JPS6299836A - Fault detection and selection circuit - Google Patents

Abstract

PURPOSE:To detect a fault with attachment/detachment of a small number of minimum units and to improve the maintainability of a fault detection/ selection circuit, by selecting test data in the following machine cycle when a fault is detected and also holding the presence or absence of generation of a fault at that time point. CONSTITUTION:A selection means 6a of a minimum unit 3a selects a minimum unit 1 or 2 via a selection condition changing means 10 which works in response to a selection circuit 300. The output of the selected unit 1 or 2 is supplied to a fault detecting circuit 7 for detection of the generation of a fault. The result of this detection is held by the 1st holding means 8. Then the means 10 is controlled by the means 8 and the means 6a selects the output of a test data setting means when a fault is detected. In the same way, the result of detection given from the circuit 7 is held by the 2nd holding means 9. Thus the fault of the unit 1, 2 or 3 is decided by the contents of both means 8 and 9. This attains the detection of faults with attachment/detachment of a small number of minimum units and improves the maintainability of a fault detection/ selection circuit.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a fault detection and selection circuit for detecting and selecting a minimum unit for maintenance and replacement of a fault in an electronic device, such as an information processing device.

〔overview〕

The present invention provides a failure detection selection device comprising means for selecting monitoring signals from a plurality of minimum units in an electronic device, detecting a failed minimum unit when a failure occurs in the device, and retaining the detection signal in a first holding means. In the circuit, by including means for selecting preset test data instead of the above-mentioned monitoring means and holding the detection signal in the second holding means when a fault is detected, it is possible to easily connect and disconnect in the event of a fault detection. This is intended to improve maintainability by reducing the minimum number of units that can fail and making it easier to select the minimum number of failed units.

[Conventional technology]

FIG. 2 shows an example of a failure detection and selection circuit in a conventional information processing device. In FIG. 2, 1.2.3 is the minimum unit to be replaced for maintenance when a failure occurs, and is a large-scale integrated circuit mounted on a single printed circuit board or socket, which varies depending on the device. In FIG. 2, register 4 in minimum unit 1 and register 5 in minimum unit 2 are input to selection means 6 in minimum unit 3, and the output of selection means 6 is connected to fault detection circuit 7. The presence or absence of a disorder can be checked. Further, the output of the fault detection circuit 7 is held in the first holding means 8.

In the figure, only two minimum units 1.2 are input to the selection means, but in reality, a plurality of N units are shown.
(N is a natural number of 2 or more) are connected.

[Problem that the invention seeks to solve]

The conventional fault detection and selection circuit described above can only detect the occurrence of a fault when the fault is detected by the fault detection circuit 7, but cannot point out the location where the fault has occurred. All of the minimum units 1.2.3 are likely to be faulty locations. In the case of a fixed failure, it is possible to find the minimum failed unit by replacing one unit at a time, but in the case of an intermittent failure, it is necessary to predictably replace all the related minimum units, or to find the most likely failure unit even at the risk of further failure. There was a drawback that only a certain minimum unit had to be replaced.

SUMMARY OF THE INVENTION An object of the present invention is to provide a fault detection and selection circuit that eliminates the above-mentioned drawbacks and enables fault detection to be performed by inserting and removing a smaller number of minimum units, thereby improving maintainability.

Means for Solving Problem C] The present invention selects data from each of a plurality of minimum units constituting an electronic device, detects the failed minimum unit when the electronic device has a failure, and detects the detected signal. In the fault detection selection circuit, the fault detection selection circuit is equipped with a means for holding in the first holding means, the test data set in advance is selected instead of the data from the minimum unit, and when a fault is detected, the detection signal is held in the second holding means. The device is characterized in that it includes a means for holding the device.

[Effect]

The present invention performs failure detection by selecting preset test data in addition to conventional monitoring signals from each minimum unit, and when a failure is detected, the detection signal is transferred to another second holding means. to hold. Therefore, if the second holding means holds the detection signal, it means that the smallest unit including the selection means is out of order, and the faulty smallest unit can be isolated.

〔Example〕

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

FIG. 1 is a block diagram showing one embodiment of the present invention. In FIG. 1, 1.2.3 is the smallest unit that can be replaced for maintenance when a failure occurs. Minimum unit 1.
The contents of register 4.5 in 2 are data lines 100.200
is input to the minimum unit 3a including the selection means 6a. The output of the selection condition changing means 10 is inputted to the selection condition input of the selection means 6a via the signal line 302. At this time, the selection condition changing means 10 outputs the same selection condition as in the conventional example shown in FIG. 2, which is input manually through the signal line 300 out of the two inputs. As a result, the selection means 6a selects the data line 10 designated by the signal line 302 providing the selection condition.
Select and output any data of 0.200. This output is sent to the fault detection circuit 7 via the data line 303. When a fault is detected in the fault detection circuit 7, the signal line 3
01 to the first holding means 8, and displays and holds the fact that a failure has occurred. The occurrence of a failure is notified to the selection condition changing means 10 via the signal line 304.

The selection condition changing means 10 generates selection conditions for selecting data to be input to the selection means 6 via the data line 305 set in advance by the data setting means 11 in the next machine cycle after the failure occurs. Further, this selection condition is input to the selection means 6 via the signal line 302.

The selection means 6 outputs test data set in advance by the test data setting means 11. This selection result is sent to the fault detection circuit 7 via the data line 303, and if a fault is detected by the fault detection circuit 7, it is notified to the second holding means 9 via the means line 301 and held. If there is only one test data, it is possible that some failures may not be detected, so in this embodiment, two test data are used.

In this case, in the next machine cycle, the data line 306 is set in advance by the test data setting means 11.
A selection condition for selecting another test data inputted to the selection means 6 via the selection means 6 is generated by the selection condition changing means 10, and the same operation as described above is performed.

The fault information held by the first and second holding means as described above is retrieved by a conventionally known method and used to point out the least faulty unit that is the cause of the fault.

The feature of the present invention is that the second holding means 9, the selection condition changing means 10, and the test data setting means 11 are numbered 1 in FIG.

Next, analysis of data obtained by this example will be explained. The relationship between the contents of the retained data held in the first holding means 8 and the second holding means 9 after performing the above series of detection and selection operations and the failure of the smallest unit 1 or 2 and the smallest unit 3 is shown in the following table. It is given as follows.

Here, an X mark indicates a failure in the smallest unit and data retention of the failure in the holding means, and an O mark indicates no failure in the smallest unit and data non-retention in the holding means.

That is, if data is held in the first holding means and data is not held in the second holding means, the minimum failure unit is the minimum unit 1 or 2, and both the first and second holding means hold data. In this case, the minimum unit is either 1 or 2 or the minimum unit 3. In this case, by removing the smallest unit 1 or 2 and measuring it, the smallest failed unit can be isolated.

Note that, if the information processing apparatus in which the present invention is implemented is firmware controlled, which input of the minimum unit 1 or 2 was selected in the selection means 6a at the time of the failure indicates the firmware address at the time of the failure. It can be examined by preserving it in a different way.

Furthermore, even in the case of pure hardware control, it is possible to check by saving the control information.

〔Effect of the invention〕

As explained above, the present invention selects test data in the subsequent machine cycle when a fault is detected in the selection circuit, and also retains the presence or absence of a fault at this time, thereby reducing the number of test data. By inserting and removing the smallest unit, the smallest failed unit can be detected and selected, which has the effect of improving maintainability.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention. FIG. 2 is a block diagram showing a conventional example. 1.2.3.3a...Minimum unit, 4.5...Register, 6.6a...Selection means, 7...Fault detection circuit, 8...First holding means, 9... -Second holding means, 1
0... Selection condition changing means, 11... Test data setting means, 100.200.300-306... Signal line.

Claims (1)

[Claims] (1) comprising means for selecting data from each of the plurality of minimum units constituting the electronic device, detecting the failed minimum unit when the electronic device fails, and retaining this detection signal in the first holding means; The fault detection selection circuit according to the present invention is characterized in that it includes means for selecting preset test data instead of the data from the minimum unit and for holding the detection signal in the second holding means when a fault is detected. fault detection selection circuit.