JPH04351747A - Disk control system - Google Patents

Abstract

PURPOSE:To shorten the waiting time at the time of read and write by constituting the system in such a way that the execution of an MPU command and the control on a disk device are performed by a disk controller, and a reference signal is counted by a counter, which is reset by an index signal, while its counted value is latched by a register. CONSTITUTION:A read/write processing command, etc., are written by the disk controller 1 with the MPU via a system bus 101, and after the above execution, its result is informed to the MPU. The reference signal 105 is counted by a fixed value by the counter 2, which is reset by the index signal 103. A value of the counter 2 is latched by the register 3, where read by the MPU is mediated via the bus 101, and the device is controlled by a control signal 102 and the signal 103. By this method, a physical position of each sector on a recording medium corresponds proportionally to a value of the register 3, and hence the latency time at the time of read/write of a sector can drastically be reduced.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a disk control system, and more particularly to a disk control system that controls reading and writing of data in a disk device.

0002

2. Description of the Related Art In conventional disk control systems, when reading and writing data on a disk device, a host MPU sends commands corresponding to read and write operations to a disk controller for controlling the disk device. Then, the target sector information consisting of cylinder, head, sector number, etc. is written, and this disk controller first writes the ID (IDENTITIY) of the target sector according to this command and sector information.
) field is found, and after this ID field is found, the following DAT
A control method is used to read or write data in the A field.

[0003] As such a controller,
For example, for floppy disk devices, there is μPD765AC manufactured by NEC Corporation, and for hard disk devices, there is μPD7261A manufactured by NEC Corporation [Published by NEC Corporation: "μPD765A, 765B, 7265 FD"
C User's Manual" (ITM-782J), and "μPD7261, 72061 Read Disk Controller User's Manual" (IEM-5005A)
)].

[Problems to be Solved by the Invention] In the control method in the conventional disk control system described above,
When the disk controller starts the process of finding the ID field of the target sector, it happens that the ID field of the target sector has just passed the read/write head. Because of the rotation, the target sector has to wait for another revolution to be read out via the read/write head, thus it takes up to However, there is a disadvantage in that there is a waiting time for one rotation, and an extra time is required to correspond to this waiting time.

0004

[Means for Solving the Problems] The disk control system of the present invention includes a disk controller that receives and executes read/write processing commands from an MPU that controls read/write processing sequences, and a disk controller that controls a predetermined reference signal at a constant level. a counter that counts only a value and is reset by a predetermined INDEX signal; a register that latches the counter value and mediates reading of the counter value by the MPU via a system bus; and a predetermined control signal and the INDEX signal. and a disk device controlled by the disk controller via signals.

[0005]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, the present invention will be explained with reference to the drawings.

FIG. 1 is a block diagram showing a first embodiment of the present invention. As shown in FIG. 1, this embodiment uses an MPU (not shown) that controls the sequence of read/write processing and a ROM that contains control firmware, etc.
The above-mentioned M
A disk controller 1 to which commands for read and write processing are written by the PU and after execution reports the results to the MPU, and a counter 2 which counts the reference signal 105 by a certain value and is reset by the INDEX signal 103. a register 3 that latches the value of the counter 2 and mediates reading by the MPU via the system bus 101; and a disk device 4 that is controlled by the disk controller via the control signal 102 and the INDEX signal 103. It is composed of:

In FIG. 1, the INDEX signal 103 output from the disk device 4 is a signal output every time the storage medium in the disk device 4 rotates once, and is used when formatting the storage medium in sectors. This is a reference signal. When this INDEX signal 103 is input to counter 2, counter 2 receives this INDEX signal 103.
After the EX signal 103 is input, the reference signal 105 is sequentially counted. Therefore, the counter value at counter 2,
That is, according to the value of the register 3, the physical position of each sector on the storage medium corresponds proportionally to the INDEX signal 103. For example, when the frequency of the reference signal 105 is fr and the number of sectors existing in each track of the storage medium is N, the relationship between the value in the counter 2 and the position information of each sector is expressed as shown in the following table.

[0008]

[Table 1]

[0009] Here, MCNT is given by the following formula.

[0010]MCNT=(trpm*fr)/N
trpm: period required for one rotation of the INDEX signal Next, the operation of reading from the storage medium of the disk device 4 will be explained.

When the MPU reads a plurality of sectors of the disk device 4, the MPU first reads the counter value of the register 3 and selects a sector number that can be read in the shortest time according to this value. is calculated. A read command and sector information are written to the disk controller 1. In the disk controller 1, detection of the ID field of the target sector is started from the written sector information, but since the ID field is found in the shortest time from the value of the register 3, The DATA field of the target sector can be read and written in a short time without waiting for rotation. Similarly, for other sectors, processing for the disk controller 1 is continued while reading the register 3.

[0012] Here, the operation of reading all sectors on a track will be explained.

For example, if the value of register 3 is read and this counter value indicates the second sector value in the above table, the MPU sends a read command and a read command to disk controller 1. The third sector information, which is the start, is written. This makes it possible to read from the third sector to the Nth sector without waiting for rotation. Furthermore, for the first sector to the second sector, as in the case of the read operation described above, after reading register 3 and detecting the sector that can be written in the shortest time, the disk controller By writing the write operation and start sector information to 1, it is possible to continue processing.

FIG. 3 is a diagram showing the correspondence between the counter value in the counter and the corresponding sector number on the storage medium, and it is clear that the relationship between the two is a linear proportional relationship. ing.

Next, a second embodiment of the present invention will be described. FIG. 2 is a block diagram showing this embodiment. Figure 2
As shown in , in this embodiment, read and write processing commands and the like are written by an MPU (not shown) in correspondence with the system bus 106, and after execution, the results are reported to the MPU. The disk controller 5 and the reference signal 111 are counted by a certain value,
O to which the INDEX signal 110 and selection signal 108 are input
Similarly, the counter 8, which is reset by the output of the R circuit 6, counts the reference signal 111 by a certain value and outputs the IN signal.
a counter 10 that is reset by the output of the OR circuit 7 to which the DEX signal 110 and the selection signal 109 are input; a register 9 that latches the value of the counter 8 and mediates reading by the MPU via the system bus 106; Similarly, the value of counter 10 is latched and the value of system bus 10 is latched.
6, a register 11 that mediates reading by the MPU, a control signal 107 and an INDEX signal 110,
and disk devices 12 and 13 controlled by the disk controller 5 via selection signals 108 and 109 for selecting disk devices.

The difference between this embodiment and the first embodiment is that two disk devices are provided, and along with this, two OR circuits each associated with a counter and a register are provided. That is what we are doing. In this case, by gating the INDEX signal 110 in the OR circuits 6 and 7 via the selection signals 108 and 109 corresponding to the disk drives 12 and 13, Physical read/write position information within the disk device can be obtained. Needless to say, INDEX signal 11
As mentioned above, 0 is a signal that is output every time the storage medium in the disk drives 12 and 13 rotates once, and is a reference signal when formatting the storage medium in units of sectors. Therefore, in the corresponding counters 8 and 10, after the input of the INDEX signal 110,
The reference signal 105 is counted sequentially, and the physical position of each sector on the storage medium is determined according to the counter values in counters 8 and 10, that is, the values in registers 9 and 11, as in the first embodiment. , specified for each disk device. The operations of the disk controller 5, counters 8 and 10, registers 9 and 11, etc. corresponding to the disk devices 12 and 13 are exactly the same as in the first embodiment described above, and the operations are different for each disk device. By having counters and registers,
It becomes possible to handle each disk device separately.

As a specific example, when the present invention is applied to a floppy disk device, the number of revolutions of the storage medium is 3.
In the case of 00 rpm, the time required for one rotation is 200 m.
s (milliseconds), and 1 m as the reference signal 102.
If a signal with a period of s is input and the full count value of the counter is set to 200, each track has 10 sectors, and a range of 20 units of the counter value corresponds to each sector.

Furthermore, when applied to a hard disk device, if the rotation speed of the storage medium is 3600 rpm,
The time required for one rotation is 16.7ms, and the reference signal 1
If a signal with a period of 0.1 ms is input as 02 and the full count value of the counter is set to 167, then 1
With a 33 sector per track configuration, a range of 5 units of counter value will correspond to an individual sector.

Therefore, when a read or write operation is executed using a counter, it is possible to predict in advance the sector in which read/write can be executed in the shortest time. It becomes possible to reduce the time required to access one rotation time sector. In the above example, in the case of a floppy disk device, it is approximately 200ms, and in the case of a hard disk device, it is approximately 16.7ms.
It is s.

As explained above, the present invention has the effect that when reading and writing sectors of a disk device, the rotation waiting time caused by the rotation of the storage medium of the disk device can be significantly reduced. be.

[Brief explanation of drawings]

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

FIG. 2 is a block diagram showing a second embodiment of the invention.

FIG. 3 is a diagram showing the relationship between a counter value and a sector number in a counter.

[Explanation of symbols]

1, 5 Disk controller 2, 8, 10 counter 3, 9, 11 Register 4, 12, 13 Disk device 6, 7 OR circuit

Claims (1)

[Claims] Claim 1: A disk controller that receives and executes a read/write processing command from an MPU that controls a read/write processing sequence, and a predetermined reference signal that is counted by a certain value and reset by a predetermined INDEX signal. a register that latches the value of the counter and mediates reading of the counter value by the MPU via a system bus, and a predetermined control signal and the INDEX signal, which are controlled by the disk controller. A disk control system comprising a disk device.