JP3043738B1 - Fault occurrence identification system and its identification method - Google Patents

Abstract

An object of the present invention is to quickly identify a failure without interrupting system processing. A bus control device (2) controls a central processing unit (1).
The use status of the bus 10 is monitored by a plurality of input / output control devices 3, 4, and 5 connected via the bus 10, and when a failure occurs, a failure device report signal 8 which is failure information.
In addition to outputting 0, the central processing unit 1 disconnects any of the input / output control units 3, 4, and 5 in which a failure has occurred from the bus based on the failure unit report signal 80.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a fault occurrence specifying system for specifying a fault in a system for transferring data using a bus and a method for specifying the same.

[0002]

2. Description of the Related Art Generally, in a system for transferring data using a bus, for example, a central processing unit and a plurality of input / output control units are connected via a shared bus. If a failure occurs in any of the input / output control devices, the use of the shared bus becomes impossible, so that the use status of the bus is monitored.

In order to monitor the use state of such a bus, for example, Japanese Patent Application Laid-Open No.
When a failure occurs, the monitoring test circuit outputs bus error information to the CPU. As a result, the faulty input / output control device is opened, and the shared bus can be used.

[0004]

However, in the above-mentioned prior art, although a shared bus can be used by opening a faulty input / output control device, a fault occurs. In such a case, it is necessary to perform a test on another input / output control device that does not cause a failure.

For this reason, not only takes time to specify the I / O control device in which a failure has occurred, but the processing of the system is interrupted while the I / O control device in which the failure has occurred is specified. There is a problem.

The present invention has been made in view of such circumstances, and provides a failure occurrence identification system and a method for identifying failures that can quickly identify a failure without interrupting the processing of the system. Is what you can do.

[0007]

According to a first aspect of the present invention, there is provided a fault occurrence identifying system, comprising: a central processing unit; a plurality of input / output control units connected to the central processing unit via a bus; monitoring Then Tomo the use state of the bus by
If the data transfer on the bus ends normally,
Outputs an end report signal, and if a failure is confirmed, the
A bus monitoring circuit that outputs an occurrence report signal and an input / output control device
Use the bus use request signal from the
Bus permission corresponding to any of the I / O controllers
A bus arbitration circuit that transmits an enable signal and a bus arbitration circuit
Takes in the bus use permission signal received
Information on which of the buses are used,
A failure occurrence report signal from the bus monitoring circuit is fetched and the
That identifies one of the I / O control units that are generating the error
And a bus control device having a device specifying circuit , wherein the central processing unit disconnects any of the input / output control devices specified by the faulty device specifying circuit from the bus. Further, the faulty device specifying circuit specifies any of the input / output control devices in which a fault has occurred, and outputs a faulty device report signal to the central processing unit. Then, the central processing unit issues an input corresponding to the faulty device report signal. It is possible to output a bus connection invalidation signal for invalidating the bus connection to any of the output control devices. In addition, the faulty device identification circuit may include a plurality of set / reset flip-flops that hold any of the bus use permission signals until the normal end report signal or the faulty device report signal is output, and any one of the set / reset flip-flops An AND circuit for specifying any of the input / output control devices in which a fault has occurred by taking the logical product of the information held in the step S1 and the fault occurrence report signal;
An encoder that encodes a logical value by the D circuit and uses the encoded value as a failure device report signal. According to a fourth aspect of the present invention, there is provided a fault occurrence specifying method, wherein a first step of performing data transfer between a central processing unit and a plurality of input / output control units and a use state of a bus by the input / output control units are monitored .
In both cases, when data transfer on the bus ends normally,
Outputs a normal end report signal and if a failure is confirmed
A second step of outputting a failure occurrence report signal, and input / output control
Obtain bus use right using bus use request signal from device
Bus corresponding to any of the principal to the input-output control unit
A third step of transmitting a permission signal;
Which of the capture and I / O control devices is using the bus
Information and a fault report signal
And identify any I / O control unit that has failed
And a fifth step of disconnecting any of the specified input / output control devices from the bus. The fourth step includes a step of specifying one of the input / output control units in which a failure has occurred and outputting a failure unit report signal to the central processing unit, and a step of input / output control corresponding to the failure unit report signal. Outputting a bus connection invalidation signal that invalidates a bus connection to any of the devices. Further, the fourth step includes a step of holding any one of the bus use permission signals until the normal end report signal or the faulty device report signal is output; The step of specifying any one of the input / output control devices in which a failure has occurred by taking the logical product with the occurrence report signal, and the step of encoding the logical value of the logical product by the AND circuit to generate a failure device report signal Can be included. In the fault occurrence specifying system and the specifying method according to the present invention, the bus control unit monitors the use state of the bus by a plurality of input / output control units connected to the central processing unit via the bus, and the fault occurs. In this case, the failure information is output, and the central processing unit disconnects any of the I / O control units having a failure from the bus based on the failure information.

[0008]

Embodiments of the present invention will be described below.

FIG. 1 is a block diagram showing an embodiment of a fault occurrence specifying system according to the present invention, and FIG. 2 is a block diagram showing details of the fault occurrence specifying circuit of FIG.

The fault occurrence identification system shown in FIG. 1 comprises a central processing unit 1, a bus control unit 2, an input / output control unit 3,
4 and 5 are provided. The bus controller 2 and the input / output controllers 3, 4, and 5 are connected to the central processing unit 1 by a bus 10. The bus 10 is used for data transfer between the central processing unit 1 and the input / output controllers 3, 4, and 5.

The input / output control devices 3, 4, and 5 provide bus use request signals 20, 30, 40 for obtaining the right to use the bus 10.
And the bus use permission signals 50, 60, and 70 are individually transmitted to and received from the bus control device 2.

The bus control device 2 includes a bus monitoring circuit 100,
A bus arbitration circuit 200 and a faulty device identification circuit 300 are provided.

The bus monitoring circuit 100 monitors the normality of the bus 10. When the data transfer on the bus 10 ends normally, a normal end report signal 400 is output, and when a fault is confirmed, a fault occurrence report signal 500 is output.

The bus arbitration circuit 200 includes a bus use permission signal 50 among the input / output control units 3, 4, and 5 that intends to obtain the right to use the bus 10 using the bus use request signals 20, 30, and 40. , 60, 70 is transmitted.
Thus, the right to use the bus is given to any one of the input / output control devices 3, 4, and 5, and the use of the bus 10 is permitted.

The faulty device identification circuit 300 receives the bus use permission signals 50, 60, 7 transmitted from the bus arbitration circuit 200.
0 is acquired, and information as to which of the input / output control devices 3, 4, and 5 is currently using the bus 10 is obtained.

Also, a normal termination report signal 400 and a failure report signal 500 from the bus monitoring circuit 100 are taken in, and information on the presence or absence of a failure is obtained. Further, the central processing unit 1 sends a faulty device report signal 80 for notifying information of the input / output control devices 3, 4, and 5 specified by the acquired information.
Output to

Also, the bus connection invalidation signals 90, 91 and 92 indicating that the connection of the bus 10 is invalidated from the central processing unit 1 to each of the input / output control units 3, 4 and 5 are individually output. . When any one of the bus connection invalidation signals 90, 91, 92 is transmitted from the central processing unit 1, one of the corresponding input / output control devices 3, 4, 5 is logically disconnected from the bus 10.

FIG. 2 shows details of the faulty device identification circuit 300.

The faulty device identification circuit 300 includes set / reset flip-flops 301, 302, 303, AND circuits 304, 305, 306, OR circuits 307, 308,
309 and an encoder 310.

The set / reset flip-flops 301,
Reference numerals 302 and 303 hold one of the bus use permission signals 50, 60 and 70 until the normal end report signal 400 or the failure device report signal 80 is output.

AND circuits 304, 305, and 306 are set / reset flip-flops 301, 302, 303
And the failure report signal 5
By performing AND operation with 00, the input / output control devices 3, 4, and 5 in which the failure has occurred are specified. The result is encoded by the encoder 310 and transmitted to the central processing unit 1 as a faulty device report signal 80 and reported.

Next, the operation of the fault occurrence specifying system having such a configuration will be described.

First, each of the input / output control units 3, 4, and 5 sends a bus use request signal 20, 30, 40 to the bus arbitration circuit 200 to use the bus 10.

The bus arbitration circuit 200 arbitrates upon receiving any of the bus use request signals 20, 30, and 40.
After the arbitration, any one of the bus use permission signals 50, 60, and 70 for permitting use of the bus 10 to only one of the input / output control devices 3, 4, and 5 is transmitted. At the same time, the fault identifying circuit 300 outputs the bus use permission signals 50 and 6
Of the signals 0 and 70, the transmitted signal is held.

The input / output control devices 3, 4, 5 for which use has been granted
Starts data transfer to the central processing unit 1 using the bus 10. When the data transfer ends normally, the bus monitoring circuit 100 transmits a normal end report signal 400.

When a failure is detected by the bus monitoring circuit 100 while any of the input / output control devices 3, 4, and 5 is transferring data, the failure is reported to the central processing unit 1 by the failure identification circuit 300. Is done.

That is, the AND circuits 304, 305, and 306 of the fault identification circuit 300 determine whether any one of the bus use permission signals 50, 60, and 70 held by any of the set / reset flip-flops 301, 302, and 303 has a fault. The logical product with the occurrence report signal 500 is taken.
As a result, the input / output control devices 3, 4,
5 is specified, and the faulty device report signal 80 encoded by the encoder 310 is transmitted to the central processing unit 1.

When the failed device report signal 80 is transmitted, the set / reset flip-flops 301, 302, 303
Is reset.

The central processing unit 1 that has received the failure report keeps the devices in which no failure has occurred among the input / output control devices 3, 4, and 5 running. Any one of the corresponding bus connection invalidation signals 90, 91, and 92 is transmitted to any of the failed input / output control devices 3, 4, and 5, and the failed input / output control devices 3, 4, and 5 are transmitted. , 5 are disconnected from the bus 10, and the processing of the system is continued.

As described above, in the present embodiment, the use state of the bus 10 by the plurality of input / output control units 3, 4, 5 connected to the central processing unit 1 via the bus 10 by the bus control unit 2. And when a failure occurs, outputs a failure device report signal 80 as failure information,
The central processing unit 1 disconnects any of the input / output control units 3, 4, and 5 that have failed from the bus based on the failure unit report signal 80, so that the processing of the system is not interrupted. , The fault can be quickly identified.

In this embodiment, the bus controller 2
Has described the case where any one of the input / output control devices 3, 4, and 5 in which a fault has occurred has been described. However, the present invention is not limited to this example, and a fault is reported from the central processing unit 1 to a terminal device (not shown), You can let them know. In this case, it is easy to confirm the failure, so that the repair or replacement of the failed portion can be quickly performed.

[0032]

As described above, according to the fault occurrence specifying system and the specifying method according to the present invention, the bus control unit controls the bus by the plurality of input / output control units connected to the central processing unit via the bus. The usage status is monitored, and if a failure occurs, the failure information is output, and the central processing unit disconnects any of the failed input / output control devices from the bus based on the failure information. Therefore, the failure can be quickly identified without interrupting the processing of the system.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an embodiment of a failure occurrence identification system according to the present invention.

FIG. 2 is a block diagram showing details of a failure occurrence specifying circuit in FIG. 1;

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 Central processing unit 2 Bus control device 3,4,5 Input / output control device 10 Bus 20,30,40 Bus use request signal 50,60,70 Bus use permission signal 80 Faulty device report signal 90,91,92 Bus connection invalid Signal 100 Bus monitoring circuit 200 Bus arbitration circuit 300 Faulty device identification circuit 301, 302, 303 Set reset flip-flop 304, 305, 306 AND circuit 307, 308, 309 OR circuit 310 Encoder 400 Normal end report signal 500 Failure report signal

Claims (6)

(57) [Claims] A central processing unit; a plurality of input / output control units connected to the central processing unit via a bus; and a use state of the bus by the input / output control unit is monitored .
And the data transfer on the bus ends normally.
Output a normal end report signal and confirm the failure
A bus monitoring circuit for outputting a fault occurrence report signal in the case;
Using the bus use request signal from the input / output control device,
What kind of I / O control unit is trying to obtain the right to use the bus
Bus arbitration that sends a bus use permission signal corresponding to
And the bus license transmitted from the bus arbitration circuit.
Signal, and any of the input / output control devices
Obtain information on whether the bus is being used and
Signal from the road to generate a fault
Faulty device that identifies any of the I / O control devices
And a bus controller and a constant circuit, said central processing unit, especially by the impairment device specific circuits
A fault occurrence specifying system, wherein any one of the specified input / output control devices is disconnected from the bus. 2. The faulty device specifying circuit specifies any of the input / output control devices having a fault and outputs a faulty device report signal to the central processing unit. The fault occurrence identification system according to claim 1 , wherein a bus connection invalidation signal for invalidating the bus connection is output to any of the input / output control devices corresponding to a device report signal. 3. A plurality of set / reset flip-flops each holding one of the bus use permission signals until the normal end report signal or the failed device report signal is output, An AND circuit for specifying any one of the input / output control devices in which a fault has occurred, by taking a logical product of the information held in one of the set / reset flip-flops and the fault occurrence report signal; An encoder for encoding a logical value by an AND circuit and using the encoder as the faulty device report signal.
The fault occurrence specifying system according to claim 1 . 4. A first step of performing data transfer between a central processing unit and a plurality of input / output control units, and monitoring a use state of the bus by the input / output control units .
And the data transfer on the bus ends normally.
Output a normal end report signal and confirm the failure
A second step of outputting a failure occurrence report signal, and a previous step using a bus use request signal from the input / output control device.
What kind of I / O control unit is trying to obtain the right to use the bus
A third step of transmitting a bus use permission signal corresponding to any of the above.
And the input / output control device.
Information on which of the buses are used,
The failure occurrence report signal is fetched and a failure has occurred.
A fault occurrence specifying method comprising: a fourth step of specifying any one of the input / output control devices; and a fifth step of disconnecting any of the specified input / output control devices from the bus. . 5. The fourth step includes: identifying one of the input / output control devices having a failure and outputting a failure device report signal to the central processing unit; 請 characterized to include the steps of outputting a bus connection invalid signal to disable the connection of the bus to one of the input-output control unit corresponding to the signal
A method for identifying a fault occurrence according to claim 4 . 6. The fourth step includes: holding either the bus use permission signal until the normal end report signal or the faulty device report signal is output; Identifying any one of the input / output control devices in which a failure has occurred by taking the logical product of any of the information and the failure occurrence report signal; and encoding a logical value based on the logical product by the AND circuit. And
5. The fault occurrence specifying method according to claim 4 , further comprising the step of setting the faulty device report signal.