JPH11143800A - Communication flow control system for computer and method therefor and recording medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To continue recessing of an application program without increasing the load of memory resources of this system even when the amount of requests for output instruction of the application programs exceeds a specified level. SOLUTION: In a communication flow control system of a computer in which an application program 1, a communication control driver 2, and a communication board 4 are asynchronously operated, an output instruction request queue 3 is provided in a kernel space, and a request holding file 7 is provided in a user space for accepting an output instruction from the application program 1. When more than a certain amount of output instructions are stored in the output instruction request queue 3, subsequent output instructions from the application program 1 are held in the request holding file 7, and when less than a certain amount of output instructions are stored in the output instruction request queue 3, the output instructions held in the request holding file 7 are shifted to the output instruction request queue 3.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a computer communication flow control method, method and recording medium, and more particularly to a computer communication flow control method, method and recording when an application program operates asynchronously with a communication control driver. Regarding the medium.

[0002]

2. Description of the Related Art In a computer equipped with a conventional communication board, an output command to the communication board may not be completed for a long time. During this time, when a request for an output command is continuously generated from the application program, the communication control driver manages the request in a queue. However, if the number of requests becomes enormous, flow control works to prevent the kernel memory resources from being squeezed, and requests exceeding a certain amount cannot be accepted. Therefore, a request for an output command from the application program to the communication control driver may be an error or a wait, and the application program operation may end abnormally or stop.

[0003] In communication between computers or between a computer and a terminal, a communication flow control system for transmitting a control code and controlling data transmission from a transmitting side when received data is stagnant is disclosed in Japanese Patent Laid-Open Publication No. Hei. -78057. The communication flow control method disclosed in this publication transmits a control code a plurality of times in order to more reliably suppress transmission. In this case, the other computer whose transmission has been suppressed waits until transmission output processing becomes possible again by the communication board or the communication control driver. Requests from the application program during that time can be accumulated up to a certain amount that does not compress the memory resources of the entire system.

[0004]

In the above-mentioned prior art, when a request for an output command from an application program is accumulated for a certain amount or more, the application program which operates asynchronously with the communication control driver is accumulated. However, there is no choice but to wait until the requested request is processed or to terminate abnormally, which affects the business processing of the application program.

An object of the present invention is to continue processing of an application program without increasing the load on the memory resources of the system even when the request for an output instruction from the application program exceeds a certain amount.

[0006]

According to a first aspect of the present invention, there is provided a communication flow control method for a computer in which an application program, a communication control driver, and a communication board operate asynchronously. In order to receive an output instruction from a program, a first output instruction request queue is provided in a kernel space, and a second output instruction request queue is provided in a user space. When the data is stored as described above, a subsequent output command from the application program is held in the second output command request queue.

The second computer communication flow control system of the present invention is a computer communication flow control system in which an application program, a communication control driver, and a communication board operate asynchronously, and receives an output command from the application program. Therefore, a first output instruction request queue is provided in the kernel space, and a second output instruction request queue is provided in the user space, and when a predetermined amount or more of output instructions are accumulated in the first output instruction request queue, A subsequent output command from the application program is held in the second output command request queue, and when the number of output commands in the first output command request queue becomes smaller than a certain value, it is a difference from the certain value. A free space is calculated, and output commands corresponding to the free space are transferred from the second output command request queue to the first output command request queue.

A third communication flow control method for a computer according to the present invention is a communication flow control method for a computer in which an application program, a communication control driver, and a communication board operate asynchronously. Queuing a first output instruction request queue provided in a kernel space, detecting whether an output instruction is queued in the first output instruction request queue in response to an output instruction from the application program, and If the instruction did not exist, issue an output instruction to the communication board,
If it is less than a certain amount, the output command received from the application program is queued in the first output command request queue. A communication control driver provided in the kernel space, which is to be passed to the flow control support module, and a data board that is provided in the kernel space and outputs data to the outside in accordance with an output command issued from the communication control driver; and If the amount is equal to or more than a fixed amount, the output instruction received from the communication control driver is notified to the flow control support daemon, a flow control support module provided in the kernel space, and the output instruction received from the flow control support module. A flow control support daemon provided in the user space, which is held in the second output instruction request queue; It holds the output command received from the serial flow control support module, and having a second output instruction request queue provided in the user space, the.

A fourth communication flow control method for a computer according to the present invention is a communication flow control method for a computer in which an application program, a communication control driver, and a communication board operate asynchronously. Queuing a first output instruction request queue provided in a kernel space, detecting whether an output instruction is queued in the first output instruction request queue in response to an output instruction from the application program, and If the instruction did not exist, issue an output instruction to the communication board,
If it is less than a certain amount, the output command received from the application program is queued in the first output command request queue. First processing to be passed to the flow control support module, and when an output request is completed in the communication board, it is detected whether an output instruction is queued in the first output instruction request queue, and the output instruction is present. In the case, after the first output instruction is dequeued and the output instruction is issued to the communication board, if the output instruction of the first output instruction request queue is less than a certain amount, the output instruction is provided in the user space. A second process for instructing the flow control support module to read the output command held in the second output command request queue, and a communication provided in a kernel space for performing the second process. And your driver, the data output to the outside in accordance with published output instruction from the communication control driver, a communication board provided in the kernel space, the first
In the processing of (1), when the number of output instructions queued in the first output instruction request queue is equal to or more than a certain amount, the output instruction received from the communication control driver is notified to the flow control support daemon, and the second When the number of output instructions queued in the first output instruction request queue is less than a certain amount, a free space that is a difference from the certain amount is calculated, and the number of output instructions corresponding to the free space is calculated. To the flow control support daemon to read from the second output instruction request queue, and queue the output instruction received from the flow control support daemon in the first output instruction request queue. The flow control support module and the first processing,
An output instruction received from the flow control support module is held in a second output instruction request queue, and in the second processing, an output instruction corresponding to the free space is read from the head of the second output instruction request queue. A flow control support daemon provided in a user space to be passed to the flow control support module, and a second command provided in a user space for holding an output instruction received from the flow control support module.
And an output command request queue.

According to a fifth aspect of the present invention, there is provided a communication flow control method for a computer according to the first to fourth computers, wherein the predetermined amount is the number of output instructions. .

According to a sixth aspect of the present invention, in the communication flow control method of the first to fourth computers, the predetermined amount is a total memory amount of queued output instructions. There is a feature.

A first communication flow control method for a computer according to the present invention is a communication flow control method for a computer in which an application program, a communication control driver, and a communication board operate asynchronously. Detecting whether an output instruction is queued in the first output instruction request queue provided in the kernel space, and if there is no output instruction, the communication board can receive the output instruction. A first step of issuing an output instruction to the communication board after confirming that the state is a state, and when the number of output instructions queued in the first output instruction request queue is less than a certain amount, A second step of queuing an output instruction received from the application program in the first output instruction request queue; The output queued in queue command when was a certain amount or more, provided the output command received from the application program to the user space 2
A third step of holding an output instruction in the output instruction request queue, and detecting whether an output instruction is queued in the first output instruction request queue when the processing of the communication board is completed, In this case, in the fourth step of dequeuing the first output instruction and issuing the output instruction to the communication board, in the fourth step,
If the number of output instructions queued in the first output instruction request queue is less than a certain amount, a free space, which is a difference from the certain amount, is calculated, and the head of the second output instruction request queue is A fifth step of reading out the output instructions corresponding to the free space and queuing them in the first output instruction request queue.

[0013] In a second communication flow control method for a computer according to the present invention, in the first communication flow control method for a computer according to the first invention, the predetermined amount is the number of queued output instructions. And

According to a third communication flow control method for a computer according to the present invention, in the first computer communication flow control method according to the first invention, the fixed amount is a total memory amount of queued output instructions. It is characterized by.

The recording medium of the present invention, when receiving an output instruction from an application program, detects whether the output instruction is queued in a first output instruction request queue provided in the kernel space, and determines whether the output instruction is queued. If not, a first process for confirming that the communication board is ready to receive the output command and issuing the output command to the communication board; and queuing the first output command request queue. If the number of output instructions being executed is less than a predetermined amount, a second process of queuing the output instructions received from the application program in the first output instruction request queue; If the number of output instructions queued in the queue is equal to or more than a certain amount, the output instruction received from the application program is stored in the second space provided in the user space.
A third process that is held in the output command request queue and, when the process of the communication board is completed, whether the output command is queued in the first output command request queue is detected. If so, a fourth process of dequeuing the first output command and issuing the output command to the communication board, and in the fourth process, the command is queued in the first output command request queue. If the number of output instructions is less than the predetermined amount, the empty amount which is a difference from the predetermined amount is calculated, the output instructions corresponding to the empty amount are read from the head of the second output instruction request queue, and the first instruction is read out. A program for causing a computer to execute the fifth process of queuing in the output command request queue.

[0016]

Embodiments of the present invention will be described with reference to the drawings.

First, a first embodiment of the present invention will be described with reference to the drawings.

FIG. 1 is a block diagram showing a first embodiment of the present invention. An application program 1 such as a business application and a communication control driver which receives an output request from the application program 1 and controls a communication board 4 are shown. 2, an output command request queue 3 for queuing unprocessed requests among output commands from the application program 1, a communication board 4 connected to a business application monitoring terminal, It is composed of a flow control support module 5 for notifying the user space, a flow control support daemon 6 for reading and writing the same in the reception request holding file 7, and a request holding file 7 for holding output requests. Here, it is assumed that the application program 1 operates asynchronously with the communication control driver 2 and the communication board 4 irrespective of the state of the communication board 4.

Next, the operation of the first embodiment of the present invention will be described in detail with reference to FIGS.

First, an operation when an output request is generated from an application program will be described with reference to FIGS. FIG. 2 is a flowchart showing an operation when an output request is generated from the application program 1.

When an output is generated to a monitoring terminal (not shown) connected to the communication board 4, the application program 1 issues an output request to the communication control driver 2 (step S1). The communication control driver 2 receives an output request issued from the application program 1. At this time,
In the kernel space where the communication control driver 2 exists,
Requests from application programs are handled as request blocks. Further, the application program operates asynchronously with the communication control driver 2, so that even if the application program 1 issues an output request to the communication control driver 2, the processing can proceed without waiting for the output completion (step S1). S2).

The communication control driver 2 receiving the request checks whether an output instruction can be issued to the communication board 4. Here, the criterion as to whether or not it can be issued is based on the output instruction request queue 3
Does not exist in the communication board 4 and the communication board 4 is ready to receive the output command (step S
3).

If possible, the communication control driver 2 issues an output command to the communication board 4 (step S).
4). If the output instruction cannot be issued, it is next checked whether the output instruction request queue 3 can be queued. The criterion of the possibility is that queuing is possible when a request for an output instruction of a certain amount or more has not already been queued (step S5).

Here, if queuing is possible, it is queued in the output instruction request queue 3 (step S6),
If queuing is not possible, the request block is passed to the flow control support module 5 (step S7).

The flow control support module 5 copies the image of the request block as it is to a memory included in the flow control support daemon 6 in the user space and passes it to the flow control support daemon 6 (step S8). Then, the flow control support daemon 6 writes and holds the request block in the request holding file 7 (step S9).

Next, the operation when the processing of the communication board 4 is completed will be described with reference to FIGS. FIG.
5 is a flowchart showing the operation when the processing of the communication board 4 is completed.

When the output request is completed in the communication board 4 (step S11), the communication control driver 2 first checks whether or not a request block exists in the output command request queue 3 (step S12). The first request block is dequeued and an output instruction is issued to the communication board 4 (step S13).

Next, the communication control driver 2 checks whether or not the request block can be queued in the output instruction request queue 3. As a criterion of whether or not it is possible, it is determined that queuing is possible when a request block in the queue is less than a certain amount (step S14).

If the amount is equal to or less than the certain amount, the flow control support module 5 accesses the output command request queue 3 to calculate a free amount that is a difference from the certain amount, and sends a request to the flow control support daemon 6. An instruction is issued to read the required blocks for the free space from the beginning of the holding file 7 (step S15). The flow control support daemon 6 passes the request block read from the request holding file 7 to the flow control support module 5 (Step S16). The flow control support module 5 queues the received request block in the output instruction request queue 3 (Step S17).

Next, a second embodiment of the present invention will be described with reference to the drawings.

FIG. 4 is a block diagram showing a second embodiment of the present invention, which comprises a computer and a recording medium 8. The computer is basically the same as that described in the first embodiment of the present invention in FIG. Recording medium 8
Records a computer communication flow control program. The recording medium 8 may be a magnetic disk, an optical recording disk, a semiconductor memory, or another recording medium.
The communication flow control program of the computer is read from the recording medium 8 into the computer, and controls the same operation as in the first embodiment of the present invention.

In the above description, it has been described that it is checked whether or not a request of a predetermined amount or more of output instructions is queued in the output instruction request queue 3. However, this fixed amount is determined by the number of queued output instructions. Or may be defined by the total memory amount of output instructions.

[0033]

The first effect of the present invention is that even if the request for an output command from an application program operating asynchronously with the communication board exceeds a certain amount, the processing of the output command of the communication board is delayed or stopped. Since the output request is continuously received without being affected by abnormal termination or waiting, the application program can continue the processing as it is.

The second effect is that the above control is realized without increasing the load on the memory resources of the system.

[Brief description of the drawings]

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

FIG. 2 is a flowchart illustrating an operation when an output request is generated from an application program 1;

FIG. 3 is a flowchart showing an operation when processing of the communication board 4 is completed.

FIG. 4 is a block diagram showing a second embodiment of the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Application program 2 Communication control driver 3 Output command request queue 4 Communication board 5 Flow control support module 6 Flow control support daemon 7 Request holding file 8 Recording medium

Claims (10)

[Claims] 1. A communication flow control method for a computer in which an application program, a communication control driver, and a communication board operate asynchronously, wherein a first output instruction is provided in a kernel space for receiving an output instruction from the application program. A request queue is provided with a second output command request queue in the user space, and when an output command is accumulated in the first output command request queue in a predetermined amount or more, a subsequent output command from the application program is transmitted. A communication flow control method for a computer, wherein the communication flow control method is held in the second output instruction request queue. 2. A communication flow control method for a computer in which an application program, a communication control driver, and a communication board operate asynchronously, wherein a first output instruction request is issued to a kernel space in order to receive an output instruction from the application program. A queue is provided with a second output command request queue in the user space, and when the output commands are accumulated in the first output command request queue in a predetermined amount or more, the output commands from the application program thereafter are sent to the first output command request queue. When the number of output instructions in the first output instruction request queue is less than a certain amount, a free amount that is a difference from the certain amount is calculated and held in the second output instruction request queue. A computer communication flow control method, wherein instructions are transferred from the second output instruction request queue to the first output instruction request queue. 3. A communication flow control method for a computer in which an application program, a communication control driver, and a communication board operate asynchronously, wherein a first command provided in a kernel space for queuing an output instruction from the application program is provided. And an output command request queue for receiving the output command from the application program, detecting whether the output command is queued in the first output command request queue, and, when no output command is present, the communication board An output instruction is issued to the first application if the output instruction is less than a predetermined amount.
And a communication control driver provided in a kernel space for passing an output instruction received from the application program to a flow control support module when the output instruction request queue is equal to or more than a predetermined amount. A communication board provided in a kernel space for outputting data externally in accordance with an output instruction issued from a driver; and, when the number of the output instructions is equal to or more than a predetermined amount, controlling an output instruction received from the communication control driver. A flow control support module provided in a kernel space for notifying a support daemon; and a flow control support daemon provided in a user space for holding an output command received from the flow control support module in a second output command request queue. And a user space for holding an output command received from the flow control support module. A communication flow control method for a computer, comprising: a second output instruction request queue provided in the computer. 4. A communication flow control method for a computer in which an application program, a communication control driver, and a communication board operate asynchronously, wherein a first command provided in a kernel space for queuing an output instruction from the application program. And an output command request queue for receiving the output command from the application program, detecting whether the output command is queued in the first output command request queue, and, when no output command is present, the communication board An output instruction is issued to the first application if the output instruction is less than a predetermined amount.
If the output command request queue is queued in the output command request queue and the output command request queue is equal to or more than a predetermined amount, a first process of passing the output command received from the application program to the flow control support module, and the output request is completed in the communication board At that time, it is detected whether or not an output instruction is queued in the first output instruction request queue. If an output instruction is present, the first output instruction is dequeued and the output instruction is issued to the communication board. Thereafter, when the number of output instructions in the first output instruction request queue is less than a certain amount, the flow control support is performed so as to read the output instruction held in the second output instruction request queue provided in the user space. A communication control driver provided in a kernel space for performing a second process for instructing the module, and an external device according to an output instruction issued from the communication control driver. A communication board provided in a kernel space for outputting data; and in the first processing, when the number of output instructions queued in the first output instruction request queue is equal to or more than a predetermined amount, the communication is performed. The output command received from the control driver is notified to the flow control support daemon. In the second process, if the number of output commands queued in the first output command request queue is less than a certain amount, Calculates the free space, which is the difference from the free space, instructs the flow control support daemon to read out the output commands for the free space from the second output command request queue, and outputs A flow control support module provided in a kernel space for queuing an instruction in the first output instruction request queue; and The output command received from the row control support module is held in a second output command request queue, and in the second process, the output commands corresponding to the free space are read from the head of the second output command request queue and read. A flow control support daemon provided in user space to be passed to the flow control support module; and a second output command request queue provided in user space for holding output commands received from the flow control support module. A communication flow control method for a computer. 5. The communication flow control method for a computer according to claim 1, wherein the predetermined amount is the number of queued output instructions. 6. The computer communication flow control method according to claim 1, wherein the predetermined amount is a total memory amount of queued output instructions. 7. A communication flow control method for a computer in which an application program, a communication control driver, and a communication board operate asynchronously, wherein a first command provided in a kernel space when an output command is received from the application program. Detects whether an output instruction is queued in the output instruction request queue, and if no output instruction is present, confirms that the communication board is ready to receive the output instruction, A first step of issuing an output instruction, and if the number of output instructions queued in the first output instruction request queue is less than a certain amount, the output instruction received from the application program is sent to the first output instruction request queue. A second step of queuing in the output instruction request queue of the first step, and a fixed amount of output instructions queued in the first output instruction request queue. If the above is the case, a third step of holding the output command received from the application program in a second output command request queue provided in the user space, and when the processing of the communication board is completed, Detecting whether an output instruction is queued in the first output instruction request queue, and if an output instruction is present, dequeues the first output instruction and issues an output instruction to the communication board; And in the fourth step, if the number of output instructions queued in the first output instruction request queue is less than a certain amount, calculate an empty amount that is a difference from the certain amount. A fifth step of reading output instructions corresponding to the free space from the head of the second output instruction request queue and queuing the output instructions in the first output instruction request queue. Communication flow control method for a computer. 8. The communication flow control method for a computer according to claim 7, wherein said predetermined amount is the number of queued output instructions. 9. The method according to claim 7, wherein the predetermined amount is a total memory amount of queued output instructions. 10. When an output instruction is received from an application program, it is detected whether an output instruction is queued in a first output instruction request queue provided in a kernel space, and no output instruction is present. A first process for confirming that the communication board is ready to receive an output command and issuing an output command to the communication board; and an output command queued in the first output command request queue. Is less than a certain amount, a second process of queuing the output command received from the application program in the first output command request queue, and queuing in the first output command request queue If the number of output instructions is equal to or more than a predetermined amount, the third output instruction request queue provided in the user space holds the output instruction received from the application program. Processing, when the processing of the communication board is completed, it is detected whether or not an output instruction is queued in the first output instruction request queue. A fourth process of dequeuing and issuing an output instruction to the communication board; and in the fourth process, when the number of output instructions queued in the first output instruction request queue is less than a certain amount. Calculates the amount of free space, which is the difference from a certain amount,
A fifth process of reading out the output commands corresponding to the free space from the head of the second output command request queue and queuing the output commands in the first output command request queue; A recording medium characterized by the above-mentioned.