JPH0479074A - Formating system for magnetic disk device - Google Patents

Abstract

PURPOSE:To always enable a normal alternate processing by recording the address information of a sector and the address information of an alternating destination to the ID parts of respective sectors and recording these plural ID parts to the area of a defective sector in the case of the defective sector. CONSTITUTION:When a sector 51 is reported in defective sector information, first of all, a write gate signal 2 is turned on only in the case of the defective sector 51, and an (n) number of ID are written in the defective sector 51. Further, the cylinder number, head number and sector number of an alternating destination sector 61 are set to an alternate destination cylinder number, head number and sector number 21-23. Then, when access to the defective sector is requested, the first ID part of the defective sector 51 is read and error is detected. When error is not detected, the alternate sector 61 is accessed according to alternate destination address information in the ID part, and the data is written in or read out. Thus, even when medium defect is at any position of the defective sector, the normal alternate processing can be executed.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a formatting method for a magnetic disk device, and particularly to a formatting method for a magnetic disk disk such as a fixed magnetic disk device.

[Prior Art] FIG. 7 shows sector allocation for one track with alternative sectors provided in a conventional formatting method such as that published in the October 25, 1982 issue of Nikkei Electronics. FIG. 8(a) shows the internal allocation of each sector, FIG. 8(b) shows the internal allocation of the ID part in it, and FIG.
The figure shows an alternative processing method for defective sectors using the conventional formatting method and the processing results. In figure 6, (1) is the index pulse signal, (2) is the write gate signal, and (3) is the write signal. Data, (4) is gap 1, (51) to (5n) are sectors, (61),
(62) is alternative sector, <7) is gap 4, (8)
is the synchronization button, (9) is the ID section, (10) is the gap 2, (11) is the data address mark, (12) is the data, (13) is the CRC code, (14) is the gap 3, ( 15) is an ID address mark, (16) is a cylinder number, (17) is a head number, (18) is a sector number, and (19) is an ID CRC code.

Next, the operation will be explained. Generally, medium defects occur on magnetic disk disks during the manufacturing process, and it is extremely difficult and expensive to produce a magnetic disk disk free of such medium defects. Therefore, the manufacturer of the magnetic disk controller will use a disk disk with this media defect, but the magnetic disk controller will be designed to ensure that the user will not experience write/read errors due to the media defect during use. At the time of formatting, it is necessary to exclude the area containing the medium defect (hereinafter referred to as a defective sector) from the area used for writing/reading. Conventionally, formatting has been widely performed on a track-by-track basis, and a plurality of sectors are set for each track. Usually this sector is 256.512 or 1024
They are set to a size that can record bytes of data, and several dozen are provided for each track. Some of these sectors are used as replacement sectors to replace defective sectors. During formatting, defective sector information is reported to the magnetic disk controller from the host computer, which is a normal user. Therefore, the magnetic disk controller erases the reported sector as defective and replaces the sector with a replacement sector. Sector erasure and alternative methods will be explained with reference to FIGS. 7, 8, and 9.

Figure 7 shows the allocation of sectors for one track using the conventional formatting method, which is synchronized with an index pulse (1) generated from a disk device (not shown) once every rotation of the disk. Oh, light.
By turning on the gate signal (2) and sending one track's worth of write data (3) to the disk device, one track's worth of cells is allocated. In this diagram, there is a leading gap (4) and a final gap (7) to provide some extra space at the beginning and end of each track.
, and n sectors (51) to (5n) and two alternative sectors (61) and (62) are allocated. FIG. 8(a) shows the internal allocation of each sector, and FIG. 8(b) shows the internal allocation of the ID section therein. When writing data to a sector or reading data from this sector, it is necessary to specify the sector.
), the accompanying information is written as shown in . This accompanying information includes 1) a synchronization pattern (8) for synchronizing with the rotation of the disk, 2) an ID section (9), 3) data (1)
2) Data address mark (11
L4) There is a CRC code (13) for detecting or correcting errors that occur when reading and writing data (12), and the ID section (9) contains 1) an ID address for recognizing the ID section. Mark (15), 2) Head number 7), The most significant bit (bit 7) of this byte is used as a flag bit to indicate that the sector is a bad sector, and the head is specified using only the lower 4 bits. .

3) Cylinder number one (16L4) which is the track number
) Sector number (18) in the rack, 5) ID section There is a CRC code (19) for detecting or correcting errors that occur when reading or writing.

Therefore, formatting of one track is performed in the order of the leading gap, multiple sectors, multiple alternative sectors, and the final gap.

In one track formatted in this way, for example, if sectors (51) and (53) have been reported to the magnetic disk controller as defective sector information from the host computer,
Deletion and alternative methods of 1) and <53) will be explained with reference to FIG.
) by sending the write sector (3) whose flag bit is on to the magnetic disk device, the sector is rewritten to one that differs only in the flag bit. As a result, the sector is deleted from the write/read use area as a bad sector. Furthermore, during the same index pulse, the write gate signal (2) is turned on at the time of the alternative sector (61), and when the sector (51)
Write data (3) having the same contents as before the sector was set as a bad sector is sent to the magnetic disk device (the flag bit is off). As a result, the sector (51) is replaced by the 0th-order index index (61).
Between pulses, the defective sector (53) is replaced by a replacement sector (62) using the same method as described above.

By the above method, even if there is a medium defect on the magnetic disk, the defective sector is replaced with a defect-free replacement sector, so that the user can use the magnetic disk as if it were a defect-free disk.

[Problems to be Solved by the Invention] Since the formatting method of the conventional magnetic disk device was performed as described above, there is no problem if the defect in the defective sector is located outside the ID section.
If there is a defect in the part, normal replacement processing may not be possible. This is because by turning on the flag bit in the ID section, the sector is set as a bad sector and erased from the write/read area.If this bit is defective and the flag bit is turned on, However, if this bit is off at the time of reading, it becomes impossible to distinguish between the current sector and the alternative sector to which the current sector is being replaced. As mentioned above, in the conventional formatting method of a magnetic disk device, there is a problem in that if a defect is located in the ID section, normal replacement processing may not be possible.

The present invention has been made to solve the above-mentioned problems, and its object is to provide a formatting method for a magnetic disk device that can normally carry out alternative processing no matter where a medium defect is located.

[Means for Solving the Problem] The formatting method of the magnetic disk device according to the present invention includes address information of the sector (the cylinder number, the head number, and the sector number) in the ID section of each sector. Address information of the replacement destination (replacement destination cylinder 1 number, replacement destination head number, and replacement destination sector 1 number) is recorded, and in the case of a defective sector, a plurality of the above ID parts are recorded in the area of the defective sector. It is.

[Operation] In the present invention, when a host computer requests access to a defective sector, first the first ID part of the defective sector is read to detect an error. If an error is detected, the next ID section is read and error detection is performed. As long as an error is detected,
Perform the above processing. If no error is detected, I
The alternative sector is accessed and data is written or read based on the alternative destination address information in part D.

As a result, normal replacement processing can be performed no matter where the medium defect is located in the defective sector.

[Example] Hereinafter, an example of the present invention will be described with reference to the drawings. 1st
The figure shows an example of the sector allocation for one truck when an alternative sector is provided according to the present invention.
Figure (a) shows the internal allocation of each sector of the present invention, Figure 2 (b) shows the internal allocation of the ID part therein, and Figure 2 (c) shows the internal allocation of each sector. The bit allocation for FLA in Fig. 2(b) is an example, Fig. 3 shows an alternative processing example in the present invention using flow lights, and Fig. 4 shows an example of defective processing. This figure shows an example of formatting a part of a sector. FIG. 5 shows the result of performing alternative processing for a defective sector according to the present invention. FIG. 6 is a flowchart showing an example of write/read processing according to the present invention. This is shown in . In the figure, (1) is the index pulse signal, (2) is the write gate signal, (3) is the write data, (4) is the gap 1, (51) to (5n) are the sectors, (61), (6
2) is the alternate sector, (7) is the gap 4, (8) is the synchronization pattern, (9) and (91) to (9n) are the ID
(10) is gap 2, (11) is data address mark, (12) is data, <13) is CRC code, (14) is gap 3, (15) is ID address/
Mark, (16) is the cylinder number, (17) is the head number, (18) is the sector number, (19) is the ID part CR
C code, (20) is the FLAG, (21) is the replacement destination head number, and (23) is the replacement destination sector number.

Next, a method for formatting one track when this magnetic disk device receives a formatting command from a host computer will be explained in detail with reference to FIGS. 1 to 5. In synchronization with the index pulse signal (1) generated from the disk device once per rotation of the disk, the write signal (2) is turned on and the write data for one track (3) is sent to the disk device. By doing this, sectors for one track are allocated.In Figure 1, there is a leading gap (4) and a final gap (7) at the beginning and end of each track to provide some margin, and, n sectors (51
) to (5n) and two alternative sectors (61), (62
) is assigned. Figure 2 (a) shows the internal allocation of each sector, and since it is necessary to specify the sector when writing data to or reading data from this sector, each sector has a In addition to (12), additional information as shown in FIG. 2 (,) is written. This accompanying information includes 1) synchronization information (8) for synchronizing with the rotation of the disk;
2) ID part (9), 3) Pig address mark (11) for recognizing data (12), 4) Data (12)
) consists of an error code (13) for detecting or correcting errors that occur when reading or writing. Figure 2 (
b) shows the internal allocation of the ID section in Fig. 2(a); 1> FLAf for recognizing whether a normal sector is a defective sector; (20); 2) Track number. Cylinder number (16), 3) Head number (
17), 4) 1 to sector number in rack (18), 5
) It consists of a CRC code (19) for detecting or correcting errors that occur when reading and writing the ID section. FIG. 2(c) shows an example of the bit assignment of the FLAG section in FIG. 2<b>. Binary data 0 indicating that each sector is a normal sector is written in the FLAG portion of each sector (51) to (5n) in FIG. Therefore, formatting of one track is performed in the order of the leading gap, a plurality of sectors, a plurality of alternative sectors, and a final gap. For example, in one track formatted in this way, defective sector information is sent from the host computer.
FIG. 3 is a flowchart of an alternative processing example and FIG. 4 shows an alternative method for erasing sectors (51) and (53) when sectors (51) and (53) have been reported to the magnetic disk device.

This will be explained using FIG.

In FIG. 3, first, the write gate signal (2) is turned on only when there is a defective sector (51), and n IDs are written in the defective sector. As shown in FIG. 4, the n IDs are composed of n ID parts and n1 gap parts. The value of n is at least 2 or more, and the writing of the ID does not exceed the area of the defective sector. At this time, in the FLAG in each ID section (91) to (9n), as shown in FIG. 2(c), the first ID (91) is set to binary data 1, and the next ID (92) is set to Similarly, ID (9n) is set to binary data 2, and ID (9n) is set to binary data n to distinguish the sector from a normal sector without a defect that has binary data 0. In addition, the cylinder number (16) is the cylinder number of the sector (51), and the head number (17) is the cylinder number of the sector (51).
1), and the sector 1 number of the sector (51) is set in the sector 1 number (18). Furthermore, the replacement destination cylinder number (21) and the replacement destination head number (22) are added.
), the replacement destination sector number (23) contains the cylinder number, head number, and head number of the replacement destination sector (61), which is the replacement destination.
Set sector number. As a result, the sector is deleted from the write/read area as a bad sector.

Next, during the same index pulse as above, the write gate signal (2) is turned on only for the alternative sector (61), and the ID of the alternative sector is rewritten. This rewriting is performed as follows. I of the said alternative sector (61)
FLAG (20) in part D (9) is set to binary data 0 indicating a normal sector, and cylinder number one (
16) is the cylinder number of the alternative sector (61), and head number (17) is the cylinder number of the alternative sector (61).
The sector 1 number of the alternative sector (61) is set in the sector 1 number (18). Also, the replacement destination cylinder number (21) and the replacement destination head number (22)
, the cylinder number, head number, and sector number of the defective sector (51) that is the replacement destination are set in the replacement destination sector number (23). This allows us to know which sector the alternative sector is replacing. However, this
It doesn't matter if you set it or not. In addition, values that do not exist in the relevant disk device are set for the replacement destination cylinder number 1 (21), replacement destination head number (22), and replacement destination sector 1 number (23) in a normal sector that does not undergo replacement processing. If you set this value, it will not be confused with the value of the sector to be substituted.However, this is also an additional function. by this,
The sector (51) is replaced by an alternative sector (61) of 0 or more.The method of replacing the sector (51) with an alternative sector (61) using FIG. 3 was explained using the same method.・Defect sectors (5
3) is replaced by an alternative sector (62). FIG. 5 shows how defective sectors (51) and (53) are replaced by the above method.

In disk devices that have undergone such alternative processing,
The processing method when a write or read request is made from the host computer will be explained using FIG.

First, an ID whose FLAG value is O is searched for in the requested sector. If the sector is a normal sector that has not been subjected to alternative processing, this ID is found, writing or reading is performed in the sector, and the target processing is completed.

If the sector is a defective sector that has been replaced, this ID cannot be found, so perform the following process 1) F
Increase the LAG value by one and search for the ID again.

2) If there is no error in the ID section, enter the replacement destination cylinder number (
21), writes or reads to the alternative sector specified by the alternative destination head number (22) and the alternative destination sector number (23) to complete the target process.

3) If there is an error in the ID section and the value of FLAG is less than or equal to n, return to 1) and search for the next ID.

If it exceeds n (considering the error rate, it is extremely unlikely that a reading error will occur for all IDs),
The host computer is notified of the failure to read the defective sector and the process ends.

As described above, if this alternative processing method is used, normal writing or reading can be performed no matter where the medium defect is located in the defective sector.

Note that the formatting example of one track in the above embodiment is just an example; for example, it is not always necessary to set the alternative sector on the same track as the defective sector, and it is not necessary to set the alternative sector on the same track as the defective sector. Similar effects can be obtained even if the length is appropriately selected within the gist of the present invention.

[Effects of the Invention] As described above, according to the present invention, when formatting a defective sector specified by a host computer, the address information of the sector (the cylinder number , the relevant head number, and the relevant sector number) and the replacement destination address information (the replacement destination cylinder number, the replacement destination head number, and the replacement destination sector number), etc. are recorded, so that media defects can be avoided. This has the effect that normal replacement processing can be performed no matter where the defective sector is located. In addition, since the address information of the replacement destination is added to the ID section, even if the defective sector and its replacement sector are not on the same track (even if the cylinder number or head number is different)
This has the effect of enabling alternative processing.

[Brief explanation of drawings]

Figure 1 shows a case where an alternative sector according to the present invention is provided.
An example of sector allocation for tracks is shown in Figure 2 (
a) is a diagram showing the internal allocation of each sector of the present invention, FIG. 2(b) is a diagram showing the internal allocation of the ID section, and FIG. FIG. 3 is a flowchart showing an example of alternative processing according to the present invention, FIG. 4 is a diagram showing an example of formatting a part of a defective sector, and FIG. 6 is a flowchart track showing an example of write/read processing in the present invention; FIG. 7 is a track in which a substitute sector is provided in the conventional formatting method. Figure 8(a) is a diagram showing the internal allocation of each sector, Figure 8(b) is a diagram showing the internal allocation of the ID section, and Figure 9 is a diagram showing the internal allocation of each sector. FIG. 2 is a diagram illustrating an alternative processing method for defective sectors using a conventional formatting method and its implementation results. In the figure, (1) is the index pulse signal, (2
) is the write gate signal, (3) is the write data, (4)
is gap 1, (51) to (5n) are sectors, (61
) (62) is an alternative sector, (7) is a gap 4, (8
) are synchronization patterns, (9) and (91) to (9n)
is the ID section, (10) is the gap 2, and (11) is the data section.
Address mask, (12) is data, (13) is CRC
code, (14) is gap 3, 15) is ID address mark, (16) is cylinder 1 number, (17) is head number, (18) is sector 1 number, (19) is ID part CRC code, ( 20) is FLAG, (21> is the replacement destination cylinder number, (22) is the replacement destination head number, (23)
) is the replacement destination sector number. In addition, in the figures, the same reference numerals indicate the same or equivalent parts. Fig. 3, Fig. 11, Fig. 2 (P/f-), Fig. 4, Fig. 5, Fig. 6, Fig. 7, Atsushi, Fig. 8.

Claims (1)

[Claims] A formatting method in which a plurality of sectors and alternative sectors are set in each track on a disk surface by formatting, and the formatting method sets a plurality of sectors and alternative sectors in each track on a disk surface, and in accordance with a defective sector specified by a host computer, the defective sector is A formatting method for a magnetic disk device, characterized in that at least two or more ID sections and a gap section are embedded, and the ID section is provided with address information of the sector and address information of an alternative destination.