JPS5987527A - Initial program loading system - Google Patents

Abstract

PURPOSE:To raise reliability of an initial program loading system by monitoring the stop of an initial program loading request signal in a main control device, and confirming the end of a normal operation of a subprocessing device. CONSTITUTION:An initial program load IPL is sent out separately each to each sub-processing device 20 from a main processing device 10, or each device 20 generates its own IPL start indication, by which a necessary IPL request signal is sent out continuously and separately each to the device 10 from the device 20. On the other hand, the device 10 detects its request signal and sends out an IPL data to the device 20. When the device 20 receives it, a signal stop of the device 10 is detected by executing the stop of the IPL request signal after receiving its final data, and the normal end of its IPL can be confirmed.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Application of the Invention] The present invention comprises a main processing unit and 1. Or, in a distributed processing information processing system (e.g. electronic exchange, electronic computer) consisting of multiple slave processing units, the initial program load of each slave processing unit is performed independently of the others, and the slave processing unit and main processing unit are The present invention relates to an initial program loading method that allows the program to be executed from any location and allows the main processing unit to confirm that the program has completed successfully.

[Prior art]

This type of conventional initial program loading method includes, for example: 1. Initial program loading of the main processing unit (hereinafter referred to as
It's called IPL. ), and after completing its own IPL, the IPL of each slave processing device is simultaneously activated.

Therefore, the IP address from the main processing device to each slave processing device is
Since it was not possible to start L at any time, and also independently from the main processing unit and the song sub-processing unit, it was not possible to start up the main processing unit and the song sub-processing unit, so after completing repairs, expansions, tests, etc. of a specific sub-processing unit, or in the event of a failure, etc. In the case of multiple IPLs, a problem arises in that IPL startup must be executed for the main processing unit and other slave processing units each time.

Note that, as described above, since all six slave processing devices undergo IPL activation at the same time, it is difficult and rare to individually check whether or not the IPL has ended normally.

[Purpose of the invention]

It is an object of the present invention to eliminate the drawbacks of the prior art described above, to enable IPL of each slave processing unit to be executed independently of the others, from either the slave processor or the main processor, and to complete the IPL normally. can be confirmed by the main processing unit.
The objective is to provide a PL system.

[Summary of the Invention] The 14 configurations of the IPL method according to the present invention are as follows: ■ The file memory for storing PL data is exclusive to the main processing unit;
A secondary processing device is provided in a distributed processing type information processing system configured to receive the IPL data via the main processing device, and the main processing device individually sends an IPL activation signal to each of the secondary processing devices. or each slave processor issues its own IPL activation instruction, so that each of the six slave processors sequentially transmits the required IPL request signal to the main processor. , the main processing device detects the IPL data request signal and sends a predetermined IPL data request signal to the slave processing device, and the slave processing device receives the IPL data request signal and sends the IPL data request signal after receiving the final data. By completely stopping the transmission of the signal, the main processing unit detects the stop of the signal and confirms that the IPL has ended normally.

[Embodiments of the invention]

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

FIG. 1 is a system configuration diagram of an embodiment of the IPL method according to the present invention, and FIGS. 2 and 3 are flowcharts of the main processing device side and the slave processing device side, respectively.

Here, 10 is the main processing unit CM'P TJ ), 11
is its central control unit (cc), 12 is its memory (M
), 13 is the same IPL control device (IPLC), 14 is,
The same timer (TM), 20 indicates each slave processing unit (SPUI to 5PUn) of n11JA, 21 indicates its central control unit (CC), 22 indicates the same memory (M), and 23 indicates the same I
The PL control device (IPLC) 3o is a file memo IJ (FM
), 40 is an individual IPL start signal path (, S T ) for each slave processing device 20, 41 is an individual IPL data request signal path (几Q), and 42 is a common IPL data signal path ( DATA).

First, in the case of IPL startup from the main processing unit 10 side (second
Flow starting from start A in FIG. 3 will be explained.

When the software of the central control device 11 transmits an IPL activation instruction for the slave processing device 20 to the IPL control device 13 via the control line 10A, the PL control device 13 instructs the slave processing device 20 (for example, SP [ The IPL control device 23 sends an IPL start signal to the IPL control device 23 via the IPL start signal path 40, and starts the timer 14 for monitoring reception of the IPL data request signal via the control line 10C. The reception time of the IPL data request signal to be returned from 23 via the IPL data request signal path 41 is monitored.

Here, the IPL data request signal is sent to the IPL data request signal on the main processing unit side.
If the timer 14 normally overflows for a predetermined period of time (with some margin added to the time required for normal IPL) before receiving it at the PL control device 13, the timer 14 is sent via the control line 10D. 14 to I l)
The L control device 13 is notified of this, so the I
The PL control device 13 notifies the software of the central processing unit 11 via the control line 10B that the IPL data request signal cannot be received, and ends all operations.

On the other hand, the IPL control device 23 on the slave processing device 20 side, which has received the IPL start signal from the IPL control device 13 via the IPL start signal path 40, requests the IPL control device 13 via the IPL data request signal path 41. The IPL control device 1 continuously sends out IPL data request signals.
3 via IPL data signal path 42.
Monitors reception of PL data.

This goes back and forth, but ①When the PL control device 13 receives the IPL data request signal, it controls the timer 14? l5
central controller 1 as well as reset via lOc.
The reception of the above-mentioned IPL data request signal is notified to the software of No. 1 via the control line 10B.

The central controller 110 software is 0, after which the IPL
An IPL data sending command is transmitted to the control device 13 via the control line 10A, and the IPL control device 13 receives the command.
starts the file memory 30 via the control line 30A.

IPL stored in the activated file memory 30
Data is temporarily stored in the memory 12 via the data line 30B.

The IPL control device 13 sequentially reads the IPL data from the memory 12 via the data line 10E, ! p, IP
The IPL data is further sent to the IPL control device 23 via the L data signal path 42, and during this time, the stop timing of the IPL data request signal that is continuously sent out from the IPL control device 23 is monitored.

When detecting the stop of the above IPL data request signal, the IPL
The control device 13 notifies the central control device 11 that the IPL data request signal has stopped via the control line 10B, and ends the series of operations.

On the other hand, the IPL control device 23 sequentially sends the IPL data sent from the IPL IIJ output device 13 via the IPL data signal path 42 to the memory 2 via the data line 2OA.
After storing the final data, the central controller 21 is activated via the control line 20B, and the data is stored in the memory 22.
The operation is performed based on the IPL data stored in the IPL data.

The activated central control unit '1f21 is connected to the control line 20C.
A command to stop the IPL data request signal is transmitted to the IPL control device 23 via the IPL control device 23.
stops sending the IPL data request signal, and ends the series of operations.

Next, when an IPL start request is issued from the slave processing device 20 side, the operations in the flow after start B in FIGS. 2 and 3 are performed, and the desired IPL is completed in the same way as described above.
Each operation is exactly the same as that described above, so a description thereof will be omitted.

As described above, according to this embodiment, it is possible to perform IPL of the slave processing device 20 without performing IPL of the main processing device 10 itself, and by monitoring the stop of the IPL data request signal, There is also the effect of improving reliability in that the processing device 10 can confirm that the IPL of the slave processing device 20 has completed normally.

Although FIG. 1 shows a case where there are a plurality of slave processors 20, it is of course possible to have only one slave processor 20, and the main processor 10 and the slave processor 20 may both be duplicated. may have been done.

Additionally, IPL activation signal path 40. IPL data request signal path 41. The IPL data signal path 42 may be a PCM line or the like in addition to a normal signal line, and especially in the case of an electronic exchange, it may be composed of, for example, a time switch or a space switch of a time division exchange controlled by a main processing unit. It is also possible to use a specific channel of the communication path.

〔Effect of the invention〕

As described in detail above, according to the present invention, IPL of each slave processing device can be executed independently and from either the main or slave processing device, and the main processing device can confirm the successful completion of the IPL. This improves the efficiency of distributed processing information processing systems.

Significant effects can be obtained in improving reliability and operability.

[Brief explanation of the drawing]

FIG. 1 is a system configuration diagram of an embodiment of the IPL method according to the present invention, and FIGS. 2 and 3 are flowcharts of the main processing device side and the slave processing device side, respectively. 10.20...Main processing unit, sub-processing unit, 11°21
...Central control device, 12.22...Memory, 13゜23...IPL control device, 14...Timer, 30
...File memory, 40...IPL start signal path,
41...IPL data request signal path, 42...IPL
data signal path.
1 Agent Patent attorney Kosaku Fukuda; ゛-: (1 other person)

Claims (1)

[Claims] 1. A distributed processing type in which a file memory for storing initial program load data is exclusive to the main processing unit, and each slave processing unit is configured to receive the initial program load data via the main processing unit. In an information processing system, the main processing unit sends an initial program load start signal to each slave processing unit individually, and each slave processing unit issues its own initial program load start instruction, so that each slave processing The device successively sends each required initial program load request signal to the main processing device, and the main processing device detects the initial program load request signal and sends the necessary initial program load request signal to the slave processing device. By sending predetermined initial program data, the slave processing device receives it, and stops the initial program load request signal after receiving the final data, the main processing device stops the signal. An initial program loading method characterized by detecting and confirming normal completion of the initial program loading.