JP2888019B2 - Magnetic disk controller - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a magnetic disk controller, and more particularly to a magnetic disk controller having a function of preventing a read error from occurring from a magnetic disk.

[0002]

2. Description of the Related Art A conventional magnetic disk controller performs a head offset when a magnetic disk is misaligned, attempts to read by shifting a head in both directions from a center position of a track, and reads data in either direction. When it was read at the head offset of, the end of axis misalignment correction was reported to the host processor.

In the case of a magnetic medium defect in which a defect has occurred in a part of a track on a magnetic disk, it is determined that a read error has been corrected by an error correction code (ECC) recorded at the end of each part. To report to the end. The maintenance staff checks the error log registered by the host processor, and performs recovery processing from the host processor at the discretion of the maintenance staff.

[0004]

In this conventional magnetic disk control device, since the axis deviation of the magnetic disk and the medium defect of the magnetic disk are not recovered, the maintenance process requires the intervention of maintenance personnel. If the processing by the maintenance person is delayed, there is a problem that data recorded on the magnetic disk medium cannot be read.

An object of the present invention is to solve the above-mentioned problems,
It is an object of the present invention to provide a magnetic disk control device having a function of recovering the axis deviation of a magnetic disk and a medium defect.

[0006]

A magnetic disk control device according to the present invention has first and second directors respectively connected to a plurality of magnetic disks and performing access to the same magnetic disk by different paths. A control device, comprising: track data storage means for storing data of one track of a magnetic disk in the first and second directors; and a track for diagnosing one track.
Truck diagnostic data storage means for storing diagnostic data
Reads and stores track diagnostic data recorded on the rack
Diagnostic read data storage means for performing the same , the first and second directors are shared, and the axis deviation read error counting / recording means for counting and recording the number of read errors due to the axis deviation of the magnetic disk, and the medium defect of the magnetic disk By
Media defect reading that counts and records the number of read errors
An error counting / recording means is provided , and when an axis deviation occurs a certain number of times, data of one track is read, stored in the track data storage means, and the track is reformatted .
A certain number of reading errors occur due to body defects
And data on the track are stored in the track data storage means.
After that, truck diagnostic data storage means and diagnostic reading
Based on defect position information obtained using data storage means
Data stored in the track data storage means
It has a configuration for reformatting above .

[0007]

[0008] Another magnetic disk controller that by the present invention, in the magnetic disk control device described above, storage occlusion information storing occlusion information prohibiting writing to disk write failure is detected in reformatting the track Means and write data storage means for storing write data for a track when the write failure is detected is provided commonly to the first and second directors.

[0009]

Next, the present invention will be described with reference to the drawings.

FIG. 1 is a block diagram showing the configuration of one embodiment of the present invention. In FIG. 4, directors 3-1 and 3-2 as first and second directors in the disk controller 2 directly related to the present invention are connected to the host processor 1 via different paths 100 and 101, respectively. , 6-n, etc., via the paths 120 and 121. A nonvolatile memory 4 is provided between the two directors 3-1 and 3-2 via signal lines 110 and 111, respectively. Furthermore, directors 3-1 and 3
-2 are provided with data buffers 5-1 and 5-2 for storing write and read data for the magnetic disks 6-1 to 6-n performed via the paths 120 and 121, respectively.

FIGS. 2A and 2B are block diagrams showing the structures of the data buffers 5-1 and 5-2 in FIG. The data buffers 5-1 and 5-2 are used for testing track data storage units 7-1 and 7-2 for storing read data of one track of the magnetic disks 6-1 to 6-n, and for track defects. Track diagnostic data storage units 8-1 and 8-2 for storing the track diagnostic data of
And diagnostic read data storage units 9-1 and 9-2 for storing diagnostic data recorded and read on the track.

FIGS. 3 to 6 show the nonvolatile memory 4 of FIG.
The contents of the configuration including are shown.

FIG. 3 shows an axis deviation reading error counting recording section 4a of the nonvolatile memory 4.

When the director 3-1 detects a read error due to an axis shift on the magnetic disk 6-1, for example, the director 3-1 checks the axis shift read error counting and recording section 4a of the nonvolatile memory 4, and checks the disk number 202, cylinder number 203 and head. If the information with the same number 204 is not registered, the disk number 202, cylinder number 203 and head number 204 where the error is detected are stored in the area where the registration flag 201 is not set, and an axis misalignment reading error occurs. The number 205 is set to 1.

If the information has already been registered, 1 is added to the number of times of occurrence of axis deviation reading error 205.

When a write command is issued from the host processor 1 to a track registered in the axis deviation reading error counting and recording section 4a, an axis deviation reading error registered in the axis deviation reading error counting recording section 4a occurs. When the number of times 205 reaches a certain number, the director 3-1 reads one track of the disk indicated by the registered disk number 202, cylinder number 203 and head number 204, and stores the data in the track data storage unit 7 in the data buffer 5-1. -1 is stored.

The director 3-1 is connected to the host processor 1
The writing of one track is performed by the write data transferred from the CPU and the data stored in the track data storage unit 7-1.

When the writing is completed normally, the director 3
"-1" resets the registration flag 201 of the axis deviation reading error counting and recording unit 4a and deletes the axis deviation information.

FIG. 4 shows the contents of the medium defect read error count recording section 4b of the nonvolatile memory 4 of FIG.

When the directors 3-1 and 3-2 detect a reading error due to a medium defect on the magnetic disks 6-1, 6-2,..., They register the error in the medium defect reading error count recording unit 4b. Here, the magnetic disk 6 of the director 3-1 is used.
1 will be described as an example.

The director 3-1 has a magnetic disk 6-1.
When a read error due to a medium defect is detected in reading from the medium, the medium defect read error counting and recording unit 4b of the nonvolatile memory 4 is checked.

The director 3-1 has a registration flag 301 set, and has a disk number 302 and a cylinder number 303.
When the information that matches the head number 304 is not registered, the disk number 3 of the disk 6-1 that has detected the error in the area where the registration flag 301 has been reset.
02, the cylinder number 303 and the head number 304 are stored, and after setting the registration flag 301, 1 is set to the number of times of occurrence of a medium defect reading error 305.

If the registration flag 301 has already been set and the information in which the disk number 302, cylinder number 303, and head number 304 match has been registered, the director 3-1 sets the medium defect read error occurrence count 305 to 1
Is added.

When a write command is issued from the host processor 1 to a track registered in the medium defect reading error counting recording section 4b, when the number of medium defect reading error occurrences 305 reaches a certain number, the director 3-1
Reads one track from the magnetic disk 6-1 and stores it in the track data storage unit 7-1 of the data buffer 5-1.

The director 3-1 has a data buffer 5-
After storing the track diagnostic data in one track diagnostic data storage section 8-1, the diagnostic data stored in the track diagnostic data storage section 81 is written with the track in which the error occurred as one record.

Next, after reading one record of data from the magnetic disk 6-1 and storing it in the diagnostic read data storage 9-1, the director 3-1 stores the diagnostic data in the track diagnostic data storage 8-1. The data is compared with the data storage unit 9-1, and if they do not match, defect position information is created.

When the creation of the defect position information is completed, the director 3-1 converts the write data transferred from the host processor 1 and the data stored in the track data storage section 7-1 based on the created defect position information. One track is written by skipping the medium defect position. 1
When the writing of the track ends normally, the director 3-1
Is the registration flag 3 of the medium defect reading error counting and recording unit 4b.
01 is reset and the medium defect registration information is deleted.

The directors 3-1 and 3-2 perform writing of one track on a track where a reading error has occurred due to an axis deviation and a medium defect, and report to the host processor 1 that the writing has been completed normally.

FIG. 5 shows the block information storage sections 4c and 4d of the nonvolatile memory 4 of FIG.

The block information storage section 4c stores the director 3-1.
6-n, and the block information storage unit 4d is a path between the director 3-2 and the disks 6-1, 6-2,. If a retry is not successful due to a magnetic disk write failure,
The blocking information for closing the magnetic disk in which the write failure has occurred is set, and access from the same path to the same magnetic disk is prohibited. When the failed magnetic disk is recovered and maintained, the directors 3-1 and 3-2 reset the blocked information of the maintained disk and set the access permission state.

FIG. 6 shows the write data storage section 4e of the nonvolatile memory 4 of FIG.

If the director 3-1 writes data on one track due to misalignment or a medium defect, if a write failure of the magnetic disk occurs and retry is not successful,
The block information of the written disk is set in the block information storage section 4c, the track write data stored in the track data storage section 7-1 of the data buffer 5-1 is stored in the track data 405, and a write failure occurs. After storing the generated disk number 402, cylinder number 403, and head number 404, the registration flag 401 is set.

The host processor 1 includes a director 3-1.
Issues a disk re-try command to the director 3-2.

The director 3-2 examines the write data storage section 4e and, if registered, writes one track with the data stored in the track data 405 and the write data transferred from the host processor 1. Do.

When the writing of the track is completed normally, the director 3-2 deletes the write information in the write data storage section 4e, and reads the axis deviation read error count recording section 4a.
Alternatively, the information registered in the medium defect reading error counting recording section 4b is deleted.

In this manner, the processing for recovering the axis deviation of the magnetic disk and the processing for recovering the medium defect can be automatically performed.

[0037]

As described above, according to the present invention, when a read error occurs due to a misalignment of the magnetic disk in the same track a certain number of times, when a write command is issued, the disk controller writes data on one track. Thus, there is an effect that the axis deviation recovery processing can be automatically performed.

When a write command is issued when a read error due to a medium defect of a magnetic disk occurs on the same track for a certain number of times, the disk control unit checks the defect position of the medium and generates 1 based on the medium defect position information created. There is an effect that the medium defect recovery processing can be automatically performed by writing tracks.

Further, when a write failure occurs during track rewriting due to an axis deviation and a medium defect, access to the magnetic disk from the path where the write failure has occurred is prohibited, and information on the track where the write failure has occurred is stored. By storing the data in the non-volatile memory, it is possible to prohibit the reading of the track from the alternate path and to retry the writing software.

[Brief description of the drawings]

FIG. 1 is a block diagram showing the configuration of an embodiment of the present invention.

FIG. 2A is a block diagram illustrating a configuration of a data buffer 5-1 in FIG. 1, and FIG. 2B is a block diagram illustrating a configuration of a data buffer 5-2.

FIG. 3 is a configuration diagram of an axis deviation reading error counting and recording unit 4a of the nonvolatile memory 4 of FIG. 1;

FIG. 4 is a configuration diagram of a medium defect reading error count recording unit of the nonvolatile memory 4 of FIG. 1;

FIG. 5 is a blockage information storage unit 4c of the nonvolatile memory 4 of FIG. 1;
It is a block diagram of 4d.

FIG. 6 is a configuration diagram of a write data storage unit 4e of the nonvolatile memory 4 of FIG.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Host processor 2 Magnetic disk controller 3-1 and 3-2 Director 4 Non-volatile memory 4a Axis misalignment reading error counting recording part 4b Medium defect reading error counting recording part 4c, 4d Blocking information storage part 4e Write data storage part 5- 1,5-2 Data buffer 6-1 to 6-n Magnetic disk 7-1,7-2 Track data storage unit 8-1,8-2 Track diagnostic data storage unit 9-1,9-2 Diagnostic read data storage Part 201, 301, 401 Registration flag 202, 302, 402 Disk number 203, 303, 403 Cylinder number 204, 304, 404 Head number 205 Number of times of misalignment reading error 305 Number of times of medium defect reading error 405 Track data

──────────────────────────────────────────────────続 き Continued on the front page (58) Field surveyed (Int.Cl. ⁶ , DB name) G11B 20/18

Claims (2)

(57) [Claims] 1. A magnetic disk control device having first and second directors respectively connected to a plurality of magnetic disks and performing access to the same magnetic disk by different paths, wherein the first and second directors are provided. A track data storage means for storing data of one track of a magnetic disk in the director and a track for diagnosing one track.
Truck diagnostic data storage to store rack diagnostic data
The track diagnostic data recorded on the step and the track
Diagnostic read data storage means for reading and storing , wherein the first and second directors are shared and count and record the number of read error occurrences due to axis shift of the magnetic disk; Medium
Medium for counting and recording the number of read errors due to defects
Provided a defect read error counting and recording means, reading one track of data when the axis deviation is constant number of generations to reformat the tracks stored in the track data storage means, reading the number of errors by the medium defect is one
When a certain number of occurrences occur, the data on the track is
After storing the data in the data storage means,
Storage means and the diagnostic read data storage means.
Storing the track data based on defect position information
A magnetic disk control device for reformatting data stored in a means on a track . 2. Blocking information storage means for storing blocking information for prohibiting writing to a disk in which a write failure has been detected by reformatting a track, and writing for storing write data to a track when the write failure has been detected. magnetic disk controller according to claim 1, characterized in that the data storage means provided in common to the first and second director.