JPS62125739A - Communication control system - Google Patents

Abstract

PURPOSE:To realize a communication control procedure with high reliability and high performance by providing a block exclusive for input decision, comparing a pattern possible for taken by an input with an actual input and outputting an identifier of an input pattern. CONSTITUTION:Inputted data is stored in an input data storage section 302. A communication control information part (header) of the data is inputted also to a comparison section 304 by the control of an operation execution and entire control section 301. The comparison section 304 uses the header part and a 'header identifier header pattern mask pattern conversion table' 305 to decide an identifier corresponding to the header part of the inputted data. It is a row coordinate inputted to a communication control matrix 303. On the other hand, a column coordinate is stored in a 'state register' 306. When the row and column coordinates are decided, the matrix 303 outputs an identifier of an 'operation' subroutine to be executed and gives it to an entire control section 301, which controls the entire system and executes the 'operation' subroutine.

Description

[Detailed Description of the Invention] [Field of Application of the Invention] The present invention relates to a method for realizing communication control procedures, and in particular, communication control suitable for high performance in a matrix method for verifying and realizing advanced and complicated procedures. Regarding the method.

[Background of the invention]

Regarding the description and verification of protocols, state transition matrices have been used as shown in the known example 1 "Data Communication Handbook: 7.3 Protocol Description and Verification Method J" (IEICE, 1984).

However, when realizing this, it is necessary to refer to Publicly known example 2 "6. Protocol description format and conformance" (Tadanori Mizuno, Information Processing ■○L, 26. No.

4 (April 1985), the protocol was written in a flowchart format and realized using software.

In this case, it is necessary to check the flow independently of the matrix verification, which hinders high reliability.

Additionally, there is a problem in that if a part of the protocol is modified, the entire flow needs to be reviewed.

Although there are some methods that directly execute the matrix, high performance cannot be expected because the input judgment (determination of the coordinates in the row direction of the matrix) is performed sequentially using software.

[Purpose of the invention]

An object of the present invention is to provide means for realizing communication control procedures with high reliability and high performance.

[Summary of the invention]

The matrix method has the advantage of directly linking the verification and implementation of communication control procedures, but it has the disadvantage that the overhead of input discrimination is large.

The present invention provides a block dedicated to input determination, compares possible input patterns with actual input, and outputs the input pattern identifier as a result.

[Embodiments of the invention]

An embodiment of the present invention will be described below with reference to FIGS. 1 to 7.

FIG. 1 shows an example of a simple communication control procedure between node A (101) and node B (102).

FIG. 2 shows the operation on the 102 side of the procedure shown in FIG. 1 using a communication control matrix.

FIG. 3 shows a block diagram of one embodiment of the invention.

The entire apparatus of this embodiment is controlled by a control section 301. controlled by. The input data is stored in the input data storage section 30
2. Of these, the communication control information section (hereinafter referred to as "header"; see FIG. 5(a)) is also input to the comparison section 304 under the control of 301. The comparison unit 304 uses this header section and the "header identifier→header pattern/mask pattern conversion table" 305 to determine the corresponding identifier of the header section of the input data. This becomes the row coordinate input to the communication control matrix 303. On the other hand, the column coordinates are stored in the “status register” 306.

Once the row and column coordinates are determined, a step 303 outputs the identifier of the "action" subroutine to be executed and passes it to the overall control unit 301. In addition to overall control, 301 also executes the "operation" subroutine, processes input data, responds (±R5P
)reply.

The “state” transition and the “state” register 306 are also rewritten.

FIG. 4 shows details of FIG. 3 304 (r comparison section").

The comparison section includes a control section 401, an input header register 4o2.
Header pattern register 403, mask pattern register 4041 row coordinate counter 405, comparison gate group 41
0 (the set of EXNOR411, ○R412, and A
ND413).

402 stores the header section extracted from the input data storage section 302.

403 and 404 store header and mask patterns corresponding to the row coordinates indicated by 405.

The communication control matrix uses a mass pattern d as shown in FIG. 2 in the input header.

This is because there are many "n't, care" parts (indicated by *). By matrixing the "don't care" parts and omitting the check, the number of rows in the matrix (types of patterns that the input header can have) can be reduced. can be significantly reduced.

This method is essential for advanced communication procedures where the input header spans several bytes or more.

Many "don't care" parts appear when communication procedures are layered and a matrix is created for each layer.

401 (7) gate group checks the match between 402 and 4.03 by focusing with 411 and other EXNOR gates, and checks unnecessary don and 'tcare bits by using 412 and other OR gate groups and 404 mask pattern. except for. If all bits that require checking match, AND
The output of the gate 413 becomes 1.6 The comparison control unit 401 updates the value of the row coordinate counter until this matching result is obtained.

When a matching result is obtained, step 401 inputs the value of the row coordinate counter to the communication control matrix as a comparison result (row coordinate of the matrix corresponding to the input data).

FIG. 5 shows the format of each data and table.

(a) is the format of input data. The input data includes the communication control information section (header section 511) and the data body (512).
It becomes more.

(b) is a header identifier→header/mask pattern conversion table. 521 is a header pattern, and 522 is a table for obtaining a mask pattern.

In order to facilitate implementation in hardware, it is desirable that the item addresses of each table match the row coordinates or be easily calculated by operations such as addition and multiplication.

(C) is an example of a mask pattern. The bit position of O indicates don't care. 0 Figure 6 is Figure 3 301
(Overall control, "operation" execution section) is a flowchart showing the operation.

Data is input to the data storage section 302 (601).

Extract the header part and store it in the register 402 602
), issues a comparison start instruction to the comparison unit 304.

It waits for the comparison to end (604 No), and when it has finished (604 Yes), it receives the "action" subroutine identifier output from the comparison section (605), and executes the predetermined action (input data processing, response (±RSP)) indicated by this identifier. (reply, “state” transition, rewriting the “state” register, etc.). (6) FIG. 7 is a flowchart showing the operation of FIG. 4 401 (comparison control section).

The comparison unit starts its operation upon receiving a comparison start instruction from the overall control 301. (701) First, the row coordinate counter is reset, and (702) a comparison pattern is loaded from the header/mask pattern table 305 into the registers 403/404 according to the value indicated. (703) Detect whether this pattern matches the input by the output of the AND gate 413, and if it does not match (704 No), update the row coordinate counter (705
), returns to 703 and performs the next comparison.

If they match (704Yes), set the row coordinate counter value to 301.
706) and notifies the end of the comparison (707).

According to this embodiment, high reliability can be expected because the matrix for verifying communication control procedures can be applied as is when implementing it, and input discrimination, which was a problem with the matrix method, can be performed quickly by hardware. The hardware of the input discriminator is not aware of the procedure and only compares the entire header at the bit level, so it can handle a variety of procedures.

The use of mask patterns has the effect of reducing the size of the communication control matrix.

[Effects of the Invention] According to the present invention, a matrix for verifying communication control procedures can be applied as is to implementation, and input can be determined at high speed, making it possible to realize communication control procedures with high reliability. It also has the effect of being able to perform with high performance.

[Brief explanation of drawings]

FIG. 1 is an explanatory diagram of communication control procedure signals according to an embodiment of the present invention;
Figure 2 is a display diagram of an example of a communication control matrix; Figure 3 is an overall block diagram of an embodiment of the present invention; Figure 4 is a block diagram of a comparing section; FIG. 7 is a flowchart showing the operation of the overall control section, and FIG. 7 is a flowchart showing the operation of the comparison control section. 301... Overall control unit, 303... Communication control matrix, 304... Header comparison unit, 305... Header/mask table, 306... "Status" register. 1. 1. 3. 4. 11. 5. t, -e, "r,... A person who wears a mask
Mask T-Heart No. 6 '70

Claims (1)

[Claims] 1. In a device that communicates with others according to a defined communication control procedure, the communication control procedure is shown with the communication control information part of the input data on the front side, and the "state" determined by the history of previous inputs on the front side. The part that is expressed by the arranged matrix and determines the coordinates in the row direction of the matrix corresponding to the input data, and the "state" determined from the history of previous inputs.
A communication control method characterized by storing a column and a part indicating coordinates in a column direction of a matrix, determining the coordinates of a column in the matrix for each input data, and executing the operation indicated by the column.