JP2000293946A - Skip processing method of defective position in disk device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enable the skip process in a short time along with reducing the regions holding a list for the defect information. SOLUTION: The physical track number of the defective track is preliminarily registered in a defective cylinder list as the defective cylinder number so as to treat the cylinder, to which the defective track having the defective position belongs, as the defective cylinder, and when the access is requested from a host device, the physical cylinder number objective for the access is obtained from the request for access, then by locating the defective cylinder list while making the physical cylinder number as the object, the defective cylinder being registered in this defective cylinder list is skipped, then the process according to the request for access is carried out by using only the proper cylinder wherein the defective track is not included.

Description

DETAILED DESCRIPTION OF THE INVENTION

 (Table of Contents) Technical Field to which the Invention pertains Prior Art (FIGS. 11 to 13) Problems to be Solved by the Invention Means for Solving the Problems Embodiment of the Invention [1] Description of First Embodiment (FIG. 1 to FIG. 1) (FIGS. 6 and 14) [2] Description of Second Embodiment (FIGS. 7 to 10) [3] Other Effects of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a plurality of disk-shaped recording media, such as a magnetic disk, an optical disk, and a magneto-optical disk, for recording information on each disk-shaped recording medium and reproducing information from each disk-shaped recording medium. In particular, the present invention relates to a method for accessing a disk-shaped recording medium while skipping the defective area when a defective area occurs on a recording surface of the disk-shaped recording medium.

[0002]

2. Description of the Related Art Generally, in a hard disk drive having, for example, a plurality of magnetic disks as disk-shaped recording media, these magnetic disks are rotatably provided around the same axis and a plurality of magnetic heads are provided on a recording surface of the magnetic disk. It is provided opposite. Each magnetic head records data on a magnetic disk and reads data recorded on the magnetic disk.

In a state in which the plurality of magnetic disks are driven to rotate at a constant rotational speed by a spindle motor, the plurality of magnetic heads are driven by a positioner to move integrally in a radial direction or a substantially radial direction of the magnetic disks. Each magnetic head is positioned on a predetermined track and accesses (records / reads data on) the track on each recording surface of the magnetic disk.

[0004] On each recording surface of a magnetic disk, a large number (for example, 10,000) of tracks are concentrically arranged, and each track has a physical track number (for example, a physical track number) which designates the track corresponding to a physical position. Physical address). For example, if the number of tracks is m,
The physical track numbers are “0”, “1”, “2”,..., “M” in order from the outside (or inside) of the magnetic disk.
-1 ".

A set of tracks provided with the same physical track number on a plurality of magnetic disks is called a cylinder. Then, for example, the physical track number i (i
= 0 to m-1), a physical cylinder number i is assigned to a cylinder which is a set of tracks assigned. By the way, in a region accessible by the user on each recording surface of the magnetic disk, a track where data cannot be recorded / reproduced correctly due to a scratch or the like may be generated during manufacture or actual operation of the magnetic disk (hard disk device). Such a track is called a defective track.
When a higher-level device accesses a magnetic disk, it is necessary to avoid access to a defective track as described above and to specify only a normal track as an access target.

For this reason, conventionally, a defective track list is created in a recording format as shown in FIG. 11 for each magnetic head (each recording surface of a magnetic disk), and the defective track list is stored in a predetermined area (for example, a system) of the magnetic disk. Area). As a defective track list shown in FIG. 11, an area capable of storing a maximum of N defective track numbers (physical track numbers) for each magnetic head is continuously secured in the system area. For example, information on defective tracks on the recording surface of the magnetic disk accessed by the magnetic head H1 [defect track numbers DT1 (H1), DT2 (H
1),...] Are stored in the area specified by the addresses 2N to 4N-1. The defect track numbers DT1 (H1), DT2 (H1),... Stored in these areas are the physical track numbers assigned to the defective tracks, respectively, and are stored in ascending order from the head area specified by the address 2N. You. “FFh” is stored in the unused area of the defective track list.

The registration of a defective track number in the defective track list is performed in a self-test process when the hard disk drive is manufactured. When the hard disk device receives an access request from a higher-level device (host), the hard disk device performs a skip process (alternate process) described later with reference to the defective track list, thereby avoiding the defective track and only the normal track. To perform an operation according to the access request.

In other words, the control unit in the hard disk device obtains, from the access request, the number of the magnetic head to be sought and the number of the track to be sought by the magnetic head (physical track number). Based on the obtained magnetic head number, the head address of the defective track list corresponding to the magnetic head is determined, and the defective track numbers registered in the defective track list are sequentially searched from the head. At this time, the number of searched defective track numbers is counted. The search for the defective track number is performed until the searched number is larger than the sum of the number of the defective track numbers preceding that number and the physical track number to be sought. Then, when the condition is satisfied, it is determined that the track given the value of the sum as the physical track number is the track to be sought, and the seek operation by the magnetic head is performed on the track. The process of skipping a defective track and replacing the defective track with a normal track as described above is referred to as a defective track replacement process (or a track skip process).

The above-described skip processing of a defective track will be described with reference to specific numerical examples shown in FIGS. 12 (a) to 12 (c) and FIG. Here, FIG.
2A shows a physical track number assigned to a track on the recording surface of the magnetic disk to be sought by the magnetic head H1, for example, and FIG. 12B shows the position of the defective track (the position where "x" is written). ), And FIG.
It shows a number assigned to a normal track by skip processing of a defective track.

As shown in FIGS. 12A and 12B, on the recording surface of the magnetic disk to be sought by the magnetic head H1, the tracks provided with the physical track numbers 3, 5, 6, and 15 are defective. In the case of a track, a defective track list of the magnetic head H1 secured in the area specified by the addresses 2N to 4N-1 is created in the system area in a recording format as shown in FIG. That is,
Defective track numbers DT1 (H1), DT2 (H1), DT3 (H1), DT4 (H1) in the defective track list of the magnetic head H1 in FIG.
Are the defect track lists of the magnetic head H1 shown in FIG.

Then, by performing the above-described skip processing of the defective track based on the defective track list of the magnetic head H1 shown in FIG. 13, the normal tracks are respectively assigned the track numbers 0 as shown in FIG. ,
Can be used as tracks with 1, 2,... For example, 0 to 0
If 2 is specified, the seek operation is performed on the tracks assigned physical track numbers 0 to 2 as they are, but the track assigned physical track number 3 is a defective track, and therefore the seek target When the track number 3 is designated as a track number, a normal track to which the physical track number 4 is assigned is actually treated as the track number 3. Similarly, since the tracks to which the physical track numbers 5 and 6 are assigned are also defective tracks, when 4 to 11 are designated as track numbers to be sought, the physical track numbers 7 to 14 are respectively actually set.
The normal tracks assigned with are assigned as track numbers 4 to 11.

[0012]

However, in the above-described conventional defective track skipping method, it is necessary to create and store a defective track list for each magnetic head. If the number of magnetic heads is n, then N It is necessary to secure an area capable of storing × n defective track numbers, and many areas in the system area are used for the defective track list.

Further, when performing skip processing using a defective track list, a defective track corresponding to the magnetic head must be avoided based on the number of the magnetic head to be subjected to a seek operation. When performing (continuous) access, it is checked whether or not the track to be sought is a defective track for each magnetic head, and if it is a defective track, the avoidance process is executed, so that considerable processing time is required. .

Further, the number of defective track numbers must be counted and the search must be performed in order from the head data of the defective track list. In particular, when the track number to be sought is a large value, the physical It takes a lot of processing time to determine the track number.
The present invention has been made in view of such a problem, and by enabling skip processing to be performed in a unit of a cylinder instead of a unit of a track when a defective track occurs, an area for holding a list for a defect information is provided. It is an object of the present invention to provide a method for skipping defective parts in a disk device, which can greatly reduce the number of times and can perform skip processing in a short time.

[0015]

In order to achieve the above object, a defective portion skip processing method according to the present invention (claim 1)
Is a disk device that has a plurality of disk-shaped recording media rotatable about the same axis and records information on each disk-shaped recording medium and reproduces information from each disk-shaped recording medium. A method for accessing the disk-shaped recording medium while skipping the defective portion when the defective portion occurs on the recording surface of the recording medium, wherein the cylinder to which the defective track having the defective portion belongs belongs to the defective cylinder. In order to treat as, the physical track number of the defective track is registered in advance in the defective cylinder list as a defective cylinder number, and when an access request is received from a higher-level device, the physical cylinder number to be accessed is obtained from the access request, By searching the defective cylinder list for the physical cylinder number, the defective cylinder is searched. Skip defective cylinder registered in the list, and characterized by performing a process corresponding to the access request using only normal cylinder containing no defective track.

At this time, the defective cylinder numbers are registered in the defective cylinder list in ascending order, and the addresses of the defective cylinder numbers registered in the defective cylinder list are used as element numbers when the defective cylinder list is handled as an array. Using (claim 2), the number of defective cylinders existing from the leading cylinder to each defective cylinder may be obtained based on the element number (claim 3). Further, the total number of defective cylinder numbers registered in the defective cylinder list may be registered in the head area of the defective cylinder list.

Then, a binary search of the defective cylinder list for the physical cylinder number to be accessed is performed using the total number of the defective cylinder numbers and the element numbers, so that the physical cylinders to be accessed from the first cylinder are obtained. By obtaining the number of defective cylinders existing up to the numbered cylinder and further searching the defective cylinder list based on the number of defective cylinders obtained by the binary search, Among the normal cylinders except for, recognizes a normal cylinder existing at a position corresponding to the physical cylinder number of the access target, regards the normal cylinder as the cylinder to be accessed, and Processing according to the access request may be performed (claim 5).

On the other hand, if there is a continuous defective cylinder number in the defective cylinder list, the first defective cylinder number is registered as a representative defective cylinder number, and subsequent defective cylinder numbers are deleted. A value obtained by adding the number of defective cylinder numbers following the first defective cylinder number to the number of defective cylinders existing up to the cylinder to which the representative defective cylinder number has been assigned is referred to as the number of skip cylinders. By registering in correspondence with the element number of the cylinder number, the defective cylinder list is converted into a representative defective cylinder list, and the registered total number of the defective cylinder numbers in the representative defective cylinder list, the element number of the representative defective cylinder list, and Target the physical cylinder number of the access target using By performing a binary search of the representative defective cylinder list, from among the representative defective cylinder numbers registered in the representative defective cylinder list, obtaining the number of representative defective cylinder numbers that do not exceed the physical cylinder number of the access target, Further, by performing a search of the representative defective cylinder list based on the number of representative defective cylinder numbers obtained by the binary search and the number of skip cylinders, among the normal cylinders excluding the defective cylinder, A normal cylinder existing at a position corresponding to the physical cylinder number of the access target may be recognized, the normal cylinder may be regarded as the access target cylinder, and the processing corresponding to the access request may be performed on the normal cylinder. (Claim 6).

According to the above configuration, in the method of processing for skipping a defective portion in the disk device of the present invention (claims 1 to 6), the physical track number of the defective track is registered as a defective cylinder number in the defective cylinder list. When an access request is received from the server, the defective cylinder list is referred to, the defective cylinder is skipped, and processing according to the access request is executed using only normal cylinders. That is, the cylinder to which the defective track belongs is treated as a defective cylinder, and the skip processing can be performed in a unit of a cylinder instead of a unit of a track.

Further, the defective cylinder numbers are registered in the defective cylinder list in ascending order, the addresses of the respective defective cylinder numbers in the defective cylinder list are used as element numbers (array index) of the array, and the defective cylinder numbers are registered in the leading area of the defective cylinder list. By using the total number of defective cylinder numbers (that is, the total number of defective cylinders), it is possible to perform a binary search of the defective cylinder list for the physical cylinder number to be accessed, and to speed up the skip processing. (Claims 2, 4, 5).

At this time, the defective cylinder numbers are registered in the defective cylinder list in ascending order, and the number of defective cylinders can be obtained based on the aforementioned element numbers. Is unnecessary. In other words, the number of defective cylinders corresponding to the binary search result can be immediately obtained from the element number, which also enables the binary search of the defective cylinder list to be realized.

Furthermore, since the number of registered elements is reduced in the representative defective cylinder list in which the defective cylinder list is made compact, the binary search for this representative defective cylinder list can be performed at a higher speed, and the defective cylinder list can be made to correspond to the representative defective cylinder number. By using the registered number of skip cylinders, the number of list searches to be performed after the binary search can be reduced, so that the skip processing can be further speeded up.

[0023]

Embodiments of the present invention will be described below with reference to the drawings. [1] Description of First Embodiment First, with reference to FIG.
The configuration of a general hard disk device 10 to which the embodiment and the second embodiment are applied will be described.

The hard disk device 10 of this embodiment shown in FIG. 14 has a plurality of disk-shaped recording media (FIG. 1).
4, six magnetic disks 11 are provided. By rotating a rotary shaft 12 on which the magnetic disks 11 are coaxially mounted by a spindle motor 14, a plurality of magnetic disks 11 are coaxially mounted. It is provided so as to be rotatable around a rotation shaft 12). The rotating shaft 12 is rotatably connected to the housing 20 by a bearing 13. The spindle motor 14 is driven by a spindle motor drive circuit 21 to drive the magnetic disk 11 (rotary shaft 12).
Is controlled to rotate at a constant rotational speed.

The housing 20 includes the magnetic disk 11 described above.
In addition, a magnetic head 15, a carriage 16, and a positioner (linear motor) 17 are incorporated. Case 2
0, a plurality of (ten in FIG. 14) magnetic heads 15 are respectively provided on the recording surfaces 11a of the plurality of magnetic disks 11.
It is provided so as to face. Each magnetic head 15
Records data (write information) from the write / read circuit 26 on the recording surface 11a of the magnetic disk 11 or writes / reads data recorded on the recording surface 11a of the magnetic disk 11 as read information. The data is read out to the circuit 26.

In the housing 20, a plurality of magnetic heads 15 are mounted on one carriage 16,
The carriage 16 is moved to a positioner (linear motor) 17.
And fixed at a predetermined position, thereby moving integrally with each other in the radial direction of the magnetic disk 11 and positioning it on a predetermined track.
1 to access (record / read data) a track on each recording surface 11a. Magnetic head 15 (carriage 16) by positioner 17
Is controlled by the positioning circuit 24 via the servo amplifier 25.

In order to control the hardware configuration as described above, the hard disk drive 10 includes a power supply control interface in addition to the spindle motor drive circuit 21, the positioning circuit 24, the servo amplifier 25, and the write / read circuit 26. 22, a control circuit 23 and an interface 27 are provided. The power supply control interface 22 supplies power to the spindle motor drive circuit 21, and the interface 27 supplies write information,
While receiving a control signal, address information, etc., it sends out readout information, status, etc. to the host device.

The control circuit 23 includes an interface 2
The controller 7 controls the positioning circuit 24 and the write / read circuit 26 based on the control signal and address information received from the host device in the step 7. For example, when a write instruction is sent as a control signal from a higher-level device, the control circuit 23 controls the positioning circuit 24 so as to position the magnetic head 15 on a predetermined track according to the address information. The write / read circuit 26 is controlled so that data (write information) is recorded in the write / read circuit. When a read instruction is sent as a control signal from the host device, the control circuit 23
Indicates that the magnetic head 1 is located on a predetermined track corresponding to the address information.
5 is controlled, and the write / read circuit 26 is controlled to reproduce data (read information) from the predetermined track. Data (read information) reproduced from a predetermined track of the magnetic disk 11 is transmitted to a write / read circuit 26 and an interface 27.
Is sent back to the host device via

Incidentally, each recording surface 1 of the magnetic disk 11
In 1a, as described above, a large number (for example, 10,000)
Are concentrically arranged, and each track is provided with a physical track number (physical address) for designating the track corresponding to a physical position. For example, when the number of tracks is m, the physical track numbers are “0”, “1”, “2”,..., “M−1” in order from the outside (or inside) of the magnetic disk. As described above, a set of tracks provided with the same physical track number on the plurality of magnetic disks 11 is:
A cylinder is called a physical track number i (i = 0
A physical cylinder number i is assigned to a cylinder which is a set of tracks assigned with the numbers (m-1).

As described above, the area accessible by the user on each recording surface 11a of the magnetic disk 11 is
During manufacturing or actual operation of the hard disk device 10 (magnetic disk 11), a defective track where data cannot be recorded / reproduced correctly may occur depending on a defective portion such as a scratch. In the present embodiment, the cylinder to which the defective track having the defective portion belongs is treated as a defective cylinder, and when the host apparatus accesses the magnetic disk 11, the above-described access to the defective cylinder is avoided, and only the normal cylinder is used. Is specified as an access target.

Therefore, in the present embodiment, FIGS.
Instead of the conventional defective track list shown in FIG. 1, a defective cylinder list as shown in FIG. 1 is created, and this defective cylinder list is stored in a predetermined area (for example, a system area) of the magnetic disk 11. In this defective cylinder list, the physical track number of the defective track is registered in advance as a defective cylinder number.

FIG. 1 is a diagram showing the structure of a defective cylinder list in the present embodiment. As shown in FIG. 1, the defective cylinder list contains a maximum of N defective cylinder numbers (physical track numbers of defective tracks). Are registered in ascending order, and the address of each defective cylinder number can be used as an element number when the defective cylinder list is handled as an array as shown in FIG. The total number M (≦ N) of defective cylinder numbers registered in the defective cylinder list is registered in the head area of the defective cylinder list. An area capable of storing such a defective track list is continuously secured in the system area. The number of cylinders that can be used by the user is about 10,000, of which the number of defective cylinders is 20.
It is considered that the maximum number N of defective cylinder numbers registered in the defective cylinder list is set to about 100. The unit of the address is, for example, 1 byte.

In the defective cylinder list shown in FIG. 1, the total number of registered defective cylinder numbers (that is, the total number of defective cylinders) M is stored in the head area specified by the addresses 0000h and 0001h. Information on the defective cylinder to which the defective track belongs (M defective cylinder numbers DC1, DC
2,..., DCM) are stored in an area specified by addresses 0002h to 2M + 1. The defective cylinder numbers DC1, DC2,..., DCM stored in these areas are the physical cylinder numbers (the same as the physical track numbers of the defective tracks) assigned to the defective cylinders to which the defective tracks belong, respectively, and have addresses 0002h, 0003h. Are stored in ascending order from the area specified by. The unused area of the defective cylinder list (the area specified by addresses 2M + 2 to 2N + 1 in FIG. 1) is treated as “Don't Care”.

A defective portion (defective track) is detected in a self-test process at the time of manufacturing the hard disk device 10, and the physical track number of the defective track is registered as a defective cylinder number. Here, the defective cylinder list will be described with reference to specific numerical examples shown in FIG. 4 (a) to FIG. 4 (d), FIG. 5 and FIG.

FIG. 4A shows physical track numbers / physical cylinder numbers assigned to tracks / cylinders on the recording surface 11a of the magnetic disk 11, and FIG.
FIG. 4 shows a defective track position (a position where “x” is written) of the magnetic disk 11 to be sought by the magnetic head Hi.
(C) shows a defective track position (a position where “x” is written) of the magnetic disk 11 to be sought by the magnetic head Hj (i ≠ j). Then, FIG. 4D shows FIG.
FIG. 5B shows the position of the defective cylinder to which the defective track shown in FIG. 4C belongs (the position where "x" is written).

As shown in FIGS. 4A and 4B, on the recording surface 11a of the magnetic disk 11 to be sought by the magnetic head Hi, physical track numbers 3, 5, 6,
The track assigned 15 is a defective track, and as shown in FIGS. 4A and 4C, the recording surface 11 of the magnetic disk 11 to be sought by the magnetic head Hj.
a, physical track numbers 7, 8, 15, 16, 1
Tracks assigned with 9 and 22 are defective tracks.

In this case, as shown in FIGS. 4A and 4D, the physical cylinder number (physical track number)
The cylinders assigned 3, 5, 8, 15, 16, 19, and 22 are defective cylinders, and a defective cylinder list is created in the system area in a recording format as shown in FIG. That is, in the head area of the defective cylinder list, 9 is registered as the total number M of defective cylinder numbers, and 3, as defective cylinder numbers DC1 to DC9, respectively.
5, 6, 7, 8, 15, 16, 19, and 22 are registered.

When the skip processing method of this embodiment is executed, the defective cylinder list created as shown in FIG. 5 is handled as an array as shown in FIG. That is, the defective cylinder list is handled as an array having the array name “DftCyl” and the number of elements M (= TotalDftCyl), and prior to executing the skip processing method of the present embodiment, for example, the array declaration “int DftCyl [ TotalDftCyl] "has been done. In this embodiment, the address of the defective cylinder list is used as an element number (list index) of the array, and the number of defective cylinders (ClySkipped) existing from the leading cylinder to each defective cylinder is “ It is given as element number +1 ".

Next, a description will be given, with reference to FIG. 2 and FIG. 3, of the defective site skip processing method of the first embodiment having the above-described defective cylinder list. here,
FIG. 2 is a flowchart (steps S1 to S6) for explaining a defective portion skip processing method according to the first embodiment of the present invention, and FIG. 3 is a flowchart of a binary search (binary search) executed in step S2 of FIG. It is a flowchart (step S11-S24) for demonstrating a procedure. The descriptions in FIGS. 2 and 3 are in C language.

The processing according to the flowcharts shown in FIGS. 2 and 3 is executed by the control circuit 23. That is, the control circuit 23 reads the defective cylinder list from the system area of the magnetic disk 1, and handles the defective cylinder list as an array “DftCyl [TotalDftCyl]” as shown in FIG. When receiving an access request from the host device, the control circuit 23 obtains a physical track number to be accessed from the access request (address information) as a search target cylinder number “CylNo”, and then obtains the cylinder number. Using “CylNo” and the total number of defective cylinders M (= TotalDftCyl) obtained from the defective cylinder list, a search process according to the flowcharts shown in FIGS. 2 and 3 is executed.

As shown in FIG. 2, first, the number (integer) of defective cylinders to be skipped (integer) "CylSkipped" is declared as an initial value 0 (step S1), and the result obtained by the binary search process shown in FIG. Bsearch (DfCyl, TotalDftCyl, CylNo) ”
Is set as “CylSkipped” (step S2). Here, in the binary search process shown in FIG. 3, the cylinder number “Cy
Performs a binary search of the array “DftCyl [TotalDftCyl]” for “lNo”, and starts with the cylinder number “CylN” from the first cylinder.
Obtain the number of defective cylinders existing up to the cylinder to which "o" is assigned. That is, as shown in FIG. 3, lower limit lo, upper limit hi, and upper limit hi, as parameters (integer values) for the binary search.
After declaring the binary value mid (step S11), it is determined whether or not the total number of defective cylinders M (= TotalDftCyl) is 0 (step S12). If the total number of defective cylinders is 0 (YES route from step S12), Bsea
While “0” is output as the value of rch (DfCyl, TotalDftCyl, CylNo) (step S13), the total number of defective cylinders is 0.
Otherwise (NO route from step S12), initial values of the lower limit value lo and the upper limit value hi are lo = 0, hi = To
talDftCyl-1 is set (step S14).

Thereafter, it is determined whether or not lo <hi (step S15). If lo ≧ hi (step S15)
It is determined whether or not DftCyl [lo] ≦ CylNo (step S16). Where “DftCyl [lo]”
Is the element number in the array "DftCyl [TotalDftCyl]"
Means the defective cylinder number corresponding to lo. And Df
If tCyl [lo] ≦ CylNo (YE from step S16)
On the other hand, “lo + 1” is output as the value of Bsearch (DfCyl, TotalDftCyl, CylNo) (step S17),
If DftCyl [lo]> CylNo (from step S16 to N
“Lo” is output as the value of Bsearch (DfCyl, TotalDftCyl, CylNo) to the O route) (step S18).

If it is determined in step S15 that lo <hi (YES route from step S15),
Mid = (lo + hi) / 2 is calculated as the binary value mid (step S19). At this time, values after the decimal point are rounded down.
Then, it is determined whether or not DftCyl [mid] = CylNo (step S20). Here, “DftCyl [mid]” means a defective cylinder number corresponding to the element number mid in the array “DftCyl [TotalDftCyl]”. And DftCyl
If [mid] = CylNo (YES route from step S20), “mid + 1” is output as the value of Bsearch (DfCyl, TotalDftCyl, CylNo) (step S21), while Df is output.
If tCyl [mid] = CylNo is not satisfied (NO from step S20)
In the route, it is determined whether or not DftCyl [mid]> CylNo (step S22).

If DftCyl [mid]> CylNo (YES route from step S22), Bsearch (DfCyl, TotalDft)
While “mid-1” is output as the value of (Cyl, CylNo) (step S23), if DftCyl [mid]> CylNo is not satisfied (NO route from step S22), the lower limit lo is replaced with “mid + 1” (Step S24), and the process returns to Step S15.

As described above, the result of the binary search Bsearch
When (DfCyl, TotalDftCyl, CylNo) is obtained, “Bsearch (DfCyl, TotalDftCyl, CylNo)” is set as “CylSkipped” in step S2 of FIG. 2 as described above.
Next, the number of defective cylinders obtained by the binary search [that is, CylSkipped = Bsearch (DfCyl, TotalDftCyl, CylN
o)] and the cylinder number "CylNo".
By searching for tCyl [TotalDftCyl] "(steps S3 to S6), a normal cylinder number existing at a position corresponding to the physical cylinder number" CylNo "to be accessed among the normal cylinders excluding the defective cylinder is obtained. .

That is, it is determined whether or not CylSkipped <TotalDftCyl (Step S3), and CylSkipped <TotalDftCyl is determined.
If it is ftCyl (YES route from step S3), it is determined whether or not CylSkipped + CylNo <DftCyl [CylSkipped] (step S4). Where DftCyl [Cyl
Skipped] is the sequence “DftCyl [TotalDftCyl]”
Means the defective cylinder number corresponding to the element number CylSkipped. If CylSkipped <TotalDftCyl is not satisfied (NO route from step S4), “CylSkipped”
Is incremented by 1 (step S5), and the process returns to step S3. On the other hand, in step S3, CylSkipped <
If it is determined that it is not TotalDftCyl, or if it is determined in step S4 that CylSkipped + CylNo <DftCyl [CylSkipped], "CylSkipped + CylNo" is obtained as the final target cylinder number (step S4).
6).

In other words, the above steps S3 to S6
The processing of the detected defective cylinder number DftCyl [CylSkip
ped] is the sum of the number CylSkipped of defective track numbers existing before the defective cylinder number DftCyl [CylSkipped] and the physical cylinder number CylNo to be accessed CylSkipped +
It is performed until it becomes larger than CylNo. Then, when the condition is satisfied, it is determined that the cylinder given the value of the sum CylSkipped + CylNo as the physical cylinder number is the cylinder having the track to be sought.

In this way, a usable target cylinder number CylSkipped + CylNo corresponding to the physical cylinder number CylNo to be accessed is obtained, and the control circuit 23 determines that the track belonging to the cylinder to which the target cylinder number CylSkipped + CylNo is assigned is accessed. Track, and the magnetic head 1
The positioning circuit 24 is controlled so as to perform the seek operation of Step 5. As described above, the process of skipping a defective cylinder and replacing a defective track with a track belonging to a normal cylinder is called a cylinder skip process. Upon receiving an access request from a higher-level device, the hard disk device 10 of the present embodiment searches for a defective cylinder list and executes a cylinder skip process, thereby avoiding defective cylinders and only normal cylinders that do not include defective tracks. To perform an operation according to the access request.

Next, the cylinder skip processing of the first embodiment will be described with reference to specific numerical examples shown in FIGS. 4 (a) to 4 (e), 5 and 6. When the defective cylinder list shown in FIG. 5 is handled as an array “DftCyl [9]” as shown in FIG. 6, for example, when the search target cylinder number CylNo = 5, first, FIG.
The binary search is performed as follows according to the flowchart shown in FIG. That is, since TotalDftCyl = 9,
In step S12, the determination is NO, and in step S14,
lo = 0, hi = 9-1 = 8 are set. At this time, since lo (= 0) <hi (= 8), naturally, step S1
5 is YES, and in step S19, the binary value mid =
(0 + 8) / 2 = 4 is calculated. Then, as shown in FIG. 6, since DftCyl [4] = 8, DftCyl [4] ≠ CylNo
(= 5), and a NO determination is made in step S20. Also, since DftCyl [4]> CylNo, a YES determination is made in step S22, and the upper limit hi is set to mid-1 in step S23.
= 4-1 = 3. Thereafter, step S1
5, since lo = 0 and hi = 3, step S15
Is determined as YES, and the binary value mid =
(0 + 3) / 2 = 1 is calculated. At this time, since DftCyl [1] = 5 as shown in FIG. 6, DftCyl [1] = CylN
o (= 5), a YES determination is made in step S20, and mid + 1 = 1 + 1 = 2 is output as the value of Bsearch (DfCyl, 9, 5) in step S21.

As a result of the binary search, Bsearch (DfCyl, 9,
5) When 2 is obtained, the final target cylinder number is obtained in the following manner according to the flowchart shown in FIG. That is, first, in step S2, CylSkipped
= Bsearch (DfCyl, 9,5) = 2 is set. At this time, since CylSkipped (= 2) <TotalDftCyl (= 9), a YES determination is made in step S3. Also, CylSkipp
Since ed + CylNo = 2 + 5 = 7 and DftCyl [2] = 6 as shown in FIG. 6, a NO determination is made in step S4, CylSkipped is incremented to 3 in step S5, and the process returns to step S3. Again, Cy
Since lSkipped (= 3) <TotalDftCyl (= 9), a YES determination is made in step S3, and CylSkipped + CylNo =
Since 3 + 5 = 8 and DftCyl [3] = 7, step S
In step S5, the determination is NO. In step S5, CylSkipped is incremented to 4, and the process returns to step S3. Also at this time, CylSkipped (= 4) <TotalDftCyl (=
9), a YES determination is made in step S3, and Cy is determined.
Since lSkipped + CylNo = 4 + 5 = 9 and DftCyl [4] = 8, a NO determination is made in step S4, and step S5 is performed.
Then, CylSkipped is incremented to 5 and the process returns to step S3 again. Again, CylSkipped (=
5) <TotalDftCyl (= 9), so a YES determination is made in step S3, but CylSkipped + CylNo = 5 + 5 = 1.
Since 0 and DftCyl [5] = 15, a YES determination is made in step S4, and CylSkipped + CylNo = 5 + 5.
= 10 is obtained as the final target cylinder number.

For the array "DftCyl [9]" shown in FIG. 6, that is, the specific numerical examples shown in FIGS. 4 (a) to 4 (d) and FIG. FIG. 4E shows the result (number after the skip processing) obtained by performing the processing. By skipping defective cylinders, normal tracks can be used as tracks assigned with track numbers 0, 1, 2,..., Respectively, as shown in FIG.

For example, when 0 to 2 are specified as the physical cylinder numbers to be accessed, the cylinders assigned with the physical cylinder numbers 0 to 2 are accessed as they are, but the physical cylinder number 3 is assigned. Since the assigned cylinder is a defective cylinder, if 3 is specified as the physical cylinder number to be accessed, the normal cylinder assigned the physical cylinder number 4 is actually handled as the cylinder number 3. Similarly,
Since the cylinders assigned with the physical cylinder numbers 5 to 8 are also defective cylinders, if 4 to 9 are specified as the physical cylinder numbers to be accessed, the normal physical track numbers 9 to 14 are respectively assigned. Tracks are handled as track numbers 4 to 9. Less than,
Similarly, the cylinders provided with the physical cylinder numbers 15, 16, 19, and 22 are also defective cylinders. Therefore, when 10 to 14 are specified as the physical cylinder numbers to be accessed, the physical track numbers 17 , 18,
Normal tracks assigned 20, 21, 23 are handled as track numbers 10 to 14.

As described above, according to the first embodiment of the present invention, the cylinder to which the defective track belongs is treated as a defective cylinder, and the skip processing can be performed not in the track unit but in the cylinder unit. There is no need to store the defective track list for each head, and the area for holding the defective cylinder list (defect information list) is greatly reduced. Further, since it is not necessary to determine the start address of the defective track list based on the head number of the magnetic head as in the related art, the skip processing can be performed in a very short time.

In particular, when the read operation and the write operation are sequentially performed, if the skip processing of the defective track is performed for each magnetic head as in the related art, the track skip processing must be performed every time the magnetic head is changed. No. However, if the cylinder skip processing irrespective of the head number of the magnetic head is performed as in the present embodiment, each magnetic head performs a read operation or a write operation on the track (cylinder) of the same number, The skip processing is performed only once, and the magnetic disk 11
The time required for access to the URL can be greatly reduced.

In the present invention, since a cylinder having at least one defective track is treated as a defective cylinder, a track which belongs to the same cylinder and which can be actually used is disabled, and the user area is reduced. However, as described above, the number of cylinders that can be used by the user is about 10,000 at present, of which the number of defective cylinders is 20.
Since it is considered to be 考 え 30, the reduction of the user area as described above does not cause any problem, and the merit of reducing the list area / reducing the processing time is sufficiently large.

In the present embodiment, a binary search (binary search) of a defective cylinder list for a physical cylinder number to be accessed can be performed without performing a search from the top of the list as in the related art. Since the number of defective cylinders corresponding to the result of the binary search can also be obtained immediately from the element number, the speed of the skip processing can be greatly increased as compared with the related art.

That is, by performing the binary search, the number of searches required to reach the target can be reduced. For example, if 128 defective cylinders are registered in the list, and if the binary search is not performed, the search processing must be performed 128 times at the worst. it can. Where α is
After the binary search, the number of searches performed to obtain the final target cylinder number, that is, steps S3 to S in FIG.
This corresponds to the number of repetitions of 5.

[2] Description of Second Embodiment In the second embodiment of the present invention, for example, when a defective cylinder list as shown in FIG. 5 is handled as an array (see FIG. 6), the array is changed to FIG. Are compacted (converted) from the state shown in FIG. 10 to the state shown in FIG. 10, and a search process is performed using the compacted array (representative defective cylinder list).

The processing for compacting the defective cylinder list is performed by the control circuit 23, for example, on the hard disk drive 1
8 is performed only once according to the procedure shown in FIG. 8 when the power is turned on, and the obtained representative defective cylinder list is stored in the system area of the magnetic disk 11, the memory of the microcomputer forming the control circuit 23, and the like. The compacting process is performed according to the following two criteria.

If there are consecutive defective cylinder numbers in the defective cylinder list, the defective cylinder numbers are registered as representatives of the first defective cylinder number, and the other defective cylinder numbers are deleted. For example, in the example shown in FIG. 6, “5-8”, “15,
16 is represented by the first defective cylinder numbers "5" and "15", and "6 to 8" and "16" are deleted from the list.

The number of defective cylinders existing between the head cylinder and the cylinder to which the first defective cylinder number (representative defective cylinder number) is assigned continues after the first defective cylinder number (representative defective cylinder number). A value obtained by adding the number of defective cylinder numbers is registered as a skip cylinder number corresponding to the element number of the first defective cylinder number (representative defective cylinder number). For example, in the example shown in FIG.
A value 5 obtained by adding the number 3 of the defective cylinder numbers consecutive after the representative defective cylinder number 5 to the number 2 of the defective cylinders existing up to the cylinder to which the number of the defective cylinders is added is set as the representative defective cylinder number 5 as the skip cylinder number. Is registered in correspondence with the element number (list index) 1 of. Similarly, for the representative defective cylinder number 15, the skip cylinder number 7 is registered corresponding to the element number (list index) 2.

Here, the procedure of creating the representative defective cylinder list (the process of compacting the defective cylinder list) in the second embodiment will be described with reference to the flowchart (steps S41 to S47) shown in FIG. The description in FIG. 8 is in the C language. This FIG.
In the processing shown in the figure, the array of defective cylinder list “DftCyl [T
otalDftCyl] ”and the total number of defective cylinders M (= TotalDftCyl)
Is input, and the array of the representative defective cylinder list “cmpDft
Cyl [TotalCmpDftCyl] ", the total number of registered defective cylinder numbers in the representative defective cylinder list TotalCmpDftCyl,
Array "dftSkip [TotalCmpDf" of the number of defective cylinders to be skipped up to the representative defective cylinder (number of skipped cylinders)
tCyl] "is obtained as a result (output) of the compacting process.

As shown in FIG. 8, first, a dummy cylinder number dmmyCyl and element numbers (list index) i, j of initial value 0 are declared as parameters (integer value) for compaction processing, and then ( Step S41), i <
It is determined whether or not TotalDftCyl (Step S4
2). If i <TotalDftCyl (YES route from step S42), cmpDftCyl [j] = DftCyl [i] (step S43). That is, the value of the element number i in the array of the defective cylinder list is set as the value of the element number j in the array of the representative defective cylinder list. At the same time, the dummy cylinder number dmmyCyl is set to cmpDftCyl [j] (step S43).

Thereafter, it is determined whether or not i + 1 <TotalDftCyl (step S44), and i + 1 <TotalDftCyl.
(YES route from step S44), dmmy
It is determined whether or not Cyl + 1 = DftCyl [i + 1] (step S45). If dmmyCyl + 1 = DftCyl [i + 1] (YES route from step S45), dummy cylinder number
After incrementing dmmyCyl and index i by 1 (step S47), the process returns to step S44.

On the other hand, when i + 1 <TotalDftCyl is not satisfied (NO route from step S44) or when dmmyCyl + 1 ≠ DftCyl
yl [i + 1] (NO route from step S45)
Is set to dftSkip [j] = i + 1 (step S46).
That is, “i + 1” is set as the value of the element number j in the array of the number of skip cylinders. At the same time, the indexes i and j are incremented by 1 (step S46).

Thereafter, the flow returns to step S42, and the processing of steps S43 to S47 is repeatedly executed until it is determined in this step S42 that i ≧ TotalDftCyl. When i ≧ TotalDftCyl (Step S4
(NO route from 2), “j” is set as the total number of defective cylinder numbers registered in the representative defective cylinder list TotalCmpDftCyl (step S48), and the compacting process ends.

For example, when the compacting process of the procedure shown in FIG. 8 is performed on the array shown in FIG. 6, the array shown in FIG.
That is, the total number of registered defective cylinder numbers TotalCmpDftCyl =
The sequences "cmpDftCyl [5]" and "dftSkip [5]" are obtained. The array on the upper side (defective cylinder number) in FIG. 10 corresponds to “cmpDftCyl [5]”, and the array on the lower side (number of skip cylinders) in FIG. 10 corresponds to “dftSkip [5]”.

Next, a description will be given, with reference to FIG. 7 and FIG. 9, of a defective portion skip processing method according to the second embodiment, which is performed using the representative defective cylinder list obtained as described above. Here, FIG. 7 is a flowchart (steps S31 to S39) for describing a defective portion skip processing method according to the second embodiment of the present invention.
FIG. 9 is a flowchart (steps S51 to S64) for explaining the procedure of the binary search (binary search) executed in step S32 in FIG. The descriptions in FIGS. 7 and 9 are in C language.

The processing according to the flowcharts shown in FIGS. 7 and 9 is also executed by the control circuit 23. That is, the control circuit 23 reads the representative defective cylinder list from the system area of the magnetic disk 1 and stores the representative defective cylinder list in an array “cmpDftCyl [TotalCyl] as shown in FIG.
mpDftCyl] "and the control circuit 23
Receives an access request from the host device, obtains a physical track number to be accessed from the access request (address information) as a cylinder number "CylNo" to be searched, and then represents the cylinder number "CylNo" as the representative. Total number of registered defective cylinder numbers in the defective cylinder list Tota
The search processing according to the flowcharts shown in FIGS. 7 and 9 is executed using lCmpDftCyl.

As shown in FIG. 7, first, the number of defective cylinders to be skipped (integer) “CylSkipped” and the index (integer) of the representative defective cylinder list “cmpDftCylInd”
ex ”is declared as an initial value 0 (step S3).
1), the result "Bsearc" obtained by the binary search process shown in FIG.
h (cmpDfCyl, TotalCmpDftCyl, CylNo) ”to“ cmpDftCylInd
ex ”(step S32).

Here, in the binary search processing of the first embodiment shown in FIG. 3, a defective cylinder list (array) as shown in FIG. 6 is searched, while the binary search processing of the second embodiment shown in FIG. In the binary search process, a representative defective cylinder list (array) as shown in FIG. 10 obtained by the compacting process described above is searched. However, the binary search process of the second embodiment shown in FIG. 9 is substantially the same as the binary search process shown in FIG. 3 only by changing the form of the search target list. That is, “DftCyl [TotalDftCyl]” in FIG.
Is replaced with “cmpDftCyl [TotalCmpDftCyl]”, and “TotalDftCyl” in FIG. 3 is replaced with “TotalCmpDftCyl”.
FIG. 9 shows a case where the binary search processing is actually performed. In the first embodiment and the second embodiment, the same software for the binary search processing is used.

Accordingly, steps S51 to S64 in FIG.
Correspond to steps S11 to S24 in FIG. 3 respectively, and the detailed description of the binary search processing of the second embodiment shown in FIG. 9 is omitted.
Of the representative defective cylinder numbers registered in the representative defective cylinder list, the physical cylinder number “Cy to be accessed”
lNo ”, the number of representative defective cylinder numbers not exceeding“ Bsea
rch (cmpDfCyl, TotalCmpDftCyl, CylNo) ".

When “Bsearch (cmpDfCyl, TotalCmpDftCyl, CylNo)” is obtained by the binary search process of the second embodiment shown in FIG. , Tot
alCmpDftCyl, CylNo) ”. Then, the value obtained by the binary search [cmpDftCylIndex = Bsearch (cmpDfCyl, T
otalCmpDftCyl, CylNo)], the array of skip cylinder numbers “dftSkip [TotalCmpDftCyl]” and the cylinder number “CylNo
And the array “cmpDftCyl [TotalCmpDftCyl]
(Steps S33 to S3).
9) Among the normal cylinders excluding the defective cylinder, a normal cylinder number existing at a position corresponding to the physical cylinder number “CylNo” to be accessed is obtained.

That is, first, it is determined whether or not cmpDftCylIndex ≠ 0 (step S33), and cmpDftCylIndex ≠ 0.
(YES route from step S33), the number of defective cylinders to be skipped is set as “CylSkipped” as “df
tSkip [cmpDftCylIndex-1] "is set (step S3
4). Here, dftSkip [cmpDftCylIndex-1] is an element number cmpD in the array “dftSkip [TotalCmpDftCyl]”.
It means the number of skip cylinders corresponding to ftCylIndex-1. On the other hand, if cmpDftCylIndex ≠ 0, that is, cmpD
If ftCylIndex = 0 (NO route from step S33), the number of defective cylinders to be skipped “CylSkipped”
Is set to 0 (step S35).

Thereafter, cmpDftCylIndex <TotalCmpDftCyl
Is determined (step S36), and cmpDftCylI is determined.
If ndex <TotalCmpDftCyl (YES route from step S36), CylSkipped + CylNo ≧ cmpDftCyl
It is determined whether or not [cmpDftCylIndex] (step S37). Here, cmpDftCyl [cmpDftCylIndex] is the element number in the array “cmpDftCyl [TotalCmpDftCyl]”
It means the representative defective cylinder number corresponding to cmpDftCylIndex.

Then, CylSkipped + CylNo ≧ cmpDftCyl
If [cmpDftCylIndex] (Y from step S37)
ES route) has "dftSkip [cm
pDftCylIndex] ”and“ cmpDftCylInd
ex ”is incremented by 1 (step S3
8) Return to step S36. On the other hand, in step S36
If it is determined that cmpDftCylIndex <TotalCmpDftCyl, or in step S37, CylSkipped + CylNo ≧
If it is not cmpDftCyl [cmpDftCylIndex],
“CylSkipped + CylNo” is obtained as the final target cylinder number (step S39).

In this way, the target cylinder number Cy
When lSkipped + CylNo is obtained, in the second embodiment, as in the first embodiment, the control circuit 23 determines that the track belonging to the cylinder to which the target cylinder number CylSkipped + CylNo is assigned is the track to be accessed. The positioning circuit 24 is controlled so that a seek operation is performed on the track by the magnetic head 15. As described above, also in the second embodiment, when the hard disk device 10 receives an access request from the host device, the hard disk device 10 searches the defective cylinder list and executes the cylinder skip processing, thereby avoiding the defective cylinder and avoiding the defective track. The operation according to the access request is performed using only the normal cylinders not including the.

Next, the cylinder skip processing of the second embodiment will be described with reference to specific numerical examples shown in FIGS. 4 (a) to 4 (e) and FIG. As shown in FIG. 10, the array "cm
In the case of performing a search targeting, for example, cylinder number CylNo = 5 using pDftCyl [5] ", first, a binary search is performed as follows according to the flowchart shown in Fig. 9. That is, since TotalCmpDftCyl = 5, , A NO determination is made in step S52, and in step S54, lo = 0,
hi = 5-1 = 4 is set.

At this time, of course, lo (= 0) <hi (= 4)
Therefore, a YES determination is made in step S55, and a binary value mid = (0 + 4) / 2 = 2 is calculated in step S59. Then, as shown in FIG. 10, cmpDftCyl [2] = 15
Therefore, cmpDftCyl [2] ≠ CylNo (= 5), and a NO determination is made in step S60. Also, DftCyl [2]> Cy
lNo, so a YES determination is made in step S62,
In step S63, the upper limit value hi is replaced with mid-1 = 2-1 = 1.

Thereafter, the flow returns to step S55, where lo = 0,
Since hi = 1, a YES determination is made in step S55, and a binary value mid = (0 + 1) / 2 = 0 in step S59.
Is calculated. At this time, as shown in FIG.
Since [0] = 3, DftCyl [1] ≠ CylNo (= 5), and a NO determination is made in step S60. Also, DftCyl
[1] <CylNo, so a NO determination is made in step S62, and in step S64, the lower limit lo is set to mid + 1 = 0 + 1 = 1
Is replaced by

Thereafter, the flow returns to step S55, and lo
= 1, hi = 1, so a NO determination is made in step S55. At this time, since DftCyl [1] = 5 as shown in FIG. 10, cmpDftCyl [1] = CylNo (= 5), and a YES determination is made in step S56, and a determination is made in step S57.
Lo + 1 = 1 + 1 as the value of Bsearch (cmpDfCyl, 5, 5)
= 2 is output.

As a result of the binary search, Bsearch (cmpDfCyl,
When (5, 5) = 2 is obtained, the final target cylinder number is obtained as follows according to the flowchart shown in FIG. That is, first, in step S32, cmpD
ftCylIndex = Bsearch (cmpDfCyl, 5, 5) = 2 is set. At this time, since cmpDftCylIndex ≠ 0, a YES determination is made in step S33, and CylS
dftSkip [cmpDftCylIndex-1] as kipped = dftSkip [1]
Is set. Here, as shown in FIG. 10, dftSkip [1]
= 5, CylSkipped = 5 is set.

Then, cmpDftCylIndex (= 2) <TotalC
Since mpDftCyl (= 5), YES in step S36
Judgment. Also, CylSkipped + CylNo = 5 + 5 = 10
And cmpDftCyl [2] = 15 as shown in FIG.
Therefore, a YES determination is made in step S37, and CylSkipped + CylNo = 5 + 5 = 10 is obtained as the final target cylinder number in step S39.

As described above, by using the array “cmpDftCyl [5]” of the representative defective cylinder list of the second embodiment shown in FIG. 10, the concrete examples shown in FIGS. 4 (a) to 4 (d) and FIG. When the cylinder skip processing is performed on a typical numerical example, the same skip processing result as in the first embodiment [FIG.
(E) is obtained. As described above, according to the second embodiment of the present invention, the same operation and effect as those of the first embodiment can be obtained, and a representative defective cylinder list (see FIG. 10) in which the defective cylinder list (see FIG. 6) is compacted. Is used, the following operation and effect can be further obtained.

That is, in the representative defective cylinder list, the number of registered elements is smaller than in the defective cylinder list of the first embodiment, so that the binary search for this representative defective cylinder list can be performed at higher speed. Also, by using the skip cylinder number [dftSkip [TotalCmpDftCyl]] registered corresponding to the representative defective cylinder number, the number of list searches to be performed after the binary search can be reduced, so that the skip processing is further speeded up. be able to. That is, in the first embodiment, when the defective cylinder numbers exist continuously (for example, 5 to 8), it is necessary to repeat the processing of steps S3 to S5 four times in the processing shown in FIG. In the embodiment, the final target cylinder number can be obtained without repeating the processing as described above, and the number of list searches after the binary search can be greatly reduced.

[3] Others Note that the present invention is not limited to the above-described embodiment, and can be implemented with various modifications without departing from the spirit of the present invention. For example, the first and second
In the embodiment, the method of each embodiment is described with reference to FIGS.
(E), a case where the present invention is applied to numerical examples as shown in FIGS. 5, 6, and 10 has been described, but the present invention is not limited to these numerical examples.

In the first and second embodiments described above, the case where the disk-shaped recording medium is a magnetic disk, that is, the case where the disk device is a hard disk device (magnetic disk device) is described. The present invention is not limited to this, and is applied similarly to the above-described embodiment even when the disk-shaped recording medium is a magneto-optical disk, an optical disk, or the like, and obtains the same operation and effect as the above-described embodiment. be able to.

[0088]

As described above in detail, the defective portion skip processing method in the disk device of the present invention (claims 1 to 3)
According to 6), the cylinder to which the defective track belongs is treated as a defective cylinder, and when a defective track occurs, skip processing can be performed not in units of tracks but in units of cylinders. Therefore, a defective cylinder list (defect information list) Is significantly reduced. In addition, since there is no need to perform a process of avoiding a defective track for each magnetic head when performing sequential access, there is an effect that the skip process can be performed in a short time.

Further, it is possible to perform a binary search of the defective cylinder list for the physical cylinder number to be accessed without performing a search from the beginning of the list as in the conventional case, and to perform a defective cylinder corresponding to the result of the binary search. Can also be obtained immediately from the element number, so that the skip processing can be performed much faster than in the past.
~ 5).

Further, in the representative defective cylinder list, the number of registered elements is small and the binary search can be performed at higher speed.
By using the number of skip cylinders, it is also possible to reduce the number of list searches to be performed after the binary search,
Further speed-up of the skip processing can be realized (claim 6).

[Brief description of the drawings]

FIG. 1 is a diagram showing a configuration of a defective cylinder list used when executing a defective portion skip processing method in a disk device of the present invention.

FIG. 2 is a flowchart for explaining a defective part skip processing method in the disk device as the first embodiment of the present invention.

FIG. 3 is a flowchart illustrating a procedure of a binary search (binary search) executed in the first embodiment.

FIGS. 4A to 4E respectively show a physical track number and a physical track number to explain the method of the first embodiment in detail.
It is a figure which shows the specific example of a physical cylinder number, the defective track position of the magnetic head Hi, the defective track position of the magnetic head Hj, a defective cylinder position, and the number after a skip process.

FIG. 5 is a diagram showing a defective cylinder list obtained corresponding to the specific examples shown in FIGS. 4 (a) to 4 (d).

FIG. 6 is a diagram showing an array of a defective cylinder list shown in FIG. 5;

FIG. 7 is a flowchart for explaining a defective portion skip processing method in a disk device as a second embodiment of the present invention.

FIG. 8 is a flowchart illustrating a procedure for creating a representative defective cylinder list according to the second embodiment.

FIG. 9 is a flowchart illustrating a procedure of a binary search (binary search) performed in the second embodiment.

10 is a diagram showing an arrangement of a representative defective cylinder list obtained by compacting the arrangement of the defective cylinder list shown in FIG. 6;

FIG. 11 is a diagram showing a configuration of a general defective track list.

FIGS. 12A to 12C are diagrams showing specific examples of a physical track number, a defective track position, and a number after skip processing, respectively, in order to specifically describe a conventional track skip processing.

FIG. 13 is a diagram showing a defective track list obtained corresponding to the specific examples shown in FIGS. 12 (a) and 12 (b).

FIG. 14 is a diagram illustrating a configuration of a general hard disk drive to which an embodiment of the present invention is applied.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 10 Hard disk drive 11 Magnetic disk (disk-shaped recording medium) 11a Recording surface 12 Rotary shaft 13 Bearing 14 Spindle motor 15 Magnetic head 16 Carriage 17 Positioner (linear motor) 20 Housing 21 Spindle motor drive circuit 22 Power control interface 23 Control circuit 24 Positioning circuit 25 Servo amplifier 26 Write / read circuit 27 Interface

Claims (6)

[Claims] 1. A disk device which comprises a plurality of disk-shaped recording media rotatable about the same axis and records information on each disk-shaped recording medium and reproduces information from each disk-shaped recording medium. A method for accessing a disk-shaped recording medium while skipping a defective portion on a recording surface of a disk-shaped recording medium, the cylinder including a defective track having the defective portion. In order to treat a defective cylinder as a defective cylinder, the physical track number of the defective track is registered in advance in the defective cylinder list as a defective cylinder number, and when an access request is received from a higher-level device, the physical cylinder number to be accessed is determined from the access request. Then, by searching the defective cylinder list for the physical cylinder number, Skip defective cylinder that is registered in the cylinder list Recessed, and performs a process corresponding to the access request using only normal cylinder containing no defective track, the defect site skip processing method in a disk device. 2. The defective cylinder number is registered in the defective cylinder list in ascending order, and an address of each defective cylinder number registered in the defective cylinder list is used as an element number when the defective cylinder list is handled as an array. 2. The method according to claim 1, wherein the defective portion is skipped. 3. The defective portion skipping process in a disk device according to claim 2, wherein the number of defective cylinders existing from the head cylinder to each defective cylinder is obtained based on the element number. Method. 4. The disk device according to claim 2, wherein a total number of defective cylinder numbers registered in the defective cylinder list is registered in a head area of the defective cylinder list. Defect part skip processing method. 5. A binary search of the defective cylinder list for the physical cylinder number of the access target using the total number of the defective cylinder numbers and the element numbers, so that the physical cylinder of the access target from the first cylinder is obtained. By obtaining the number of defective cylinders existing up to the numbered cylinder, and further searching the defective cylinder list based on the number of defective cylinders obtained by the binary search, In the normal cylinder excluding the cylinder,
Recognizing a normal cylinder existing at a position corresponding to the physical cylinder number of the access target, treating the normal cylinder as the cylinder to be accessed, and performing processing according to the access request on the normal cylinder. 5. The method according to claim 4, wherein the defective portion is skipped. 6. When there is a continuous defective cylinder number in the defective cylinder list, the first defective cylinder number is registered as a representative defective cylinder number, and subsequent defective cylinder numbers are deleted. A value obtained by adding the number of defective cylinder numbers following the first defective cylinder number to the number of defective cylinders existing up to the cylinder to which the representative defective cylinder number has been assigned is referred to as the number of skip cylinders. By registering the defective cylinder list in correspondence with the element number of the cylinder number, the defective cylinder list is converted into a representative defective cylinder list, and the registered total number of the defective cylinder numbers in the representative defective cylinder list, the element number of the representative defective cylinder list, and Using the physical cylinder number of the access target By performing a binary search of the defective cylinder list, the number of representative defective cylinder numbers not exceeding the physical cylinder number of the access target among the representative defective cylinder numbers registered in the representative defective cylinder list is obtained. By performing a search of the representative defective cylinder list based on the number of representative defective cylinder numbers obtained by the binary search and the number of skip cylinders, the access is performed in the normal cylinders excluding the defective cylinder. Recognizing a normal cylinder existing at a position corresponding to the target physical cylinder number, regarding the normal cylinder as the access target cylinder, and performing a process according to the access request on the normal cylinder. 5. The method according to claim 4, wherein the defective portion is skipped.