JPS58105362A - Data transfer control system - Google Patents

Abstract

PURPOSE:To save the data of each station to an external memory with high efficiency, by carrying out the control between user programs using a semaphore process and the semaphore count control by means of a device different from a CPU. CONSTITUTION:When the input of a work station WS101A is over, the decision is given to whether or not the corresponding semaphore is idle or not by the semaphore count of a USA. If the semaphore is not idle, the interruption is queued. If the semaphore is idle, the corresponding semaphore switch is turned on to set the data of the WS101A to a main memory 102. Then whether a UPA is ready or not is decided. If the UPA is not ready, +1 is given to a semaphore counter. Then the end of the next data input is queued. If the UPA is ready, +1 is given to the count value. Then the interruption is applied, and an input end state is reset from the station WS.

Description

TECHNICAL FIELD OF THE INVENTION The present invention relates to a data transfer control method for saving input data from a plurality of work stages to a common external memory based on user semaphore processing in an information processing system.

Technical background of the invention and its problems In an information processing system, when input data from multiple work stages is stored in a common external memory based on user semaphore processing, all of the processing has conventionally been performed using O8 (0 perity). ■System) IIC
However, in the past, the burden placed on the O8 was large, the software network became complicated, and the processing load on the processing unit that controlled the entire system increased, reducing the processing capacity of the entire system. This was one of the factors leading to the decline.

Purpose of the Invention The present invention provides a processing device that controls the entire system in a system that has a function of saving data of each of a plurality of work stages to a common external memory based on user semaphore processing and associated glodarum processing. and O
It is an object of the present invention to provide a data transfer control system capable of executing user semaphore processing without imposing a burden on the work stage 8 and efficiently transferring and saving input data from the work stage 8 to the external memory.

Summary of the Invention The present invention provides a processing device that controls the entire system by controlling switches between user programs and controlling semaphore counts by user semaphore processing when input data from multiple work stages is saved to a common external memory. This is done in a processor unit with a separate hardware configuration, based on the semaphore count value.

Embodiment of the Invention An embodiment of the invention will be described below with reference to the drawings. 1st
The figure is a block diagram showing one embodiment of the present invention. In the figure,
100 is a processing unit (hereinafter H-
(referred to as CPU), 101mu.

101m and 101c are Works Steel” w 7 (wS
A, WSi.

WSc), 102 is this O'7 station, 101h
*103 is the main memory, which is an external memory (for example, MT) in which the input data from 101m and 101@ is saved. This main memory XOX contains control! program (O8
), and the user communication area (hereinafter referred to as UCA
(hereinafter referred to as USA), a user semaphore area (hereinafter referred to as USA) used for processing the store function between user programs;
Each of the above work stages, 101h, 101me10
A plurality of user programs UPJ, UPj # UPJ, for performing data input/output processing of the self-specific work stage 1011 provided corresponding to 1c.
An area for a user program A (UPA) and the like for transferring data transferred onto the UCAC to the external memory 102 is secured. 104 is a user program UPI, UPj based on the semaphore count value of USA, independent of the H-CPU 100;

4) Rounding processor unit for 4t mafau processing (hereinafter referred to as US-CPU) that performs switch control between UPj and UPA and semaphore count control
It is.

FIG. 2 is a diagram showing the order of receiving and passing data and control in the system shown in FIG. 1 above.

O2, . . . indicate data order, and C1, C2, . Figure 3 shows workstation 1
4 shows an example of the processing operation of the user programs UPI, UPj, UPJ- corresponding to the program 1101Be101C, and FIG. 4 shows the processing operation of the user program A (UPA) for saving data to the external memory 102. The processing shown in FIG. 3 and the processing shown in FIG. 4 are performed in parallel.

The process indicated by the broken line frame in the figure is the user semaphore process. Here, the semaphore count value for user program A (UPA) is called A semaphore count value, and its counter part is called A semaphore count, and the user program LIPJ.

UPj, seven muff count value for UPj (1
” or “O”) is the semaphore count, and its kakunta is called the semaphore switch.The initial value of this person's semaphore counter is ``0'', and the initial value of semaphore processing is ``''
0' (off).

Here, the operation of one embodiment will be explained. When either of the work stages 101m and 101c completes data input, semaphore processing as shown in FIGS. 3 and 4 is executed under the control of the U8-CPU J o 4, The data from the work stage (for example, 101m) that has completed data input is transferred to the path D shown in Figure 2.
I, O2, . . . are saved in the external memory 102. The specific operation will be explained below. First, the processing operations of the user programs UPI, UF4, and UPS corresponding to the work stages 101kT101m and 101c will be explained with reference to FIG. Data input for work stage deviation (for example, 101 mm) is completed (Figure 2 DI
, Fig. 3 a), first check whether the corresponding semaphore is empty according to the semaphore count of USA, that is, whether the count content of the semaphore switch corresponding to 101 workstations is "0# (off)". (b) in Figure 3), if it is not vacant, that is, "l" (on), it is set as an interrupt waiter (Figure 3C), and if it is vacant, that is, 601 (off), it is applicable. By setting the semaphore switch to "1" (on) (Fig. 3d), data from the workstation 101 is transferred to the main memory 101.
2 (Figure 2 02, Figure 3 e)
. Next, it is determined whether the user program A (UPA) is in the ready state, that is, whether the counter value of the A semaphore counter is "0" [7 (Fig. 3 f); , that is, if it is not "0", the count value of the A semaphore counter is +1 (Fig. 3g) -
Then, the program returns to the initial Stella f (FIG. 3a) and waits for the next data input to be completed. Also, if it is in the ready state, that is, if it is "0", the count value of the A semaphore counter is incremented by 1 and an interrupt is placed in the wait B shown in FIG. 4 (FIG. 3 h), and the first step (step Return to Figure 3 a).

Next, the processing operation of user program A (UPA) is
This will be explained with reference to the figures. Here, first, the count value of the A semaphore counter is checked (Figure 4a), and if the count value is "0", interrupt wait B (Figure 4b) is executed.
If it is not "0", input the data set in UCA to be processed (D3) in Figure 2 and write it to the external memory 102 (D4 in Figure 2 and D in Figure 4). .

Next, set the semaphore switch corresponding to the data written to the external memory 102 to "0" (off) (Fig. 4 e),
After that, it is checked whether each semaphore switch is "θ" (off) (Fig. 4 f), and if it is "0", the process returns to the first step (Fig. 4 a), and "If not, the waiting person shown in FIG. 3 is interrupted (FIG. 4g) and the process returns to the first step (FIG. 4a).

Such semaphore processing is performed by the H-CPU1. o.

US-CPU1 independently of and also using O8
04 user program mutual switch control, and accordingly, the user program geography is executed sequentially,
Multiple work stages, 101m, 101m, 101
The data written in c is saved in a common external memory 102.

Effects of the Invention According to the present invention, in a system having a function of saving data of each of a plurality of work stages to a common external memory based on user semaphore processing and accompanying datagram processing, knee f
By having a configuration in which switch control and semaphore count control between f logs and semaphore counts are performed independently by a processor unit having a hardware configuration separate from the processing device that controls the entire system, the processing device and OS
It is possible to provide a data transfer control system that can efficiently save input data from each of the work stages to an external memory without placing a burden on the work stages.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing one embodiment of the present invention, FIG. 2 is a diagram showing the flow of data and control in the above embodiment, and FIGS. 3 and 4 are for explaining the operation of the above embodiment. 0 101m, 101m, jolc... workstation (wsmuw WSI e WSc), 102...
External memory (MT), 103...main memo I), 104
・・・Flosser unit (US-CPU), UCA・
...User comments. Two cages. area, USA...User semaphore area, UP
I, UPI, UPI, UPA...User program. Applicant's agent Patent attorney Takehiko Suzue Figure 1 112 ■ $3! l! ! 1 Shin 4! ! l

Claims (1)

[Claims] an external memory for storing input data of a plurality of workstations, a plurality of workstations, and a plurality of user programs and control programs corresponding to each of the workstations and the external memory; A main memory allocated with a user semaphore area serving as a communication area for receiving and passing human output data between the user programs, and a switch area between the user programs, and a system-wide switch process between the user programs. It has a processor unit that performs control based on the semaphore count value of the user semaphore area independently of a processing device that controls the semaphore count value, and the processor unit controls addition and subtraction of the semaphore count value and its counting. performing switching processing between the user programs according to the contents, and controlling data transfer to the work stage, the communication, the second cage, and the area, and the data transfer from the communication, the second cage, and the area to the external memory; A data transfer control method featuring: