JPH0437924A - Auxiliary storage control method - Google Patents

Abstract

PURPOSE:To reduce the seeking frequency by designating the corresponding alternate recording area and giving a write or read instruction to the designated area after all nondefective recording areas undergone the write or read operations. CONSTITUTION:A control means 114 decides whether plural recording areas designated by the write or read requests have the defects respectively or not. If so, the address of the corresponding recording area is held by a holding means 113 together with the holding information on the alternate area. Then the write or read operations are applied to the nondefective recording areas and then to the alternate recording area. Therefore the alternate area is never processed during the processing of a series of recording areas in a data area when the continuous recording areas are designated in the data area. Thus no seeking operation is needed when the processing is returned to the data area from the alternate area. Then the seeking frequency is reduced.

Description

[Detailed Description of the Invention] [Table of Contents] Overview Industrial Application Fields Conventional Technology Problems to be Solved by the Invention Means for Solving the Problems Action Examples Effects of the Invention [Summary] Transfer of data to the auxiliary storage device Regarding the auxiliary storage control method that controls writing and reading, the purpose is to reduce the number of seek operations, and the auxiliary storage device has a data area consisting of multiple recording areas and a replacement area, and one of the auxiliary storage devices. a storage means comprising a plurality of storage areas having a capacity corresponding to the recording area; and an address generation means for generating an address indicating a storage area of the storage means corresponding to each of the storage areas specified in a write request or a read request. , determines whether or not there is a defect in each of the specified recording areas, and if there is no defect, specifies the supplied address and the corresponding recording area, issues a write instruction or read instruction, and confirms that there is a defect. In some cases, a control means for holding reservation information regarding an address and a corresponding replacement recording area in a storage means and for suspending a write instruction or a read instruction for this recording area, and an auxiliary memory for storing reservation information regarding an address and a corresponding replacement recording area, and for holding a write instruction or a read instruction for this recording area; and an access means for performing a write operation or a read operation to a designated recording area of the device, and after the write operation or read operation for all recording areas determined to be free of defects is completed, the corresponding replacement recording area The configuration is such that a write instruction or a read instruction is given by specifying .

[Industrial Application Field] The present invention relates to an input/output control method between an auxiliary storage device such as a magnetic disk device and a processing device.

In magnetic disk drives, reliability is important, and if a defective part of the recording medium is detected during formatting, the disk control device that controls the magnetic disk drive will, for example, A sector in a spare track secured as a replacement area is allocated as a replacement sector in place of a sector.

[Conventional technology]

A magnetic disk, which is the recording medium of a magnetic disk device, consists of multiple cylinders. For example, one of these cylinders is reserved as a replacement area, and each track included in the other cylinders is used as a data area. ing.

For example, as shown in FIG. 5, if a defect is detected in sector 2 of trunk A in a data area consisting of n sectors, the disk controller will replace this sector 2 with a sector in the replacement area described above. One (for example, sector a) is allocated. Further, at this time, information indicating that sector 2 is defective and information indicating the allocated replacement sector (sector a) are recorded, for example, in the fault information recording area of the corresponding track.

In addition, when starting up the magnetic disk device, the disk control device reads the fault information recorded in the fault information recording area of each track, and records the defective sector and the corresponding replacement in the replacement information table provided in the memory. The correspondence relationship with sectors is stored. Furthermore, when an instruction is given to access a defective sector, access is made to sector a designated as a replacement sector based on this replacement information table.

Therefore, for example, when reading the data recorded in sectors 1 to 3 in FIG. 5, the disk controller first reads the data recorded in sector 1, and
Next, the data in sector a, which is a replacement sector for sector 2, is read out, and then the data in sector 3 is read out.

Also, when data is written to these sectors, the writing is performed in the same order.

[Problem to be solved by the invention]

By the way, in the conventional method described above, for example, the fifth
After the access to sector 1 in the figure is completed, seek processing is performed to search for sector a, and after the access to sector a is completed, seek processing is performed to search for sector 3. In this way, when accessing consecutive sectors including the sector to which the replacement sector has been allocated, seek processing is performed multiple times, resulting in the problem that the time required for the access processing is long.

An object of the present invention is to provide an auxiliary storage control method that reduces the number of seek processes.

[Means to solve the problem]

FIG. 1 is a block diagram of the principle of the present invention.

In the figure, an auxiliary storage device 101 has a data area made up of a plurality of recording areas and a replacement area made of a replacement recording area allocated to a defective recording area included in this data area.

The storage means 111 consists of a plurality of storage areas having a capacity equivalent to one recording area of the auxiliary storage device 101.

The address generating means 112 generates addresses that sequentially designate a plurality of storage areas of the storage means 111, corresponding to each of the plurality of recording areas specified by an input write request or read request.

The control means 114 determines whether or not each of the plurality of recording areas specified by the write request or the read request has a defect, and when it is determined that there is no defect, the address generation means 1
When a write or read instruction is given by specifying the address generated by step 12 and the corresponding recording area, and it is determined that there is a defect, the address corresponding to the corresponding recording area and reservation information regarding the replacement recording area are retained. Means 113
In addition, any write or read instructions for this recording area are held on hold.

The access means 115 accesses the storage means 1 according to the write instruction.
The data stored in the corresponding storage area of 11 is written to the corresponding storage area of the auxiliary storage device 101, and the data read from the corresponding storage area is stored in the corresponding storage area of the storage means 111 in response to the read instruction. .

After the write operation or read operation for all the recording areas determined to be defect-free as a whole is completed, the control means 114 controls the corresponding replacement recording area and It is configured to issue a write instruction or a read instruction by specifying an address.

[For production]

Address generation means 1 corresponds to each of a plurality of recording areas specified by an input write request or read request.
12, addresses that sequentially designate a plurality of storage areas of the storage means 111 are generated.

Further, the control means 114 determines whether or not each of the plurality of recording areas specified in the above-mentioned write request or read request has a defect, and when it is determined that there is no defect, the control means 114 A write instruction or a read instruction is given by specifying the address generated by the address generating means 112 described above and the corresponding recording area. When a writing instruction is given by this control means 114,
In response to this write instruction, the access means 115 writes the data stored in the corresponding storage area of the storage means 111 to the corresponding recording area of the auxiliary storage device 101. Further, when a read instruction is given, the access means 115 reads data from the corresponding recording area in accordance with the read instruction, and the read data is stored in the corresponding storage area of the storage means 111. Stored.

On the other hand, when it is determined that there is a defect, the address corresponding to the corresponding recording area and suspension information regarding the replacement recording area are held in the holding means 113, and write or read instructions for this recording area are suspended. Further, after the write operation or read operation for all the recording areas determined to be free of defects is completed, the control means 114 specifies and writes the corresponding replacement recording area and address based on the above-mentioned suspension information. A write instruction or a read instruction is issued, and in response, the access means 115 performs a write operation or a read operation.

In the present invention, a write operation or read operation is first performed on a recording area without defects among a plurality of designated recording areas, and then a write operation or read operation is performed on a replacement recording area in one line. be exposed.

Therefore, when consecutive recording areas within the data area are specified, processing of the alternate recording area is not performed in the middle of processing a series of recording areas within the data area. This eliminates the need for seek processing when returning from the alternate area to the data area, making it possible to reduce the number of seek processes.

[Example] Hereinafter, an example of the present invention will be described in detail based on the drawings.

FIG. 2 shows a configuration diagram of a disk control device according to an embodiment of the present invention.

In FIG. 2, 201 represents a host computer (host), 202 represents a magnetic disk device, and 210 represents a disk control device.
0 controls data exchange between the host 201 and the disk device 202.

Further, in the disk control device 210, 211 is a microprocessor (MPU), 212 is a memory, and 21 is a microprocessor (MPU).
3 indicates an interface control circuit, 214 a disk control circuit, 215 a data buffer, and 216 a buffer control circuit. MPU211 mentioned above
, memory 212, interface control circuit 213, disk control circuit 214, and buffer control circuit 216 are interconnected via a bus.

The magnetic disk device 202 is equipped with a magnetic disk consisting of a plurality of tracks, one of the above-mentioned plurality (D tracks) is secured as a replacement area, and the other tracks are used as a data area. Each of the above-mentioned tracks is composed of n sectors, and data is read and written in sector units.In addition, each sector has “11
They are numbered from number to rnj.

The data buffer 215 is the magnetic disk device 2 described above.
It is divided into m storage areas each having a storage capacity corresponding to each sector of 02, and each storage area has rl to r
They are numbered up to mJ.

The buffer control circuit 216 controls writing and reading of data to and from a storage area of the data buffer 215 designated by an instruction from the MPU 211.

The interface control circuit 213 processes communication between the host 201 and the disk control device 210, and the disk control circuit 214 processes the communication between the host 201 and the disk control device 210.
In response to an instruction from 11, data writing and reading operations to and from the disk device 202 are started.

Furthermore, the data read from the disk device 202 is
The data is temporarily stored in the data buffer 215 by the disk control circuit 214 via the buffer control circuit 216 mentioned above, and then read out by the interface control circuit 213 via the buffer control circuit 216, and the data is stored in the data buffer 215 via the buffer control circuit 216 described above.
1. At this time, MPU2
11 indicates a device pointer P to the buffer control circuit 216.
.

The host pointer Ph is supplied to indicate the storage area of the data buffer 215 that stores data from the disk control circuit 214, and the host pointer Ph is supplied to indicate the storage area to be read by the interface control circuit 213. There is.

Further, the data supplied from the host 201 is sent to the buffer control circuit 216 by the interface control circuit 213.
The data is temporarily stored in the data buffer 215 via the buffer control circuit 216, and then read out by the disk control circuit 214 via the buffer control circuit 216 and written to the disk drive W2O2. At this time, MPU211
supplies the above-described device pointer P and host pointer Ph to the buffer control circuit 216 to control the storage operation and read operation to the data buffer 215.

In addition, if a sector included in a track in the data area is defective, a sector in the above-mentioned replacement area is allocated to that sector, and the defective sector and its sector are recorded in the fault information recording section of the corresponding track. Failure information regarding sectors of the allocated replacement area is recorded.

For example, the number of the defective sector and the number of the sector in the replacement area allocated to that sector may be recorded as failure information.

For example, when starting up the magnetic disk device 202, the MPU 211 of the disk control device 210 starts an operation to read failure information (described later) recorded on the magnetic disk.
The replacement information table 221 in the memory 212 stores the information for each track.

When formatting a magnetic disk, etc., the number of the defective sector and the number of the replacement sector assigned to that sector are recorded in a predetermined area of each track, and this information is read out as failure information. Just do it.

In the following, the input/output control operation by the disk controller 210 will be explained separately into an operation for reading data from the disk device 202 and an operation for writing data.

FIG. 3(a) is a flowchart showing the operation of reading data from the magnetic disk device 202, and FIG. 3('b) is a flowchart showing the operation of transferring data from the data buffer 215 to the host 201. .

FIG. 4(a) shows data from the host 201 to the data buffer 21.
A flowchart representing the data transfer processing operation to 5 is shown in Figure 4 (
c) shows a flowchart representing the operation of writing data to the magnetic disk device 202.

For example, the host 201 selects four consecutive sectors rIJ, f2j, r3J of track “Aj.

The operation of reading data from the magnetic disk device 202 and the operation of transferring data to the host 201 will be described in the case where reading of data with "r4" specified is instructed.

In response to instructions from the host 201, the disk controller 2
The MPU 211 of No. 10 first uses the device pointer Pa.
Initial values are set for and host pointer P. for example,
An initial value r indicating the storage area number rIJ of the data buffer 215 is set in the device pointer P4 and the host pointer P1.
It is sufficient to set lJ. At this time, the failure pointer P (described later) also has an initial value r1. is set (step 301. step 321).

Next, the MPU 211 selects the sectors r11 to “4” mentioned above.
The replacement information table 221 in the memory 212 is sequentially referred to for each sector, and it is determined whether each sector is a normal sector with no defects (step 302). For example, if the above-mentioned replacement information table 221 does not store the above-mentioned number rl indicating the sector rlJ,
The MPU 211 determines that this sector "IJ" is a normal sector.

In this way, if the determination in step 302 described above is affirmative, the MPU 211 controls the disk control circuit 2.
14 to perform a read operation for the designated sector (step 303). According to this instruction,
The disk control circuit 214 activates the read operation by the disk device 202, and the buffer control circuit 216 stores the read data in the storage area fIJ of the data buffer 215 indicated by the device pointer P. .

After that, the MPU 211 uses the device pointer P mentioned above.
The next number (for example, number r2j) of the storage area indicated by 4 is set in the device pointer P (step 304).

Further, in step 305, the MPU 211 determines whether or not a defective sector is included among the sectors processed so far, and if the determination is negative, the MPU 211 returns the fault pointer in the same manner as in step 304 described above. In step 3o6), if the determination is affirmative, the failure pointer P0 is not updated.

Therefore, when normal sectors are specified consecutively, each sector is sequentially read by the disk control circuit 214 and stored in the data buffer 215, and the device pointer P4 and fault pointer P are sequentially read. Updated.

On the other hand, for example, sector r3. If there is a defect in sector r3J and sector ra in the replacement area is allocated as a replacement sector to this sector r3J, the number r3J indicating sector 31 is stored in the replacement information table 221. In this case, the MPU 211 It is determined that sector r3j is defective, resulting in a negative determination in step 302 described above.

In this case, the MPU 211 provides a fault pointer storage section 222 in the memory 212, and stores the fault pointer P at this time.
, and the sector determined to be defective (for example, sector r3
Step 3: Store the number of the replacement sector corresponding to J).
07), and thereafter, similarly to step 304, the device pointer P is updated (step 308), and the next sector (for example, sector r41) is processed.

In this way, for a defective sector, the value of the fault pointer P is held in the fault pointer storage unit 222, the read process for the corresponding replacement sector is skipped, and the read process for the normal sector is performed first. conduct. The value of the fault pointer P described above indicates the storage area of the data buffer 215 corresponding to the sector determined to be defective.

Here, in step 305, the MPU 211 refers to the above-mentioned failure pointer storage unit 222, and based on whether or not the failure pointer P0 is stored, determines whether there is a sector that has been determined to be defective and has skipped the read process. What is necessary is to determine whether or not this is the case.

Furthermore, the MPU 211 repeats the processing from step 302 to step 308 described above for all sectors designated by the host 201, and when it is determined in step 309 that the processing for all sectors has been completed, the MPU 211 returns the fault pointer described above. Fault pointer P stored in storage unit 222.

The replacement sector is processed based on the information regarding the replacement sector and the replacement sector (step 310). After that, MPU21
1 is the device pointer P to the fault pointer P0 mentioned above.
The dO value is set (step 311), and the process ends.

By the way, in parallel with the data storage operation by the disk control circuit 214 described above, the interface control circuit 21
3, data is being read from the data buffer 215.

In step 306 described above, while the failure pointer P is being sequentially updated, the MPU 211 determines that the host pointer PbO value described above is smaller than the failure pointer P, resulting in an affirmative determination in step 322.

In this case, the buffer control circuit 216 supplies the contents of the storage area of the data buffer 215 addressed to the host pointer P to the interface control circuit 213,
In response, the interface control circuit 213
Data transfer processing to the host 201 is performed (step 323).

At this time, the MPU 211 updates the host pointer P1 and determines whether or not the transfer processing for all sectors in step 324L has been completed (step 325), and continues until an affirmative determination is made in step 325, as described in steps 322 to 324L. Repeat step 325.

On the other hand, in the case of a negative determination in step 322 described above, step 322 is repeated.

Therefore, in this case, the transfer operation by the interface control circuit 213 is temporarily stopped, and then restarted when the replacement sector processing is completed in step 310 described above and the failure pointer P0 is updated. The data in each sector after r3J is stored in the host 201.
will be forwarded to.

In this way, the read data corresponding to sectors "11 to r4J" are sequentially transferred to the host 201.

The operation of writing data from the host 201 to sectors rla to sector r41 of track rAJ of the above-mentioned magnetic disk device 202 will be described below.

First, the MPU 211 performs step 301.32 described above.
Similarly to 1, host pointer Pb.

An initial value r14 is set for each of the device pointer Pd+failure pointer P (steps 401 and 421).

The data supplied from the host 201 is sequentially sent to the buffer control circuit 216 by the interface control circuit 213.
The buffer control circuit 216 stores the data in the storage area of the data buffer 215 indicated by the host pointer P (step 402). Furthermore, after the processing in step 402, the MPU 211 uses the host pointer P
h is updated (step 403), and the above-described steps 402 to 404 are performed until it is determined in step 404 that the processing for all sectors has been completed.

Furthermore, when the data storage operation to the data buffer 215 described above is started, the device pointer P becomes smaller than the host pointer Ph, so an affirmative determination is made in step 422, and the data reading process from the data buffer 215 and Writing processing to the magnetic disk device 202 is started.

Based on the replacement information table 221, the MPU 211
Step 42: Determine whether or not the sector designated as the write destination for the data stored in the storage area corresponding to device pointer P4 is a normal sector with no defects.
3) If the determination is negative, write processing for the corresponding sector is skipped. In this case, similar to step 307 described above, the fault pointer P1 and the replacement sector number are stored in the fault pointer storage unit 222 (step 424), and only the device pointer P4 is updated similarly to step 308 ( step 425).

On the other hand, in the case of an affirmative determination in step 423 described above, the buffer control circuit 216 supplies the data stored in the storage area corresponding to the device pointer Pd to the disk control circuit 214, and this data is written to the specified sector (step 4
26). Furthermore, after the process of step 426 is completed, M
The PU 211 updates the device pointer Pa (step 427).

Furthermore, if it is determined in step 428 that there is no sector for which write processing has been skipped up to that point (in the case of a negative determination in step 428), the MPU 211 updates the failure pointer P (step 429), and In the case of an affirmative determination in 428, the failure pointer P
, performs processing on the next sector without updating.

Further, when it is determined that the processing of steps 422 to 429 for all sectors has been completed, an affirmative determination is made in step 430, and the failure pointer P stored in the failure pointer storage unit 222 and the number of the replacement sector are Based on this, write processing is performed for the replacement sector (step 431), and the device pointer PdO value is set in the failure pointer P1 (step 432), after which the processing ends.

As mentioned above, based on the substitution information table 221,
It is determined whether a sector designated as a target for a read operation or a write operation is a defective sector, and read processing and write processing for the defective sector are skipped. Also, at this time, the fault pointer P indicating the storage area of the data buffer 215 corresponding to the sector containing the defect and information regarding the replacement sector corresponding to the sector described above are held in the fault pointer storage unit 222.

Further, after all the processing for normal sectors is completed, read processing or write processing for replacement sectors is performed based on the failure pointer storage unit 222.

Therefore, even if a defective sector is included in consecutive sectors, all specified sectors can be read or written without performing seek processing when returning from the replacement area to the normal data area. It is possible to carry out processing.

This makes it possible to reduce the number of seek processes and shorten the time required to access the magnetic disk device 202.

Note that there is no limit to the number of fault pointers P8 stored in the fault pointer storage section 222 described above, and a plurality of fault pointers P0 may be stored.

In this case, step 309 and step 4 described above
In step 29, all replacement sectors stored in the failure pointer storage unit 222 are sequentially processed, and after these processes are completed, the value of the device pointer P is set to the failure pointer P0.

〔Effect of the invention〕

As described above, according to the present invention, when consecutive recording areas are specified, processing of the replacement area is not performed between processing of the consecutive recording areas, and processing is performed from the replacement area to the original continuous recording area. By eliminating the need for a seek process when returning to the previous page, it is possible to reduce the number of seek processes, and it is possible to reduce the time required to access the auxiliary storage device.

[Brief explanation of drawings]

Fig. 1 is a block diagram of the principle of the present invention, Fig. 2 is a configuration diagram of a disk control device according to an embodiment of the present invention, Figs. 3 and 4 are flowcharts showing the operation of the embodiment, and Fig. 5 is a conventional FIG. 2 is an explanatory diagram of access processing. In the figure, 101 is an auxiliary storage device, 111 is a storage means, 112 is an address generation means, 113 is a holding means, 114 is a control means, 115 is an access means, 201 is a host computer (host), 202 is a magnetic disk device, 210 is a disk control device, 211 is a microprocessor (MPU), is a memory, is an interface control circuit, is a disk control circuit, is a data buffer, is a buffer control circuit, is a replacement information table, and is a failure pointer storage section. 1. Block diagram of the principles of the present invention. FIG. 1. Block diagram of the magnetic disk control device according to the embodiment. Flow chart showing the operation of the embodiment. Diagram

Claims (1)

[Claims] (1) Auxiliary storage device (
101), and a storage means (11) comprising a plurality of storage areas having a capacity equivalent to one recording area of the auxiliary storage device (101).
1), and the storage means (11) corresponding to each of the plurality of recording areas specified in the input write request or read request.
1) address generation means (112) for generating addresses that sequentially designate a plurality of storage areas; and determining whether each of the plurality of storage areas specified in the write request or read request has a defect. However, when it is determined that there is no defect, a write instruction or a read instruction is given by specifying the address generated by the address generating means (112) and the corresponding recording area, and when it is determined that there is a defect, a control means (114) for holding in a holding means (113) reservation information regarding an address corresponding to a corresponding recording area and a replacement recording area, and suspending a write instruction or a read instruction for this recording area; The data stored in the corresponding storage area of the storage means (111) is transferred to the auxiliary storage device (111) according to the storage means (111).
access means (115) for writing in the corresponding recording area of 101) and storing data read from the corresponding recording area in the corresponding storage area of the storage means (111) in response to the read instruction; After the write operation or read operation for all recording areas determined to be free of defects is completed, the control means (114) performs the corresponding replacement based on the reservation information held in the holding means (113). An auxiliary storage control method characterized by being configured to issue a write instruction or a read instruction by specifying a recording area and an address.