JPS60107149A - High speed initializing and processing system of file - Google Patents

Abstract

PURPOSE:To speed up the initializing processing by reading an initialized data for one page's share of a data area at each write instruction to apply initializing processing of a file of the said size thereby decreasing a processor running time required for the initializing processing. CONSTITUTION:An initializing data area IND of a main memory is provided with a capacity for one page's share and the initializing data IS for one page's share is stored. On the other hand, instructions are arranged in an instruction area CCW of the main memory depending on the size of the file initialized by the processor. That is, n-set of write instructions (WRITE) and data chaining flags CDF are arranged in succession to a normal seek instruction (SEEK) and a command chaining flag CCF. The processor identifies the file size of the size area of the main memory in the initializing processing and forms automatically the instruction groups.

Description

DETAILED DESCRIPTION OF THE INVENTION [Technical Field of the Invention] The present invention relates to a high-speed file initialization processing method for initializing files on magnetic disks, floppy disks, etc. using initialization data from a processor. The present invention relates to a high-speed initialization processing method for files that can reduce the amount of initialization and perform initialization processing at high speed.

[Prior art and problems]

In a computer system, various storage devices and input/output devices are connected to a processing device (processor), and the processor uses these devices to execute desired processing. Various files necessary for this processing are provided in a storage device, and a magnetic disk device, a floppy disk device, etc., which are inexpensive and have stable performance, are used as the storage device.

In the storage device that stores these various files.

Since the processor uses this to perform predetermined processing, the larger the capacity, the more convenient it is to be able to store various files. Figure 1 is a configuration diagram of a general computer system, and in Figure 9, l is a central processing unit that includes at least an arithmetic unit (hereinafter referred to as a processor) 10 and its main memory 11, and 2 is a file system. The device is composed of a magnetic disk device, and includes a magnetic disk dust roller 21. Head access mechanism 22. In such a configuration that has a magnetic disk 23 and a magnetic head 24, the processor IO gives a seek command to the magnetic disk controller 21 as necessary, drives the head access mechanism 22, and operates the magnetic head 24. The magnetic head 24 is positioned at a desired track on the disk 23, and the magnetic head 24 writes or reads from the desired track. Write data is sent from the processor IO to the magnetic disk controller 21 and further to the magnetic head 24, and conversely, read data is sent from the magnetic head 24 to the processor 10 via the magnetic disk controller 21. On the other hand, in such a file device,
It is necessary to initialize files to prepare for new processing when a computer system is powered on or when performing specific processing.

Since such a file device cannot be completely erased like a semiconductor memory, it is necessary to send initialization data from the processor 10 and write it to a file on the magnetic disk 23 that requires initialization processing.

In accordance with this initialization process, the initialization data area I stores initialization data in the real memory area on the main memory 11.
ND, a command area CCW that stores a group of instructions necessary for writing initialization data, and a size area FS that stores the size of a file that requires initialization processing. processing was being performed. That is, initialization data is entered and the QS (Qptra
The maximum capacity (for example, 8192 bytes) that can be transferred by the ``l'' instruction of ``TingSystem'' is prepared in the data area END. This initialization data is EOD (EN) in sectors (for example, 256 bytes) as shown in Figure 3.
DOF DATA) is inserted.

The rest of the 256 bytes are all "work" (or 0), for example.
This is the data. The instruction area 00W contains a seek instruction (SEEK) and four write instructions (WT(,ITE)).
and the physical address of the disk to be written (D S ) d
dress). I'm here.

00FI, j: Command chaining flag, indicating that a command (instruction) continues; ODF is a data chaining flag, indicating that data continues. These four write commands are necessary to write 8192 bytes of data, which is initialization data, because file data is managed page by page in the virtual storage system, assuming that one page is a maximum of 2048 bytes. This is because instructions are required.

After making preparations in this way, when the processor IO issues the 108TART command to the magnetic disk controller 21 as shown in Figure 4, the magnetic disk controller 21 directly reads the contents of the command area COW of the main memory 11 and Execute the initialization process ■. That is, it reads out the seek command, which is the first command in the command area COW, and identifies that it is a seek operation for the magnetic disk 23.

Next, read the disk physical address DS Adclrtss.

The magnetic head 24 is positioned on the corresponding track of the magnetic disk 23 by the head access mechanism 22, and the OOF
'eIdentify and read the data l5-1''lr of the first page of initialization data in the initialization data area IND by the next command, the write command.

The next write command is written to the magnetic head 24.■
When the data l5-2 of the second page of the initialization data is read out and written in the same manner, and the write command (2) is executed in the same manner, there is no ODF, so the process ends. This completes the initialization process for four pages of file data. Next, the processor JO checks the file size in the file size area FS, and if there is more file capacity, the processor JO checks the disk physical address DS kcldre in the instruction area 00W.
sz is rewritten to the next address, the IO5TART command is issued again, and the magnetic disk controller 21 is made to perform a similar initialization operation, and this is repeated until the initialization operation for the capacity of the file to be initialized is completed.

According to such a conventional initialization processing method, for example, in order to initialize a 1 megabyte file, the processor 10 must send the IO5TAR,T command and rewrite the physical address 128 times. processing, which imposes a burden on the processor 10, that is, requires a long running time;
In addition to the problem of not being able to perform other processes, there was also the problem that accesses had to be made 128 times on the magnetic disk side, which increased the access time and the initialization process. Furthermore, since the initialization data is huge, it occupies a large area of the precious main memory, causing the problem that the main memory cannot be used effectively.

[Purpose of the invention]

An object of the present invention is to shorten the running time of a processor required for initialization processing, reduce the number of device accesses,
An object of the present invention is to provide a high-speed file initialization processing method that can speed up initialization processing.

[Structure of the invention]

To achieve the above object, the present invention provides a system having a processing device including a processor and a main memory, and a file device for storing files, a data area for storing one page of initialization data in the main memory. and an instruction area for storing a group of instructions necessary for the initialization process.

The processor sets a number of write instructions in the instruction area according to the number of writes of the one page of initialization data necessary to initialize the size of the file to be initialized, and the file device initializes the processor. Directly reading the contents of the command area of the main memory when receiving a processing start command, reading one page of initialization data from the data area for each write command, and initializing a file of the corresponding size. It is characterized by

[Embodiments of the invention]

The present invention will be explained in detail below using examples.

Fig. 5 is a detailed explanatory diagram of the present invention, and Fig. 6 is an explanatory diagram of its operation. As shown in Figure 5.

In the present invention, main memory 11. The initialization data area IND in (Figure 1) has a capacity of one page (for example, 2048 bytes), and the initialization data ■S for one page, that is, the unit data of 256 bytes in Figure 73 is 2048/256= Eight pieces are stored in series. on the other hand.

In the instruction area OOW of the main memory 11 (Figure 1),
Instructions are listed according to the size of the file to be initialized by the processor 10 (Figure 1). That is, nine sets of write commands (WRITE) and data chaining flags ODF are enumerated following nine normal seek commands (SEEK) and command chaining flags OOF. This number is the size of the file to be initialized It: 204B×n
If it is a byte, the write commands from ■ to ■ are listed,
The processor 10 uses the main memory 11 during initialization processing.
(Figure 1) These command groups are automatically created by identifying the file size of the size area FS.

Next, an explanation will be given of the operation according to the structure shown in Fig. 6 and Fig. 5. First, in the initialization process, the processor 10 stores a size area F in the instruction area OCW on the main memory 11.
The required number of write commands are set as described above from the SO size, and (n-1) data chaining flags CDF are set accordingly, and the disk physical address DSAddrt, ? The start address of the file is set as s. Next, the processor IQ issues an IO5TART command to the magnetic disk controller 21 as a processing start command. At this time, the magnetic disk controller 21 directly reads the contents of the command area OCW of the main memory 11, first reads the seek command (SEEK), identifies that the seek operation is for the magnetic disk 23, and then determines the seek destination. Disk physical address DSA
dd and rtss are read, and the magnetic head 24 is positioned on the corresponding track of the magnetic disk 23 by the head access device jl122. Furthermore, the magnetic disk controller 21 sets the command chaining flag c following the seek command.

The next instruction, write instruction ■, is read from F, thereby initializing data IS'f in initializing data area IND.
: Read and write operations are performed on the magnetic head 24, and the next write command ■ is read by the data chaining flag CDF following the write command, thereby reading the initialization data Is of the initialization data area IND, and writing to the magnetic head 24. .

The RAD 24 is caused to perform a write operation. Thereafter, this process is repeated in the same manner until the data chaining flag ODF is exhausted, that is, the same operation is performed for write instructions ■ to ■, and the data chaining flag OD is
Since there is no F, this will detect the end of the initialization process,
A completion notification is issued to the processor 10. As a result, the corresponding file on the magnetic disk 23 contains initialization data Is.
is written, and the initialization operation ends. Therefore, as shown in Figure 6, it does not depend on the file size.

The processor 10 completes the initial M initialization process with one access to the device, and since initialization data is sent continuously to the file device side, there is no rotational wait. On the other hand, in the conventional method, if the file size is m bytes, 9 m bytes/8192 bytes of device access are required, and processing is divided for each access on the file side.

It is necessary to wait until a new corresponding truck arrives. Also, the capacity of the initialization data area IND on the main memory 11 is small (2 in the example shown).
, the increase in the instruction area CCW is slight, so the main memory can be used effectively.

In the above explanation, IQi2 (148 bytes) was used as an example, but other byte numbers may be used depending on the device configuration and processor, and the file device is not limited to a magnetic disk device, but may also include a floppy disk device, i-Q bubble memory, etc. It may be.Furthermore.

A channel device may be provided between the magnetic disk controller and the central processing unit, and the channel device may read the main memory.

Although the present invention has been described above using one embodiment, the present invention can be modified in various ways according to the spirit of the present invention.

These are not excluded from the present invention.

〔Effect of the invention〕

As explained above, according to the present invention, in a system including a processing device including a processor and a main memory, and a file device for storing files, the main memory includes a data area for storing one page of initialization data. and an instruction area for storing all the instruction groups necessary for the initialization process.

The processor sets in the command area a number of write commands corresponding to the number of writes of the one page of initialization data necessary to initialize the size of the file to be initialized, and the file device is initialized by the processor. When receiving a processing start command, the contents of the command area of the main memory are directly read, and for each write command, one page of initialization data of the data area is read and a file of the corresponding size is initialized. Therefore, regardless of the size of the file to be initialized, initialization can be performed with a single access to the device, which has the effect of shortening the running time of the processor and freeing the processor to perform other processing. In addition, since initialization data is sent continuously to file devices, it is possible to write to files continuously without waiting for initialization data, which reduces the time required for initialization processing. It is also effective. Furthermore, the area required for initialization processing of the main memory can be reduced, and the main memory can be used more effectively, which together greatly contribute to improving the performance of such computers.

[Brief explanation of drawings]

Figure 1 is a general computer system configuration diagram. Figure 2 is an explanatory diagram of the conventional initialization processing method, Figure 3 is an initialization data configuration diagram for initialization processing, Figure 4 is an illustration of the operation of the conventional initialization processing method, and Figure 5 is the diagram of the present invention. FIG. 6 is an explanatory diagram of an embodiment of the initialization processing method, and FIG. 6 is an explanatory diagram of the operation of the initialization processing according to the present invention. In the figure, 1... central processing unit, 2... file device. 10... Processor, ll... Main memory, 21
...Magnetic disk controller, 23...Magnetic disk. CCW...command area, IND...initialization data area. Patent Applicant Fujitsu Co., Ltd. Agent Patent Attorney Akira Yamatani; l-1 Figure 2 Figure 3 Figure 4 Figure ■ ■ −−1−−−−− ■ Phi + Li (■!2L−−− −−−−−−−−−−”
L-111111--"L-11111--" 6 Flash TART PSD size 10°

Claims (1)

[Claims] A processing device including a processor and a main memory. In a system having a file device for storing files, the main memory is provided with at least a data area for storing one page of initialization data and an instruction area for storing a group of instructions necessary for the initialization processing, The processor sets a number of write instructions in the instruction area according to the number of writes of the one page of initialization data necessary to initialize the size of the file to be initialized, and the file device receives the initialization processing from the processor. When a start command is received, the contents of the instruction area of the main memory are directly read, and for each write command, one page of initialization data of the data area is read and a file of the corresponding size is initialized. Features a high-speed file initialization processing method.