JPH0286243A - Communication protocol processing system - Google Patents

Abstract

PURPOSE:To relieve an overhead at event notice and to reduce the software and hardware of a communication equipment by devising the title system such that an event is notified not via an entry of the event. CONSTITUTION:The notice of an event from other state transition matrix to plural state transition matrixes A-C, that is, the notice of the event to a task 22 is implemented via an event entry 21. Then only one event entry is provided on the plural state transition matrixes A-C and the event among them is given to the state transition matrixes through a direct routine cell not via the event entry. Thus, the overhead at event notice is relieved to save the software and hardware of the communication equipment.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to protocol processing in a communication device, and more specifically, an event notification interface between layers in an IIm type communication protocol, and different state transition matrices within the same layer. Regarding the event notification interface between.

In recent years, with the development of communication technology, various communication methods such as facsimile communication, packet switching, and videotex communication have been proposed and put into practical use. A protocol (communication protocol) is required to organically and integrally combine these different communications to establish systematic (standard) communications. A protocol models the components of a network (communication network) and is the arrangement necessary for communication on the model (for example, message types, order of information transmission and reception).

[Conventional technology]

First, the concept of a protocol will be explained with reference to FIGS. 7 to 9. In addition, Figures 7 to 9 are as follows:
"Kawaoka, idiot" protocol format % expression %) 187-192. February 1985”.

FIG. 7 is a diagram showing a layered model of the protocol. Nodes A and B are communication devices such as terminals.

The protocol is modeled in layers such as (N+1) layer, (N) layer, etc. Entities are means for executing protocols and are provided for each layer. For example, an entity in layer (N) is layered immediately below layer (N-1).
) Receives the (N-1) services provided by the entity and executes the procedure processing routine of the (N) layer. Then, (N) is added to the (N+1) layer immediately above the (N) layer.
) to provide services. What defines the operations performed by such (N) layer entities is a protocol.

In general, various methods have been proposed for describing protocols, but here the description will be based on a state transition model.

FIG. 8 is a diagram showing an example of a state transition model that can be performed between a data provider and a data consumer.

In the figure, i+ is an internal transmission request notification, 12 is a reception permission notification from the other party, i3 is an internal data transfer end notification, i4 is a data consumption notification from the other party, and 01 is the other party. Request for transmission to the other side, 02 indicates data transmission to the other party, - indicates no input or output. Further, in i10, i indicates input, 0 indicates output, and * indicates that the output of the data provider side becomes the input of the data consumption side.

FIG. 9 is a diagram showing state transitions on the data provider side. As shown in the figure, the procedure processing routine to be executed by the entity is determined based on the input (also called an event) and the state of the data provider at this time.

The relationship between inputs and states shown in FIG. 9 is usually called a state transition matrix.

In conventional hierarchical communication protocol processing, each layer has an independent state transition matrix, and each layer has an event reception. This situation is shown in FIG. In the diagram, X, Y,
Z is an entity (hereinafter referred to as a task) in a different layer, and x, y, and z are event receivers for tasks X, Y, and Z, respectively. An event receiver is a program that connects procedural processing routines that can be performed between adjacent layers. For example, as shown in FIG. 10, an event that occurs for task X in a layer immediately below the layer of task X is connected to event receiver X (queueing), and the event is acquired through this. As will be described later, Task Task X undergoes a state transition by executing a procedural processing routine, and creates an event for the layer immediately above the layer of task X. This event is given to task Y via event socket y for task Y. Task Y, which has acquired the event in this way, determines and executes a procedure processing routine for this event. Then, task Y queues the event created in this procedural processing routine to event receiver 2, and provides it to task Z. Task Z that acquired the event
similarly determines and executes a procedural processing routine.

Processing within each layer is performed as shown in FIG.

For example, task X is a state transition matrix a related to the white layer.
(specified by matrix number @a), and from the event number C of the event obtained from the event socket X and the state number C of the layer state (status) at this time,
Obtain the routine number d of the corresponding procedure processing routine. This routine number d is the start address (
This is given to the routine address table T+ which stores the procedure processing routine address.

Task X, which has learned the procedural processing routine address corresponding to routine number d, calls and executes the procedural processing routine starting from this address.

In particular, the control process from obtaining matrix number a to routine call is called matrix control process. 11th
This is indicated by reference number 10 in the figure. During the execution of the procedure processing routine corresponding to routine number d, select the Fl that you want to call.
An event is created for (eg, layer of task Y). Then, the routine call returns to the matrix control processing 10, and the task Y is notified of the event by gaging the event reception port y.

The above processing is normally performed by a program. This program has independent subprograms for each layer and a main program that organically combines the subprograms.

[Problem to be solved by the invention]

However, in the conventional hierarchical communication protocol processing method described above, tasks (in other words, subprograms or entities) are provided independently for each layer, and events between tasks (in other words, between state transition matrices) are Since an event can be obtained only through the receiver, there is a problem in that the overhead of the program that constitutes the event receiver at the time of event notification becomes large.

Therefore, the present invention solves the above problems and makes it possible to notify an event without going through an event receiver, thereby reducing the overhead at the time of event notification and thereby reducing the software and hardware of a communication device. purpose.

[Means to solve the problem]

FIG. 1 is a diagram for explaining the present invention in detail. The multiple state ail transition matrices of the protocol (if A
, B, and C), one event receiver 2 is used for the A matrix corresponding procedure processing, the B matrix corresponding procedure processing, and the C matrix corresponding procedure processing respectively obtained from
1 will be provided.

In addition, other state transition matrices (for example, state transition matrix B for state transition matrix A) are included in the A matrix compatible procedure process, B matrix compatible procedure process, and C matrix compatible procedure process defined in each state transition matrix. Receives control information for directly calling a routine to notify an event.

The plurality of state transition matrices may be between different layers (for example, corresponding to the above-mentioned FIG. 10), and
It may be between the same layers.

Furthermore, since a plurality of state transition matrices directly notify events by routine calls without going through an event receiver, by providing one task 22 for these, matrix-compatible procedure processing can be executed.

[Effect]

notification of events from other state transition matrices (e.g. in lower layers) to multiple state transition matrices;
In other words, the event notification for the task 22 is sent via the event receiver 21. For example, the state g3 transition matrix A obtains event 1- via the event socket 21 (by queue). Then, the A matrix corresponding procedure process corresponding to the acquired event is executed.

An event generated by this execution is notified by a routine call that directly calls another state transition matrix B according to the control information during processing of the A matrix corresponding procedure. B that responds to the event notified in this way
Matrix-enabled procedural processing is performed. Events that occur during this execution are notified to other directly called state transition matrices C according to the control information, and the corresponding C matrix corresponding procedure processing is executed.

In this way, there is only one event receiver for multiple state transition matrices, and notifications of events between these matrices are given directly to the corresponding state transition matrices by routine calls without going through the event receivers, which solves the conventional problem. The points are cleared.

〔Example〕

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

FIG. 2 shows a configuration example of a communication device implementing the present invention.

In the figure, the communication device includes a main body system device 30, a communication control processing device 34, and a peripheral system device 35. The main system device 30 includes a central processing unit (CPU) 31 and a main storage device (
MEM) 32 and a data transfer device (DCH) 33. The central processing unit 31 performs control specified by the protocol according to a flowchart described later. The main storage device 32 stores a state transition matrix, a corresponding procedure processing routine, a routine address table, etc., which will be described later. The data transfer device 33 controls data transfer between the central processing device 31, the communication control processing device 34, and the peripheral devices 35. The communication control processing device 34 connects the main body system device 30 and the line 3.
6, and controls modulation/demodulation, line connection/cutoff, etc. Peripheral devices 35 include a CRT, printer, keyboard, and the like.

FIG. 3 shows the central processing unit 31 and main storage device 3 in FIG.
2 is a functional block diagram of protocol control performed in step 2. FIG.

The event input section 40 is connected to the event receiving port 21 shown in FIG.
This corresponds to receiving event notifications for the task 22 from lower layers and the like.

The 7-trix storage unit 41 stores a plurality of state transition matrices. The routine storage unit 42 calls the state transition matrix of the matrix by a direct routine call and notifies the matrix corresponding procedure routine corresponding to the state number defined in each state R transition matrix and the event that occurs in this routine. (This is called a secondary matrix control processing routine).

The routine address table 43 is a table that defines the correspondence between the state numbers found in each state transition matrix and the routine storage unit 42.

The control section 44 controls each of the illustrated sections according to the operation flowchart of FIG. 4, which will be described later. The procedure processing unit 45
The corresponding procedure processing routine specified by the control unit 44 is executed.

Next, the operation of this embodiment will be explained with reference to FIGS. 4 to 6 in addition to FIG. 3. Here, FIG. 4 is a flowchart showing the operation of this embodiment, FIG. 5 is a diagram showing the processing procedure of this embodiment, and FIG. 6 is a diagram showing the processing of this embodiment step by step.

In the following explanation, a procedure for controlling two consecutive layers (hereinafter, the lower layer is referred to as layer A and the upper layer is referred to as 8 layers) in hierarchical communication protocol processing will be taken as an example.

First, the device is in an event holding state ((
a)). In this state, it is assumed that an event (2) for layer A and an event (2) for B1 occur consecutively in a lower layer of layer A (FIG. 6(b)). Event (2) is given to the event input unit 40 in FIG. 3 by queuing by another program. Then, the event input unit 40 obtains the first event ■ (step 101 in FIG. 4; FIG. 6 (
C)).

The control unit 44, which receives the notification of the event ■ from the event input unit 40, determines to which layer this event ■ corresponds (step 102; FIG. 6(d)).

The control unit 44 knows that the event ■ corresponds to the state transition matrix A, and the matrix storage unit 4
The state transition matrix a1 shown in FIG. 1 to FIG. 5 is read out (step 103: FIG. 6(e)). Assuming that the status number is CI and the event number is b+, the control unit 44 determines the corresponding routine address d1 (step 104; FIG. 6(f)). Then, the control section 44 refers to the routine address table 43, learns the address of the 81-compatible procedure processing routine corresponding to the state number d1, and calls the routine (step 105; FIG. 6(g)). Then, the control unit 44
1 corresponding procedure processing routine is provided to the procedure processing section 45. As a result, the procedure processing unit 45 a1 corresponding to event ■
1 Execute the corresponding order processing (step 106: Fig. 6 (h
)).

In FIG. 5, the process from taking matrix number a1 to the corresponding routine call is shown as -th order matrix control processing.

Depending on whether an event occurs during execution of the a11 corresponding order processing obtained in this way, the processing is divided into two types as described below.

First, it is assumed that event O for the state transition matrix 8 occurs during the al corresponding procedure processing (Fig. 6 (i2)).
. Detecting this, the control unit 44 follows the secondary matrix control processing routine address acquired when the routine storage section d1 of the routine address table 43 is accessed earlier, and from the end of the a11 corresponding order processing, as shown in FIG. In the same manner as the next matrix sub-part processing, the program continues to obtain the routine number corresponding to event G] ("YES" in step 107: FIG.
j2)). This secondary matrix control processing routine address is from step 107 to step 102 in FIG.
, and the secondary matrix control process is thus condensed from step 102 to step 106.
This means the processing up to.

After determining the layer corresponding to event 0 (step 102), the control unit 44 determines the state transition matrix a related to this layer.
2 from the matrix storage section 41 Step 1
03) , 811 status and event θ at this time
The routine number corresponding to the event number is known (step 104). Now, as shown in FIG. 5, if the routine address d2 is obtained from the event number b2 of event 0 and the state number C2 at this time, the control unit 44 retrieves the routine address d2 from the routine storage unit 42 by a routine call. Step 105) reads the a2-compatible procedure processing routine corresponding to
does this (step 106). In this embodiment, the 8th layer is the highest layer processed by one task, so the routine address table 43 is based on the state transition matrix a.
Regarding the status address specified in 2, it has only a procedure processing routine address and does not have a secondary matrix control processing routine address.

On the other hand, if no event occurs during the a1 corresponding procedure processing in step 107 of FIG. 4, the control unit 44 determines whether there is a next event in the event input unit 40 (step 108). As shown in FIG. 6 (11), since the event ■ follows the event ■, the control unit 44
obtains the event ■ from the event input section 40. Then, according to steps 102 to 106, a corresponding procedural processing routine is determined and executed ((j+) in FIG. 6).

(kl)). Then, the initial state (Fig. 6(a))
Return to

One embodiment of the present invention has been described above. In the above actual flow example, one event receiver (event input section 40) is provided for state transition matrices A and B, and events that occur during processing related to state transition matrix A are notified to the state transition matrix by a routine call. There is. As a result, the overhead of the interlayer interface of the event reception can be reduced.

In addition, since it is sufficient to have one task (corresponding to the procedure processing unit 45 in FIG. 3) common to the state transition 7 tricks A and B,
Each set of software and hardware that executes the protocol can be reduced.

Furthermore, in the above embodiment, the state transition matrices at, a
2 has been described as corresponding to different layers, but the state transition matrix can be divided into multiple parts and managed within the same layer. For example, one state 171 transition matrix is created with the relationship between states and events that should not occur in protocol processing, and when this is accessed, the necessary response processing is immediately read out and executed as a failure. can do.

(Effects of the Invention) As explained above, according to the present invention, there is only one event receiver for a plurality of state transition matrices, and notification of events between these matrices can be directly performed via a routine without going through the event receivers. - Since it is given to the corresponding state transition matrix in a call, the overhead at the time of event notification can be reduced and the software and hardware of the communication device can be reduced.

[Brief explanation of the drawing]

FIG. 1 is a diagram for explaining the present invention in detail, FIG. 2 is a block diagram showing a configuration example of a communication device to which the present invention can be applied, FIG. 3 is a functional block diagram of an embodiment of the present invention, and FIG. 4 is an operation flowchart of the embodiment of the present invention, FIG. 5 is a diagram showing the processing procedure of the embodiment of the present invention, FIG. 6 is a diagram showing the processing of the embodiment of the present invention step by step, and FIG. 7 is a layered model of the protocol. , FIG. 8 is a diagram showing an example of a state transition model, FIG. 9 is a diagram showing state transition on the data provider side, FIG. 10 is a diagram for explaining the conventional method, and FIG. 11 is a diagram showing each layer. It is a figure which shows the processing procedure performed in FIG. 21...Event reception, 22...Task, 30.
...Main system device, 31...Central processing unit (CPU),
32... Main memory device (MEM), 33... Data transfer device (DCH), 34... Communication control processing device, 3
5... Peripheral system device, 36... Line, 40... Event input section, 41... Matrix storage section, 42...
- Routine storage unit, 43... Routine address table, 44... Control unit, 45... Procedure processing unit. Fig. 3 Functional block diagram of the non-inventive embodiment Fig. 3 Processing for detailed explanation of the invention Fig. 1 Applicable to the invention) I! Block 0 showing an example of training on i-communication equipment
Figure 2 shows an example of the invention, an operation flow chart, and the fourth diagram 91! Transition matrix a Diagram showing the clam filling procedure of the non-inventive embodiment Node A Node B Diagram showing the hierarchical model of the global protocol FIG. 7 (Data provider side) (Data consumer side) An example of a state transition model Figure 8 Diagram showing the state transition on the provider side

Claims (1)

[Claims] In protocol processing in a communication device, one event receiver is provided for notifying events for a plurality of state transition matrices of the protocol, and matrix-compatible procedure processing specified by each state transition matrix is performed. A communication protocol processing method characterized in that control information for directly calling another state transition matrix by a routine call and notifying an event is provided therein.