JPH09320209A - Disk storage device and its accessing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To prevent the reduction in data accessing speed by shortening a disk rotation standby time and thus shortening a time needed for access processing in a system, for executing retrial processing according to the occur rence of an accessing error in access processing. SOLUTION: This disk storage device is a hard disk device(HDD) having a function to execute retrial processing, if accessing errors occur when accessing processing such as sequential reading or writing of a series of sectors is performed in a target track N on a rotating disk 1. A disk controller (HDC) 11 saves information regarding a sector in which an accessing error occurs in an internal register 11a and continues normal access processing from a next sector. After the end of access processing, by using the information saved in the internal register 11a, retrial processing is performed for the error sector.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a disk storage device having a retry function, which is applied to a magnetic disk device, for example, and performs an access process again when an access error occurs.

[0002]

2. Description of the Related Art Conventionally, in a magnetic disk device such as a hard disk device (HDD), a read access (data read operation) or a write access (data write operation) is performed in units of sectors formed on a disk which is a recording medium. Be done. The sector is a recording area when a large number of concentric tracks formed on the disk are divided.

The HDD is provided with a function of executing so-called retry processing when an access error (read error or write error) occurs during read access or write access (collectively referred to as access processing below). There is. This retry process is a process of re-executing the access process for the sector in which an error has occurred (which is an error sector). Normally, if you can not access normally even if you repeat it a certain number of times,
The disk controller (HDC) notifies the host computer that an access error has occurred and access processing cannot be performed.

Normally, when the error sector is a defective sector such as a scratch that cannot be recovered, even if the retry process is repeated a certain number of times, normal access cannot be performed. Further, in the case of a temporary access error caused by the presence of dust or the like at the position of the error sector, there is a high probability that normal access will be possible by the retry processing.

Here, the HDC has a function called a sequencer for sequentially and automatically executing access processing.
When the address (sector number) of the start sector to be accessed is set, the access process is automatically executed up to the last sector.

As a concrete example, as shown in FIG. 5, assume that the sectors 1 to 5 of the target track N on the disk are sequentially accessed. When the HDC detects an access error in the sector 3 when sequentially accessing the sector 3 starting from the sector 1, as described above,
Move to retry processing. That is, the access processing for the error sector 3 is performed again.

In the retry processing, since the access processing is repeated a preset number of times, a waiting time for one rotation of the disk is generated each time the retry processing is executed. In the specific example shown in FIG. 5, for convenience, it is assumed that the retry process for the error sector 3 is successful at the second rotation of the disk and the process proceeds to the access process for the remaining sectors 4 and 5 to be accessed. There is.

[0008]

As described above, in the conventional HDD, when an access error occurs during the access process, the retry process is repeated for the error sector. After executing this retry processing, access processing to the sectors after the error sector is executed. In the retry process, a waiting time for one rotation of the disk is generated every time the access process is performed.
For this reason, in the access processing for a series of sectors, if the retry processing is executed midway, the processing time corresponding to the disk rotation waiting time increases, and as a result, the time required for the access processing increases. In particular, if the number of times of retry processing increases, the access processing will be significantly slowed down, and the data processing speed of the entire system will be lowered.

It is an object of the present invention to shorten the disk rotation waiting time in the method of executing the retry processing according to the occurrence of an access error during the access processing, and consequently shorten the time required for the access processing to access the data. It is to prevent the speed from decreasing.

[0010]

DISCLOSURE OF THE INVENTION The present invention relates to a disk storage for performing a read access or a write access access processing on a rotating disk in a predetermined order of recording areas such as sectors as a predetermined access unit. The device is applied to the device and has a function of executing a retry process when an access error occurs. The present apparatus stores information for identifying a recording area (error sector) in which an access error has occurred when an access error occurs, and continues normal access processing from the next sector. After the access process is completed, the retry process for the error sector is executed.

With such a configuration, when an access error occurs during the access process, the normal access process is continued from the next sector after the error process such as saving the information specifying the error sector. That is, the access process other than the error sector is completed before the retry process is executed. Therefore, the access process can be completed without requiring the disk rotation waiting time associated with the retry process. Since the retry process is executed after the completion of this access process, the disk rotation waiting time can be minimized. In other words, the error sector is temporarily put on hold, and after the normal access process is completed, the error sector is subjected to the access process as the last sector to be accessed.

[0012]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing a system configuration related to this embodiment, FIG. 2 is a timing chart for explaining the operation of this embodiment, and FIG. 3 shows error sector information related to this embodiment. FIG. 4 is a conceptual diagram for explaining, and FIG. 4 is a flowchart for explaining the operation of the present embodiment. (System Configuration) In this embodiment, as shown in FIG. 1, a hard disk drive (HDD) using a disk 1 which is rotated at a high speed by a spindle motor 2 as a recording medium.
Is assumed. In the HDD, the head 4 moves in the radial direction of the disk 1 and is positioned on a target track (a track including a sector to be accessed) on the disk 1,
A data read operation (read access) or a data write operation (write access) is executed.

The head 4 is held by a head actuator 5 and is sought in the radial direction of the disk 1 by the driving force of a voice coil motor (VCM) 6. VCM
6 is driven by the VCM driver 7. The spindle motor 2 is driven by a spindle motor driver (SPM driver) 3. VCM driver 7 and SPM
The driver 3 is controlled by the CPU 8.

The head amplifier 9 amplifies the read signal read from the head 4 at the time of reading data, and performs read / write.
Output to the write (R / W) signal processing circuit 10. At the time of writing data, the write signal output from the R / W signal processing circuit 10 is amplified and output to the head 4. R / W
The signal processing circuit 10 is a read / write channel that performs signal processing of a read / write signal, and has a function of converting a read signal from the head 4 into reproduced data and converting write data into a write signal.

Further, the R / W signal processing circuit 10 reproduces the servo information (positional information) read by the head 1 and outputs it to the CPU 8. The servo information includes track address during seek control (movement control to the target position) and position error information during position control (track following control). The CPU 8 controls the positioning of the head 4 on the target track on the disk 1 based on the servo information. CPU
8 uses the memory 14 consisting of ROM and RAM,
It is a main control device that controls the drive of the DD.

The disk controller (HDC) 11 is
Data transfer control between the HDD and the host computer and data recording / reproduction control are performed via the host interface 12. The buffer memory 13 is access-controlled by the HDC 11 and has a sector buffer function for temporarily storing write data to be written to the disk 1 and reproduced data reproduced from the disk 1, and a disk memory 13.
It is a buffer RAM having a function of holding defect information (defect information) regarding the above defect (defective sector).

Further, the HDC 11 of this embodiment has a function of detecting an access error at the time of access processing and a function of identifying an error sector in which the access error has occurred and performing a retry processing. HDC11 is shown in FIG.
As shown in, the address information (sector number SN) for specifying the error sector, the buffer address BA (storage area secured in the buffer memory 13) for storing the data of this error sector, and the retry process are executed. The retry flag information RF for controlling is set in the internal register 11a. The retry flag information RF is
“1” is a flag that indicates that the retry process is necessary, and “0” means a flag that indicates that the retry process is unnecessary. (Access Processing of This Embodiment) The access processing of this embodiment will be described below with reference to FIGS. 2 and 3.

As shown in FIG. 2, the HDC 11 of the present embodiment sets a target track N by an access command (read command or write command) given from a host computer, and accesses a sector number (here, 1). The sequencer function is to automatically access each sector sequentially.

Specifically, as shown in the flow chart of FIG. 4, the CPU 8 seeks and positions the head 4 to the target track N on the disk 1 in response to an access command from the host computer (steps S1 and S).
2). The HDC 11 starts access processing (read access or write access) from the designated sector (sector 1) in the target track N by the sequencer function (step S3).

Here, as shown in FIG. 2, it is assumed that an access error (read error or write error) has occurred in the next sector 3 after accessing sector 2. When detecting the access error, the HDC 11 sets the address information (sector number SN) of the error sector (sector 3) and the buffer address BA of the buffer memory 13 for storing the data of the error sector in the internal register 11a ( YES in step S4, S
5). Further, retry flag information RF (“1”) for instructing the retry process for the error sector is set in the internal register 11a.

After executing such error processing (processing for storing information of error sectors), the HDC 11 checks whether or not there is a sector to be accessed in the current target track N (step S10). ). That is,
It is checked whether the error sector 3 is the last sector in the track N (the last sector to be accessed).

Here, the error sector 3 is the last sector,
That is, if there is no sector after that sector, H
DC11 is the retry flag information R of the internal register 11a.
F is cleared (“0” is set), and retry processing for the error sector is executed (NO in step S10, S
7, S8). That is, the sector number 3 of the internal register 11a
And the buffer address BA are set, and the access processing for the error sector is executed. For read access,
The data read operation is executed by the head 4 for the error sector 3, and the reproduced data obtained by the R / W signal processing circuit 10 is stored in the buffer address BA of the buffer memory 13. The data is transferred from the buffer memory 13 to the host computer via the host interface 12. In the case of a write access, after the write data from the host computer is stored in the buffer address BA of the buffer memory 13, the write signal is output to the head 4 via the R / W signal processing circuit 10 to the error sector 3. Write the data.

When such an error sector is the last sector, the retry process is repeated a certain number of times until it is normally accessed, so that a disk rotation waiting time occurs as in the conventional case, but the same. The probability that the last sector in a track is always the error sector is extremely low.

On the other hand, when there are sectors (4, 5) to be accessed after the error sector, the HDC 11
The retry process for the error sector 3 is suspended, and the access process for the sector (4,5) to be accessed is continued (YES in step S10, S11). At this time, the HDC 11 secures a storage area (saved buffer address BA) in which the data of the error sector 3 of the buffer memory 13 should be stored, and stores the data obtained by the access processing in the other storage areas. That is, HDC1
In the case of read access, 1 sequentially stores the respective reproduction data of the sectors 1, 2, 4, 5 other than the error sector 3 in the buffer memory 13 and transfers them to the host computer via the host interface 12. Therefore, the data transfer to the error sector 3 is temporarily excavated. The same applies to the case of the write access. Write data is written in the sectors 1, 2, 4, 5 other than the error sector 3, and the data writing to the error sector 3 is suspended.

In the target track N, if the sector to be accessed is only the error sector 3, that is, the error sector 3
When the access processing other than the above is completed, the HDC 11 reads the stored information (SN, BA, RF) regarding the error sector 3 from the internal register 11a. The HDC 11 sets the sector number S of the error sector 3 as the target of the next access process.
N and buffer address BA are set, and error sector 3
Access processing, that is, retry processing is executed (steps S7 and S8). At this time, the HDC 11 clears the retry flag information RF. That is, when the retry flag information RF is cleared to "0", it means that the retry process is completed or unnecessary.

As described above, according to the present embodiment, as shown in FIG. 2, when an access error occurs during the access process, information (SN,
(BA, RF) is saved, and the access processing to the sectors after the error sector is continued. That is, the retry process for the error sector is suspended, and the access process for all sectors other than the error sector in the target track N is continuously executed. Therefore, the disk rotation waiting time does not occur until the access processing for the last sector to be accessed is completed.

After the access processing is completed, the retry processing for the error sector is executed, so that a disk rotation waiting time occurs as shown in FIG. Therefore, as compared with the conventional method, since the access processing to the sectors other than the error sector is executed during one rotation of the disk, the disk rotation until the access processing (including the retry processing) of all the sectors is eventually ended. It is possible to reduce the waiting time (see FIG. 5).

In other words, according to the present embodiment, when an error sector occurs in the case where a series of sectors are sequentially accessed, the access sector is temporarily suspended as the last sector of the access process. To do. Then, when the processing of the last sector (here, sector number 5) of the normal access processing is completed, access processing (that is, retry processing) for the error sector is executed. In order to put this retry process on hold and execute it at the end,
Information on the error sector (S
N, BA, RF).

In this embodiment, the case where the HDC 11 saves the information (SN, BA, RF) regarding the error sector in the internal register 11a has been described, but the present invention is not limited to this, and the CPU 8 secures the register in the memory 14. You may save it. The method in which the CPU 8 saves is applied in the case of the non-automatic mode of the access processing by the sequencer function of the HDC 11 instead of the automatic mode, as described above. (Effects of this Embodiment) Based on the following conditions, this embodiment, when a plurality of sectors are accessed by the head 4 in the same track N, is compared with the conventional method.
It is possible to shorten the disk rotation waiting time by the head transit time corresponding to the number of A sectors.

As the first condition setting, in the non-minimum rotation waiting time mode (No Minimum latency)
Assume Here, in all sectors to be accessed,
The start sector number is S, the total number of sectors to be accessed is B,
When the error sector number (error sector position) is E and S> 0, "A = (S + B-1) -E"
The formula is established. However, (S + B-1)> E. That is, there is a difference in the disk rotation waiting time for the number of A sectors at the time of the end of access from the conventional case. If S = 0,
The expression "A = (S + B) -E" is established. However, (S +
B)> E.

As a second condition setting, a minimum rotation waiting time mode (minimum latency) is assumed. Here, when the access start head position is G and S> 0, “A = (S + B−1) − (G +
1) ”is established. However, G + 1 = (S + B-1)
It is. In the case of S = 0, “A = (S + B) −
The expression “(G + 1)” holds. However, (G + 1) = (S
+ B-1).

[0032]

As described in detail above, according to the present invention, in a disk storage device having a function of executing a retry process when an access error occurs, a retry process is performed when an access error occurs during the access process. By executing the process as a post-process, it is possible to shorten the disk rotation waiting time associated with the retry process. Therefore, as a result, the time required for the access processing can be shortened, and the reduction in the data access speed can be suppressed.

[Brief description of drawings]

FIG. 1 is a block diagram showing a system configuration of an HDD related to an embodiment of the present invention.

FIG. 2 is a timing chart for explaining the operation of this embodiment.

FIG. 3 is a conceptual diagram for explaining information of an error sector related to this embodiment.

FIG. 4 is a flowchart for explaining the operation of the embodiment.

FIG. 5 is a timing chart for explaining a conventional method.

[Explanation of symbols]

 1 ... Disk 2 ... Spindle motor 3 ... Spindle motor driver 4 ... Head 5 ... Head actuator 6 ... Voice coil motor (VCM) 7 ... VCM driver 8 ... CPU 9 ... Head amplifier 10 ... Read / write signal processing circuit 11 ... Disk controller (HDC) 11a ... Internal register 12 ... Host interface 13 ... Buffer memory 14 ... Memory (ROM, RAM)

Claims (7)

[Claims] 1. A disk storage device for performing a read access or a write access access processing on a rotating disk in a set order of recording areas of a predetermined access unit, wherein an access error during the access processing Occurrence detected,
A means for performing an error process for specifying the recording area in which the access error has occurred, and, after the error process is completed, the access process is continued when a recording area as a target of the access process exists, And a means for executing a retry process for the recording area in which the access error has occurred after the completion of the above step. 2. A disk storage device for performing a read access or a write access access processing on a rotating disk in a set order of recording areas of a predetermined access unit, wherein an access error during the access processing Occurrence detected,
Means for performing error processing for storing address information for specifying a recording area in which the access error has occurred and flag information for instructing retry processing, and a recording area for the access processing exists after the error processing is completed. Means for determining whether or not there is a recording area that is the target of the access processing, the access processing is continued, and the flag information and the address information are read out after the completion of the access processing, A disk storage device comprising means for executing a retry process for a recording area designated by the address information based on information. 3. A means for executing the retry process when the access process is completed after the error process is completed and the recording area which is the target of the access process does not exist. The disk storage device according to claim 1 or 2. 4. A disk storage device that performs read access or write access processing in sector units on a rotating disk in a set order, wherein an error sector in which an access error occurs during the access processing is executed. Means for detecting and storing the address information of this error sector and flag information for instructing retry processing; means for determining whether or not there is a sector to be accessed after the error sector; When the target sector exists, a means for continuing the access processing until the access processing for all the sectors to be accessed is completed, and after the access processing is completed, the flag information and the address information are read out to obtain the flag information. Retry on the error sector of the address information based on And a retry process for the error sector of the address information when a retry process is instructed by the flag information or when there is no sector to be accessed after the error sector. A disk storage device comprising means for executing. 5. A disk having a large number of concentric tracks, each track comprising a plurality of sectors, starting from a designated sector in the same track to the last sector to be accessed. A disk storage device that sequentially performs read access or write access access processing, and when a retry processing is instructed by the flag information, an error sector specified by the address information is accessed in the same track. 5. The disk storage device according to claim 4, wherein the retry process is executed as the last sector of the. 6. An access method applied to a disk storage device for executing a read access or a write access access process in a set order for a recording area of a predetermined access unit with respect to a rotating disk, the access method comprising: A step of starting the access process from the first recording area, a step of detecting the occurrence of an access error during the access process, and address information specifying the recording area in which the access error has occurred and flag information indicating a retry process. A step of saving, a step of determining whether or not there is a recording area to be accessed after the recording area in which the access error has occurred, and a final step when there is a recording area to be accessed according to the determination result, Access to the recording area is completed. And a step of reading the flag information and the address information after the access processing is completed, and executing a retry processing for a recording area designated by the address information based on the flag information. An access method characterized by that. 7. A disk is provided with a large number of concentric tracks, each track comprising a plurality of sectors, and read access is made to the rotating disk in sector order in a set order. Alternatively, the method is an access method applied to a disk storage device that executes an access process of write access, which comprises a step of starting the access process from a designated sector in a track to be accessed, and an error sector in which an access error occurs during the access process A step of detecting, a step of storing address information for specifying the error sector and flag information for instructing a retry process, a step of determining whether or not there is a sector to be accessed after the error sector, If there is a sector to be accessed according to the judgment result, the maximum A step of continuing the access process until the access process for the sector is completed, and after the access process is completed, the flag information and the address information are read and a retry process for the error sector of the address information is performed based on the flag information. A step of determining whether to execute, and a retry process for the error sector of the address information when the retry process is instructed by the flag information or when there is no sector to be accessed after the error sector An access method comprising the steps of: