JPS62235662A - Magnetic disk controller - Google Patents

Abstract

PURPOSE:To make data release processing at a central processor unnecessary by reading out automatically the same data from a magnetic disk device, in case an error is detected at the time of reading out from a disk cache. CONSTITUTION:When an instruction for reading out a data from a magnetic disk device 30 is outputted to a magnetic disk control device 20 from a central processor 10, an instruction interpreting and executing means 21 issues an instruction for reading out a data from a disk cache 22, to a disk cache controller 27. As a result, the data which is read out is inputted to a data selecting means 23. In such a case, when an error is detected by a disk cache error detecting circuit 24, the controller 27 reads out the same data from the magnetic disk device 30, and the data which is read out is transferred to the processor 10 through a data buffer 26, and the selecting means 23. On the other hand, in case no error is detected, the selecting means 23 transfers the data from the disk cache 22, to the processor 10.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a magnetic disk control device, and particularly to automatic correction means when reading data.

[Conventional technology]

Conventionally, in an information processing system consisting of a central processing unit, a magnetic disk unit, and a magnetic disk control unit, damage to the magnetic disk medium or malfunction of the magnetic disk control unit occurs during the process of reading data from the magnetic disk unit and transferring it to the central processing unit. In some cases, the correct data could not be transferred to the central processing unit. In such a case, the central processing unit cannot continue the process, so the process is continued after recovering the file using a previously stored file on another medium.

For example, if files in a magnetic disk device are stored on a magnetic tape before processing starts, and updated data during the processing process is sequentially stored on other magnetic tapes, an unrecoverable failure of the magnetic disk device occurs. In this case, the above two magnetic tapes are used to reproduce the state of the file before the failure occurred on a normal magnetic disk device.

[Problem that the invention seeks to solve]

However, the above-mentioned conventional method has the drawback that the recovery process takes a considerable amount of time υ, and the user is forced to wait until the recovery process is complete.

An object of the present invention is to automatically read the same data from the magnetic disk device and automatically send it to the central processing unit when a failure occurs in the disk cache medium in a magnetic disk control device with a built-in disk cache. The system sends correct data that has been corrected, eliminating the need for recovery processing.

[Means for solving problems]

The magnetic disk control device of the present invention includes a disk cache error detection circuit that detects an error when reading data from a disk cache, and a disk cache error detection circuit that detects an error when reading data from a disk cache. - A disk cache controller that performs automatic head start correction, and an automatic correction mode storage means that stores whether or not automatic correction is to be performed based on a command from the central processing unit.

〔Example] Next, the present invention will be explained with reference to the drawings.

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

A command interpretation/execution unit 21 built into the magnetic disk control device 20 interprets and executes commands from the central processing unit 10 to the magnetic disk control device 20 .

The disk cache 22 is a medium-sized Yasshi memory that enables high-speed reading of magnetic disk data.

The data buffer 26 is a buffer that temporarily stores read data from the magnetic disk device 20.

The data selection means 23 selects either the data of the disk cache L22 or the data buffer 26 as the read data to be transmitted to the central processing unit 10, and selects the data from the disk cache controller 23. 27.

The disk cache error detection circuit 24 is a circuit that detects errors when reading data from the disk cache 22.

The automatic correction mode storage means 25 stores a mode of whether or not to automatically read the same data from the magnetic disk device 30 when the disk cache 22 is detected in response to a read command from the central processing unit 10. The entire instruction interpretation/execution means 21 is set/reset by an instruction from the central processing unit 10.

The disk cache controller 27 receives the cache & read instruction from the instruction interpretation/execution means 21 and reads the data from the disk cache 22 if there is data in the cache 22 in the disk. The magnetic disk drive 30 also performs control to read data from the magnetic disk drive 30 when the automatic correction mode storage means 25 is on when an error is detected by the disk carriage and error detection circuit 24.
Control is performed to output data f:R.

Next, the operation of this embodiment will be explained.

When the central processing unit 10 issues an automatic correction mode setting command to the magnetic disk control bag [20], the automatic correction mode storage means 25 is set via the command interpretation and execution means 21. In this state, when the central processing unit 10 issues a command to the magnetic disk control device 2o to read data from the magnetic disk device 30, the command interpretation execution means 2
1 gives the controller 27 an instruction to read data from the disk cage &22. If there is data you are looking for in Disc Cage J-22.

The data is read out and input to the data selection means 23.

At this time, when an error is detected by the disk error detection circuit 24, the disk cache controller 27 detects an error because the automatic correction mode storage means 25 is set.
continuously outputs the same data from the magnetic disk device 30. The successive data from the magnetic disk drive 30 passes through the data buffer 26 and is input to the data selection means 23. The data selection means 23 selects the correct data from the data buffer 2 under the control of the disk storage controller 27.
6 is selected and all data is transferred to the central processing unit 10.

If the disk cache error detection circuit 24 does not detect all errors, data is not read from the magnetic disk group [30], and the data selection circuit 23 selects data from the disk cache 22 and transfers the data to the central processing unit 10. transfer.

〔Effect of the invention〕

As explained above, in a magnetic disk control device with a built-in disk cache, if an error is detected when reading the disk cache, the central processing unit can automatically read the same data from the magnetic disk device. This has the effect of completely eliminating the need for recovery processing.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing one embodiment of the present invention. DESCRIPTION OF SYMBOLS 10...Central processing unit, 20...Magnetic disk control device, 21...Instruction interpretation execution means, 22. Old disk cache, 23...
Data selection means, 24...Disk cache error detection circuit, 25...Auto correction mode storage means, 26...Data buffer, 27.=-f
disk cache controller, 28...Magnetic disk unit #, a output control circuit, 30...Magnetic disk unit. -1,

Claims (1)

[Claims] a disk cache; a disk cache error detection circuit that detects an error when reading data from the disk cache; and when the detection circuit detects an error, automatically reads data corresponding to the data in which the error has been detected from the magnetic disk device. 1. A magnetic disk control device comprising: a disk cache controller that performs correction; and automatic correction mode storage means that stores whether or not automatic correction is to be performed based on a command from a central processing unit.