JPS58109954A - Magnetic disc controller - Google Patents

Abstract

PURPOSE:To minimize the number of times of re-reading as a whole and to improve the processing speed, by a means which stacks the information of a readout error sector and summarizes re-readout of a sector having an error. CONSTITUTION:In writng data of all sectors of an arbitrary track 2 to a main storage device 7, when a sector 2a of the track 2 is written to the device 7, the production of readout error is detected in the sector 2a. Then, the address of the device 7 writing the sector 2a is stacked to a readout error sector information stack area 10. A magnetic disc controller 3 writes data from a sector 2B to the device 7 continuously. When a sector 2b is written, and the production of readout error to the sector 2b is detected, the similar operation is repeated as the operation of the sector 2a. When a magnetic disc storage medium 1 turns once, the sectors 2a, 2b only are written in the said area of the device 7 again and afterwards, as for the re-readout, the said operation are repeated.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a magnetic disk control device that controls data transfer between a main storage device and a magnetic disk device.

Figure 5 is a block diagram showing an example of a conventional magnetic disk control device. In the figure, IIJ is a magnetic disk device recording medium, (2I is any track on the storage medium, (2A) t (2B), (2a), (2b
) is each sector on this track, (31 is the magnetic disk control training, (41 is the memory address color, +51t
161 is an error detection circuit that detects errors in the data recorded in this data register; 71 is a data register sent from the magnetic disk controller (3: This is the main storage device that stores stored data.

A flowchart of data transfer in the magnetic disk control device configured as described above is shown in FIG.

Next, the operation will be explained. Magnetic disk control device 13
+ is preset and next memory address counter (
The data storage address indicated by 41 is stored in the main memory (71
Then, one word of data from an arbitrary track (21) of the magnetic disk device recording medium fil is transferred.

Assume that a continuation error occurs in sector (2a) after sending sector (2A). A read error in sector (2) is detected by the error detection circuit 16) after sending out the last word of data in sector (2a), and the magnetic disk control device (3) corrects the transfer from sector (2B) onward. Then, the read operation for sector (2a) is executed again. If sector (2a) can be read, transfer from sector (2B) onwards is continued.

FIG. 3 is a diagram showing the relationship between the number of rotations of the medium and the transfer sector number in this case, and shows the case where both sectors (2a) and (2b) can be continued for the first time after three read operations each. As shown in the figure, the magnetic disk drive recording medium requires five revolutions until the data of track (2) is completely transferred to the main storage device (7+) by the above method.

In other words, as mentioned above, a rotational waiting time of one revolution is required each time a read error occurs, and if a read error occurs in multiple sectors in one track, the rotational waiting time is proportional to the number of sectors in which an error occurs. The disadvantage was that it increased.

This invention was made in order to eliminate the above-mentioned drawbacks of the conventional device, and even if a sector in which a read error occurs in one track, the rotational waiting time is reduced by the number of sectors in which the read error occurs. The purpose of this invention is to obtain a magnetic disk control device that does not depend on the magnetic disk.
9, fil~171, (2A)-(28)e
(2a) and (2b) indicate the same or corresponding parts as the same reference numerals in FIG. 110 is a read error sector information stack area. The fifth factor is a flowchart of data transfer by the device in FIG.

In the magnetic disk control device configured as described above, when data of all sectors of an arbitrary track + 2) is stored in the main storage device (7), the data of the sector (2a) is continuously transferred from the first sector of track (2). Suppose that it is detected that a read error has occurred in sector (2a) at the time when the data up to and including It is stacked in the error sector information stack area IIG.Magnetic disk i control device (31 is a sector (
2B) to the main memory (7). When writing to sector (2b), it is detected that a read error has occurred in sector (2b), and the same operation as for sector (2a) is repeated. Magnetic disk device recording medium (
When the track 1) rotates once, all the data on the track 21 is written to the main storage device (7), and the magnetic disk controller +31 writes the data to the main storage device (the area indicated by the line on 71, that is, the sector (2a), It is recognized from the error information stack area IIG that the data in the area where (2b) is written is invalid, and only sectors (2a) and (2b) are read out again and written to the corresponding area of the main storage device (7). For re-reading, the above-described operation is repeated. FIG. 6 is a diagram showing the relationship between the number of rotations of the medium and the transfer sector number.

In the figure, if both sector (2a) and sector (2b) are sectors that can be read normally after retrying the read operation three times, all sectors of track (2) will be The data can be normally written to the main storage device (7). - In the above embodiment, a magnetic disk control device was shown that writes data from the magnetic disk device to the main storage device, but the data is attached to each sector. By utilizing the present invention when reading address information, it can be similarly utilized for a device that writes data in a main storage device to a magnetic disk device.

As explained above, the present invention has the advantage that by stacking the information of the read error sectors and rereading the error sectors all at once, it is possible to minimize the number of rereadings in a comprehensive manner and improve the processing speed. There is.

[Brief explanation of the drawing]

Fig. 1 is a block wiring diagram showing an example of conventional magnetic disk control electric manufacturing, Fig. 2 is a flowchart of data transfer by the device shown in Fig. 1, and Fig. 3 is a diagram of the magnetic disk device recording medium of the device shown in Fig. 1. FIG. 4 is a block diagram showing an embodiment of the present invention; FIG. 5 is a flowchart of data transfer by the device shown in FIG. 2; FIG.
This figure is a diagram showing the relationship between the rotational speed of the magnetic disk drive recording medium of the apparatus shown in FIG. 2 and the transfer sectors. In the figure, (1) is a magnetic disk device recording medium, (2)
is an arbitrary track on the recording medium, (31 is a magnetic disk controller, +j+ is a memory address counter, (5) is a data register, 161 is an error detection circuit, (7) is a main UL device, tl (l is a readout This is the error sector information stack area. The same reference numerals in each figure indicate the same or corresponding parts. Agent Shin Kuzuno -

Claims (1)

[Scope of Claims] (11) In a magnetic disk control device that continuously checks data spanning multiple sectors on a magnetic disk device into a main storage device, the main device reads data track by track from the magnetic disk device. means for writing each sector into a corresponding area in a storage device; an error detection means for determining the acceptability of each sector of the above-mentioned read data; a means for stacking information of sectors in which an error has been detected by the error detection means; and this stack. means for rewriting only the information of the sector read out from the magnetic disk and rewriting the area corresponding to the sector on the main storage device; 121. A magnetic disk control device characterized by comprising a means for repeating the above-mentioned readout and detecting the data of the next track if no error is detected. The means for stacking the information of the detected sectors stores the information in a magnetic disk control device.
The magnetic disk control device described in . (3) The magnetic disk control device according to claim 1, wherein the means for stacking nine sectors of information after an error is detected by the error detecting means stores the information in a main storage device.