JPS6242259A - Method for transferring data between processors - Google Patents

Abstract

PURPOSE:To reduce the hardware quantity of a master processor by putting and sending the control data on the final byte when data are sent to the master processor from a slave processor. CONSTITUTION:The slave processor 1 writes the control data to the final byte of a buffer 3 for transmission and also transfers the data during the final time. Here the communication is performed in time assignment and therefore data are sent from the buffer 3 at a prescribed time point. These data are written to a buffer 4 for the reception set at the side of a master processor 2. The reception data are already written to the buffer 4 when the control data is written to the buffer 4. Thus validity of the control data can be confirmed simply by checking the final byte of the buffer 4 through the processor 2. Then the processor 2 can immediately receive the accurate data.

Description

[Detailed Description of the Invention] [Summary] In a digital exchange that adopts and uses a distributed processing method,
A mask processor (CPR) assigns multiple slave processors (LPR) using a time assignment method.
), when transmitting data from the slave processor to the mask processor, the hardware of the mask processor is reduced by transmitting control data on the last byte.

[Industrial application field]

The present invention relates to a digital exchange using a distributed processing method, in which data is transferred from a slave processor installed in a subscriber circuit, a trunk circuit, etc. to a mask processor that controls switching processing using a time assignment method. This relates to a sending method.

In conventional data transfer methods, the mask processor reads the transferred data at a timing completely different from the data transfer, so the receive buffer's peripheral circuitry must check the reception timing and validity and notify the mask processor. However, there was a strong need for improvement.

[Conventional technology]

In a digital exchange that uses a distributed processing method, it is necessary to transfer data between a mask processor and slave processors. In cases like this, conventionally the demand assignment method was used to transfer data while repeating transmissions and responses between both processors, but recently, in order to simplify the data transfer procedure, data is transferred using the time assignment method. method has been used.

FIG. 3 is an explanatory diagram of an example of a conventional inter-processor data transfer method.

In the figure, 1 is a slave processor, 2 is a mask processor, 3 is a transmission buffer of the slave processor 1, 4 is a reception buffer of the mask processor 2, 5 is a control data buffer, and 6 is a control circuit.

When transferring data from slave processor 1 to mask processor 2, the transfer data is first transferred from slave processor 1 to transmission buffer 3, and the transfer data stored in transmission buffer 3 is transferred to reception buffer 4 of mask processor 2. Then, the master processor 2 takes steps to read the data in the reception buffer 4.

The time assignment method referred to here is the buffer 3-
This is the data transfer method of buffer 4.

In the conventional data transfer method using the time assignment method, the slave processor 1 sends transfer data and control data for transfer, and the mask processor 2 side transfers the control data stored in the control data buffer 5 to the control circuit 6. When it is determined that the received data in the buffer 4 is valid data, the mask processor 2 takes in the received data in the buffer 4 and transfers the data.

[Problem that the invention seeks to solve]

However, since the mask processor 2 reads the transferred data at a timing completely different from the data transfer, the receive buffer 4
There is a drawback in that a control circuit 6 must be provided around the mask processor, and the control circuit 6 must check the reception timing and validity and notify the mask processor.Therefore, there has been a strong demand for an improvement in this area.

[Means for solving problems]

The above problem is that when the mask processor 2 communicates with multiple slave processors 1 using the time assignment method in a digital exchange that employs a distributed processing method, control data is stored in the last byte of the transmission buffer 3 of the slave processor 1. The solution is to write and send the last byte last in time.

[Effect]

According to the present invention, the validity of the data can be confirmed immediately when the receiving buffer 4 receives the final byte, so the mask processor 2 can read the data with a simpler operation than in the past. .

〔Example〕

FIG. 1 is a diagram showing an embodiment of the inter-processor data transfer method according to the present invention.

FIG. 2 is a diagram illustrating a data transfer method according to the present invention.

When performing communication, slave processor 1 writes communication data to sending buffer 3 as in the conventional system.

In the present invention, if the transfer data is configured in units of 4 bytes as shown in FIG. 2, the slave processor 1 writes control data to the last byte of the sending buffer 3,
Moreover, the data was transferred at this final time.

Since communication is performed by time assignment, at a predetermined time, data is sent out from the sending buffer 3 and written into the receiving buffer 4 on the mask processor 2 side.

The mask processor 2 reads the transfer data from the slave processor 1 by accessing this reception buffer 4.

In this case, the mask processor 2 accesses the reception buffer 4 at any time. The master processor 2 first reads the control byte, checks its validity, and then takes in the data. For this reason, when the control byte is input to the reception buffer 4, the remaining data must already be determined.

In the present invention, as described above, the control data is written in the final byte and the data is transferred at this final time, so that by the time the control data is written to the reception buffer 4, the reception data has already been written to the buffer 4. Therefore, the mask processor 2 can easily confirm the validity of the control data by checking the last byte of the reception buffer 4, and the mask processor 2 can immediately receive accurate data.

In the conventional method, the control circuit 6 of the reception buffer 4 confirms the reception timing and validity, then notifies the mask processor 2, and upon receiving this notification, the mask processor 2 reads the reception data of the buffer 4. The peripheral circuitry of the receiving buffer 4 is greatly simplified.

〔Effect of the invention〕

As described in detail above, according to the present invention, there is no need for a peripheral circuit for a receiving buffer, and the amount of hardware can be reduced.

[Brief explanation of drawings]

FIG. 1 is a diagram showing an embodiment of the inter-processor data transfer method according to the present invention. FIG. 2 is a diagram illustrating a data transfer method according to the present invention. FIG. 3 is an explanatory diagram of an example of a conventional inter-processor data transfer method. In the figure, 1 is a slave processor, 2 is a mask processor, 3 is a transmission buffer of the slave processor 1, 4 is a reception buffer of the mask processor 2, 5 is a control data buffer, and 6 is a control circuit.

Claims (1)

[Claims] When a master processor (2) communicates with a plurality of slave processors (1) using a time assignment method in a digital exchange that employs a distributed processing method, An inter-processor data transfer method, characterized in that control data is written in the last byte of a buffer (3), and the last byte is sent temporally last.