TWI798536B - Method and computer program product for analyzing failures on flash data - Google Patents

Abstract

The disclosure describes a method for analyzing failures on flash data, which is performed by a processing unit of an analysis host when loading and executing software program code. The processing unit is coupled to a peripheral storage interface to connect to a malfunctioned storage device. The method includes obtaining data that can be used in failure analysis, and backing-up the obtained data in a local storage device, rather than backing-up the whole data stored in the storage device, to facilitate preliminary analysis.

Description

Flash memory data fault analysis method and computer program product

The invention relates to a storage device, in particular to a method for analyzing faults of flash memory data and a computer program product.

As the demand for storage space and circuit miniaturization of memory devices (such as DRAM, flash memory, etc.) increases, memory manufacturers seek to increase chip performance without sacrificing precious motherboard space. Instead of using traditional single-die or multi-die packages, chip designers are using multi-die stacking solutions. Multi-die (Dies, such as multiple memory chips) stacking not only reduces the overall area of the package, but also consumes less energy when driving signals and obtains faster transmission speeds, thereby effectively improving the performance of electrical signals. The wire bonding technology of stacking die is used to connect the stacked memory chip and the printed circuit board to each other, so that the controller is coupled or connected to the memory chip through the substrate. Afterwards, the stacked memory chip, controller and substrate are packaged again to form a single integrated circuit (Integrated Circuits, IC). However, when a major error occurs in the data stored in the memory chip and failure analysis is required, it may be necessary to destroy the packaged chip. Therefore, the present invention proposes a flash memory data failure analysis method and a computer program product, which are used for efficiently backing up data before destructive analysis, which is beneficial to future failure analysis.

In view of this, how to alleviate or eliminate the deficiencies in the above-mentioned related fields is a problem to be solved.

This specification relates to a method for fault analysis of flash memory data, by analyzing the processing unit of the host Elements are implemented when the software's code is loaded and executed. The processing unit is coupled to the peripheral storage interface, and is used for connecting the storage device on which a failure occurs. The method includes: obtaining the data used for failure analysis in the storage device and backing up the data in the local storage device, but not completely backing up all the data in the storage device, so as to assist in preliminary analysis.

The present specification also relates to a computer program product used for fault analysis of flash memory data, including program codes loaded and executed by the processing unit of the analysis host. The processing unit is coupled to the peripheral storage interface, and is used for connecting the storage device on which a failure occurs. Its program code includes: obtaining the data used for fault analysis in the storage device and backing it up in the local storage device, but not fully backing up all the data in the above storage device, thereby assisting in preliminary analysis.

One of the advantages of the above-mentioned embodiment is that it avoids spending a lot of time to completely back up all data through the technical means of acquisition and backup as described above.

Another advantage of the above-mentioned embodiment is that through the above-mentioned acquisition and backup technical means, the user's private or confidential information in the storage module is prevented from being stolen by illegal personnel from the local storage device used for backup.

Other advantages of the present invention will be explained in more detail with the following description and drawings.

10: Fault analysis system

100: storage device

110: storage IC

120: Controller

130: storage module

140: Substrate

150: printed circuit board

160: connector

200: computer

210: Host

211: processing unit

212: Memory

213: display interface

215: input interface

217:Peripheral storage interface

219:Local storage device

230: Display

250: input device

310: processing unit

320: host interface

340: Flash memory interface

360: RAM

S410~S460: method steps

S510~S580: method steps

S610~S680: method steps

FIG. 1 is a schematic diagram of a fault analysis system according to an embodiment of the present invention.

FIG. 2 is a block diagram of an analysis host according to an embodiment of the present invention.

FIG. 3 is a block diagram of a storage device according to an embodiment of the invention.

FIG. 4 is a flowchart of a fault analysis method according to an embodiment of the present invention.

FIG. 5 is a flowchart of a data backup method according to an embodiment of the present invention.

FIG. 6 is a flowchart of a fault analysis method according to an embodiment of the present invention.

The following description is a preferred implementation mode of the invention, and its purpose is to describe the basic spirit of the invention, but not to limit the invention. For the actual content of the invention, reference must be made to the scope of the claims that follow.

It must be understood that words such as "comprises" and "including" used in this specification are used to express It shows that there are specific technical characteristics, values, method steps, operation processes, components and/or components, but it does not exclude the addition of more technical characteristics, values, method steps, operation processes, components, components, or any of the above combination.

Words such as "first", "second", and "third" used in the claims are used to modify the elements in the claims, and are not used to indicate that there is a priority order, a pre-relationship, or an element An element preceding another element, or a chronological order in performing method steps, is only used to distinguish elements with the same name.

Referring to FIG. 1 , the failure analysis system 10 may at least include a storage device 100 (object to be analyzed) and a computer 200 . The storage device 100 can be a memory card (Memory Card), a portable disk (Portable Disk), a solid state disk (Solid State Disk, SSD), etc., and at least includes a storage integrated circuit (Integrated Circuits, IC) 110, a printed circuit board (Printed Circuit Board (PCB) 150 and connector 160. The storage device 100 can be configured in various electronic products, such as personal computers, notebook computers (Laptop PC), tablet computers, mobile phones, digital cameras, digital video cameras and other electronic products. The computer 200 includes at least a host 210 , a display 230 and an input device 250 . The display 230 may be a thin film transistor liquid crystal panel (Thin Film Transistor-Liquid Crystal Display, TFT-LCD panel), an organic light-emitting diode panel (Organic Light-Emitting Diode, OLED panel), etc., for displaying the stored content in the representative storage device 100 The text, numbers, symbols, patterns, or the screen provided by the application program to the analysis engineer. The input device 250 may include a keyboard, a mouse, a touch panel, etc., and is used for an analysis engineer to input information or control actions, so as to complete downloading of data stored in the storage device 100 and fault analysis.

The storage integrated circuit 110 may at least include a controller 120 , a storage module 130 and a substrate 140 . The controller 120 and the storage module 130 can be mounted on the substrate 140 , so that the controller 120 can be coupled or connected to the storage module 130 through the circuit in the substrate 140 . The storage module 130 may include multiple NAND flash memory dies stacked on the substrate 140 to provide a large amount of storage space, usually hundreds of gigabytes (Gigabytes, GB), or even several terabytes (Terabytes, TB), used to store user data, such as high-resolution pictures, videos, etc. The storage cells in each NAND grain can include single-level cells (Single-Level Cells, SLCs), multi-level cells (Multi-Level Cells, MLCs), triple-level cells (Triple Level Cells, TLCs), quadruple Formula units (Quad-Level Cells, QLCs) or any combination of the above. The controller 120 can communicate with the storage module 130 using Open NAND Flash Interface ONFI Sync, Open NAND Flash Interface ONFI Async, Double Data Rate Switch (DDR Toggle) or other communication protocols. The controller 120, the storage module 130 and the substrate 140 can be packaged into a single integrated circuit 100 and installed on the printed circuit board 150, so that the controller 120 can be coupled to the host 210 through the circuit in the printed circuit board 150 and the connector 160 .

Referring to FIG. 2, the host (or analysis host) 210 may at least include a processing unit 211, a memory 212, a display interface 213, an input interface 215, a peripheral storage interface (Peripheral Storage Interface) 217 and a local storage device 219. The processing unit 211 can be implemented in various ways, such as using general-purpose hardware (for example, a single processor, a multi-processor with parallel processing capability, a graphics processor, or other processors with computing capabilities), and when executing a fault analysis application program When the code of the , provides the functions described below. The peripheral storage interface 217 can be Universal Serial Bus (Universal Serial Bus, USB), Advanced Technology Attachment (Advanced Technology Attachment, ATA), Serial Advanced Technology Attachment (Serial Advanced Technology Attachment, SATA), fast peripheral component interconnection (Peripheral Component Interconnect Express, PCI-E), Universal Flash Storage (UFS), Fast Non-Volatile Memory Express (NVMe), Open-Channel SSD (Open-Channel SSD), Embedded Multimedia Card (Embedded MultiMedia Card, eMMC) or other interfaces for connecting to the storage device 100 . When the processing unit 211 loads and executes the fault analysis application program, it sends a command to the storage device 100 through the peripheral storage interface 217 for reading the storage device 100. Store the data in the device 100 for further fault analysis. The memory 212 can be a dynamic random access memory (Dynamic Random Access Memory, DRAM), providing a volatile storage space for caching the data required by the processing unit 211 during the execution of the fault analysis application program, for example, variables , data tables, etc., and the data read from the storage device 100 . The local storage device 219 can be a hard disk (Hard Disk), a solid state disk, etc., providing non-volatile storage space for permanently storing data read from the storage device 100 . The processing unit 211 can also be connected to the display 230 and the input device 250 through the display interface 213 and the input interface 215 respectively.

Referring to FIG. 3 , the controller 120 includes at least a processing unit 310 , a host interface 320 , a flash memory interface 340 and a random access memory (Random Access Memory, RAM) 360 . The processing unit 310 can be implemented in a variety of ways, such as using general-purpose hardware (for example, a single processor, a multi-processor with parallel processing capability, a graphics processor or other processors with computing capabilities), and when executing data backup applications When the code of the , provides the functions described below. The host interface 320, such as USB, ATA, SATA, PCI-E, UFS, NVMe, Open-Channel SSD, eMMC or other interfaces, is connected to the peripheral storage interface 217 of the analysis host 210 for enabling the processing unit 310 to pass through the host interface 320 Start and execute the backup method after receiving a command from the analysis host 310 (more specifically, the processing unit 211 ), and reply the data and related information stored in the storage module 130 to the analysis host 310 through the host interface 320 . The flash memory interface 340 can be ONFI Sync, ONFI Async, DDR Toggle or other interfaces. The RAM 360 can be a dynamic random access memory (Dynamic Random Access Memory, DRAM) or a static random access memory (Static Random Access Memory, SRAM), which is used to store the program code of the data backup application program and the slave storage module 130 read the data. In other embodiments, the RAM 360 may be disposed on the printed circuit board 150 outside the storage integrated circuit 110, and the processing unit 310 accesses the RAM 360 through a memory controller (Memory Controller, not shown in FIG. 3 ). , the present invention is not limited thereby.

When the storage device 100 fails or is judged to be a defective product, a failure analysis is required, It is used to find out the cause of the failure and improve the manufacturing process of specific components in the storage device 100 accordingly, or to modify the program code of the firmware loaded and executed by the processing unit 310 . Before performing destructive analysis (such as hardware damage, firmware damage, stored data damage, etc.), the analysis host 210 can first back up the data stored in the storage module 130, and perform preliminary analysis based on the backup data. In some embodiments, the analysis host 210 can back up all data stored in the storage module 130 . However, the capacity of the storage module 130 is huge, ranging from several GB to several TB, and it will take a lot of time to completely back up all the data, which may take 6-12 hours, which is not conducive to quickly generate a preliminary report. In addition, a full backup of all the data in the storage module 130 will occupy a large amount of space in the local storage device 219 , and the user's private or confidential information in the storage module 130 is also easy to be stolen by criminals.

In order to solve the above-mentioned problems, the present invention proposes a flash memory data fault analysis method for obtaining and backing up key data used for fault analysis in the storage device 100, instead of completely backing up all the data in the storage device 100, thereby Assist in initial analysis.

In one embodiment, the analysis host 210 is responsible for determining the key data that needs to be backed up in the storage device 100, and integrates the data sent by the storage device 100 to generate fault analysis code that can be executed by engineers, computers, or both. The parsed file. Referring to the fault analysis method implemented by the program code of the fault analysis module loaded and executed by the processing unit 211 shown in FIG. 4, the detailed description is as follows:

Step S410: Obtain device information of the storage device 100 . The processing unit 211 can send one or more management commands (Administration Command or Management Command) defined by the standard through the peripheral storage interface 217 or one or more commands defined by the manufacturer of the storage IC 110 (also referred to as supplier commands, Vendor Command) to the storage device 100, and receive the reply device information from the storage device 100, such as the factory identifier (Factory Identifier, FID) of the storage IC 110, the identifier of the controller 120, the identifier of the storage module 130, the NAND equipped therein Flash memory type, NAND flash memory structure (such as plane, channel, number of interleaves, etc.), length of each page, number of pages contained in each physical block, etc.

Step S420 : Generate a device description file and a device basic parameter file according to the device information of the storage device 100 , and store them in the local storage device 219 . For the convenience of interpretation, the content of the device description file can be encoded as an Extensible Markup Language (XML document), including the identification of the controller 120, the identification of the storage module 130, the type of NAND flash memory equipped therein, and the NAND flash memory structure, the length of each page, and the number of pages contained in each entity block. The device basic parameter file can be a 512-byte or 1024-byte binary file (binary file), including the FID of the storage IC 110 dumped from the storage device 100 .

Step S430: Generate empty backup data list files and data files and store them in the local storage device 219 for storing key data for fault analysis in the storage device 100 (specifically, the storage module 130 ). The content of the backup data list file can be encoded as a text file (Text File), which is beneficial to engineers and/or code interpretation. It contains multiple lines (Rows), and the length of each line is fixed. Bad Block) Metadata for each page of each block. The data files can be binary files. If a page in the storage module 130 stores system data, these system data will be stored in a continuous area in the data file, and the specific field (Field) in the corresponding row of the backup data list file will store the location information of this area , such as the start position and length of the message.

Step S440: Set the variable i to (or call it initially) 0. The storage module 130 includes multiple physical blocks, and each physical block includes multiple pages. Although the hardware of the storage module 130 is configured as multiple channels, each channel is connected to one or more dies, and each die contains multiple physical blocks (which can be identified by using logical unit numbers, block numbers, etc.) , but the processing unit 310 can logically arrange all the entity blocks into a sequence according to the rules set by the entity, and use the variable i to indicate which entity block is currently being processed. It should be noted here that in most current standards, in order to reduce the complexity of operating the storage device 100, in the normal mode (Normal Mode, that is, not in the factory mode), the host-side execution is used to drive the storage device. The code for device 100 is not saved Knowledge of the entity architecture of the module 130 . However, since the code of the fault analysis application is provided by the manufacturer of the storage IC 110 to the manufacturer of the storage device 100 and runs in factory mode, only the knowledge of operating the storage module 130 is available.

Next, the processing unit 211 repeatedly executes a loop (loop, including steps S451 to S459), for obtaining key data that can be used for fault analysis in each physical block of the storage module 130, until all physical blocks are Until the process has been completed ("Yes" path in step S457).

Step S451: Determine whether the i-th physical block is an original bad block (Original Bad Block). If yes, the processing unit 211 executes step S457, otherwise, executes step S453. The storage device 100 will test all the physical blocks in the storage module 130 in the card opening procedure before leaving the factory to detect which physical blocks are bad blocks (herein referred to as original bad blocks), and these bad blocks The physical location information of is recorded in the original bad block table (Original-Bad-Block Table), and stored in the specific physical block in the storage module 130 . At the beginning of the backup process, the processing unit 211 can first send a specific supplier command to the storage device 100 (in detail, the controller 120) through the peripheral storage interface 217 to read the original bad block table from a specific physical block, and store the reply. The original bad block table is in the memory 212 for quick lookup. In this step, the processing unit 310 can search the content of the original bad block table and determine whether the physical location information of the i-th physical block is stored in the original bad block table. If yes, it means that the i-th physical block is an original bad block, and data backup may not be performed to save time for data reading. In general, through step S451, the processing unit 211 can skip the original bad block without processing according to the content of the original bad block table, so as to save time for data backup.

Step S453: Obtain type information of the i-th entity block. The processing unit 310 can send a specific supplier command to the storage device 100 (in detail, the controller 120) through the peripheral storage interface 217, for reading the 0th page of the i-th physical block, and from the 0th page Get the type information of the i-th entity block from Meta Data. In addition to the original bad block, each physical block can be identified as a system block (System Block), link block (Link Block, or called Mapping block, Map Block), cache block (Cache Block, or user data block User-data Block), mother block (Mother Block), child block (Child Block), etc., and use another metadata in the 0th page A byte to identify whether it is a new bad block (New Bad Block).

Each page in the system block can be used to store metadata and system data (System Data), for example including the profile data (Profile Data) of the controller 120 and the storage module 130, the original bad block table, the original bad row table (Original Bad Column Table), section start table, In-System Programming (ISP Code), etc., or any combination of the above, wherein the metadata is used to describe the system data in the same page.

Each page in the link block can be used to store metadata and system data, such as high-level group table (High-level Group Table), logical entity location mapping table (Logical-To-Physical, L2P Table), entity logical location mapping Table (Physical-To-Logical, P2L Table), host flash memory location mapping table (Host-To-Flash, H2F Table), flash memory host location mapping table (Flash-To-Host, F2H Table), or any combination of the above, wherein Metadata is the data used to describe the system data in the same page.

Each page in the cache block except the End-Of-Block (EOB Page) can be used to store metadata and user data (User Data), where metadata is used to describe user data in the same page material. The user data cannot be interpreted by the controller 120 , but it is data that can be interpreted by the operating system, driver program or specific application software executed by the host 210 , and these user data may include the user's private or confidential information. The block end page of the cache block is a special system page, indicating that the cache block ends here, and the subsequent pages do not store any useful data, and are used to store metadata and system data, where metadata is used to describe the same page The data in the system data.

The parent block and the child block are cache blocks generated in a garbage collection program (Garbage Collection, GC Process), and the data stored therein are similar to the cache blocks. The garbage collection program collects the broken user data in multiple pages, and writes the collected user data into the new Pages (that is, pages in the parent block or child block) are used to make these released pages available for use by other user data after erasing. The parent block means that the pages in it are full of user data in the garbage collection process; and the child block means that the pages in it are not full of user data in the garbage collection process.

A new bad block is a physical block that has an error check and correction (Error Check and Correction, ECC) error when the user uses it. It may be a system block, a link block, a cache block, a parent block or a child block before an ECC error occurs. wait. When an ECC error occurs in a certain block during a read operation, the specific byte of metadata in the 0th page stores the information that this block is a new bad block.

Step S455: Read complete or partial data of each page in the i-th block according to the type information, and add content to the backup data list file and data file according to the received data. The processing unit 211 can send a specific supplier command to the storage device 100 (in detail, the controller 120) through the peripheral storage interface 217 to read the complete or partial data of all or some pages in the i-th block. In addition, The processing unit 211 may further receive the outline information of each page of the i-th block from the reply of the storage device 100 , such as the location and length of the entity in the storage module 130 . The policy for reading data per page is described as follows: For a system block, read metadata and system data for all pages within it.

For link blocks, read metadata and system data for all pages within it.

For a cache block, read metadata for all pages within it except the end page of the block. If the cache block contains the end page of the block, the metadata and system information of the end page of the block are also read.

For the parent block or child block, read the metadata of page 1.

For the new bad block, according to the above-mentioned read policy and the type information of this block, read the complete or partial data of all or part of the pages therein. For example, when this new bad block was a system block or a link block before the error occurred, read the metadata and system data of all pages in it. The reading policy of other types of new bad blocks can be deduced in the same way, and will not be repeated for the sake of simplicity.

For each page of the physical block, if it contains system data, the processing unit 211 may add the system data of this page after the current end position of the data file in the local storage device 219 . The processing unit 211 can also add a new line to the data list file in the local storage device 219, which is used to store the metadata of this page, including a plurality of fields, respectively storing the entity location information of this page (such as the logical unit number number, entity block number, page number, etc.), page type (such as user data page, ISP page, mapping table page, etc.), the serial number of this block, page content index, ECC error flag, system information pointing to the data file storing this page The region index information (such as offset and length), etc. It should be noted that the meaning of page content index will vary with different page types. For example, for a data page, the page content index represents a logical block number (Logical Block Number, LUN), and for a mapping table page, the page content index represents a mapping table number. The ECC error flag is used to indicate whether there is an error bit in the system data or user data of this page but cannot be corrected. If this page is a user profile page, the area indicator info field stores a false value (NULL).

Step S457: Determine whether all necessary physical blocks in the storage module 130 have been processed. If yes, the processing unit 310 executes step S460, otherwise, executes step S459. Those skilled in the art know that all necessary physical blocks in the storage module 130 refer to physical blocks that have been used to store data, excluding empty blocks that have not been used yet.

Step S459: The variable i is incremented by 1.

Step S460: Perform fault analysis according to the content of the generated file. The processing unit 211 can display on the display 230 a user interface (User Interface, UI) that is helpful for engineers to interpret through the display interface 213, including the contents of the above-mentioned device description file, device basic parameter file, backup data list file and data file, It enables engineers to conduct preliminary analysis based on professional experience. Or, the processing unit 211 executes the program code of the fault analysis application program, and uses various algorithms to analyze the contents of the device description file, device basic parameter file, backup data list file and data file in the local storage device 219, The preliminary analysis is completed, and the analysis result is displayed on the display 230 through the display interface 213 .

In another embodiment, the controller 120 of the storage IC 110 and the analysis host 210 perform different divisions of labor. The controller 120 of the storage IC 110 is responsible for determining which key data needs to be backed up, and the analysis host 210 is responsible for integrating the data transmitted by the storage device 100. The uploaded data is generated into a fault analysis code that can be executed by an engineer, a computer, or a file that can be parsed by both. The data backup method implemented with reference to the program code of the data backup module loaded and executed by the processing unit 310 shown in FIG. 5 is described in detail as follows:

Step S510 : Receive a message to start data backup from the analysis host 210 through the host interface 320 , and the message can be included in a reserved byte in a standard-defined read command, or in a provider command.

Step S520: Set variable i to 0 (or call it initial). Other technical details are similar to step S440, and will not be repeated for simplicity.

Next, the processing unit 310 repeatedly executes a loop (loop, including step S530 to step S570), which is used to obtain key data that can be used for fault analysis in each physical block of the storage module 130, until all physical blocks are Until the process has been completed ("Yes" path in step S560).

Step S530: Determine whether the i-th physical block is an original bad block. If yes, the processing unit 310 executes step S560, otherwise, executes step S540. Some technical details are similar to step S451, and will not be repeated for simplicity. Different from step S451, the processing unit 310 can read the original bad block table from the specific physical block through the flash memory interface 340, and store the returned original bad block table in the RAM 360 for quick search.

Step S540: Obtain type information of the i-th physical block. The processing unit 310 can issue a command to the storage module 130 through the flash memory interface 340 to read the 0th page of the i-th physical block, and obtain the type information of the i-th physical block from the metadata in the 0th page . For the type information of the physical block, please refer to the description of step S453, which will not be repeated for simplicity.

Step S550: Send a command to the storage module 130 through the flash memory interface 340 according to the type information to read the complete or partial data of all or part of the pages in the i-th block, and pass the master The machine interface 320 replies to the analysis host 210 . In addition to the read data, the processor 310 can also return the outline information of each page of the i-th block, such as the physical location and length in the storage module 130 . For the policy of reading the data of each page, please refer to the relevant part of the description of step S455, and will not repeat it for simplicity.

Step S560: Determine whether all necessary physical blocks in the storage module 130 have been processed. If yes, the processing unit 310 executes step S580, otherwise, executes step S570. Those skilled in the art know that all necessary physical blocks in the storage module 130 refer to physical blocks that have been used to store data, excluding empty blocks that have not been used yet.

Step S570: The variable i is incremented by 1.

Step S580: Reply to the analysis host 210 with a message that data backup is complete.

It should be noted here that since the storage module 130 can be designed as a multi-channel access architecture, multiple rounds in the loop described in FIG. 5 can be executed in a time-sharing multitasking environment (Time-sharing Multitasking Environment) or multi-core processor, and the execution of some steps between two or more rounds can overlap in time, for example, when the i-th round executes step S550, the i+1th round executes step S540, etc., to optimize the reading efficiency, The invention should not be limited thereby.

In addition, with reference to the fault analysis method implemented by the program code of the fault analysis module loaded and executed by the processing unit 211 shown in FIG. 6, the detailed description is as follows:

Step S610: The technical details are similar to step S410, and will not be repeated for simplicity.

Step S620: The technical details are similar to step S420, and will not be repeated for simplicity.

Step S630: The technical details are similar to step S430, and will not be repeated for simplicity.

Step S640 : Send a message to start data backup to the storage device 100 . The processing unit 211 can send a standard-defined read command or a supplier command through the peripheral storage interface 217 to notify the controller 120 to start data backup, so that the processing unit 310 starts to execute the data backup method as shown in FIG. 5 . This step corresponds to step S510 in FIG. 5 .

Next, the processing unit 211 repeatedly executes a loop (including step S650 to step S670), which is used to obtain from the storage device 110 the The key data used for fault analysis, and the content of the backup data list file and the data file are generated until all the physical blocks have been received and processed (the path of "Yes" in step S670).

Step S650: Receive information about a physical block from the storage device 100 . This step corresponds to step S550 of FIG. 5 . The received physical block may be the above-mentioned system block, link block, cache block, parent block, child block, etc. In addition, the received physical block may be a new bad block.

Step S660: Add content to the backup data list file and data file according to the received data. The technical details are similar to the relevant parts of step S455, and will not be repeated for simplicity.

Step S670 : Determine whether a message that data backup is complete is received from the storage device 100 through the peripheral storage interface 217 . If yes, the processing unit 211 executes step S680, otherwise, executes step S650. This step corresponds to step S580 of FIG. 5 .

Step S680: Perform fault analysis according to the content of the generated file. The technical details are similar to the content of step S460, and will not be repeated for simplicity.

It should be noted that although steps S510 to S580 in FIG. 5 are completed by the processing unit 310 in the controller 120 as described above, since the above method is triggered by the processing unit 211 of the analysis host 210 when executing step S640, the analysis The processing unit 211 of the host 210 can be regarded as a technical means that indirectly completes steps S510 to S580 as described above.

Through the execution of the method in the embodiments described in FIGS. 4 to 6 , in summary, the data obtained and backed up by the processing unit 221 of the analysis host 210 from the storage device 100 does not include user data. To summarize from another aspect, the data acquired and backed up by the processing unit 221 of the analysis host 210 from the storage device 100 includes metadata and system data of each page in each block except the original bad block.

All or part of the steps in the method of the present invention can be realized by computer instructions, such as the operating system of the computer, the driver program of the specific hardware in the computer, or the software program. In addition, it can also be implemented in other types of programs. Those skilled in the art can write the methods of the embodiments of the present invention into computer instructions, and no further description will be made for the sake of brevity. According to the embodiment of the present invention Computer instructions for implementing the method can be stored on a suitable computer-readable medium, such as DVD, CD-ROM, USB drive, hard disk, or placed on a network (for example, the Internet, or other suitable vehicle) Accessed web server.

Although the elements described above are included in FIGS. 1 to 3 , it is not excluded that more additional elements may be used to achieve better technical effects without violating the spirit of the invention. In addition, although the flow charts in FIG. 4 to FIG. 6 are executed in a specified order, those skilled in the art can modify the order of these steps under the premise of achieving the same effect without violating the spirit of the invention. Therefore, The invention is not limited to using only the sequence described above. In addition, those skilled in the art may also integrate several steps into one step, or perform more steps sequentially or in parallel in addition to these steps, and the present invention is not limited thereby.

Although the present invention is described using the above examples, it should be noted that these descriptions are not intended to limit the present invention. On the contrary, the invention covers modifications and similar arrangements obvious to those skilled in the art. Therefore, the claims of the application must be interpreted in the broadest manner to include all obvious modifications and similar arrangements.

S410~S460: method steps

Claims (11)

A method for analyzing flash memory data faults, implemented by a processing unit of an analysis host when loading and executing software program codes, wherein the processing unit is coupled to a peripheral storage interface for connecting to a storage device where a fault occurs The above-mentioned method includes: the above-mentioned processing unit of the above-mentioned analysis host obtains the data used for fault analysis in the above-mentioned storage device through the above-mentioned peripheral storage interface and backs it up in a local storage device, instead of completely backing up all the data in the above-mentioned storage device, thereby To assist in preliminary analysis, wherein the storage device includes a storage module, the storage module includes multiple physical blocks, and each of the above physical blocks includes multiple pages for storing metadata, system data and user data, wherein, The above-mentioned user data includes the user's private or confidential information, and the above-mentioned data obtained and backed up include the above-mentioned metadata and the above-mentioned system data of each of the above-mentioned pages in each of the above-mentioned physical blocks except the original bad block, but do not include The above-mentioned user data of each of the above-mentioned pages in each of the above-mentioned physical blocks other than the original bad block. The method for fault analysis of flash memory data as described in claim 1, comprising: reading an original bad block table from the above-mentioned storage module in the above-mentioned storage device; and skipping the above-mentioned storage module according to the content of the above-mentioned original bad block table The original bad blocks in the file are not processed. The method for fault analysis of flash memory data as described in claim 1, comprising: obtaining the type information of each of the above-mentioned physical blocks except the original bad block from the above-mentioned storage module in the above-mentioned storage device; and according to each of the above-mentioned physical blocks Information of the above type is read from the above storage device by the same entity Complete or partial data of each of the above pages in the block, and store the above read data to the above local storage device. A computer program product for failure analysis of flash memory data, including program code loaded and executed by a processing unit of an analysis host, wherein the processing unit is coupled to a peripheral storage interface for connecting to a storage device where a failure occurs device, the above-mentioned program code includes: the above-mentioned processing unit of the above-mentioned analysis host obtains the data used for fault analysis in the above-mentioned storage device through the above-mentioned peripheral storage interface and backs it up in a local storage device, instead of completely backing up all the data in the above-mentioned storage device , so as to assist in preliminary analysis, wherein the storage device includes a storage module, the storage module includes a plurality of physical blocks, and each of the physical blocks includes a plurality of pages for storing metadata, system data and user data, Wherein, the above-mentioned user data includes the user's private or confidential information, and the above-mentioned data obtained and backed up include the above-mentioned metadata and the above-mentioned system data of each of the above-mentioned pages in each of the above-mentioned physical blocks except the original bad block, but not Contains the above-mentioned user data of each of the above-mentioned pages in each of the above-mentioned physical blocks except the original bad block. The computer program product as described in claim 4, including the program code loaded and executed by the above-mentioned processing unit of the above-mentioned analysis host: reading the original bad block table from the above-mentioned storage module in the above-mentioned storage device; and according to the above-mentioned original The content of the bad block table skips the original bad blocks in the above-mentioned storage module without processing. The computer program product as described in claim 4, including the program code loaded and executed by the above-mentioned processing unit of the above-mentioned analysis host: Obtain the type information of each of the above-mentioned physical blocks except the original bad block from the above-mentioned storage module in the above-mentioned storage device; and read each of the above-mentioned physical blocks in the same physical block from the above-mentioned storage device according to the above-mentioned type information of each of the above-mentioned physical blocks Complete or partial data of the page, and store the above-mentioned read data to the above-mentioned local storage device. The computer program product as described in claim 6, including the program code loaded and executed by the above-mentioned processing unit of the above-mentioned analysis host: the above-mentioned system for storing each of the above-mentioned pages of each of the above-mentioned physical blocks read from the above-mentioned storage module Data to the data file in the above-mentioned local storage device; store the above-mentioned metadata of each of the above-mentioned pages of each of the above-mentioned entity blocks read from the above-mentioned storage module to the backup data list file in the above-mentioned local storage device, wherein the above-mentioned metadata It is the data describing the above-mentioned system data or the above-mentioned user data in the same page; and when any of the above-mentioned pages of any of the above-mentioned entity blocks read from the above-mentioned storage module contains the above-mentioned system data, the system is recorded in the above-mentioned data file The location information of the data to the specific field of the corresponding line in the above backup data list file. The computer program product as described in claim item 7, wherein the above-mentioned data file is a binary file, and the above-mentioned backup data list file is a text file. The computer program product as described in claim 6, wherein the policy of reading and backing up all or part of the above-mentioned data includes: for the system block, reading and backing up the metadata and system data of all pages in the above-mentioned system block; for the link block, read and backup metadata and system data of all pages in the above link block; For the cache block, read and back up the metadata of all pages in the above cache block except the block end page, if the above cache block contains the block end page, also read and backup the metadata and system data of the above block end page; And for the new bad block, according to the above-mentioned type information of the new bad block and the above-mentioned read and backup policy, read the complete or partial data of all or part of the pages in the above-mentioned new bad block. The computer program product as described in claim 6, wherein the policy of reading and backing up the above-mentioned complete or partial data includes: for the parent block or the child block, reading and backing up the element of the first page of the above-mentioned parent block or child block material. The computer program product as described in claim 4, including the program code loaded and executed by the processing unit of the analysis host: obtaining the device information of the storage device from the storage device; and the device information based on the storage device Generate a device description file and a basic device parameter file, and store the device description file and the basic device parameter file in the local storage device, wherein the basic device parameter file records the manufacturer identification of the storage integrated circuit in the storage device.

Images