JP2011129221A - Disk drive supporting interface by data sector different in size, and backup method for user data - Google Patents

Abstract

In a disk drive that supports an interface with data sectors of different sizes, non-rewritten data is prevented from being lost when power is turned off in write processing.
In one embodiment of the present invention, the HDD supports two different size data sectors at the interface. The data sector on the magnetic disk is a large size data sector. When the HDD receives a write command with a small data sector from the host and the boundary of the designated address (rewrite area) does not match the boundary of the large data sector on the magnetic disk, the HDD Perform backup processing. The backup area is composed of one or a plurality of data tracks.
[Selection] Figure 8

Description

  The present invention relates to a disk drive that supports an interface with data sectors of different sizes and a method for backing up user data thereof.

  As a disk drive, an apparatus using a disk of various modes such as an optical disk, a magneto-optical disk, or a flexible magnetic disk is known. Among them, a hard disk drive (HDD) is an external storage device of a computer. As widely used. Further, it is used in many other electronic devices such as a moving image recording / reproducing apparatus or a car navigation system.

  A magnetic disk used in the HDD has a plurality of data tracks and a plurality of servo tracks. The servo track is composed of a plurality of servo sectors having address information. In addition, a plurality of data sectors including user data are recorded on the data track. Data sectors are recorded between servo sectors spaced apart in the circumferential direction.

  The head slider is fixed on the suspension of the actuator. As the actuator swings about the swing axis, the head slider moves to the target data track and is positioned on the track. The head element part of the head slider supported by the oscillating actuator accesses the desired data sector according to the servo data address information, thereby writing data to the data sector and reading data from the data sector. It can be performed.

  The HDD accesses data on the magnetic disk in units of data sectors. In a typical HDD, the data sector size is 512 bytes. For this reason, many current programs (hosts) are written based on 512-byte data sectors. On the other hand, the data size handled by software has increased more and more in recent years, and it is considered that it is preferable to increase the data size of the access unit of the HDD accordingly. As an example, a data sector size of 4 Kbytes (4096 bytes) has been proposed. Further, increasing the size of the data sector, which is a unit of access, can relatively reduce the additional redundant data required in addition to the user data and increase the capacity of the recording surface.

  One problem in changing the data sector size of the HDD is the inconsistency of the data sector size in the program (host). Even if large-sized data sectors become widespread in the future, it is unlikely that all software (host) will be written on the basis of large-sized data sectors immediately. It can be written based on the data sector. The programs used by the user include new programs and old programs. Therefore, the HDD is required to support the conventional 512-byte data sector in addition to the new large size data sector.

  In order to solve such a problem, an HDD having an interface that supports both a conventional 512-byte data sector and a new large-size data sector (4 Kbyte data sector) has been proposed (for example, , See Patent Document 1). This HDD writes and reads data to and from a magnetic disk with 4 Kbyte data sectors. The command specifies the start address and data length of the access destination. When a command with a 512-byte sector size is received, the HDD converts the designated address into an address of 4 Kbyte sector and accesses the magnetic disk.

  The boundary address of data (access area) specified in 512-byte units may not match the boundary address in 4 Kbytes. FIG. 13 shows an example of an area specified by three 4 Kbyte data sectors and a command by 512 byte data sectors. The command specification area is 5 Kbytes, and both the head address and the end address do not match the boundary address of the 4 Kbyte data sector. In such a case, the HDD needs to access all three 4K data sectors including the designated address. In other words, in addition to the data sector specified by the command (center data sector), the data sector including the address area specified for only a part of the data sector (first and last data sector) Also access.

  Specifically, in reading, the HDD reads all data sectors including data of a specified address from the magnetic disk to a buffer. Some data sectors also contain data other than addresses specified by the host. Therefore, the HDD extracts only necessary data from the data read from the magnetic disk and transfers it to the host. In writing, the HDD reads a data sector partially including a specified address, generates new sector data from a part of the read data sector and data acquired from the host, and uses the data as data on the magnetic disk. • Write to the sector.

JP 2005-63441 A

  An unexpected power cut may occur while data is being written to the magnetic disk according to the write command. In such a case, the data sector in the rewrite area designated by the command does not store the latest user data, and part or all of the user data acquired from the host is lost. However, the data sector in error is the data sector at the address specified by the host using the write command, and the only data that may be lost is the user data subject to the write command. . The designer can design the software (host) after understanding this point, and this data loss can be handled as an allowable event.

  However, when the data sector size at the interface with the host is different from the data sector size for accessing the magnetic disk in the HDD, the power-off in the write processing causes a new problem. As described above, when the read / write of the magnetic disk is a large size data sector of 4 Kbytes and the interface with the host is a small size data sector of 512 bytes, in the write process, on the magnetic disk In some cases, only a part of the data sector at the beginning and end of the rewrite area is rewritten.

  If the power is cut off during the writing of the first or last data sector, data not to be rewritten in the data sector is lost. That is, user data different from the current write command data is lost. This lost data is user data of the host (program) that issued the write command, or user data of a different host (program).

  In any case, this loss of user data is data loss that is not scheduled by the host that issued the write command or the host that owns the lost data. Therefore, the HDD is required to have a technique for preventing such data loss. Furthermore, it is necessary to reliably and efficiently perform processing for preventing data loss.

One aspect of the present invention is a disk drive having a host interface that supports both large and small data sectors and accessing the disk with the large data sectors. The disk drive includes a head that accesses the disk, a moving mechanism that moves the head on the disk, and a controller that controls the head and the moving mechanism.
The controller receives a write command and determines a data sector size of the write command. When the data sector size of the received write command is the small size data sector, each of the beginning and end of the rewrite area specified by the write command is a large size on the disk. Determine if it matches the data sector boundary. If one of the head and tail does not match the boundary of the large data sector, the backup data including the large data sector including the one is included, and if both do not match, both of the data are included. Backup data including a large size data sector is stored in a backup area consisting of one or more data tracks on the disk. As a result, it is possible to prevent data outside the data rewrite area from being lost when the power is turned off in the write process, and to reduce the delay in writing backup data to the backup area.

  In a preferred configuration, after the seek to the backup area is completed, the controller starts writing backup data from the latest data sector where writing can be started. Thereby, the delay in writing backup data to the backup area can be further reduced.

  In a preferred configuration, the controller starts writing the backup data from a preset predetermined data sector after completion of seek to the backup area. As a result, the delay in writing the backup data to the backup area can be reduced by a simple process.

  In a preferred configuration, the backup data includes data indicating that it is the latest backup data. As a result, the latest backup data can be found quickly and easily in the recovery process from the power interruption.

  The data indicating the latest backup data is preferably a counter value of a counter. As a result, it is possible to indicate that the backup data is the latest with a simple configuration. Furthermore, it is preferable that the controller initializes the backup area when the counter reaches a predetermined value. Thereby, the latest backup data can be surely indicated by the counter of a predetermined size.

  The backup data preferably includes a backup source address. As a result, the backup source can be specified in the recovery process from the power shutdown without referring to other data.

  Preferably, the controller stores the backup data in a continuous area in the backup area. Thereby, the backup processing time can be shortened.

  In a preferred configuration, the backup data includes a flag stored in a data sector before and after user data, and the controller updates the flag after writing new data to the write area, and the controller Uses the flag, the user data of the backup source, and the data in the rewrite area to execute a recovery process corresponding to power shutdown in data writing. As a result, the recovery process corresponding to the power shutdown can be executed with the backup data and the backup source user data without referring to other data.

  In a preferred configuration, the controller temporarily holds address information of a data sector storing the backup data, and specifies the address of the flag by referring to the address information in updating the flag. Thereby, the flag which should be updated easily and rapidly can be specified.

  In a preferred configuration, the controller is such that both the head and the tail do not coincide with the boundary of the large data sector, and the tail data sector including the tail is preceded by the head data sector including the head. When the time from the tail data sector to the head data sector is smaller than a threshold, the head data sector is read after the tail data sector for storage in the backup area. Thereby, the backup processing time can be shortened.

  Another aspect of the present invention is a disk drive having a host interface that supports both large data sectors and small data sectors, wherein the disk is accessed by the large data sectors. This is a method for backing up user data. This method receives a write command. The data sector size of the write command is determined. When the data sector size of the received write command is the small size data sector, each of the beginning and end of the rewrite area specified by the write command is a large size on the disk. Determine if it matches the data sector boundary. If one of the head and tail does not match the boundary of the large data sector, the backup data including the large data sector including the one is included, and if both do not match, both of the data are included. Backup data including a large size data sector is stored in a backup area consisting of one or more data tracks on the disk. As a result, it is possible to prevent data outside the data rewrite area from being lost when the power is turned off in the write process, and to reduce the delay in writing backup data to the backup area.

  ADVANTAGE OF THE INVENTION According to this invention, in the disk drive which supports the interface by the data sector of a different size, it can prevent that the data outside a data rewriting area | region is lose | disappeared at the power-off in write processing, and backup processing for that is efficient Can be done automatically.

In this embodiment, it is a block diagram which shows typically the whole structure of HDD. In this embodiment, it is a block diagram which shows typically the logical structure which performs the read / write process according to different data sector size. 6 is a flowchart showing an overall flow of read / write processing according to different data sector sizes in the present embodiment. In the present embodiment, an example is shown in which the boundary between the head address and the tail address of the read command designation address does not coincide with the data sector of the magnetic disk. In this embodiment, it is a flowchart which shows the flow of the write process containing a backup process. In this embodiment, it is a schematic diagram explaining the backup process in a data write process. In this embodiment, it is a schematic diagram explaining the backup process in a data write process. FIG. 6 is a diagram schematically illustrating information included in backup management data when the boundary between the head address and the tail address of the write command designation address does not coincide with the data sector of the magnetic disk in the present embodiment. is there. In this embodiment, it is a flowchart which shows the flow of the whole write process including a backup process. In this embodiment, it is a schematic diagram for demonstrating the write processing by the small size data sector including a backup process. In this embodiment, it is a table which illustrates the state stage in the write process including a backup process. In this embodiment, it is a flowchart which shows typically the flow of a process from starting of HDD to power-off. In this embodiment, it is a schematic diagram for demonstrating the process which reads a tail data sector first for backup. In this embodiment, it is a schematic diagram for demonstrating the process which reads a tail data sector first for backup. FIG. 10 is a diagram schematically illustrating an example of a data sector and a rewrite area in an HDD that supports data sectors of different sizes in the related art.

  Hereinafter, embodiments of the present invention will be described. For clarity of explanation, the following description and drawings are omitted and simplified as appropriate. Moreover, in each drawing, the same code | symbol is attached | subjected to the same element and the duplication description is abbreviate | omitted as needed for clarification of description. In the following, an embodiment of the present invention will be described in a hard disk drive (HDD) which is an example of a disk drive.

  The HDD according to the present embodiment supports both a large data sector and a small data sector in the host interface. In other words, the host can issue read and write commands to the HDD in both the large data sector and the small data sector. The size of the small size data sector is, for example, 512 bytes, and the size of the large size data sector is, for example, 4K (4096) bytes.

  The data sector on the magnetic disk is a large size data sector. The HDD always accesses (reads or writes) the magnetic disk with large data sectors. As described above, by increasing the data sector size on the magnetic disk, it is possible to reduce redundant data required for the data sector and increase the data capacity of the recording surface.

  The HDD of this embodiment receives a write command with a small size data sector from the host, and if the boundary of the designated address (rewrite area) does not coincide with the boundary of the large size data sector on the magnetic disk, -Perform data backup processing. As a result, even if an unexpected power-off occurs during the write process, it is possible to prevent the loss of user data outside the rewrite area and unrelated to the write command. This embodiment is characterized by this data backup processing. Before describing the write processing of this embodiment in detail, the overall configuration of the HDD will be described.

  FIG. 1 is a block diagram schematically showing the overall configuration of the HDD. The HDD 1 has a magnetic disk 11 that is a disk for storing data in the enclosure 10. A spindle motor (SPM) 14 rotates the magnetic disk 11 at a predetermined angular velocity. Corresponding to each recording surface of the magnetic disk 11, a head slider 12 for accessing the magnetic disk 11 is provided. Access is a superordinate concept of read and write. Each head slider 12 includes a slider that flies over the magnetic disk, and a head element unit that is formed on the slider and converts between a magnetic signal and an electric signal.

  Each head slider 12 is fixed to the tip of the actuator 16. The actuator 16 is connected to a voice coil motor (VCM) 15, and moves in the radial direction on the magnetic disk 11 that rotates the head slider 12 by rotating about a rotation axis. The actuator 16 and the VCM 15 are moving mechanisms for the head slider 12.

  A circuit is mounted on the circuit board 20 fixed to the outside of the enclosure 10. The motor driver unit 22 drives the SPM 14 and the VCM 15 according to control data from the HDC / MPU 23. The RAM 24, which is a volatile memory, functions as a buffer that temporarily stores read data and write data. The arm electronic circuit (AE) 13 in the enclosure 10 selects the head slider 12 that accesses the magnetic disk 11 from among the plurality of head sliders 12, amplifies the reproduction signal, and reads / writes the channel. (RW channel) 21. Further, the recording signal from the RW channel 21 is sent to the selected head slider 12. The present invention can be applied to an HDD having only one head slider 12.

  In the read process, the RW channel 21 extracts data from the read signal acquired from the AE 13 and performs a decoding process. The data to be read includes user data and servo data. The decoded read user data and servo data are supplied to the HDC / MPU 23. In the write process, the RW channel 21 code-modulates the write data supplied from the HDC / MPU 23, further converts the code-modulated write data into a write signal, and supplies the write signal to the AE 13.

  The controller HDC / MPU 23 controls read / write processing, command execution order, positioning control (servo control) of the head slider 12 using servo signals, interface control with the host 51, defect management, error. For example, necessary processing related to data processing such as error handling processing in the case of occurrence of the error and overall control of the HDD 1 are executed.

  The host interface included in the HDC / MPU 23 of the HDD 1 of this embodiment supports both small size data sectors and large size data sectors. The magnetic disk 11 stores data in large size data sectors. When the command from the host 51 performs addressing by a large size data sector, the HDD 1 may read / write data from / to the magnetic disk 11 according to a normal processing flow. However, when a command from the host 51 requests access by a small size data sector, the HDD 1 needs to emulate access by a small size data sector.

  The present embodiment is characterized by data backup processing in write processing by small size data sectors. First, the overall flow of read / write processing according to different data sector sizes is shown in FIG. This will be described with reference to the block diagram and the flowchart of FIG. FIG. 2 is a block diagram schematically showing a logical configuration for performing processing according to the data sector size. The HDC / MPU 23 controls all accesses to the host interface and the magnetic disk 11.

  The HDC / MPU 23 includes a data size detection unit 231, a large size data sector control unit 232, and a small size data sector control unit 233. The HDC / MPU 23 includes an HDC and an MPU that are hardware, and each functional element described above is realized by the hardware or the MPU operating according to the firmware. Therefore, the HDC / MPU 23 has the above functional elements as separate circuit configurations, or functions as functional elements at different timings.

  When a command is received from the host 51 (Y in S11), the data size detection unit 231 specifies the data sector size of the data handled by the command (S12). Typically, the command has a field for storing information indicating the data sector to be used, and the data size detection unit 231 refers to the field to thereby determine the data sector assumed by the command.・ Detect the size.

  When the detected data sector size is a large size (Y in S12), the large size data sector control unit 232 executes a user data read / write process. Since the data sector size specified by the command and the data sector size of the magnetic disk 11 are the same, the processing of the large size data sector control unit 232 is the same as the processing of a general HDD.

  In the write process, the large size data sector control unit 232 receives user data in units of large size data sectors from the host 51 (S13) and stores the user data in the buffer 241. Further, the data from the buffer 241 is transferred to the RW channel 21 in units of large size data sectors. The transferred data sector is written to the magnetic disk 11 by the head slider 12 via the RW channel 21 and the AE 13. The RW channel 21, the actuator 16, and the AE 13 are controlled by the large size data sector control unit 232 (S14).

  In the read process, the large size data sector control unit 232 controls the RW channel 21, the actuator 16, and the AE 13 to read the data sector at the address specified by the command from the magnetic disk 11, and stores it in the buffer 241. Store (S14). Further, the read data sector is transferred from the buffer 241 to the host 51 (S15).

  When the detected data sector size is a small size (N in S12), the small size data sector control unit 233 executes a user data read / write process. In the write processing, the small size data sector control unit 233 receives user data from the host 51 in units of small size data sectors and stores the user data in the buffer 241 (S16). Thereafter, data writing processing to the magnetic disk 11 including emulation control of a small size data sector is executed (S17). Details of this method will be described later.

  In the read process, the small size data sector control unit 233 controls the RW channel 21, the actuator 16 and the AE 13 to read out the data sector at the address specified by the command from the magnetic disk 11, and stores it in the buffer 241. Store (S17). Reading from the magnetic disk 11 includes emulation control of small data sectors. Details of this method will be described later. Thereafter, the small size data sector control unit 233 further transfers the read data from the buffer 241 to the host 51 in units of small size data sectors (S18).

  The data reading process from the magnetic disk 11 by the small size data sector will be described with reference to FIG. When the data sector size in the host interface and the data sector size on the magnetic disk are different, a mismatch between the address boundary specified by the command from the host and the data sector boundary on the magnetic disk occurs. Therefore, the small size data sector control unit 233 reads data other than the data at the address indicated by the read command from the magnetic disk 11.

  FIG. 4 shows a case where the boundary between the head address and the tail address of the read command designation address does not coincide with the data sector boundary of the magnetic disk 11. The data sectors in the read area 112 on the magnetic disk are six large size data sectors 111a to 111f, and the read area 112 specified by this command is the center of the four data sectors. It occupies all of the data sectors 111b to 111e and a part of each of the front and rear data sectors 111a and 111f.

  The small size data sector control unit 233 calculates the address of the data sector in the large size data sector unit from the address specified by the read command in the small size data sector unit, and the corresponding data sector 111a. To 111f are read from the magnetic disk 11 and stored in the buffer 241. Further, the small size data sector control unit 233 extracts necessary data from the read data, and transfers the data to the host 51 in units of small size data sectors.

  Next, the flow of write processing including backup processing will be described with reference to the flowchart of FIG. When the write command is received from the host 51, the data size detection unit 231 specifies the data sector size and determines whether to perform the emulation process of the small size data sector (S21). If the emulation is not performed (N in S21), the large size data sector control unit 232 executes the data writing process to the magnetic disk 11 by the large size data sector (S22).

  When the write command specifies the address in the small size data sector (Y in S21), the small size data sector control unit 233 writes the data to the magnetic disk 11 by the small size data sector. At this time, it is determined whether the head and tail address boundaries designated by the write command match the data sector address boundaries on the magnetic disk 11 (S23).

  If the address boundary matches at both the head and the tail (Y in S23), the small size / data / sector controller 233 writes data to the magnetic disk 11 without performing backup. Specifically, the small size data sector control unit 233 collects the small size data sector into the large size data sector, and, like the large size data sector control unit 232, the large size data sector. Thus, the data writing process to the magnetic disk 11 is executed (S22).

  If the address boundary does not match at least one of the head and the tail (N in S23), the small size data sector control unit 233 writes data to the magnetic disk 11 including the backup (S24). The backup process in the data writing process will be described with reference to FIGS. 6A and 6B. In the data writing process, when the boundary of at least one of the start address and the end address of the write command designation address does not coincide with the data sector of the magnetic disk 11, the small size data sector control unit 233 of the present embodiment Perform data backup processing. The present embodiment is characterized by a backup process in this data writing process.

  In the following, processing when the boundary between both the start address and the end address of the write command designation address does not coincide with the data sector of the magnetic disk 11 will be described. If either one of the boundaries does not match, backup processing is performed only for the data sector that does not match. The write processing methods are basically the same except that the data sectors to be saved are different. When the boundary of both coincides with the data sector of the magnetic disk 11, the backup process is unnecessary and is omitted.

  FIG. 6A is a diagram schematically illustrating backup processing to the backup area 115 on the magnetic disk 11. In the example of FIG. 6A, the small size data sector control unit 233 stores some sectors in the data write area 113 composed of a plurality of sectors in the partial area 116 of the backup area 115. FIG. 6B schematically shows the data write area 113, the rewrite area 114 in which new data is written in the area 113, and the area 116 in which backup user data and backup management data are written in the backup area 115. ing. Hereinafter, backup user data and backup management data written in the backup process are referred to as backup data.

  As described above, in this example, the boundary between the start address and the end address of the write command designation address does not coincide with the data sector of the magnetic disk 11. The data sectors in the data writing area 113 on the magnetic disk 11 are six large size data sectors 111a to 111f. The area 114 specified by this command includes all of the four central data sectors 111b to 111e and a part of each of the front and rear data sectors 111a and 111f in the six data sector areas. Occupy space.

  In the write process, the small size / data / sector control unit 233 writes the data acquired from the host 51 to the magnetic disk 11 and performs a backup process of necessary data. As a feature of this embodiment, the backup area 115 is an area composed of one data track or a plurality of continuous data tracks.

  In the backup process, the small size data sector control unit 233 ignores the existing data in the backup area 115 and writes new data after the seek to the backup area 115 is completed. As a result, the small size data sector control unit 233 can start writing in the backup area 115 as soon as possible after completion of the seek.

  Typically, the small-size data sector control unit 233 sets the backup data (the backup management data and the backup data from the specified data sector (preferably the first)) set in advance with reference to the completion of seek. Starts writing user data. It is preferable that the small size data sector control unit 233 starts writing backup data in the shortest time from the completion of seek as a function of the HDC / MPU 23 to the start of data writing. Preferably, the small size data sector control unit 233 designates a data sector that can be written most recently from the data sector position during the following processing after completion of the seek, and performs backup processing from the nearest data sector. Start writing data.

  The seek completion condition depends on the design of the HDD 1. For example, the position error signal, which is the difference between the target position and the current head position, is within the specified range for the specified number of consecutive sectors, and the change speed of the position error signal is within the specified range for the specified number of consecutive sectors. As a condition, the seek process is completed. The HDC / MPU 23 enters the following process after the seek process is completed. Preferably, the small size data sector control unit 233 predicts a data sector position that can be written most recently before entering the following process, and calculates backup data from the first data sector that has entered the following process. Write. By specifying the first data sector as a write data sector, the processing time can be shortened.

  In this way, by preparing an area consisting of one data track or a plurality of continuous data tracks for backup, the small-size data sector control unit 233 can perform backup / backup from an arbitrary position in the backup area 115. Data writing can be started. Thereby, the processing time of backup processing can be shortened.

  One backup area may be formed in the entire recording area of the HDD 1, or one backup area may be formed for each recording surface. Alternatively, each recording surface may have a plurality of backup areas at different radial positions. For example, since each zone has a backup area, the processing time of backup processing can be effectively reduced. In the example described in detail below, each recording surface has one backup area.

  Data to be backed up is a large size data sector in which host-specified data occupies a part. That is, in the data sectors 111a and 111f in which the boundary (start address and end address) of the host designated address (area) is located at the center in the data sector (address position other than the start address and end address of the data sector). is there.

  Such a data sector includes user data (data unrelated to the write command) outside the rewrite area 114 specified by the write command. Therefore, by backing up the data sectors 111a and 111f in which the boundary of the data sector and the boundary of the host designated area in the data sector do not coincide with each other, the power is cut off when data is written to the magnetic disk 11. In addition, it is possible to protect user data in an area where rewrite is not designated by the write command.

  The small size data sector control unit 233 writes backup management data in the backup area 115 in addition to user data in the backup process of user data. As a result, the necessary user data can be accurately recovered from when the power is turned off. In a preferred configuration, the small size data sector control unit 233 writes the backup management data in the data sectors 111g and 111j before and after the data sectors 111h and 111i for recording the backup user data.

  FIG. 7 is a diagram schematically showing information included in the backup management data in this configuration. The backup management data and the backup user data (user data backup) are preferably written in successive data sectors. Thereby, the backup processing time can be shortened. When defective sectors exist in the backup area 115, they are skipped. In the processing by the small size data sector control unit 233, the defective sector is treated as a nonexistent data sector. When a defective sector exists between data sectors storing data, the data sectors storing data are considered to be continuous.

  The backup management data is stored in the start sector 111g and the end sector 111j. The start sector 111g stores a counter value, a backup start address, a backup end address, and a backup start flag. The end sector 111j stores a counter value, a backup start address, a backup end address, and a backup end flag. The counter value, backup start address, and backup end address of the start sector 111g and end sector 111j are the same value.

  The small size data sector control unit 233 can determine which data is the target data of the last backup process by referring to this in the recovery process from the power shutdown. The small size data sector control unit 233 increments or decrements the counter for each backup process, generates a new counter value, and includes the latest counter value in the backup management data.

  The counter value is data indicating the latest (last) backup user data. The area indicating the minimum or maximum counter value stores the latest (last) backup user data. The backup start address is the start address of the saved user data (in this example, the address of the data sector 111a). The backup end address is the end address of the saved user data (in this example, the address of the data sector 111f). By referring to these, the small size data sector control unit 233 can know the address of the backup user data in the recovery processing from the power shutdown.

  When only one data sector is to be saved, an address corresponding to a data sector that has not been saved may be indicated. For example, when only the head data sector is saved, the backup end address indicates that the tail data sector is not saved.

  The backup start flag and the backup end flag are data indicating a stage (state) when the power is turned off in the backup process. The small size / data / sector control unit 233 can specify at which stage the power shutdown occurs in the recovery process from the power shutdown by writing and rewriting these flags in the backup process. The backup start flag is stored in the start sector 111g, and the backup end flag is stored in the end sector 111j. The counter value is stored in both sectors. Other data may be stored in any data sector, or may be written in both data sectors.

  Writing with small size data sectors including backup processing of this embodiment (S24) will be described with reference to the flowchart of FIG. 8 and the schematic diagram of FIG. The state stage shown in FIG. 9 will be described later. In response to the write command from the host 51, the small size data sector control unit 233 stores the write data from the host 51 in the buffer 241 (S241). The small size data sector control unit 233 controls the actuator 16 to seek to the target track including the data sector at the head address of the address specified by the write command (S242). In the example of FIG. 9, the rewrite area exists on one data track.

  The small size data sector control unit 233 controls the RW channel 21 to read the head data sector and the tail data sector whose address boundaries do not match, and store them in the buffer 241 (S243). In the example of FIG. 9, the boundaries do not coincide in both data sectors, and the data sector 111a and the data sector 111f are read.

  The small size data sector control unit 233 moves the head slider 12 to the backup area 115 (seeking process) (S244). After the seek is completed, the small size data sector control unit 233 writes the backup user data and the backup management data in the backup area 115 (S245). In the example of FIG. 9, data is written to the start sector 111g to the end sector 111j. Data to be written is the same as described with reference to FIG. Further, as described above, in a preferred configuration, the small size data sector control unit 233 starts writing backup data from the first data sector after completion of seek.

  The small size data sector control unit 233 stores the address data indicating the area (sector 111g to sector 111j) in which the backup data is stored in the RAM 24, and temporarily holds it (S246). Further, the small size data sector control unit 233 moves the head slider 12 to the target track on which data is written (seek processing) (S247).

  The small size data sector control unit 233 creates a data string to be written to the magnetic disk 11 from the write data acquired from the host 51, the head data sector 111a, and the tail data sector 111f (S248). The small size / data / sector controller 233 writes the created data string to the magnetic disk 11 (S249). In the example of FIG. 9, data is written to the data sectors 111a to 111f.

  Thereafter, the small size data sector control unit 233 returns to the backup area 115 (seek processing) (S250), and rewrites the backup start flag and the backup end flag in the backup management data (S251). At this time, the small size data sector control unit 233 refers to the address temporarily held in the RAM 24 and specifies the address of the backup data to be rewritten.

  In FIG. 9, the backup management data is stored in two data sectors 111g and 111j. Data sectors 111h and 111i for storing user data to be backed up are sandwiched between these data sectors. The data sector 111h stores data of the data sector 111a before rewriting, and the data sector 111i stores data of the data sector 111f before rewriting.

  The status flag stored in the first data sector 111g is a backup start flag. The status flag stored in the last data sector 111j is a backup end flag. In this example, the two status flags are 1-bit data, one of 0 and 1 indicates execution and the other indicates completion.

  In writing the backup user data (S245), the small size data sector control unit 233 writes a backup start flag indicating “execution” and a backup end flag indicating “execution”. In the backup management data rewriting step (S251) after rewriting user data, the small size data sector control unit 233 rewrites each of the backup start flag and the backup end flag from “execute” to “complete”.

  It is important to write the backup start flag immediately before writing for user data backup (S245), and it is important to write the backup end flag immediately after writing. For this reason, as shown in FIG. 9, there are data sectors for storing backup user data between data sectors for storing backup management data, and these data sectors (four data sectors). ) Is preferably continuous. As described above, “continuous” means that usable data sectors are continuous and does not count defective sectors.

  FIG. 10 shows the state stage transition in the write process accompanied by the backup process. The power flag is a flag indicating whether the power of the HDD 1 is normally turned off. The power flag is prepared separately from the backup management data. In the example of FIG. 10, “FAIL” means that a power shutdown occurred during the process.

  In the example of FIG. 10, the state of the write process accompanied by the backup process is composed of 13 stages 01 to 13. The state 00 is a normal state, and is a state when the HDD 1 is normally terminated without power-off. The table in FIG. 10 is a table for explaining data recovery processing for power shutdown, and the power flag of each stage in the write processing indicates “FAIL”.

  FIG. 10 illustrates four consecutive data sectors in the backup area 115. It is not determined whether the data stored in each data sector in the backup area 115 is user data or backup management data. In the following, as an example, the data change of four specific data sectors will be described, and the data types stored in each data sector are assumed to be the same before and after rewriting.

  For example, the specific four data sectors are the data sectors 111g to 111j, and before the specific backup processing, the data sectors 111g and 111j store backup management data, respectively, and the data sectors 111h and 111i are the heads. Stores user data sector and end user data sector. In the new backup process, new backup management data is stored in the data sectors 111g and 111j, and new backup user data is stored in the data sectors 111h and 111i.

  As described with reference to FIGS. 8 and 9, the write process accompanied by the backup process starts from the state stage of stage 01 and sequentially changes to state stages 01 to 13. Thereafter, when the normal operation of the HDD 1 ends, the state changes from the stage 13 to the stage 00.

  The HDC / MPU 23 uses the backup user data to recover data when the power is cut off during the write process including the backup process. In order to perform the recovery accurately, it is necessary to know in which state stage of the write process accompanied by the backup process the process is interrupted. FIG. 10 is a table showing the state stages of 00-13. The stage number is given for convenience, and the HDC / MPU 23 does not manage the state stage by this number.

  The difference between the stage 00 and other stages is the state indicated by the power-off flag. As described above, the power-off flag is data indicating whether the HDD 1 has ended normal operation or abnormal operation has ended due to power-off. Since stage 00 is in a normal state, the power-off flag indicates the end of normal operation. In another state stage, the power-off flag indicates abnormal termination.

  Before describing each of the state stages shown in FIG. 10, the flow of processing from the start of the HDD 1 to the power-off will be described with reference to the flowchart of FIG. When the HDD 1 is activated, the HDC / MPU 23 starts an activation process (S31). In the startup process, after normal processing such as operation confirmation and parameter setting is completed, the previous power-off state is specified with reference to the power-off state flag (S32). The power-off state flag is stored in a predetermined area on the magnetic disk 11.

  When the power-off state flag indicates normal power-off (SAFE in S32), the HDC / MPU 23 initializes the backup area 115 (S33). In the initialization, prescribed data is written (overwritten) in the backup area 115. Further, the HDC / MPU 23 sets the power-off state flag to FAIL (S34), and ends the activation process (S35).

  After startup, the HDC / MPU 23 (HDD 1) executes normal processing (S36). When the power is turned off after that, if it is a normal power-off (Y in S37), the HDC / MPU 23 sets the power-off state flag to SAFE (S39) and performs a normal power-off process (S40). The HDD 1 stops. If the power is turned off abnormally (N in S37), the HDD 1 stops as it is (abnormal power off: S38). The power-off state flag remains FAIL.

  In the startup process, when the power-off state flag indicates abnormal power-off (FAIL in S32), the HDC / MPU 23 searches the backup area 115 for data corresponding to the last backup process (S41). Specifically, the HDC / MPU 23 refers to the count value of the data sector that stores the backup management data, and specifies the data sector of the count value indicating the last backup process. In the configuration in which the counter is incremented for each backup process, the data sector with the maximum count value is the data sector for the last backup process, and in the configuration for decrementing, the data sector with the minimum count value is the last backup process. Data sectors.

  The HDC / MPU 23 uses the backup user data and the status flag to specify the status stage when the power is turned off (S42). The HDC / MPU 23 further executes a recovery process according to the state stage (S43). A method for specifying the state stage and the recovery method when the power is turned off will be described later. After the recovery process, the HDC / MPU 23 executes the steps after step S33.

  As described above, the power-off state flag is set to FAIL at the time of startup, and is set to SAFE before the HDD 1 is normally stopped. For this reason, when an unexpected power-off during the operation of the HDD 1 occurs, the power-off state flag remains FAIL, and the HDC / MPU 23 can know the previous abnormal stop at the subsequent startup.

  As described with reference to the flowchart of FIG. 11, the HDC / MPU 23 performs a recovery process according to the state stage at the time of abnormal power-off. In the following, with reference to the table of FIG. 10 and FIG. 9, each state stage, a method for specifying the state stage, and the contents of the recovery process corresponding to each state stage will be described in detail.

  As shown in FIG. 9, when the HDC / MPU 23 receives a write command, the state becomes stage 01. In stage 01, the power-off state flag indicates abnormal termination. Since this is a process after startup, the power-off state flag indicates abnormal termination as described with reference to FIG. The same applies to the other state stages.

  Furthermore, since it is before writing to the data sectors 111a to 111f, the head data sector 111a and the tail data sector 111f are data before rewriting. The backup start flag and the end flag respectively indicate completion (completion of the past (previous or previous) backup processing). The count value is m, indicating the number of past backup processing.

  Next, as shown in FIG. 9, the state changes from stage 01 to stage 02 by the start of writing of the backup management data (start sector 111g). Since the writing of the start sector 111g has not been completed, the backup start flag is unreadable as shown in FIG. The rest is the same as stage 01. When the writing of the start sector 111g is completed, the state shifts from stage 02 to stage 03.

  In stage 03, the backup start flag indicates that it is being executed, and the count value is n indicating the current backup process. Others are the same as stage 02. As shown in FIG. 9, in stage 03, the user data in the data sectors 111a and 111f are written into the data sectors 111h and 111i.

  When the backup of the user data is completed and the writing of the end sector 111j is started, the state shifts from the stage 03 to the stage 04. Since the writing of the end sector 111j is not completed, the backup end flag and the count value are unreadable. The rest is the same as stage 03. When the writing of the end sector 111j is completed, the state shifts from stage 04 to stage 05. In stage 05, the backup end flag indicates that execution is in progress, and the count value indicates n indicating the current backup processing. Others are the same as stage 04.

  Next, data rewriting (writing of new data) in the first data sector 111a in the writing area starts, and the state shifts from stage 05 to stage 06. Since rewriting of the top data sector 111a has not been completed, the top data sector 111a cannot be read in stage 06. The rest is the same as stage 05. The state shifts from stage 06 to stage 07 upon completion of rewriting of the head data sector 111a. In stage 07, the data in the head data sector 111a is the rewritten new data. The rest is the same as Stage 06. In stage 07, the data sectors 111b to 111e in the write area are rewritten.

  Data rewriting in the tail data sector 111f starts, and the state shifts from stage 07 to stage 08. Since it is before completion of rewriting of the tail data sector 111f, the tail data sector 111f cannot be read in stage 08. The rest is the same as Stage 07. When the rewriting of the tail data sector 111f is completed, the state shifts from the stage 08 to the stage 09. In stage 09, the data in the tail data sector 111f is new data that has been rewritten. The rest is the same as stage 08.

  Next, in the start sector 111g of the backup area 115, rewriting of the backup start flag starts, and the state shifts from stage 09 to stage 10. This is a rewrite of the backup start flag from execution to completion. Other data does not change. The HDC / MPU 23 stores backup management data in the current backup process in the RAM 24. The HDC / MPU 23 refers to them and generates new data for the start sector 111g. In the new data, only the backup start flag has changed from the original data.

  Since rewriting of the start sector 111g has not been completed, the backup start flag is unreadable. The rest is the same as stage 09. When the rewriting of the start sector 111g is completed, the state shifts from the stage 10 to the stage 11. In stage 11, the backup start flag indicates completion, and the others are the same as in stage 10.

  When rewriting of the end sector 111j is started, the state shifts from the stage 11 to the stage 12. Since the rewriting of the end sector 111j has not been completed, the backup end flag is unreadable. The rest is the same as stage 11. When the rewriting of the end sector 111j is completed, the state shifts from the stage 12 to the stage 13. In stage 13, the backup end flag indicates completion, and the others are the same as in stage 12.

  The HDC / MPU 23 performs a recovery process in the startup process after the abnormal power supply is cut off (step S43 in FIG. 11). The recovery process is controlled and executed by the HDC / MPU 23, and the HDC / MPU 23 functions as a recovery processing unit. In this process, the HDC / MPU 23 needs to specify in which state stage the power is cut off. The HDC / MPU 23 uses the status flag of the backup area, the backup user data, and the user data of the data sector to be backed up to specify the status stage at the time of power-off.

  As described above, by using the backup user data (data sectors 111h and 111i) and the backup source user data (data sectors 111a and 111f) for specifying the state stage, the necessary management data amount can be reduced. Can be reduced. The previous power-off state flag indicates whether or not there is an abnormal power-off, and is different from the one indicating the state of the writing process in the abnormal power-off.

  As described with reference to FIGS. 9 and 10, the state stage of the write process including the backup process can be specified by four points. The data of the first data sector 111a to be backed up, the data of the last data sector 111f, the backup start flag, and the backup end flag. The HDC / MPU 23 specifies the state stage at the time of abnormal power-off based on these four points, and performs recovery processing according to the stage.

  Specifically, when the power flag indicates FAIL, the HDC / MPU 23 searches the backup management data in the backup area 115 for the backup data for the final backup process. The backup management data has a count value. The HDC / MPU 23 refers to the count value in the backup area 115 to identify the backup data (management data and user data) of the last backup process.

  Further, the HDC / MPU 23 uses the backup management data, the backup user data, and the backup source user data of the backup data to specify the state stage at the time of power shutdown. The backup management data has a backup source address (a backup start address and a backup end address in FIG. 7). The HDC / MPU 23 can read backup source user data from the magnetic disk 11 by referring to these data. When the HDC / MPU 23 cannot accurately read one of the data sectors 111a and 111f, the state stage when the power is turned off is the stage 06 or the stage 08.

  Comparing the data sectors 111a and 111f in the rewrite area with the data sectors 111h and 111i in the backup area, the HDC / MPU 23 determines whether the data sectors 111a and 111f are new data or old data before rewriting. It can be determined whether there is. If the data in the rewrite area matches the data in the backup area, the data in the rewrite area is old data, and if not, it is new data. When one of the data sectors 111h and 111i in the backup area cannot be read accurately, the state is stage 03, and the data in the data sector in the rewrite area is old data.

  The HDC / MPU 23 can read the backup start flag and the backup end flag from the data sectors 111g and 111j in the backup area. Even when one of the data sectors cannot be read accurately, as shown in FIG. 10, it is a condition for specifying the state stage.

  When the HDC / MPU 23 has not started rewriting of both the data sectors 111a and 111f in the rewrite area and is old data (state stage 01 to state stage 05), the HDC / MPU 23 stores the backup management data as The data has the same contents as stage 01. There is no processing for user data in the rewrite area and the backup area. Data update in the rewrite area is resumed from the write command from the host 51.

  When one of the data sectors 111a and 111f in the rewrite area is new data or unreadable, and the backup start flag and the backup end flag indicate execution (state stage 06 to state stage 08), the HDC / MPU 23 By writing the backup user data in the backup area back to the rewrite area data sectors 111a and 111f, the data in the data sectors 111a and 111f are made old data before update. Further, the HDC / MPU 23 sets the backup management data to data having the same contents as the stage 01. Data update in the rewrite area is resumed from the write command from the host 51.

  When both data sectors 111a and 111f in the rewrite area are rewritten and new data is stored (state stage 09 to state stage 13), the HDC / MPU 23 stores the backup management data in the state stage 13 backup management. Make the data the same as the data. The data update in the rewrite area has been completed.

  As described above, the HDC / MPU 23 stores the counter value in the backup management data. If the counter overflows, the last backup process cannot be accurately identified. Therefore, it is preferable that the HDC / MPU 23 initializes the counter and the backup area before the counter overflow. Typically, when the counter value reaches a specified value, the HDC / MPU 23 initializes the counter and the backup area. In the counter initialization process, the counter value is set to an initial value. In the initialization process of the backup area, prescribed data is written in the backup area.

  As described above, the small size data sector control unit 233 writes new data in the data rewrite area after user data backup in writing user data whose boundary value does not match the data sector. Further, the backup management data is updated. As shown in FIG. 9, the head slider 12 moves from the rewrite area 113 to the backup area 115 and then returns to the rewrite area 113.

  In the process of backing up both the head data sector and the tail data sector, the small size data sector control unit 233 always reads the head data sector first and then the tail data sector for backup. May be read out. The flow of this process is as shown in FIG.

  In a preferred configuration, when the processing time (from reading data sector for user data backup to updating backup management data) is shortened, the small size data sector control unit 233 Read the last data sector before. When the specific condition is satisfied, the small size data sector control unit 233 reads the tail data sector after reading the tail data sector and writes them in the backup area.

  When the rotation angle (time) from the end data sector to the start data sector is short, the small size data sector control unit 233 reads the end data sector first for backup, thereby shortening the processing time. There are cases that can be done. An example will be described with reference to FIGS. 12A and 12B. 12A and 12B, ES is the end data sector, and SS is the start data sector. FIG. 12A shows a flow of a process of reading the tail data sector ES after the head data sector SS of the write data track, writing to the backup area, and returning to the write data track.

  Specifically, the head slider 12 reads the head data sector SS after passing the tail data sector ES in the following, and then continues the following to read the tail data sector ES. The head slider 12 moves to a track in the backup area and writes backup data. Thereafter, the head slider 12 returns to the user data track.

  In FIG. 12A, since the interval from the end data sector ES to the head data sector SS is short, the head slider 12 changes the head data sector SS (radius position) when the head slider 12 returns to the writing track. It has passed. The small size data sector control unit 233 waits for a time until the head slider 12 reaches the head data sector SS in order to write new user data.

  FIG. 12B shows the flow of processing for reading the tail data sector first. After the seek is completed, the head slider 12 arrives at the end data sector ES before the start data sector SS. The head slider 12 reads the data sectors in the order of the end data sector ES and the head data sector SS and writes them in the backup area. Thereafter, the head slider 12 returns to the user data track. Similar to the example of FIG. 12A, the small size data sector control unit 233 waits for a time until the head slider 12 reaches the head data sector SS for writing new user data.

  Compared with the process of FIG. 12B, the process of FIG. 12A requires a lot of extra processing time in reading the head data sector SS and the tail data sector ES. The time from the completion of rewriting user data to the completion of updating backup management data is shorter in the process of FIG. 12A. However, the reduction in the backup user data read processing time in FIG. 12B exceeds the increase in the backup management data update processing time.

  In this way, if it is not possible to start writing to the first data sector after reading the last data sector through user data backup in the same cycle, if the last data sector is read after completion of seek, If the data can be read before the sector, the processing time can be shortened by reading the tail data sector first (FIG. 12B).

  On the other hand, in the same cycle, reading of the first data sector and subsequent reading of the last data sector and writing of backup data to the backup area are completed, and writing to the first data sector is started in the next cycle. If it is possible (see FIG. 9), the small size data sector control unit 233 reads the head data sector before the tail data sector. If it is possible to return to the data track in the write area in the same round as the reading of the first data sector, new user data can be immediately written from the first data sector.

  Whether the backup user data can be read, the backup, and the start of writing of the new user data can be performed in the same cycle depends on the seek time between the user data track and the backup area. In a preferred configuration, as a condition for reading the tail data sector first, the distance between the tail data sector and the head data sector is registered corresponding to each data track. This conditional distance can be different for each data track, or a different value can be used for each zone. The condition distance may be common to all tracks on the recording surface.

  When the HDC / MPU 23 has a command queuing function, the small size data sector control unit 233 may collectively perform user data backup processing corresponding to a plurality of commands. For example, if the data sector to be saved can be accessed on the same track or the same round of the disk, and the seek to the backup area and the seek from the backup area to the rewrite area are possible in the same round, The data sector control unit 233 processes a plurality of commands together.

  The small size data sector control unit 233 may write backup user data corresponding to each command to successive data sectors, or may write them separately. When dividing backup user data, backup management data is recorded for each data. When recording is performed in continuous data sectors, only the data sectors before and after the data sectors store backup management data. The small size / data / sector control unit 233 stores data indicating that a plurality of commands are targeted in the backup management data so that data corresponding to the combined commands can be recovered normally.

  As mentioned above, although this invention was demonstrated taking preferable embodiment as an example, this invention is not limited to said embodiment. A person skilled in the art can easily change, add, and convert each element of the above-described embodiment within the scope of the present invention. The present invention may be applied to disk drives other than HDDs. The moving mechanism of the head slider is not limited to the rotary actuator described above, and other moving mechanisms such as a moving mechanism that moves the head slider linearly can be used. The present invention can be applied to a disk drive having a head different from the flying head slider. For example, the head slider may always be in contact with the disk.

  As described above, backup data includes management data in addition to user data, in a form that minimizes the waiting time for rotation in any data sector of the backup area provided by one or more tracks. It is preferable to be backed up. However, the present invention can also be applied to processing that does not include backup data management data. For example, the backup source address can be specified by storing a write command on the magnetic disk.

  If only one of the beginning and end of the user data does not match the boundary of the large data sector, it is preferable to back up only the data sector that includes the one, but the data sector that includes the other that matches the boundary You may also back up. The HDD preferably uses the write start flag and the write end flag to determine the shut-off state, but the status may be specified using other data.

DESCRIPTION OF SYMBOLS 1 Hard disk drive, 10 Enclosure, 11 Magnetic disk 12 Head slider, 13 Arm electronics 14 Spindle motor, 15 Voice coil motor 16 Actuator, 20 Circuit board, 21 Read / write channel 22 Motor driver unit , 23 Hard disk controller / MPU
24 RAM, 231 Data size detection unit 232 Large size data sector control unit 233 Small size data sector control unit 241 Buffer

Claims (17)

A disk drive having a host interface that supports both large and small data sectors, and accessing the disk with the large data sectors,
A head to access the disk;
A moving mechanism for moving the head on the disk;
A controller for controlling the head and the moving mechanism,
The controller is
Receive a write command, determine the data sector size of the write command,
When the data sector size of the received write command is the small size data sector, each of the beginning and end of the rewrite area specified by the write command is a large size on the disk. Determine if it matches the data sector boundary,
If one of the head and tail does not match the boundary of the large data sector, the backup data including the large data sector including the one is included, and if both do not match, both of the data are included. Storing backup data including large size data sectors in a backup area consisting of one or more data tracks on the disk;
Having a disk drive. The controller starts writing backup data from the latest data sector where writing can be started after completion of seek to the backup area.
The disk drive of claim 1. The controller starts to write the backup data from a preset specified data sector after completion of seeking to the backup area.
The disk drive of claim 1. The backup data includes data indicating the latest backup data,
The disk drive of claim 1. The data indicating the latest backup data is a counter value of a counter.
The disk drive according to claim 4. The controller initializes the backup area when the counter reaches a predetermined value.
The disk drive of claim 5. The backup data includes a backup source address,
The disk drive according to claim 4. The controller stores the backup data in a continuous area in the backup area;
The disk drive of claim 1. The backup data includes a flag stored in a data sector before and after the user data,
The controller updates the flag after writing new data to the write area,
The controller uses the flag, the user data of the backup source, and the data in the rewrite area to execute a recovery process corresponding to a power shutdown in data writing.
The disk drive of claim 1. The controller temporarily holds address information of a data sector storing the backup data,
In updating the flag, the address of the flag is specified with reference to the address information.
The disk drive of claim 9. The controller is
Both the head and the tail do not coincide with the boundary of the large data sector, the tail data sector including the tail can be read before the head data sector including the head, and the tail data When the time from the sector to the first data sector is smaller than the threshold value,
Read the head data sector after the tail data sector for storage in the backup area;
The disk drive of claim 9. A method of backing up user data in a disk drive having a host interface that supports both large data sectors and small data sectors and accessing the disks with the large data sectors. There,
Receive a write command,
Determine the data sector size of the write command,
When the data sector size of the received write command is the small size data sector, each of the beginning and end of the rewrite area specified by the write command is a large size on the disk. Determine if it matches the data sector boundary,
If one of the head and tail does not match the boundary of the large data sector, the backup data including the large data sector including the one is included, and if both do not match, both of the data are included. Storing backup data including large size data sectors in a backup area consisting of one or more data tracks on the disk;
Method. After completion of seeking to the backup area, writing of backup data is started from the latest data sector where writing can be started.
The method of claim 12. The backup data is started to be written from a preset specified data sector after completion of seek to the backup area.
The method of claim 12. The backup data includes data indicating the latest backup data,
The method of claim 12. The backup data is stored in a continuous area in the backup area.
The method of claim 12. The backup data includes a flag stored in a data sector before and after the user data,
The flag is updated after writing new data to the write area,
Recovery processing corresponding to power shutdown in data writing is performed using the flag, the user data of the backup source, and the data of the rewrite area.
The method of claim 12.

Images