page data and the second slow page data owned.
The method according to any one of claims 1 to 5, wherein the method further comprises: backing up the backup data in a physical block;

The backing up the backup data in the physical block includes:

Erase the SLC backup physical block;

Divide the TLC physical block into the target segment according to the characteristics of the flash memory;

Write physical pages into the divided target segments in sequence;

When the physical page of one target segment is written, read all the slow page data in the one target segment, and calculate all the slow page data to obtain the corresponding data of all the slow pages in the one target segment the calculation result;

The calculation result is saved in the physical page of the SLC backup physical block, and the page number of the physical page of the SLC backup physical block is the number of the target segment corresponding to the calculation result.
The method of claim 6, wherein:

Before starting to write data to the current target segment, the step of backing up the backup data in the physical block is performed on the previous target segment of the current target segment.
The method according to claim 5, wherein the calculating and restoring the second slow page data according to the backup data and the first slow page data comprises:

Performing an XOR operation on the backup data and the first slow page data, and the result of the XOR operation is the second slow page data.
A solid-state hard disk controller, characterized in that it includes:

at least one processor;

and a memory in which a program is stored;

When the program is executed by the at least one processor, the program causes the solid-state disk controller to perform the method of any one of claims 1 to 8.
A solid-state hard disk, comprising: the solid-state hard disk controller of claim 9, and at least one flash memory medium communicatively connected to the solid-state hard disk controller.
A computer program product, characterized in that the computer program product includes a computer program stored on a solid-state hard disk, the computer program includes program instructions, and when the program instructions are executed by a solid-state hard disk controller, make the solid-state hard disk The hard disk controller executes the method of any one of claims 1-8.