JPH03274847A - Communication control system - Google Patents

Abstract

PURPOSE:To shorten a time required for the division processing of transmission data by cancelling linkage among a series of buffers in which the transmission data are stored, and performing the division of the transmission data. CONSTITUTION:The linkage among a series of buffers 131 shown by designation information supplied from a high-order hierarchy are cancelled by a transmission processing means 111. Thereby, the transmission data stored in a series of buffers 131 are divided at every data unit in accordance with the transmission processing means 111, and the designation information representing respective data unit is supplied to a low-order hierarchy. For example, when the designation information representing linked four buffers 131 are supplied to the transmission processing means 111 by a transmission control means 141, the transmission data can be divided into two by cancelling the linkage between a second buffer 131 and a third buffer 131 in the above stated four buffers 131 by the transmission processing means 111.

Description

[Detailed Description of the Invention] [Table of Contents] Overview Industrial Field of Application Conventional Technology Problems to be Solved by the Invention Examples of Means and Effects for Solving the Problems ■ Correspondence between the Examples and FIG. 1 ■ , Structure of Examples ■, Operation of Examples ■, Summary of Examples ■ 0 Variations of the Invention Effects of the Invention [Summary] Concerning the communication control method of a data transmission system in which communication procedures are layered in multiple layers. , with the aim of reducing the time required for dividing the transmission data in each layer, corresponds to each different layer, and divides the transmission data indicated by the specified information supplied from the upper layer into data units in each layer. A plurality of transmission processing means that divide the data and supply specification information indicating each of these data units to a lower layer, and a storage means that stores data unit lengths corresponding to each of the plurality of layers, and are interconnected. The supplied transmission data is stored in a plurality of blocks based on the minimum value of the data unit length stored in the storage means. and a transmission control means for storing each of these blocks in each of a plurality of buffers, and supplying specification information indicating this series of buffers to a transmission processing means corresponding to the highest layer, Each of the transmission processing means is configured to disconnect the series of buffers indicated by the supplied designation information and divide the transmission data stored in the series of buffers into a plurality of transmission data. .

[Industrial Application Field] The present invention relates to a communication control method for a data transmission system.
In particular, the present invention relates to a communication control method for a data transmission system in which communication procedures are layered in a plurality of layers, such as an open system interconnection (OSI) system.

For example, in the O3I system, the procedure for communication is hierarchically divided into seven layers, and each layer has a different data length that can be handled as one message. For this reason, in each layer, processing such as division and combination of data from the upper layer is performed to obtain a message length that is easiest for communication processing.

[Conventional technology]

Normally, each layer specifies one or more buffers and requests the lower layer to transmit data stored in these buffers.

For example, as shown in FIG. 8, a case will be described in which the Nth layer requests the N-1th layer to transmit by specifying a buffer l in which 1000 bytes of data is stored.

At this time, if the maximum data length in the N-1 layer is 512 bytes, the data stored in the buffer 1 described above cannot be handled as one message in the N-1 layer. In this case, in the N1 layer, a part of the data stored in the above-mentioned buffer 1 is copied to another buffer 2, as shown with diagonal lines in FIG.
The data stored in the buffers 1 and 2 are designated respectively and a transmission request is made to the N-2 layer.

Also, the maximum data length in this N-2 layer is, for example, 2
If it is 56 bytes, part of the data stored in each of buffers 1 and 2 is copied to each of buffers 3 and 4 in the same way as the division process in the N-1 layer described above, and Each transmission request for .2 is divided into two. Thereafter, a transmission request is made to the N-3 layer by sequentially specifying these buffers 1.3.2.4.

[Problems to be Solved by the Invention] By the way, in the conventional method described above, when dividing the transmission data in each layer, a part of the data stored in the buffer specified in the transmission request is transferred to another buffer. being copied to the buffer. Since such data copying is performed, there is a problem in that the division process takes time and data transmission efficiency is reduced.

The present invention was created in view of these points, and an object of the present invention is to provide a communication control method that divides transmission data without copying the transmission data in each layer.

[Means to solve the problem]

FIG. 1 is a principle block diagram of the communication control system of the present invention.

In the figure, a plurality of transmission processing means 111 in a communication control method of a data transmission system in which communication procedures are hierarchized into a plurality of hierarchies correspond to different hierarchies, and are indicated by designation information supplied from an upper layer. The transmission data is divided into data units corresponding to each layer, and designation information indicating each of these data units is supplied to the lower layer.

The storage unit 121 stores a data unit length indicating the size of a data unit corresponding to each of a plurality of hierarchies.

The transmission data storage means 130 are interconnected,
It consists of a plurality of buffers 131 each storing data of a predetermined length.

The transmission control means 141 divides the supplied transmission data into a plurality of blocks based on the minimum value of the data unit length stored in the storage means 121, and stores each of these blocks in each of the plurality of buffers 131. At the same time, specifying information indicating a series of buffers 131 in which transmission data is stored is sent to the transmission processing means 1 corresponding to the highest layer.
11.

Overall, each of the plurality of transmission processing means 111:
It is configured to disconnect the series of buffers 131 indicated by the supplied designation information and divide the transmission data stored in the series of buffers 131 into a plurality of transmission data.

[For production]

Each of the plurality of transmission processing means 111 corresponds to a different hierarchy of the data transmission system, and a data unit length indicating the size of a data unit corresponding to each of these hierarchies is stored in the storage means 121. There is. Based on the minimum value of the data unit length stored in the storage means 121, the supplied transmission data is divided into a plurality of blocks by the transmission control means 141, and each of these blocks is divided into a plurality of interconnected blocks. are stored in each of the buffers 131. Further, the above-mentioned transmission control means 1
41, designation information indicating the series of buffers 131 in which the above-mentioned transmission data is stored is supplied to the transmission processing means 111 corresponding to the highest hierarchy as designation information from the upper hierarchy.

This transmission processing means 111 releases the connection between the series of buffers 131 indicated by the specification information supplied from the upper layer, so that the transmission data stored in the series of buffers 131 described above is Processing means 1
11, and designation information indicating each of these data units is supplied to the lower layer.

For example, when the above-mentioned transmission control means 141 supplies the transmission processing means 111 with designation information indicating four connected buffers 131, the transmission processing means 111 selects the second of the four buffers 131. buffer 1 of
31 and the third buffer 131, the above-mentioned transmission data can be divided into two.

In the present invention, the transmission data is divided by disconnecting the series of buffers 131 in which the transmission data is stored. Therefore, it is possible to perform division processing of transmission data without copying the transmission data, and the time required for division processing of transmission data can be shortened.

〔Example〕

Hereinafter, embodiments of the present invention will be described in detail based on the drawings.

FIG. 2 shows the configuration of a data transmission system to which a communication control method according to an embodiment of the present invention is applied.

FIG. 3 is a diagram showing the flow of transmission processing of the communication control method according to the embodiment.

Here, the correspondence between the embodiment of the present invention and FIG. 1 will be shown.

The transmission processing unit 111 includes a transport layer control unit 313, a network layer control unit 314 . Data link layer control unit 315. It corresponds to the physical layer control unit 316.

The storage means 121 corresponds to the data length table 331 in the adapter control unit 301 of the host computer 201.

The transmission data storage means 130 corresponds to the buffer area 215 of the memory 212.

The buffer 131 is a buffer area 215 of the memory 212.
corresponds to each buffer within.

The transmission control means 141 corresponds to the adapter control section 301.

Examples of the present invention will be described below assuming that the correspondence relationship as described above exists.

1-j (Identification 1u motion diagram) In FIG. 2, 201 is the host computer (host), and 202 is the central processing unit (CPU) of the host 201.
203 indicates a memory of the host 201, which is connected to a line 204 via a communication adapter 210, and data is transmitted via the line 204. Additionally, when transmitting this data,
The communication adapter 210 controls the procedures of each layer below the transport layer in O3I.

The communication adapter 210 includes a line control circuit 211 that controls data transmission via the line 204, a memory 212 that stores programs and data, and a C that controls each part.
PU213.

The above-mentioned memory 212 is connected to the CPU 213 and the host 201.
A common conversation area 214 that can be accessed from both the CPU 202 of the host 201 and the communication adapter 210, and a common conversation area 214 that mediates the exchange of data between the host 201 and the communication adapter 210; The buffer area 215 is made up of buffers.

Further, in FIG. 3, 301 indicates an adapter control unit, and this adapter control unit 301 outputs an interrupt signal to the communication adapter 210 described above according to instructions from the application program, and also controls the communication adapter 210. 210 is stored in the common conversation area 214 of the memory 212 mentioned above.

Further, in the figure, reference numeral 311 denotes an interrupt processing unit that analyzes instructions stored in the common conversation area 214 and performs interrupt processing in response to the above-mentioned interrupt signal, and 312 indicates an interrupt processing unit 3.
11 shows a communication processing unit that performs communication processing in response to instructions from 11.

This communication processing unit 312 includes a transport layer control unit 313 that controls communication according to transport layer procedures.
, a network layer control section 314 that controls communication according to the procedures of the network layer, and a data link layer control section 31 that controls communication according to the procedures of the data link layer.
5, and a physical layer control section 316 that controls the line control circuit 211 according to the physical layer procedure.

Furthermore, each part of the communication processing unit 312 described above responds to a connection request sent via the line 204 with a data length that can be handled as one message in each layer (hereinafter referred to as the maximum data length of each layer). and the length of the header added in each layer are added sequentially, and this response is sent to the common conversation area 21 mentioned above.
4, the adapter control unit 301 is notified.

This adapter control unit 301 stores the maximum data length and header length of each layer added to the above-mentioned response in a data length table 331 provided therein.

In addition, in response to a transmission request from an application program, the adapter control unit 301 predicts division processing of transmission data in each layer below the transport layer based on the maximum data length of each layer stored in this data length table 331. However, buffers corresponding to the expected number of divisions (hereinafter referred to as the expected number of divisions) are acquired from the buffer area 215 of the memory 212 described above. Furthermore, the adapter control unit 301 divides the transmission data stored in the area of the memory 203 specified in the transmission request described above into a plurality of blocks according to the expected number of divisions described above, and stores each block in each of the acquired buffers. It is designed to copy the transmitted data of the block.

At this time, considering the headers added at each layer,
The position (offset) at which to start copying the transmission data is determined for each buffer.

The operation of the communication control method according to the embodiment will be explained below by dividing it into the operation of the adapter control section 301 and the divided operation in each layer of the communication processing section 312.

FIG. 4 shows a flowchart representing the division prediction processing operation of the adapter control unit 301 according to the embodiment.

First, the adapter control unit 301 provides a division processing table (see Table 1) consisting of a pointer section (described later), an offset section, and a data length section in its own work area, and sets an initial value rQJ to the offset section. Set the transmission data length Lu specified in the transmission request in the data length section (step 401)
.

Table 1 Here, when a plurality of divided processing tables are provided by the processing described below, a pointer indicating the next divided processing table is set in the above-mentioned pointer section. Also,
In FIG. 4, symbol D indicates the maximum data length in the first layer (i-1 to N), and symbol H1 indicates the maximum data length in the first layer (i=1 to N).
The symbol d is a variable indicating the offset. Also, the Nth layer corresponds to the transport layer, and the N-1st layer corresponds to the network layer, and the processes described below are executed sequentially from the first physical layer to the Nth transport layer. It looks like this.

The processing for the first layer will be explained below.

First, the header length H8 of the first layer and the maximum data length of the first layer are obtained from the data length table 331 described above.

and the maximum data length D i+1 of the i+1st layer (step 402), and the header length H1 of the first layer is added to the above-mentioned variable d (step 403). In addition, the header length H□ read in step 402 and the maximum data length
D6+1 into the following equation (1), and depending on whether this equation (1) holds true, the data from the i+1th layer is
It is determined whether it is necessary to divide the layer into layers (step 404).

D. , <D, +Hi... (1) Formula (1
) holds, the above-mentioned step 404 becomes a negative decision, and steps 402 to 404 are repeated until an affirmative decision is made in this step 404, so that the header length H3 of each layer is sequentially accumulated in the variable d. It has become.

On the other hand, for example, if the maximum data length in the transport layer is 512 bytes and the maximum data length DH-1 in the network layer is 256 bytes, then when processing the N-1st layer corresponding to the network layer, Then, an affirmative determination is made in step 404 described above.

In this case, separate processing tables are provided in addition to the partition processing table provided in step 401 by the number obtained by dividing the transmission data length Lu set in the data length section of the above-mentioned partition processing table by the maximum data length DH-1. (Step 405
). For example, if the transmission data length Lu is 1000 bytes, the first
If the maximum data length in a layer (e.g. network layer) is 2
If it is 56 bytes, three new divided processing tables are provided in step 405. Hereinafter, the divided processing table that is the basis of the newly provided divided processing table in step 405 will be referred to as the original divided processing table. At this time, the value of the variable d described above is set in the offset section of each of the division processing table of the division source and the newly provided division processing table, and
Maximum data length DN in the network layer in the data length section
-□ (for example, 256 bytes).

In addition, pointers indicating the next divided processing table are set in the pointer section of each of these divided processing tables in sequence, and the above-mentioned divided processing table of the division source is set as the first one.
A chain of these divided processing tables is formed (step 406).

Next, it is determined whether or not there is a split processing table for another split source (step 407), and if the determination is affirmative, the split processing table for the other split source is determined in step 4 described above.
Processes from step 05 to step 406 are performed.

On the other hand, as described above, when a negative determination is made in step 403 for the first time in the N-2th layer, there is no other division processing table for the division source, so a negative determination is made in step 407 described above. In this case, in step 408, it is determined whether the processing for all layers up to the Nth layer has been completed, and if the determination is affirmative, the division prediction processing is terminated, and if the determination is negative in this step 408, Starting from the division processing table that is the head of the chain of division processing tables described above, the processing of steps 402 to 408 described above is performed next.

The processing from step 402 to step 408 described above is performed at the Nth step.
By performing this on all layers up to Layer N, the division processing to be performed in each layer is predicted based on the maximum data length of each layer from the Nth layer onwards, and the transmission data is divided into blocks of the size corresponding to the shortest maximum data length. The number of blocks when divided into two can be determined as the number of division processing tables mentioned above.

The adapter control unit 301 stores information in the memory 212 of the communication adapter 210 for each of the above-mentioned split processing tables.
A buffer is acquired from the buffer area 215 of , and the transmission data is copied to the corresponding buffer based on each of these division processing tables.

For example, if four division processing tables are provided in the division prediction processing operation described above, and 6 bytes are set in the offset section and 250 bytes are set in the data length section of these division processing tables, the adapter control unit 301 , 1000
Bytes of transmission data are divided into four blocks of 250 bytes each, and four buffers (for example, buffers ■, ■, ■) are acquired corresponding to these blocks.

In addition, as shown in Figure 5, each of these blocks (indicated by symbols B, , B, , B, , B, in the diagram) is copied from the position indicated by the offset part of each buffer (in the diagram, , shown with diagonal lines).

Further, the adapter control unit 301 sets pointers indicating the sequential relationship of these buffers in the control information area provided in each of the plurality of buffers into which the transmission data has been copied, and forms a chain consisting of these buffers. . For example, the starting addresses of the four buffers acquired as described above are A, , A2 . A3. A,
Then, address A2 is stored as a pointer indicating the next buffer together with the data length etc. in the control information area of the first buffer in the chain, and address A3 is similarly stored as a pointer in the control information area of the second buffer.
The address A4 may be stored in the control information area of the third buffer.

The adapter control unit 301 stores a command indicating a transmission request for the chain formed by the above-described processing in the common conversation area 214 of the memory 212 of the communication adapter 210, outputs an interrupt signal to the communication adapter 210, and sends a command to the chain. Make a transmission request for.

Note that in steps 401 and 405 described above, when providing the division processing table, the memory 21
The buffer may be obtained from the second buffer area 215. In this case, the contents of the above-mentioned division processing table can be stored in the control information area provided in the acquired buffer. shows. Further, FIG. 7 shows a flowchart representing the division processing operation according to the embodiment.

For example, as shown in FIG.
, (1), and (2), the division processing operation of the transport layer control unit 313 will first be described in the case where a transmission request is made for a chain formed from (2), (3).

In response to instructions from the interrupt processing unit 311, the transport layer control unit 313 transfers the specified four buffers ■, ■,
The data lengths set in each of the control information areas of (1) and (2) are sequentially added (step 701), and this step 70
It is determined whether the addition result in step 1 exceeds the maximum data length (for example, 512 bytes) in the transport layer (step 702).

In this case, when the data length of the buffer ■ is added, an affirmative determination is made in step 702, and the transport control unit 313 determines that division processing is necessary. In this case, the pointer set in the control information area of the buffer (for example, buffer ■) located before the buffer to which the data length was last added (for example, buffer ■) is cleared (step 703), and the pointer set in the control information area of the buffer (for example, buffer ■) is ■Remove the chain between. As a result, the chain consisting of the four buffers described above is a chain of four buffers.

It is divided into a chain consisting of (1) and a chain consisting of buffers (2) and (2) (see Figure 6).

Next, it is determined whether the buffer to which the data length was added in step 702 described above is the last buffer in the chain (step 704), and if the determination is negative, the buffer to which the data length was added last (for example, the buffer Steps 701 to 704 described above are repeated with (2) as the head of the chain.

On the other hand, in the case of an affirmative determination in step 704 described above, the division process is ended, and a transmission request is made to the network layer control unit 314 for each of the chain consisting of buffers ■ and the chain consisting of buffers ■ and ■. (See Figure 6).

In addition, in the network layer control unit 314, the transport layer control unit 313 controls the buffer.

In response to a transmission request for a chain consisting of
As shown in Figure 6, the chain between the buffer and the buffer ■ is removed and divided, and the buffer ■
A transmission request is sequentially made to the data link layer control unit 315 for each of the above. Also, buffer ■,
In response to a transmission request for the chain consisting of (2) and (2), division processing is similarly performed, and transmission requests to the data link layer control unit 315 for each of the buffers (2) and (2) are sequentially made (see FIG. 6). .

■, End As described above, the adapter control unit 301 performs the division prediction process of steps 401 to 408 described above based on the maximum data length stored in the data length table 331, and calculates the result by this division prediction process. Acquire the number of buffers. In addition, the transmission data specified in the transmission request from the application program is divided into multiple blocks having the data length determined by the above-mentioned division prediction process, each block is copied to each acquired buffer, and these blocks are A chain consisting of buffers is established, and a transmission request regarding this chain is made to the communication adapter 210.

In addition, the transport layer control unit 313 and the network layer control unit 314 perform step 701 described above.
As explained in steps 705 to 705, the transmission data is divided by clearing the pointer of the control information area provided in the buffer and removing the chain.

This allows you to do so without copying any part of the transmitted data.
It becomes possible to perform division processing in each layer, thereby reducing the time required for division processing and improving data transmission efficiency.

■ Note that in the embodiment of the present invention described above, the case where it is applied to a data transmission system in accordance with O3 ■ is not limited to this, but it is applicable to It can be applied to any transmission system.

Furthermore, in ``Correspondence between Examples and Figure 1'',
Although the correspondence between the present invention and the embodiments has been described, those skilled in the art will easily assume that the present invention is not limited to this and that there are various modifications.

【Effect of the invention〕

As described above, according to the present invention, the transmission data is divided by disconnecting a series of buffers in which the transmission data is stored, so that the transmission data can be divided without copying the transmission data. It is possible to perform data division processing, and the time required for transmission data division processing can be shortened, which is extremely useful in practice.

[Brief explanation of drawings]

Fig. 1 is a principle block diagram of the communication control method of the present invention, Fig. 2
The figure is a block diagram of a data transmission system to which a communication control method according to an embodiment of the present invention is applied, FIG. 3 is a diagram showing the flow of transmission processing according to the embodiment, and FIG. 4 is a flowchart showing the expected division operation according to the embodiment. , FIG. 5 is an explanatory diagram of the transmission data copy operation according to the embodiment, FIG. 6 is an explanatory diagram of the division processing operation according to the embodiment, FIG. 7 is a flowchart representing the division processing operation according to the embodiment, and FIG. 8 is a conventional diagram. FIG. 2 is an explanatory diagram of the division processing operation of FIG. In the figure, 111 is a transmission processing means, 121 is a storage means, 130 is a transmission data storage means, 131 is a buffer, 141 is a transmission control means, 201 is a host computer (host), 202.21
3 is a central processing unit (CPU), 203, 212 is a memory, 204 is a line, 211 is a line control circuit, 301 is an adapter control unit, 311 is an interrupt processing unit, 312 is a communication processing unit, 313 is a transport layer control 314 is a network layer control section, 315 is a data link layer control section, 316 is a physical layer control section, and 331 is a data length table. Figure 2 is a configuration diagram of the data transmission system of the embodiment.

Claims (1)

[Claims] (1) In a communication control method for a data transmission system in which communication procedures are layered into multiple layers, each layer corresponds to a different layer, and transmission data indicated by specified information supplied from a higher layer is handled in each layer. a plurality of transmission processing means (111) that divides each data unit into data units and supplies specification information indicating each of these data units to a lower layer;
and storage means (121) storing a data unit length indicating the size of the data unit corresponding to each of the plurality of hierarchies; a transmission data storage means (130) consisting of a plurality of buffers (131) for dividing the supplied transmission data into a plurality of blocks based on the minimum value of the data unit length stored in the storage means (121); , stores each of these blocks in each of the plurality of buffers (131), and transmits specification information indicating a series of buffers (131) in which the transmission data is stored to a transmission processing means (corresponding to the highest layer). 111), each of the plurality of transmission processing means (111)
A series of buffers (131
), and the series of buffers (1
31) A communication control system characterized in that the transmission data stored in 31) is divided into a plurality of transmission data.