JP2980326B2 - Disk controller - Google Patents

Description

The present invention relates to a disk control device, and more particularly, to a disk control device that controls writing and reading of a file to and from a disk device.

(Prior Art) In a computer system using a disk operating system (DOS), a file allocation table (FA) is stored in a specific location on a disk device.
T) A region is provided. This FAT indicates the use status of the data storage area of the disk device, that is, the file allocation status in the data storage area.
T includes crust information indicating the connection between the sector positions where the file is stored.

Cluster information generally has multiple clusters,
Each cluster functions as a pointer indicating the sector position on the disk where the file is stored.

Therefore, when the file is opened, the host computer refers to the cluster information in the FAT to find the sector position where the file actually exists, and instructs the disk controller to access, ie, read, the sector position. When one file is stored over a plurality of sectors, the host computer obtains a subsequent sector position based on the cluster information and repeatedly issues a read instruction to the sector position.

As described above, conventionally, it has been necessary to sequentially refer to the contents of the FAT by the host computer in order to determine the position of the sector to be accessed. For this reason, it was difficult to execute high-speed access to the disk device.

(Problems to be Solved by the Invention) Conventionally, in order to determine the position of the sector to be accessed, it is necessary to sequentially refer to the contents of the FAT by the host computer, and there is a disadvantage that the disk device cannot be accessed at high speed. Was.

SUMMARY OF THE INVENTION The present invention has been made in view of the above circumstances, and has as its object to provide a disk control device capable of executing high-speed access to a disk device.

[Structure of the Invention] (Means and Action for Solving the Problems) The present invention relates to a disk control device that controls writing and reading of data to and from a disk device.
Storage means, data storage means, means for writing in the storage means a copy of the contents of the file allocation table area on the disk device, including cluster information indicating the connection between the sector positions where the files are stored, and the disk device Means for determining, in response to a read request from the host computer for the data area, whether the read request is a read request for a new file; and
A sector position detecting unit for sequentially detecting a sector position of data constituting a file to which the sector specified by the read request belongs by referring to a file allocation table of the storage unit; A prefetch means for prefetching data at a sector position into the data storage means, so that data can be prefetched in file units.

In this disk controller, every time a read request for a new file is issued from the host computer, the sector position of data constituting the file to which the sector specified by the read request belongs is sequentially detected by referring to the file allocation table. And
Data is pre-read from each of the detected sector positions. As described above, when a sector of a file is first requested to be read, by performing a read-ahead process in file units of pre-reading each data constituting the file, even if the file to which the sector requested to read belongs is performed. Even when the constituent data is distributed and stored in a plurality of tracks, the sector data specified by the subsequent read request for the same file can be immediately transferred to the host computer.

(Example) Hereinafter, an example of the present invention will be described with reference to the drawings.

FIG. 1 schematically shows a computer system using a disk control device according to an embodiment of the present invention. The host computer 11 controls the entire system, and is connected to a disk drive device 13 via a disk controller 12. Disk controller 12
Controls the disk drive device 13 in accordance with a write instruction and a read instruction from the host computer 11. The disk drive device 13 writes and reads data to and from a disk device 14 such as a floppy disk or a hard disk device.

The disk device 14 has a loader area 141, a FAT area 142,
A directory area 143 and a data area 144 are allocated. The loader area 141 is, for example, a program area for starting up the system when the power is turned on.
The T area 142 is an information storage area indicating the connection of clusters.

Each cluster is a pointer indicating where the file is stored in the data area 144, and the cluster and the sector position correspond one-to-one. Therefore, FAT region 1
By referring to 42, the connection of the sector positions corresponding to each file can be recognized. The directory area 143 is an area for storing registration information of each file, that is, file names, file sizes, and the like.

The disk controller 12 has the FAT of the disk device 14
A FAT storage memory 121 into which the contents of the area 142 are copied, and a data buffer 122 for storing data read from the data area 144 of the disk device 14 are provided.

When detecting a file read instruction from the host computer 11, the disk controller 12 obtains a next sector position following the sector position specified by the read instruction with reference to the cluster information of the FAT area 142 copied to the memory 121. In addition, it has a function of reading the information of the next sector position and storing it in the data buffer 122.

Further, when the next sector position obtained by referring to the cluster information copied to the memory 121 belongs to a cylinder different from the current sector position, the disk controller 12 reads the information of the next sector position before the reading operation. It has a function for causing the disk drive device 13 to start a seek operation.

FIG. 2 shows an example of a specific configuration of the disk controller 12. 2, the disk controller 12 includes a microprocessor unit 21, a ROM 22, a disk control unit 23, and a host control unit 24, in addition to a FAT storage memory 121 including a RAM and a data buffer 122.

The microprocessor unit 21 controls the overall operation of the disk controller 12 according to a program stored in the ROM 22.
The storage memory 121, the data buffer 122, the ROM 22, the disk controller 23, and the host controller 24 are interconnected.

The microprocessor unit 21 is a memory 1 for FAT storage.
Processing such as obtaining the next sector position with reference to the cluster information of the FAT area 142 copied to 21 is executed. The disk control unit 23 executes a disk transfer process of reading data from the disk device 14 and storing the read data in the buffer 122. The host control unit 24 executes a host transfer process of transferring the data stored in the buffer 122 to the host computer 11.

Next, the operation of the disk controller 12 will be described with reference to the flowcharts of FIGS.

FIG. 3 is a flowchart showing the overall operation of the disk controller 12 after copying the contents of the FAT area 142 to the FAT storage memory 121. The FAT copy operation is performed when the power is turned on and when the FAT area 142 is updated by the host computer 11, that is, when the FAT area
It is automatically executed when a write instruction is detected.

In the flowchart of FIG. 3, step S1 is a step for determining whether or not a read instruction has been issued from the host computer 11, and if a read instruction from the host computer 11 has been detected, the read processing routine of step A2 is performed. Is executed. The read instruction includes
Address data for disk access is also included. The details of the reading process will be described later with reference to FIG.

Step S3 detects the “disk transfer flag” and
This is a step of determining whether data needs to be read from the disk device 14. This “disk transfer flag” indicates that the read processing for all clusters corresponding to the read instruction from the host computer 11 has not been completed, and is set in the read processing routine for the above read instruction. Therefore, when the "disk transfer flag" is set, the disk transfer processing routine of step A4 for reading the data of the disk device 14 is executed. The details of the disk transfer processing routine will be described later with reference to FIG.

Step S5 is a step in which the “host transfer flag” is detected to determine whether or not the read data specified by the host computer 11 has been sent to the host computer 11. This "host transfer flag" indicates that data has not yet been sent to the host computer 11, and is set in the above-described disk transfer processing routine. Therefore, when the “host transfer flag” is set, a host transfer processing routine for sending data read from the disk device 14 to the host computer 11 is executed. This host transfer processing routine will be described later with reference to FIG.

The above determination processing steps A1, A3, A5 can be considered as an extension of the idle processing.

Next, the read processing routine of step A2 will be described with reference to FIG.

That is, when the disk controller 12 detects an instruction to read a file from the host computer 11, first, based on a sector position designated by the host computer 11 and a predetermined parameter prepared in advance, a cluster corresponding to the sector position is determined. Is calculated by inverse calculation (step B1), and the calculated cluster is searched for from the cluster information of the FAT area 142 copied to the FAT storage memory 121 (step B2). In the process of step B2, the memory 121 is referred to based on the found cluster.
And the next sector position is calculated based on the cluster following the reference cluster.

In this way, before the next read instruction is issued from the host computer 11, the next sector position related to the file is obtained one after another.

Then, it is determined whether the read instruction from the host computer 11 is for a new file or a file that is currently being read (step B3).
This can be determined by, for example, whether or not the first cluster reference is made. If it is the first cluster reference, it is a new file, and if it is a subsequent reference, it is a read for the same file.

In the case of a new file, first, the buffer 122 is initialized (step B4), and thereafter, "disk transfer flag" is set.

Next, the disk transfer processing routine of step A4 will be described with reference to the flowchart of FIG. The disk transfer processing routine executes an operation of reading data from the disk device 14 to the disk controller 12 when the "disk transfer flag" is set in the data read processing routine. That is, in the disk transfer processing routine, first, an actual seek address is obtained from the next sector position obtained in the read processing of FIG. 4 (step C1).

Next, based on the seek address, it is determined whether or not the next sector position and the current sector position belong to the same cylinder (step C2). If not on the same cylinder, a seek operation is required.
Starts a seek operation (step C3). Next, when the end of the seek operation is detected (step C4), it is determined whether or not the buffer is full (step C5).

If there is free space in the data buffer 122, the data read operation of the disk drive 13 by the disk drive 13 is executed (step C6), and thereafter, the "host transfer flag" is set (step C7).

Next, it is determined whether or not the cluster whose seek position was obtained in the previous step C1 is the last cluster of the file (step C1).
C8), "Disk transfer flag" for the last cluster
Is reset (step C10), and then the file ends. If it is not the last cluster, the next cluster is obtained again (step C9), and the process ends without resetting the “disk transfer flag” and holding the obtained cluster. As a result, the processing of steps C1 to C9 is repeated.

In this manner, in the disk transfer processing routine shown in FIG. 5, after the read instruction is issued from the host computer 11, the buffer is kept full or the last cluster is detected until the processing is completed. The data is sequentially read from the disk device 14 and stored in the data buffer 122.

Next, referring to the flowchart of FIG.
The A6 host transfer processing routine will be described. This host transfer processing routine executes data transfer to the host computer 11 when the "host transfer flag" is set in the disk transfer processing routine.

In the host transfer processing routine, first, it is determined whether or not data has been input to the data buffer 122 (step D1). If not, the process waits until the data is input. If already input, data buffer 12
The content of 2 is the host computer 1 by the host control unit 24
Transferred to 1 (step D2). Then, the “host transfer flag” is reset (step D3), and the process ends.

In this manner, in the host transfer processing routine shown in FIG. 6, when only the data specified by the read instruction from the host computer 11 is transferred to the host computer 11, the processing is terminated, and the data buffer 122 prefetches the data. The data is transferred when a read instruction is actually issued from the host computer 11.

Therefore, with respect to the second and subsequent read instructions from the host computer 11, it is only necessary to transfer the data pre-read in the data buffer 122 to the host computer 11, so that high-speed access is possible.

[Effects of the Invention] As described above, according to the present invention, high-speed access to a disk device can be performed.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a disk control device according to an embodiment of the present invention, FIG. 2 is a block diagram showing an example of a specific configuration of the disk control device shown in FIG. 1, and FIGS. FIG. 6 is a flowchart for explaining the operation of the disk control device shown in FIG. 11 Host computer, 12 Disk controller, 13 Disk drive device, 14 Disk device, 121 FAT storage memory, 122 Data buffer.

Claims (2)

(57) [Claims] 1. A disk control device for controlling writing and reading of data to and from a disk device, wherein the disk device includes a storage unit, a data storage unit, and cluster information indicating a connection between sector positions where files are stored. Means for writing a copy of the contents of the file allocation table area to the storage means, and responding to a read request from the host computer for the data area of the disk device, determining whether the read request is a read request for a new file. And a sector for sequentially detecting the sector position of the data constituting the file to which the sector specified by the read request belongs by referring to the file allocation table of the storage unit if the read request is for a new file. Position detecting means; A disk control device comprising: a read-ahead means for pre-reading data at each sector position detected by the sector position detection means into the data storage means, so that data can be pre-read in file units. 2. The disk control apparatus according to claim 1, wherein said sector position detected by said sector position detecting means belongs to a cylinder different from a current sector position, and said data is read from said sector position detected by said sector position detecting means. 2. The disk control device according to claim 1, further comprising: means for starting a seek operation prior to a pre-read operation by the pre-read means for reading out the data.