JPS61251944A - Buffer area control system - Google Patents

Abstract

PURPOSE:To improve the use efficiency of a memory to shorten the processing time by acquiring a preliminarily fixed buffer area when it is necessary and releasing it when it is unnecessary though a control program reserves the buffer area to be used. CONSTITUTION:When an input/output instruction is issued form a processing program AP2, it is checked whether required data is stored in a buffer area 27 or not in accordance with an intra-file relative position 8 of a block #1, and the number 6 of the blocks #1 to be used is counted up if it is stored there; but if required data is not stored there, data is read into a buffer area 28, and the using number 6 of blocks #2 is counted up, and information is set to the block #2 to perform the input/output processing. When the close instruction of a file A is issued from a processing program AP1, the use number of the blocks #1 is counted down, and one unused buffer area is released, and a buffer management block total number 2 of a table 25 is counted down.

Description

DETAILED DESCRIPTION OF THE INVENTION [Summary] In a computer system, a buffer area used by a control program when executing a processing program.

It is acquired dynamically each time it is needed by the control program.

Release it when it is no longer needed. This improves memory usage efficiency.

[Industrial application field]

The present invention relates to a control method for acquiring and releasing buffer areas in a computer system.

[Problems to be solved by conventional technology and the invention] In a computer system, when a processing system is executed, a buffer area used by a control program working together with a processing program must be acquired and released after processing is completed. .

Conventionally, there are two methods for securing a buffer area for use by a control program: one is to secure it on the processing program side and then call the control program, and the other is to prepare it permanently on the system side and use it whenever it is needed by the processing program. There was a method.

These methods will be specifically explained below.

FIG. 3 shows a method in which a buffer area used by a control program is prepared on the processing program side.

fa+ in the same figure represents the processing flow, and the same figure (bl represents the memory map. In the same figure (bl, 31 is the memory, 32
denotes a control program, and 33 and 34 denote buffer areas secured by processing programs A and B, respectively.

As shown in the memory map of No. 311 (bl), when arbitrary processing programs A and B are executed simultaneously, the control program, processing program A, and processing program B are loaded onto memory, and each of processing program A and processing program B is An independent buffer area A and buffer area B are provided for each program.In other words, the number of buffer areas required is equal to the number of programs running simultaneously.

Here, as shown in the processing flow in Figure 3+al.

When a certain processing program is started, the started processing program executes a predetermined process and calls a control program when necessary, for example, to execute an input/output command. The called control program performs the requested processing on the buffer area reserved by the calling processing program.

In the system shown in FIG. 3, the buffer area secured by each processing program is independent and cannot be shared by other processing programs. Therefore, even if the necessary data is stored in other buffer areas,
It was necessary to always store and use the same data in the buffer area of each processing program.

Therefore, the processing time for storing data redundantly,
There was a problem that extra program steps, memory space, etc. were required.

Next, FIG. 4 shows a method in which the control program secures a buffer area in a fixed manner.

In the figure (bl), 41 is a memory, 42 is a control program, and 43 is a fixed buffer area.

As shown in the memory map in Figure 4 (bl), the control program predicts the startup of multiple processing programs in advance, secures a buffer area of the maximum necessary size, and responds to the needs of the processing programs. We will use them sequentially.

Then, as shown in Figure 4 (al), when a certain processing program is started and the control program is called during execution, the control program checks whether the necessary data is in the buffer area, and if there is, Processing is performed using existing data, and if there is no data, an empty buffer area is used to read and process the data.

In the method shown in FIG. 4, it is possible for a plurality of processing programs to share the buffer area.

Since a buffer area of the maximum necessary size is fixedly secured in advance, a large amount of buffer area is always occupied on the memory, resulting in a problem that memory usage efficiency decreases.

[Means for solving problems]

The present invention secures the buffer area used by the control program on the control program side, but does not secure a fixed buffer area in advance, but acquires the area when it becomes necessary. By releasing at the point,
This solves the problems of the conventional method described above.

The principle of the present invention force formula will be explained using FIG. 1(a).

Figure 1 (al) shows the power formula of the present invention in the form of a processing flow by one exemplary method. In the figure, 1 is a processing program, 2 is a control program, 3 is a disk,
to ■ are processing procedures.

In operation, after the processing program 1 is started.

■ Processing program 1 issues an open command to make the file usable.

(2) The control program 2 is called by the open command issued by the processing program 1, and acquires a buffer area of a predetermined size from the free space of the memory.

(2) The processing program 1 issues a READ command or a WRITE command in order to input/output data to/from the disk 3.

(2) In response, the control program 3 checks eight times whether the necessary data is stored in the buffer area acquired by the other processing program.

■ If a buffer area storing the necessary data does not exist, write to one of the free buffer areas that has already been acquired.

Transfers the record from the processing program, and reads data from the disk if it is READ.

■ If a buffer area containing the necessary data is found in step ■, share that buffer area, or read new data into the empty buffer area in step
If it is ITE, write data to the disk and also write the data to the disk.
If it is EAD, input/output processing is executed to transfer the record to the processing program.

■ A processing program issues a close instruction to free the file.

■ In response, the control program frees one of the buffer areas not used by any processing program.

With the above steps, the processing from file acquisition to release ends.

[Effect]

The operation of the method of the present invention will be explained using an example of the operation shown in FIG. This is what is shown.

i) Only the control program is loaded onto the memory.

ii) Processing program-1 is started, and processing program-1 issues a file open command.

Therefore, the control program acquires buffer area-1 from the free space in the memory and causes processing program-1 to use it. The number of open files is 1.

iii) Processing program-2 is then activated and issues a file open command. Therefore, the control program acquires buffer area -2.

However, the input/output instructions of processing program-2 can use the data in buffer area-1, so
1 will be shared. The number of open files is 2, and buffer area -2 is empty.

iv) The processing program-1 ends and closes the file, but the control program releases the free buffer area-2 because the buffer area-1 is being used by the processing program-2. The number of open files is 1.

In addition, in 1ii), the processing program-2 is the buffer area-
If the data of 1 cannot be shared.

Buffer area-2 is used, and in iv) buffer area-2 is used.
Release 1.

Furthermore, the number of open files and the number of acquired buffer areas always match.

〔Example〕

The details of the system according to the first embodiment of the present invention will be explained with reference to FIG.

In FIG. 2, 21 is a memory, 22 is a control program, 23 is a processing program API, 24 is a processing program AP2.25 is a file management table, 26 is a bath management block, 27 is a buffer area BFI, and 28 is a buffer area BF2.29. represents data file A.

The file management table 25 is a table for managing the usage status of files, and ■ is a pointer indicating the first buffer management block 26. ■ is the total number of buffer management blocks used for data file A, ■ is the file name, and ■ is file detailed information such as record length.

The buffer management block 26 is information that manages the usage state of the buffer area, where ■ is a pointer indicating the position of the next buffer management block, ■ is the number of uses of this buffer management block, and ■ is the starting position of the buffer area. pointer,
■ is the relative position within the file of the data in the buffer.

Next, the operation will be explained using a specific example.

a. Issue an open command for data file A from the processing program (API) a-1) Acquire buffer area BFI a-2) Point the buffer management block from the file management table 25 and count the total number of buffer management blocks ■ (set it to 1).

b. Issue an input/output command from the processing program (API) b-1) Read data into the buffer area BFI, count up the number of uses of buffer management block #1 and set it to 1), and write information to buffer management block #1. Set.

b-2) Perform input/output processing.

Step 1 Issue an open command for data file A from the C6 processing program (AP2) C-1) Acquire buffer area BF2 cJ) Point buffer management block #1 to rehabilitation sofa management block #2.

File management table buffer tube Count the total number of blocks Do (make it 2).

d. Issue an input/output command from the processing program (AP2) d-1) Check whether the necessary data is stored in the buffer area BFI according to the relative position in the file of buffer management block #1.

d2) When the necessary data is stored in the buffer area BFI, count up the number of uses of buffer management block #1 (set it to 2).

Data required for buffer area BFI is not stored, the buffer Load data into M area BF2 and Number of buffer management block #2 used■ count up as 1).

Information in buffer management block #2 Set.

d-3) Perform input/output processing.

e. Issue a close command for file A from the processing program (API) e-1) Count down the number of uses of the buffer management block #1 used by the processing program (API).

e-2) Refer to the buffer management block and release one buffer area that is not being used (that is, the number of uses is 0).

e-3) Count down the total number of buffer management blocks ■ in the file management table (set it to 1).

e-4) When the buffer managed by the first buffer management block (buffer jJIJliBF1) is released, point the first buffer management block (buffer management block #2) as the first buffer management block from the file management table. .

Buffers other than the first buffer management block Buffer management block (buffer management Block #2) the top area (buffer area BF2) If released, remove the previous buffer tube. change the pointer from the logical block.

Change the pointer to the buffer management block that was pointing to the buffer management block that was managing the released buffer area. (In the case shown in the figure, there is no ninth-order buffer management block, so the pointer of buffer management block #1 does not point anywhere.) In this way, requests from any number of processing programs can be responded to. A buffer area is acquired as needed, and is released when all programs using the buffer area complete their processing, and the size of the buffer area is dynamically increased or decreased.

〔Effect of the invention〕

As described above, according to the present invention, the buffer area is.

Memory is acquired by the control program only when it is needed, and is shared as much as possible by multiple processing programs, which improves memory usage efficiency5 and reduces the burden on programs, reducing processing time. It becomes possible.

[Brief explanation of drawings]

Figure 1 (al is the processing flow of the method of the present invention, Figure 1 (bl
2 is an explanatory diagram showing an example of the operation of the system of the present invention, FIG. 2 is a configuration diagram of an embodiment of the system of the system of the present invention, and FIGS. 3 and 4 are explanatory diagrams of different conventional systems. In FIG. ■ is a processing program, 2 is a control program. 3 is a disc. shows. Patent applicant Fumihiro Hase (and 2 others), patent attorney for Usatsuk Electronics Industry Co., Ltd.; 1. of. Power 3 Diagram Ca) Kitop Tsum L Scream'yo T 4/4 N. (b) (b) -1”I L

Claims (1)

[Claims] In a computer system that can process multiple processing programs in parallel, when a processing program issues a file open command, the control program acquires a buffer area from free space in memory, and then When a processing program issues a data input/output command, it checks whether the data necessary for processing exists in the already acquired buffer area, and if the data exists, it The buffer area is used, and if no data exists, new data is read and used in the previously acquired free buffer area to perform input/output processing, and then a processing program closes the file. A buffer area control method characterized in that when an instruction is issued, one of the buffer areas not used by any processing program is released.