JPS63149879A - Alternative processing control method - Google Patents

Abstract

(57) [Abstract] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

[Detailed Description of the Invention] [Summary] When a recording medium has a defect, a replacement portion is allocated to the defective portion, and data is recorded in the replacement portion by a disk control device that controls a disk device that records data blocks. A memory for recording data blocks is provided, and this memory can be accessed to prevent an increase in access time due to a seek operation to alternate parts of a disk device.

[Industrial application field]

The present invention relates to replacement processing based on a recording medium defect in a disk device, and particularly to a replacement processing control method that enables shortening of the access time to the disk device when performing replacement processing.

Currently, the processing speed of central processing units in computer systems has improved significantly, and the disparity in processing speed with input/output devices connected to this central processing unit is increasing. In particular, disk devices are equipped with a mechanical running system that performs head seek operations, and it is difficult to increase the speed beyond that level, and at present, speeding up of the running system that satisfies m has reached its limit.

On the other hand, the current manufacturing method for disk device recording media is
It is not possible to produce a perfect product without defects such as scratches. Furthermore, while using a disk device, defective parts may occur on the recording medium, and replacement processing in which a replacement part is allocated to the defective part and data is recorded is
This is an important process for disk devices.

However, when this alternation process is performed, seek operations are often involved in small disk devices, and the access time to the disk device becomes longer. It has been demanded.

[Conventional technology]

FIG. 2 is a diagram illustrating the allocation of conventional replacement processing.

FIG. 2(1) shows a data block recorded on a track in the user area, and FIG. 2(b) shows a data block subjected to alternation processing in an alternation area.

In FIG. 2 (8), each data block ■, ■, ■,
(2) each consists of an index (ID) and data. Then, in the ID, information indicating that the data block is a defective data block when the location where the data block is recorded is a defective portion of the medium, and information about the alternate allocation are written.

Here, if the data blocks ■ and ■ are defective data blocks, the replacement process is performed, and as shown by arrow A, the data block ■ is replaced by the data block ■ in FIG.
As shown by arrow B, data block ■ is allocated to
Allocated to data block ■.

In the IDs of the alternately processed data blocks (2) and (2), it is written that the alternate allocation has been completed and which defective data block is to be allocated first.

FIG. 3 is a diagram illustrating the access route after replacement processing.

FIG. 3(a) shows data blocks recorded on tracks in the user area, and FIG. 3(b) shows data blocks subjected to alternation processing in the alternation area.

Here, a case will be described in which data block ① to data block [phase] are read.

The disk control device, which receives a read command from data block ■ to data block [phase] from the central processing unit of the computer system, first accesses data block ■, reads the ID information of data block ■, and reads data block ■ determine whether it is normal or not.

If the data block (2) is normal, the disk controller reads the data and transfers it to the storage device of the computer system. Next, data block (2) is accessed, 10 information is read out, and it is determined whether or not it is normal.

If the location where data block ■ is recorded is a defective part, the disk control device records data block ■ as an error because it is a defective data block on the FD and the previous information has been written in the alternate allocation. Then, the alternate allocation is calculated and the data block (■) in the alternate area is accessed.

In small disk devices, the user area and alternate area may be provided in different cylinders, and the data block ■
In order to access, a seek operation of the head is required. Therefore, when the head is positioned at data block ■ in the alternate area, the disk controller continuously outputs ID information, and after confirming that alternate allocation of data block ■ comes first, reads the data in data block ■. and transfer it to the storage device.

The disk controller instructs the head to seek again in order to access the data block ■ in the user area, and when the head is positioned at the data block ■, it reads 10 of the data block ■, and if it determines that it is normal, transfers the data. and access the next data block [phase].

If the data block [phase] is abnormal, the data block @ in the replacement area is accessed in the same manner as described above. Then, read the ID of data block @, and read the ID of data block
It is confirmed that the alternate allocation of [phase] is performed first, and the data of data block 0 is transferred.

[Problem that the invention seeks to solve]

If there is no defective data block within the instruction processing range in which the disk control unit receives a read command from the central processing unit, the read processing is completed by sequentially transferring the data of each data block. If the head does not cross between cylinders, there is no need for a seek operation of the head. Therefore, it takes less time for the central processing unit to transfer the required data to the storage device.

However, if a defective data block exists within the instruction processing target range, as described above, for the defective data block,
The alternately allocated data block is accessed, but in this case, the alternate allocation calculates the previous location based on the ID information, performs a head seek, and then the alternate allocation reads [D] of the previous data block and moves to the target data block. After confirming that the data is the same and transferring the data, alternating allocation involves seeking from the previous data block to the data block for the next process, which increases the time required by the central processing unit to transfer the data. There is a problem that the processing efficiency of the computer system decreases.

[Means for solving problems]

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

This embodiment shows a circuit of a disk control device, in which 1 is an interface circuit for a host device such as a channel, 2 is an interface circuit for the disk device, 3 is a control circuit for controlling the entire disk control device by a microprogram, 4 to 6, and 9 is a multiplexer for switching the data transfer destination, and 7.8 is a register for temporarily storing data.

10 is a comparison circuit that compares whether the ID of the data block specified by the command sent from the host device matches the TD of the defective data block stored in the memory 11; 11 is the ID of the defective data block; 12 is a counter that indicates the addresses of the memory 11 and memory 13, and 13 is a memory that stores the data of the defective data block.

Based on the control of the control circuit 3, a counter 1 is stored in the memory 11.
The content of the ID of the defective data block is written at the address indicated by 2, and the address at this time is also sent to the memory 13, and the comparison circuit 10 writes the ID of the data block sent by the host device and the memory 11. When a match is detected with the ID of a defective data block stored in the memory, in the case of a write command, the data sent by the host device is written to the memory 13 at the address indicated by the counter 12, and in the case of a read command, the data sent from the memory 13 is written. The configuration is such that the data is read and sent to the host device.

[Effect]

With the above configuration, by recording the ID of the defective data block in the memory 11 in advance, the control circuit 3 compares whether the write/continue data block instructed by the host device is a defective data block. Since it is notified from the circuit 10, if it is a defective data block, the counter 12
Since it becomes possible to write data to or read data from the memory 13 at the address indicated by the data block, there is no need to write or read defective data blocks to or from the disk device, thereby increasing processing efficiency. I can do it.

〔Example〕

In FIG. 1, when the control circuit 3 detects power-on,
Controls the interface circuit 2 to read defective block information written to the disk device during manufacturing via terminal C;
When the data is fetched through the multiplexer 4, the ID of the defective data block is calculated, stored temporarily in the register 7 through the multiplexer 5, and sequentially written into the memory 11 at the address indicated by the counter 12.

The control circuit 3 stores the ID written in the memory 11 in the counter 12.
Based on the address information, the disk device is controlled via the interface circuit 2 and the terminal C, data is read from the replacement data block of the defective data block, and the multiplexer 5 and register 7
Then, the address specified by the counter 12 corresponding to the ID is written into the memory 13.

Next, the control circuit 3 connects the terminal A via the interface circuit 1.
When an instruction is received from a higher-level device, and this instruction is for reading, the ID of the data block specified by this instruction is stored in the register 7 via the multiplexer 5,
The comparison circuit 10 compares the ID read from the memory 11 at the address indicated by the counter 12. That is,
Compare each address information.

If the comparison results do not match, the counter 12 is incremented, the ID of the next defective data block is read from the memory 11, and compared with the ID stored in the register 7. When this operation is repeated and the counter 12 sends a carry signal, the control circuit 3 determines that the data block that the host device requests to read is not a defective data block in the disk device, resets the counter 12, and resets the ink block. The corresponding data block is read out from the disk device through the terminal C through the interface circuit 2, and the multiplexer 4, control circuit 3, interface circuit 1, and terminal B
The data is sent to the host device via .

If the above comparison results match, the control circuit 3 determines that the data block requested to be read by the host device is stored in the memory 13.
It is determined that the data is stored in the register 8, the value of the counter 12 at that time is fixed, the data is read from the memory 13, and is stored in the register 8 via the multiplexer 9. , terminal B
The data is sent to the host device via .

When the control circuit 3 receives a write command from the host device via the interface circuit 1, it stores the ID in the register 7 as in the case of a read command, and compares it with the ID read from the memory 11 at the address indicated by the counter 12. 10 for comparison.

If the comparison results do not match, the counter 12 is incremented, the ID of the next defective data block is read from the memory 11, and compared with the ID stored in the register 7. When this operation is repeated and the counter 12 sends a carry signal, the control? '11 The circuit 3 determines that the write position of the data block requested by the host device is not the position of a defective data block in the disk device, that is, a defective part of the recording medium, and resets the counter 12, 1, the corresponding data block is sent from the host device via terminal A, stored in register 7 via multiplexer 5, and sent to the disk device via multiplexer 6, interface circuit 2, and terminal A.

If the above comparison results match, the control circuit 3 determines that the data block that the host device requests to write needs to be written in the position of the defective data block that requires replacement processing in the disk device. , the data is temporarily stored in the register 7 via the multiplexer 5, and then the value of the counter 12 at that time is fixed and the data is written from the register 7 into the memory 13.

The above processing is executed every time a write/continue command is received from the host device, and when instructed to turn off the power, the counter 12 is incremented, the ID of the memory 11 and the data of the memory 13 are read, and the multiplexer 9 , register 8
, multiplexer 6, interface circuit 2, and terminals to the disk device, and based on the ID information in the memory 11, the data in the memory 13 is written and updated in the alternate area of the disk device, and the process ends.

〔Effect of the invention〕

As explained above, when the present invention receives a write/continue command from a host device and there is a defective data block within the processing range, the present invention performs write/continue processing in the memory provided in the disk control device. This eliminates the need to access the disk device directly, making it possible to shorten write/success processing time and prevent a decrease in processing efficiency of the computer system.

[Brief explanation of the drawing]

Fig. 1 is a schematic diagram of a circuit showing an embodiment of the present invention, Fig. 2 is a diagram explaining the allocation of conventional alternation processing, and Fig. 3 is a diagram explaining the access route after alternation processing. . In the figure, 1.2 is an interface circuit, 3 is a control circuit, 4, 5, and 6.9 are multiplexers, and 7
．． 8 is a register, 10 is a comparison circuit, 11.13 is a memory, and 12 is a counter.

Claims (1)

[Claims] If there is a defective part in a recording medium that records a data block consisting of an index and data, a replacement part is allocated to the defective part, and the data block to be recorded in the defective part is assigned to the replacement part. In a disk control device that controls a disk device for recording, a first memory (1) stores an index of a data block to be recorded in the defective portion, which is created from information indicating a defective state of the medium recorded in advance on the recording medium.
11), a second memory (13) for storing data of a data block to be recorded in the defective part, a counter (12) for simultaneously indicating the addresses of the first memory and the second memory, and an upper a comparison circuit (which compares the index of the data block sent out by the device and the index stored in the first memory (11));
10), and when the power is turned on, the first memory (11) is transferred from the disk device to the first memory (11).
) stores an index of the data block to be recorded in the defective portion, reads it at the address indicated by the counter (12), and compares it with the index of the data block sent from the host device by the comparison circuit (10); If they do not match, the counter (12) is incremented and the first memory (
11), and when the counter (12) sends out a carry signal, the counter (12) is reset, and at the time of writing, the data block sent by the host device is sent to the disk device. When reading data, the data block is read from the disk device and sent to the host device, and the comparison circuit (
When the comparison result of 10) matches, the corresponding counter (12)
The address to be sent by the second memory (
13), and when reading, the second memory (13)
Performs processing to send the data read from to the higher-level device,
Alternative processing control characterized in that when the power is turned off, the contents of the second memory (13) are written to the alternate part of the disk device based on the index of the first memory (11) to update data. method.