JPS596410B2 - Buffer area control processing method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a buffer area control processing method, particularly in a data processing system that uses a buffer area upstream of a main memory as a communication data storage unit in communication with a terminal device, etc. A buffer area control processing method that performs control so that at least one buffer area is left in the lending process of lending a buffer area, and performs the 5 processing aspects in the buffer area return process in a unified manner. It is related to.

Conventionally, for example, when transferring data to and from a terminal device, a buffer area is prepared on the main memory and the buffer area is used as a temporary storage area for communication data. .

That is, when transmitting data from the central processing unit side to a terminal device, a plurality of the buffer areas are borrowed, links are provided between the plurality of buffer areas, and the transmission data is stored. The channel device then reads the transmission data from the buffer area and transmits it to the terminal device. As a result, the previously borrowed buffer areas are returned. The case of receiving data from the terminal device side is also the reverse of the above, and when the central processing unit reads the received data stored in the buffer area, the buffer area is returned.

In managing such buffer areas, an area generally called a queue/terminal is prepared on the main memory, and the address address of the vacant buffer area is assigned to the queue/terminal. When the buffer area is rented out, the address information of the buffer area is extracted from the link and the buffer area is managed.

In this case, as will be described later with reference to FIG. 1, conventionally all of the buffer areas linked to one queue terminal are rented out. For this reason, the processing when the buffer area is returned to the relevant queue/terminal is as follows: (1) There is no remaining buffer area linked to the relevant queue/terminal;
(1;) There were two cases, depending on whether some remained, and it was difficult to perform unified processing. The present invention aims to solve the above-mentioned problems, and the buffer area control processing method of the present invention includes a plurality of buffer areas and queue terminals, and when the buffer areas are linked and used, The address information of the linked first buffer area is stored in the linked first buffer area address storage area of the queue terminal, and the address information of each linked buffer area is sequentially written to create a chain and the above buffer area is stored. 1. Store the address information of the linked final buffer area in the linked final buffer area address storage area of the terminal;
In the data processing system having a buffer area control function for managing the lending and return of buffer areas, the queue terminal/terminal manager maintains information on the number of buffer areas that are linked to each other, and performs the lending process. The requested number of buffer areas to be rented is compared with the number information on the queue terminal, and in the buffer area lending process, the number of buffer areas linked to the queue terminal is compared. The above lending is permitted within a range in which at least one remains, and lending requests exceeding the range are rejected.

This will be explained below with reference to the drawings. Figures 1A to 1E are explanatory diagrams illustrating problems in conventional processing methods;
, B is an explanatory diagram illustrating a processing method according to an embodiment of the present invention that solves the above-mentioned problems, FIG. 3 is a flowchart showing a processing mode in the case of the conventional method, and FIG. 4 is a diagram showing the processing mode in the case of the method of the present invention. FIG. 5 is a flowchart showing the processing mode of FIG. 5, which shows an example configuration in which the method of the present invention is executed.

Conventionally, for example, when there are buffer areas Ql, Q2, ..., Qm, and the vacant buffer areas are linked and registered on the queue terminal, and management is performed in response to rentals and returns. , is carried out as shown in FIG.

In Figure 1, 1 is the terminal, 2-1
2-m to 2-m are respectively buffer areas, 3 is a link number information storage area which indicates the number of buffer areas currently linked to the relevant queue/terminal 1, and 4 is a link start buffer area address storage area. Figure 1 A
In the illustrated example, address information BA of buffer area 2-1
5 is the negist queue terminal address storage area where the address information of the next queue terminal following queue terminal 1 is stored, and 6 is the link final buffer. This is the area address storage area, and in the case of the example shown in FIG. 1A, the buffer area 2-m
The address information BADm is stored. (1) The free buffer areas currently linked to one queue/terminal 1 are areas 2-1, 2-3, 2-4.
, 2-6, 2-m, as shown in FIG. Further, address information BADm is set in the storage area 6.

In addition, in FIG. 1, BAD3 etc. written in a circle frame indicate the address information of the next linked buffer area. (2) In this state, if three buffer areas are borrowed for the queue terminal 1, the buffer areas 2-1, 2-3 will change from the state shown in Figure 1A.
, 2-4 are lent, and the rewriting of the queue 1/terminal 1 is performed as shown in FIG. 1B. That is, the content of the storage area 3 of the terminal 1 is set to "2", and the content of the storage area 4 is set to BAD6. That is, the contents of storage areas 3 and 4 are changed. (3) When the two buffer areas 2-2 and 2-1 are returned in the state shown in FIG. 1B, extract the buffer area 2-m at the end of the link in the state shown in FIG. Following area 2-m, buffer areas 2-2 and 2-1
At the same time, as shown in FIG.

That is, the contents of storage areas 3 and 6 are changed. (4) Figure 1C
In the illustrated state, when four buffer areas are lent, buffer areas 2-6, 2-M, 2-2, 2-1
We will rent out.

At this time, there is no longer anything to be linked in terminal 1, and as shown in Figure 1D, the contents of storage area 3 are only set to "O," and the contents of storage areas 4 and 6 are generally will never be cleared. (5) In the state shown in FIG. 1D, when the two buffer areas 2-3 and 2-m are returned, the storage area of the terminal 1 will be returned as shown in FIG. 1E. The content of storage area 3 is changed to "2", the content of storage area 4 is changed to BAD3, and the content of storage area 6 is changed to BADm.

That is, the contents of storage areas 3, 4, and 6 are changed. Conventionally, the above-mentioned processing has been performed, but there are the following problems. That is, when some buffer areas are returned as shown in FIG. 1C while at least one buffer area remains on the link as shown in FIG. Change the contents of On the other hand, in a state where no linked buffer area remains as shown in Figure 1D, the first
When returned as shown in Figure E, the contents of storage areas 3, 4, and 6 need to be changed. In the former case, it is necessary to extract the final buffer area 2-m under the state shown in FIG. 1B and add the buffer area 2-2 as shown in FIG. In this case, there is no need to do so. For this reason,
In the buffer area return processing described above, it is necessary to check whether or not there is a buffer area that is actually linked to the queue terminal 1, and to perform two types of processing depending on the result.

That is, it is necessary to execute process X as shown in the flowchart shown in FIG. 3, and then execute either process Y or Z. This is due to the fact that the entire buffer area linked to the queue terminal 1 has been rented out, as shown in FIG. 1D. And Kiyuichi Terminal 1
As the number of objects increases, it becomes extremely complicated to carry out the above two types of processing. The present invention solves the above-mentioned problems, and when lending buffer areas as described above, at least one buffer area remains on the link,
This allows the above-mentioned processing to be performed in a unified manner.

In other words, under the situation shown in Figure 1C, if there is a request to lend four items, the request will be rejected, and the loan will only be made if there is a request to lend three or less items. .
FIG. 2 explains the process after lending out three or less items from the state shown in FIG. 1C.

That is, 6) If three items are lent under the condition shown in FIG. 1C, it is sufficient to change the storage areas 3 and 4 as shown in FIG. 2A.

This state is the same as the change from the state shown in FIG. 1A to the state shown in FIG. 1B. 7) And the second
Under the condition shown in Figure A, buffer areas 2-3 and 2-m
When the return is made, storage area 3 is returned as shown in Figure 2B.
and 6 are changed, and buffer area 2-3 is added to buffer area 2-1.

This process is the same as when the state shown in FIG. 1B is changed to the state shown in FIG. 1C. Therefore, when the processing method of the present invention is adopted, it is sufficient to prepare only the processing Y as shown in FIG. 4 for the return processing.

FIG. 5 shows an exemplary configuration in which the method of the invention is carried out.

The symbols 1 (1-1, 1-2, ...), 2-1 to 2-M, 3, 4, 5, and 6 in the figures correspond to FIGS. 1 and 2,
MEM is a main memory, 7-1 to 71m are respective next buffer area address storage areas, 8 is a buffer management function unit, 8
-1 is a buffer lending processing unit, 8-2 is a buffer return processing unit, ADl, AD2, ..., BADl, BAD2,
... each represents address information. Note that the processing function of the buffer return processing section 8-2 may be considered to have a function according to the flowchart shown in FIG. The buffer lending processing section 8-1 is provided with the following processing functions.

In other words, if there is a request to rent A buffer areas, the queue terminal 1-1 (or 1-2...) I
The content C of the storage area 3 in this storage area is checked, and A≧(0-1
・・・・・・・・・・・・・・・・・・・・・(
1) is checked.

If the equation (1) is satisfied, the lending request is rejected. In addition, in the case of the conventional method shown in FIG.
A≧(0 ・・・・・・・・・・・・・・・
・・・・・・(2) is checked, and in the case of the present invention, it is assumed that the contents of the storage area 3 of the queue terminal 1 are substantially (C)-1. The processing does not become more complicated than conventional methods. Note that the state shown in FIG. 5 corresponds to FIG. 1A.

In other words, on storage area 3 of terminal 1-2, the number of
5" is set, and address information BADl is stored in storage area 4.
is set, and address information BADm is set in the storage area 6. If a request for lending three buffer areas is received in this state, the buffer lending processing section 8-1 performs a check according to the above equation (1). Next, the buffer area 2-1 is examined using the contents BADl of the storage area 4 of the queue terminal 1-2 as a starting point, and the buffer area 2 is determined based on the contents of the storage area 7-1 within the buffer area 2-1. -3, and based on the contents of the storage areas 7-3 in the buffer area 2-3,
Find out.

Then, the contents of the storage area 7-4 in the buffer area 2-4 are set in the storage area 4 of the queue/terminal 1-2.
At the same time, the contents of storage area 3 are set to "2". Also, Figure 2A
When the buffer areas 2-3 and 2-m are returned under the illustrated state, the buffer return processing unit 8-2
Figure 1: Perform the following processing.

That is, the contents of the storage area 3 in the terminal 1-2 are changed to "3".

In addition, storage area 6 at Kyuichi Terminal 1-2
The buffer area 2-1 is checked based on the contents (BADl in the case of FIG. 2A), and the address information BAD3 is written in the storage area 7-1 in the buffer area 2-1. The information BADm is written to the storage area 7-3 in 3.

Address information BADm is then written into the storage area 6 in the key terminal 1-2. As explained above, according to the present invention, the buffer lending process and the buffer return process are performed in a unified manner regardless of whether there are no links at all or a few links remain on the queue terminal. It becomes possible to do it.

[Brief explanation of drawings]

1A to 1E are explanatory diagrams illustrating problems in the conventional processing method, FIGS. 2A and 2B are explanatory diagrams illustrating an embodiment of the processing method of the present invention that solves the above problems, and FIG. 3 is a flow chart representing the processing mode in the case of the conventional method,
FIG. 4 is a flow chart showing the processing mode of the method of the present invention, and FIG. 5 shows the configuration of an embodiment in which the method of the present invention is executed. In the figure, 1, 1-1, 1-2, . . . are queue terminals, 2-1 to 2-m are buffer areas, 3 is a link number information storage area, and 4 is a link start buffer area address storage. area, 5 is the negist queue terminal address storage area, 6 is the link final buffer area address storage area, 7-1 to 71m is the next buffer area address storage area, 8 is the buffer management function section, 8-1 8-2 represents a buffer lending processing section, 8-2 represents a buffer return processing section, and MEM represents a main storage device.

Claims (1)

[Scope of Claims] 1. A plurality of buffer areas and queue terminals are provided, and when the buffer areas are linked and used, the link head buffer area address storage area of the queue terminal is linked to the head linked buffer area. The address information of the buffer area is stored, and the address information of each linked buffer area is sequentially written to create a chain, and the linked final buffer area is stored in the linked final buffer area address storage area of the queue terminal. In a data processing system having a buffer area control function that stores address information of and manages lending and returning of the buffer area, the queue terminal maintains information on the number of buffer areas linked to itself; In the lending process, the requested number of buffer areas is compared with the number information on the queue terminal, and in the lending process of the buffer area, the buffer area linked to the queue terminal is A buffer area control processing method, characterized in that the lending is allowed within a range in which at least one area remains, and requests for lending exceeding the range are rejected. 2. The buffer area control function sets the address information of the first buffer area after lending to the link first buffer area address storage area of the queue terminal in the above lending process within the allowed number range. A buffer area control processing method according to claim 1, characterized in that: 3 In the buffer area return process, the buffer area control function links the returned buffer area to the final buffer area immediately before the return, and also links the address information of the final buffer area after the return to the queue.・
3. The buffer area control processing method according to claim 1, wherein the buffer area control processing method is set in a link final buffer area address storage area of a terminal.