JPH02170220A - Disk data reader - Google Patents

Abstract

PURPOSE:To speed up disk data reading by reading out the data at every byte with use of a pointer having a specific order of bytes set from the head of a disk and a size showing a specific number of bytes needed for an access given from the pointer when the data is read out of a 2-inch floppy disk. CONSTITUTION:A 1/4 track is defined as an access to a floppy disk 6', and a pointer is set in a buffer 30. Then the data are transferred from the buffer 30 by an amount equal to a size (byte). Thus the data can be read out only at a necessary part and it is not required to prepare the memories equivalent to one block unlike the conventional systems of 8, 5 and 3.5 inches, etc. In addition, the accesses are possible to the buffer 30 at every byte and reflected directly on the disk 6' to avoid a toothless state of data on a disk. Furthermore the accesses are carried out to the disk 6' at every 1/4 track. Otherwise only the transfer of data is carried out and therefore no problem is produced at all regardless of the continuous reading speed or the intermittent reading speed.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a disk data reading device, and particularly to a disk data reading device suitable for reading digital data recorded on a 2-inch floppy disk.

[Summary of the Invention] When reading data from a 2-inch floppy disk, the present invention reads data in bytes by using a pointer indicating the number of bytes from the beginning of the disk and a size indicating the number of bytes to be accessed from this pointer. By doing so, it is possible to increase the read speed, eliminate the need for a dedicated read buffer, and prevent the data on the disk from becoming incomplete.

[Conventional technology]

Traditionally, floppy disks were 8 inches.

5-inch, 3.5-inch, etc. are conceivable, and for recording and reproducing data on these floppy disks, a device as shown in FIG. 6, for example, is conceivable. That is, in FIG. 6, (1) is a host computer as an external device, (2) is a host raster 11 bus, (3) is a floppy disk controller (hereinafter referred to as FDC),
(4) is a drive interface, (5) is a disk drive device, and (6) is a floppy disk.

During recording, data from the host computer (1) is supplied to the FDC (3) via the host system bus,
Here, parity data generated based on the data is added and supplied to the disk drive device (5) via the drive interface (4), and the floppy disk (
6) is recorded.

Reproduction is performed according to a flowchart as shown in FIG. That is, in step (7), the host computer (1) issues a command to the disk operation system (hereinafter referred to as DO3) to read blocks from X block (l block (1 cluster)) to Y block of the disk (6). do. Next, in step (8), the host computer (1) calculates the physical track position, block position, etc., ie, calculates what track and what block, etc., based on the command from DO3. and step (
In step 9), the host computer (1) issues a command to the FDC (3) to start access according to the calculation, also based on the command from DO3. Step (lO)
At this time, the FDC (3) reads data from the specified address from the disk (6) and stores it in the main memory (not shown) of the host computer (1).

[Problem to be solved by the invention]

In this way, in 8-inch, 5-inch, 3.5-inch, etc. floppy disk systems, when reading the necessary data, the method is to read Y blocks from the X block from the beginning of the disk. If the blocks are set to be large, for example, as shown in FIG. 8A, even when reading one byte of necessary data, it is necessary to read one block, which results in reading more data than is necessary. As a result, the system must prepare at least one block of memory, which has the drawback of requiring an extra area.

In addition, since the disk can only be accessed in block size, the data on the disk becomes incomplete.

However, this can be solved by setting the blocks small as shown in FIG. 8B. However, when reading one block, there is no problem because it only needs to be accessed once, but when reading several to dozens of blocks continuously or intermittently, the number of mechanical accesses increases each time, and the reading speed decreases. There is.

The present invention has been made in view of the above, and provides a disk data reading device that can eliminate the above-mentioned drawbacks by utilizing the Z-trough 9-minute buffer memory (RAM) inherent in the 2-inch floppy disk system. It is something to do.

[Means to solve the problem]

When reading data from a disk, the disk data reading device according to the present invention reads data in bytes by using a pointer that indicates the number of bytes from the beginning of the disk and a size that indicates how many bytes to access from this pointer. It is configured as follows.

[Effect]

The disk is accessed using the z track, a pointer is provided in the buffer, and only the size (bytes) is transferred from there. This makes it possible to read only the necessary portion of data, eliminating the need to prepare one block worth of memory as in conventional 8.5-inch, 3.5-inch, etc. systems.

Furthermore, the buffer can be accessed in byte units, and this is reflected directly on the disk, so the data on the disk does not become incomplete as in the past. Furthermore, access to the disk is for each strunk, and in other cases, only transfer is performed, so there is no problem in terms of speed whether it is continuous reading or intermittent reading.

〔Example〕

Hereinafter, one embodiment of the present invention will be described in detail based on FIGS. 1 to 5.

FIG. 1 shows the circuit configuration of this embodiment. In the figure, parts corresponding to those in FIG. 6 are denoted by the same reference numerals, and detailed explanation thereof will be omitted.

In this embodiment, an advanced disk controller (hereinafter referred to as ADC) (20) is provided between the host system bus (2) and the drive interface (4).

ADC (20) is inner bus (21), FDC
(22), Format ivy (23), De Format ivy (
24) , error code correction circuit (25) and memory management unit (hereinafter referred to as MMU) (26)
has. The formatter (23) performs signal processing such as conversion between parallel data and serial data and 8-10 conversion. The deformatter (24) performs signal processing such as conversion between serial data and parallel data and 10-8 conversion.

(25) generates parity data from data during recording, and performs error correction using the parity data during reproduction. M
M U (26) is RAM as a buffer memory
Control the address for (30). In addition,
In this embodiment, a 2-inch floppy disk is used as the media. For details of this 2-inch floppy disk, please refer to Japanese Patent Application No. 111,709.

Here, it is being considered that digital data as well as color video signals can be appropriately recorded and reproduced by adopting the following format for the floppy disk (6').

That is, in FIG. 2A, (IIT) indicates an arbitrary track on the rotating magnetic disk (11), (12) indicates the center core into which the spindle of the drive mechanism (not shown) is fitted, and this track (IIT) is the disk (1
1) 90 @ 4 sections each in the length direction using the magnetic piece (13) as a reference to give the reference angular position when rotated.
It is divided into equal parts, and each divided section is a block BLC.
It is called K. Further, the block BLCK in the section including the magnetic piece (13) is the 0th block, and the 1st... Second.
This is the third block.

Then, as shown in FIG. 2B, the block BLCK is
The section of 4 degrees from the start end is defined as the gap section GAP for obtaining a margin during read/write, and the section of <1'' is defined as the burst section BRST. In this case,
In the 0th block, the center of the gap section GAP corresponds to the position of the magnetic piece (13).

In addition, the burst section BRST is a section in which i, a preamble signal ii, a signal iii indicating the recording density of the signal + a burst signal BR5T which also serves as a flag signal indicating that the recorded signal is digital data is recorded and reproduced. It is.

Furthermore, the section following the burst section BRST is the section of the index signal INDX. In this case, the index signal INDX is composed of an 8-bit flag signal FLAG and an 8-bit address signal I, as shown in FIG.
ADR and a 40-bit undefined signal RSVD.

It consists of an 8-bit check signal ICRC.

Then, the flag signal FLAG is the block BLCK.
The address signal IADR is a signal indicating the status or attribute of the track (IIT) to which the track (IIT) belongs, such as whether it is a defective track or whether it has been erased.
This signal indicates the number [1 to 50] of the IIT) and the number [0 to 3] of the block BLCK. In addition, check signal ICR (JJ:, signal FLAG, IADR, R5VD
This is the opposite CRCC.

Further, the remaining section following this section INDX is divided into 128 equal parts, each of which is a section in which a signal called a frame FRM is recorded and reproduced.

That is, as shown in the second diagram, one frame FRM is sequentially transmitted from the beginning to the 8-bit frame synchronization signal 5YNC.
, 16-bit frame address signal FADR, 8-bit check signal FCRC, 16-symbol (1 symbol = 8 bits) data DAT, 4-symbol redundant data PRTV, and another 16-symbol digital data. DATA and another 4 symbols of redundant data PRTV
and has. In this case, the check signal FCRC is the CRCC for the frame address signal PADR. In addition, data DATA is the original digital data that is accessed by the host device, but this data DAT^ is
Interleaving is performed intermittently within the digital data of one block BLCK, and the redundant data PRTY is one block worth (32 symbols x 128 symbols).
frame) for the digital data of the minimum path i! iI5
These are parity data CI and C2 generated by the Reed-Solomon encoding method.

Therefore, one block BLCK, l-rack (2
The capacity of digital data on T) and disk (6') is: 1 block: 4096 bytes (= 32 symbols x 128 frames) l track = 16 bytes (= 4096 bytes x 4 blocks) 1 disk: 800 bytes (= 16 and 50 bites).

Also, the number of bits in one frame FRM and block BLCK is as follows: 1 frame = 352 bits (8 + 16 + 8 bits + (16 + 4 symbols) x 8 bits x 2 pieces) 1 block (index section and frame section only)
:45120 bits (= 352 bits x 128 frames), but in reality, when recording and reproducing digital signals on the disk (6'), a small DSV is required, and Tm1n
/ Since it is necessary that Tmax is small and Tw is large, all the digital signals mentioned above have 'l'max=4
The signal is recorded on the disk (6') after T 8-10 conversion, and during playback, the inverse conversion is performed and then the original signal processing is performed.

Therefore, for the above data density, the actual number of bits on the disk (6') is multiplied by 10/8, 1 frame: 440 channel pits 1 block (index section and frame section only)
: 56400 channel bits. Also, as a result, the entire section of one block is 59719 channel bits < 56400 channel bits x 90°/85"
) (Actually, since the length of each section is assigned as described above from this number of channel bits, the total length of the frame section is slightly shorter than 85 degrees).

Therefore, the bit rate when accessing the digital signal (signal after 8-10 conversion) to the disk (6') is 14.32 Mbit/s (-59719 bits x 4 blocks x field frequency), which is 1 A bit corresponds to 69.8 ns (L=, 1/14.32 Mbit).

Thus, according to the format shown in Figure 2, it is possible to read and write 800 bytes of digital data on a 2-inch floppy disk (6'), which is equivalent to the capacity of a conventional 5-inch floppy disk (ffi (320)). It has a large capacity despite its small size.

In addition, since the rotation speed of the rotating magnetic disk (11) is the same as that for color video signals, when recording and reproducing a mixture of color video signals and digital data, the rotation speed of the rotating magnetic disk (11) is the same as that for color video signals.
), the frequency spectra of both signals recorded and reproduced are similar, and recording and reproduction can be performed under optimal conditions for electromagnetic conversion characteristics, hand contact, etc. moreover,
Even when recording and reproducing a mixture of two signals, the number of revolutions of the disk (11) is not changed, so there is no need to consider the time required to change the servo circuit, and the two signals can be used properly immediately. Furthermore, since the number of revolutions is the same and the mechanism such as the electromagnetic conversion system has the same characteristics or functions, it is also advantageous in terms of cost.

In this way, even though the floppy disk (6') is for analog signals, by applying the format shown in FIG. 2 to it, it has new effects as a next-generation floppy disk.

Next, the operation of the circuit shown in FIG. 1 will be explained.

First, recording will be explained.

Host system bus (2) of host I/computer (1)
) is the pure data for one block from FDC(22)
and is supplied to the RAM (30) via the inner bus (21).

At this time, continuous addresses from MMU (26) are supplied to RAM (30), and RAM (30) stores pure data at continuous addresses in the data memory area (30D) shown in Figure 3. are stored continuously in their original time series. When all the pure data for one block is stored in RAM (30), parity data C2 is first generated from this data in the error correction circuit (25), and is stored in the memory area (30C,) of RAM (30). This is stored in a continuous state.

Next, pure data and parity data C! are sent out in the sending order according to the recording pattern of the disk according to the address from MMU (26). is read from the RAM (30), and is supplied to the formatter (23) while generating and adding parity data CI in the error correction circuit (25). And in this formattuta (23), 8-
10 conversion is performed, and the parallel data is converted into serial data, which is supplied to a magnetic head (not shown) in the disk drive device (5) and is applied to 90 of the tracks (IIT) of the magnetic rotating disk (11). The data is recorded in a portion corresponding to the 10° interval, for example, in the first block.

When the reading of one block of data from RAM (30) is completed, the host computer (1) supplies the next block of pure data to RAM (30) in response to a write request signal. In exactly the same manner as above, the parity data c, , Cm are added and interleaved to another 90° section of the track (IIT) of the magnetic rotary disk (11), for example, the first block. recorded in Similarly below,
When recording for 4 blocks is completed, the program moves to another track.

In this example, during recording, the parity data CI area (30C) of the RAM (30) memory area is
, ) were not used, but after storing the generated parity data CI in this area (30Cr), the data DA was stored in the RAM (30) in the order according to the recording pattern.
Of course, it is also possible to read out TA and C0Ct.

Next, during reproduction, reproduction data is extracted from the output of the head when the head scans the track (IIT) in units of one block of data corresponding to a 90° interval, and is processed as follows.

That is, the deformater (24) via the disk drive device (5) and the drive interface (4)
The serial data is supplied to 8-bit parallel symbol data and is converted into 10-8 data. This parallel data is sent to the RA via the inner bus (21).
M (30) and R according to the address corresponding to interleaving during recording from M M U (26).
A M (30), data DATA, parity data C2 and C2 are respectively stored in the memory area (300), (
30C,) and (30C,). In other words,
The data string that was distributed by the addresses from M
(30).

When all data for one block is written, MMU
According to the address from (26), this RAM (3
0) is error-corrected by the parity data Ct and Cz in the error correction circuit (25), and then stored in the RAM.
(30) is written again. Therefore, RAM (
30), the error-corrected data DATA is finally stored.

In this way, when all the data DATA is stored in RAM (30) in the original chronological order with consecutive addresses, the r READY J signal becomes F DC (
22) and stored in the RAM (30) is transferred to the external host computer (1) according to the FDC (22).

Data reading during reproduction will be further explained with reference to FIGS. 4 and 5. In FIG. 4, in step (41), the host computer (1) issues a command to DO3 to read Y bytes from the X byte from the beginning of the disk (6'). That is, a pointer called the X-th byte from the beginning of the disk (6') is designated, and a size is set such that Y bytes from this pointer are the necessary data to be finally read. Next, in step (42), the host computer (1) calculates the physical track position and block position, etc. according to the command from the DO3, that is, calculates what track and what block the necessary data corresponds to. conduct. Then, in step (43), the host computer (1) also receives an instruction from DO3 to start accessing A according to the calculation result.
Generates a command to DC (20). Step (44
), the ADC (20) reads data from the specified address from the disk (6') to the RAM (30), and in step (45) the ADC (20) reads data from the specified address to the RAM (30).
) MMU
Store according to the address in (26).

Then, in step (46), by command to DO3, R
The necessary portion of data is transferred from A M (30) to the main memory of the computer (1). In other words, as shown in Figure 5, a pointer and size are used, the pointer determines the required data, the size determines the amount of data, and the RA
M (30) and transfer it to the main memory of the computer (1). Since the transfer speed at this time is fast, high-speed reading is possible.

In addition, although the above-mentioned embodiment was explained taking the case of reading as an example, it can be similarly applied to the case of writing.

〔Effect of the invention〕

As described above, according to the present invention, access to the media is performed as a track, a pointer is provided in the buffer memory, and only the size (bytes) is transferred from there, so it is possible to read/write only the necessary portion. This eliminates the need to prepare one block's worth of memory (one cluster's worth) as in conventional 8; 5.3.5 inch systems, and the memory capacity can be reduced. In addition, the buffer memory can be accessed in byte units, and this is reflected directly on the media, so that the data on the media does not become incomplete as in the past. Furthermore, access to the media is for each track, and in other cases it is only transfer, so both continuous read/write and intermittent read/write are all speed-wise (high-speed reading is possible without any problems). It becomes possible.

[Brief explanation of the drawing]

Fig. 1 is a system diagram showing an embodiment of the present invention, Fig. 2 is a diagram showing a recording former node of a 2-inch Flow Novy disk to which the invention is applied, and Fig. 3 shows the main parts of the invention. The memory mapping diagram, FIGS. 4 and 5 are diagrams for explaining the present invention in detail, FIG. 6 is a system diagram showing an example of a conventional device, and FIG. 7 is for explaining the operation of FIG. 6. FIG. 8 is a diagram for explaining a conventional example. (1) is a host computer, (6') is a floppy disk, (20) is an advanced disk controller (ADC), and (30) is a RAM.

Claims (1)

[Claims] When reading data from a disk, the data is read in bytes using a pointer indicating the number of bytes from the beginning of the disk and a size indicating how many bytes to access from the pointer. A disc data reading device characterized by: