JPH0634300B2 - Magnetic disk device - Google Patents

Description

TECHNICAL FIELD The present invention relates to an improved method of using a recording area in a magnetic disk device.

(Prior Art) In recent years, as the amount of information handled by an information processing system increases, the recording density of a magnetic disk device is increasing. However, the higher the recording density is, the more the recorded information is deteriorated due to the defect of the recording medium, and today the improvement thereof is desired.

FIG. 1 shows one track (cylinder number x, head number y) on the magnetic disk and is composed of a plurality of sectors. FIG. 2 (a) shows the structure of one sector and the gap portion G
And an ID field and a data field. Furthermore, the ID field is the synchronization part SYNC, the address part AD
R and check unit CRC, and the data field is composed of synchronization unit SYNC, data unit DATA and check unit CRC. The address part ADR of the ID field is shown in Fig. 2 (b).
Flag part FLG, cylinder part CY indicating cylinder number
It is composed of L, a head section HEAD showing the head number, and a sector section SEC showing the sector number. When the sector 3 in FIG. 1 is defective in the magnetic disk device configured as described above, (1) this track is replaced as a defective track for each track.

(2) Only this sector is replaced with a defective sector, and this sector is replaced.

In the case of the above (1), it is not desirable that the effective sectors other than the sector 3 become useless and the density becomes high. The case (2) is far superior to the case (1), but further improvement is desired.

In the above, a defective track or a defective sector is a constant error caused by a dropout of the recording medium, which is (1) a part of the ID field, (2) the entire ID field or (3) data.・ Part of field or (4)
Occurs in all data fields etc. But the ID of (1)
A part of the field, especially a defect due to dropout etc. in the front part or part of the data field of (3),
In particular, a defect due to a dropout or the like generated in a part of the rear part can be absorbed in the gap G in the front part of the ID field by shifting from the physical positions of the ID field and the data field.

In the conventional magnetic disk device, the above defect is simply regarded as a defect and there is no means for repairing the defect. Further, in a magnetic disk device capable of error correction by using an ECC (error correction code) instead of the CRC check section of the data field, error correction processing must always be performed, and extra time is always required for error correction. Needed.

(Object of the Invention) The present invention has been made in view of the above points, and an object of the present invention is to provide an improved magnetic disk device in which a defective area is reduced to reduce a decrease in recording capacity due to a defect of a recording medium. Is to provide.

A further object is to provide a magnetic disk drive in which the time required for the constant error correction caused by the constant error is reduced.

(Structure of the Invention) In order to achieve the above object, the present invention absorbs a constant error generated in a part of the recording area in the gap G by shifting the physical positions of the ID field and the data field, and makes this sector good. This is a sector and will be described in detail below.

(Embodiment) FIG. 3 is a block diagram of a magnetic disk device according to the present invention. In FIG. 3, reference numeral 1 is a magnetic disk controller, which is a write data register (WDR) 2, switching circuits 3, 4,
5, CRC circuit 6, error correction circuit 7, parallel-serial converter (P / S) 8, serial-parallel converter (S
/ P) 9, a read data register (RDR) 10, a read data buffer 11, and a controller 12 for controlling these. Reference numeral 13 is a magnetic disk.

In the above, the CRC circuit 6 performs a CRC check based on a well-known polynomial. The error correction circuit 7 is capable of correcting a burst error of a predetermined length. For example, Advanced Micro Devic
It is AmZ8065 manufactured by es (AMD). When writing the data shown in FIG. 4 (b) in the magnetic disk device configured as described above (hereinafter referred to as the first format write, abbreviated as F.W1), FIG. 5 (a)- Operates according to the flow in (c). The data recorded on the recording medium in the flow of FIGS. 5 (a) to 5 (c) is set in the WDR2, and when ready for writing, sent to the P / S8 for parallel-serial conversion and the magnetic disk 13 Transferred to. As the write data, first, a predetermined number of (1) data for the gap G, (2)
SYNC data, (3) ADR data, (4) CRC data output from the CRC circuit 6, (5) Gap G data, (6) SYNC
Data, (7) DATA, and (8) ECC data output from the error correction circuit 7 are transferred in this order. CR of (4) above
The C data is obtained by giving ADR data to the CRC circuit 6, and the ECC data of (8) is obtained by giving DATA of (7) to the error correction circuit 7.

When reading the data recorded above (AD
Format the transfer of the R section and DATA section to the host device.
This is called a read "FR", and the transfer of only the DATA part to the higher-level device is called a read "READ". It operates according to the flow of FIGS. 6 (a) and 6 (b). In FIG. 6, the READ operation is taken as an example, but in F / R, the ADR data and the DATA portion data of the sector are sequentially read from the "0" sector and transferred to the host device. The read data transferred from the flow magnetic disk 13 shown in FIGS. 6 (a) and 6 (b) is serial-parallel converted by the S / P 9, and is stored in the RDR 10 after being stored in a predetermined number. RDR
The read data set to 10 is provided for determination by the control unit 12 or transferred to the RB 11, and further necessary data is transferred to the error correction circuit 7.

Now, when trying to read the recorded data of a certain sector, the control unit 12 enters the monitoring state of the sector pulse signal shown in FIG. 4 (a). When the sector pulse is detected in this state, the control unit 1
2 sequentially reads the output of RDR10. When the SYNC data is detected here, the data in the ADR section is continuously read and transferred to the control section 12, and the data is also CRd through the switching circuit 3.
Take in C circuit 6. If all ADR part is read, control part 12
And transfer the data for CRC to the CRC circuit 6 continuously. When all CRC data is read, CRC circuit 6
It is checked whether or not a CRC error has occurred after completion of importing into the memory, and if a CRC error has occurred, error processing is performed because there is a medium defect due to dropout or the like and there is an abnormality in read data due to noise or the like. On the other hand, CRC
If it is determined that no error has occurred, the control unit 12 checks whether the read ADR data and the target address match, and if they do not match, the data field read operation is not performed and the next sector pulse is sent. wait.
On the other hand, if they match, the data / field reading operation is started, and the SYNC data detecting operation is performed in the same manner as above, and when the SYNC data is detected, the DATA portion is read. The data read from the DATA section is sequentially transferred from the RDR 10 to the RB 11 and taken into the error correction circuit 7 via the switching circuit 4. When the entire DATA section is read, the transfer to the RB 11 is completed and the ECC section data is continuously taken into the error correction circuit 7. When all the data in the ECC section is read, the acquisition into the error correction circuit 7 is completed, and whether the data input to the error correction circuit 7 is normal or not is checked by the state of the ER terminal of the error correction circuit 7. If there is an abnormality, it is determined that there is a medium defect such as dropout or the read data is abnormal due to noise, and error processing is performed. In this error processing, the control terminal of the error correction circuit 7 is usually operated to check whether the error can be corrected, and if the error can be corrected, the error location is checked and the read data in the RB 11 and the correction data output to the D terminal are detected. The exclusive OR of is taken and the data is corrected. On the other hand, if it cannot be corrected, it is determined that a burst error of a predetermined length or more has occurred, and necessary processing is performed.

Generally, the defect of the recording medium can be known at the time of shipment of the medium, or by repeating the above-mentioned F · W, F · R and READ and not being able to read normally even after retrying a predetermined number of times or more.

Next, the operation of repairing a constant error due to a medium defect when the defect location is known will be described. According to the invention, said F
A suitable result can be obtained by newly adding an instruction F.W2 for writing in a format different from W1. Now, for example, each sector is sequentially accessed until the medium defect is known in the sector 3 in FIG. 1 or, if it is found, the sector 3 is found. When sector 3 is found, the data field is read. Where the data field
If the SYNC part cannot be detected, it will retry the specified number of times.
If it still cannot be detected, it is treated as a bad sector. On the other hand, if the SYNC part can be detected, the DATA part and the ECC part are continuously read. When the DATA and ECC parts are read and an ECC error is detected, the ECC correction circuit 7 is operated to know the error location. Here, if it is found that the error location is only in the rear part of the data field, it can be changed by checking whether the physical positions of the ID field and the data field can be changed (that is, whether it can be absorbed by the gap G). At the next rotation, the physical positions of the ID field and the data field are changed and recorded as shown in FIG. 4 (c). This changed format may be fixed or variable according to the error location. If it is determined that the sector cannot be changed, the sector is treated as a defective sector, or the DATA portion is reduced so that the sector can be used as described later.

If the error correction circuit 7 finds that the error cannot be corrected (there is a burst error of a predetermined length or more) in the above, it will be DA in the next rotation.
The TA section is reduced and format write is performed. By repeating this, if the ECC error disappears or correction becomes possible and the error location is obtained, if the physical positions of the ID field and data field can be changed from the length of the DATA portion at that time, the change is made as described above.
On the other hand, if it cannot be changed, it should be treated as a bad sector as described above, or DA
Reduce the TA part so that it can be used. Next, in order to reduce the DATA area and use that sector, as shown in FIG. 4 (e), the FLG area has a spare sector flag and other spare sector presence / absence flags, the next spare sector address, and the length of the DATA area. Should be added. By doing so, there is an advantage that the spare sectors with the reduced DATA part are combined and the defective sectors can be effectively used. Further, as described above, it is possible to make the total length of the DATA portion of one track variable in the same manner as an ordinary sector without using the reduced sector of the DATA portion as a spare sector.

In the above description, CRC is used as the check code of the ID field part, but by using ECC instead of CRC, the error location can be known in the same manner as above and the physical positions of the ID field and data field can be shifted in the opposite direction. It is also possible to record and reduce the DATA part.

(Effect of the Invention) As described in detail above, according to the present invention, the error location is known, and the physical location of the ID field and the data field is shifted according to the error location for recording, thereby reducing defective sectors and making the recording area effective. Can be used for.

[Brief description of drawings]

FIG. 1 is a diagram showing a track structure of a magnetic disk device, FIG. 2 is a diagram showing a sector format, FIG. 3 is a block diagram of the magnetic disk device, and FIGS. 4 (a) to (e) are related to the present invention. Figures for explaining the sector format, Figures 5 (a)-(c) show F.
A flowchart of W, and FIGS. 6A and 6B are flowcharts of READ. 1 ... Magnetic disk controller, 3, 4, 5 ... Switching circuit, 6 ... CRC circuit, 7 ... Error correction circuit, 13 ...
Magnetic disk.

Claims (1)

[Claims] 1. A magnetic disk is divided into a plurality of sectors on the same circumference, and each sector is provided with an ID field in which address information is written, a data field in which data is written, and a gap portion defining boundaries between the sectors. In a magnetic disk device that is divided into and the magnetic field is formed in response to a predetermined write instruction to form the ID field, the data field, and the gap portion at a predetermined position of the sector, detection for detecting a defective position in the ID field and the data field. In response to a predetermined write instruction, the defect position is a predetermined position at the end of the ID field and the data field, and the width of the gap portion is reduced by a predetermined amount so that the ID field and the data field are defective. The ID field and data Field, a magnetic disk apparatus characterized by comprising a means for resetting the position of the gap portion.