SU962913A1 - Device for registering malfanctions of electronic computer - Google Patents

Description

The invention relates to digital computers with hardware and microprogrammed operational control means and can be used in computer diagnostics equipment malfunction.

When an error is detected in the actions of a computer by these means, information is entered into the computer error register, the contents of which are used to record errors and later, when analyzing these errors, to find the faulty element in the computer.

Each of the bits of the computer error register most often has several installation chains in a single state — in terms of the number of monitored parameters that are common in nature for a given error.

In order to speed up the process of detecting a faulty element in a computer in the event of a malfunction, it is necessary to quickly determine the source source, i.e. the first time register of the error register set to one (if several bits were set during a failure). Then, for this discharge, it is necessary to detect the circuit that established it in a single

the state, the signal of which, in relation to the signals of the rest circuits, came first in time.

A device for registering errors in a computer is known, in which several error sensors are connected to logic multiplication circuits connected to a group of triggers in a circuit that, based on the output signals

10 triggers generate a signal that is fed to the second input of logic multiplication circuits, and as a result of this signal, one of the triggers responds to the error signal

15 with the corresponding of the sensor, is transferred to the Error state, while the other triggers remain in the initial state {1.

However, this device does not perform all the functions assigned to the fault register in a computer, for example, does not allow fixing all errors in a faulty situation.

Closest to the proposed

25, by its technical nature, is a device for fixing signals from computer control circuits, containing a fault register, the bit inputs of which are the inputs of the control device,

30 whose input is connected to the output of the write strobe blocking node, and the serial outputs are device outputs, the error signal generation unit, the output of which is the device error signal output, the first input of the recording strobe block node is the first synchronization input of the device, the second the entry of the write strobe blocking node is the blocking input of the device lock. In order to improve the noise immunity, a common error trigger and an additional register are entered into this device, the bit outputs of the fault register are connected to the bit inputs of the additional register, the bit outputs of which are device outputs, the output signal error shaping blocks connected to the common error trigger input, the output of which is connected to the input of the blocking gate gate 2. However, the use of a system of synchronization signals does not allow the use of this device in computer nodes, such as error processing nodes, where synchronization systems are absent. The presence of several nodes and blocks (an error signal generation block, a common error trigger, and a recording strobe blocking node) leads to the fact that because of the delay in each of these nodes, the strobe recording signal for the recording strobe is generated with a time delay. This delay does not allow one to determine the source of a computer fault if the error signals follow each other at intervals of time that are shorter than the delay time. Thus, the complexity of the circuit causes a loss in the speed of the device. In addition, these disadvantages do not allow the device to be used as an autonomous, suitable for use in troubleshooting in any of the computer devices. The aim of the invention is to increase the reliability of the control. The goal is achieved by the fact that the device for recording failures of an electronic computer containing a fault register, the first single inputs of the bits of the register of failures form a group of information inputs of the device, the zero inputs of the bits of the register of failures are interconnected and connected to the input of the initial settings of the device , a switch and a bit indication block are introduced, and the zero outputs of each trigger of the fault register are connected to the corresponding inputs of the bit indication block and through The current contacts of the switch are connected by one-on-one cares of the remaining triggers of the fault register. The drawing shows a block diagram of a device for fixing computer failures. The device contains a fault register 1, a group of information inputs 2 of the device, an input 3 of the initial settings of the device, a block 4 of the serial indication and a switch 5. When the switch 5 is closed, the fault register 1 is covered by reverse connections and after the initial recording of information in any From the bits of the register 1 faults, the signal from the zero output of the trigger of this bit via feedback is fed to the single inputs of the other triggers of the fault register 1, prohibiting the recording of information into these triggers. The possibility of a new entry to the register 1 of failures will be ensured only after the signal of the initial settings at the input 3 of the device is given. With the open position of the switch 5, the device performs the usual functions that are assigned to the computer's fault recording devices — it records all faults that occur in the computer. The device works as follows. The switch 5 is set to the open position. The information inputs 2 of the device are connected to the outputs of the nodes that generate signals for computer malfunctions (such control points can be installation circuits in a single state of the equipment for external equipment of the computer or trigger registers for the central processor, etc.). On the device input 3, the fault register 1 is reset to the initial (zero) state by applying a zero potential, for example, using a button. In this case, the glow of the elements of the blo: ca 4 of the alternate indication, for example, of the LEDs, is absent. Then, the checked device of the computer is transferred to the working or diagnostic mode, in which a failure occurs. Since, in the open position of the switch 5, the device is a register providing the possibility of independent bit-wise recording of information, all types of errors will be reflected in register 1 and, the display unit 4. Thus, all errors of the tested computer device in a faulty situation will be revealed. The source is then determined, i.e. This error in the operation of the computer device, which is the cause of the failure, and all other identified reasons for the failure of the checked computer device will be only a consequence and in the analysis, the failure is not considered.

To determine the original source, the switch 5 is set to the closed position, and at input 3, the unit in the computer is reset to the initial state, as described above. The tested computer device is also reset.

In the closed position of the switch 5, the fault register 1 is covered by feedbacks in such a way that the output of each bit is connected to the input of the unit in the unit state of the other bits. After the initial entry of information into any of the bits of the fault register 1, the signal from the output of this bit is fed to the input of the unit in the unit state of the remaining bits and forcibly keeps these bits in the zero state.

Then, the checked computer device is switched to the mode in which the failure occurs. Visually, with the help of the elements of block 4 of the serial indication, the unit of the fault register 1 set in one state is determined, which indicates the initial state of the computer device being checked

Thus, the proposed device eliminates the disadvantages of the prototype.

The advantage of the proposed device is that, made as a separate board, it allows you to quickly connect to any control point of any of the tested devices of a computer and, allowing you to quickly determine the source source failure, saves you from wasting time on analyzing several circuits. - ryh are connections. Reducing the time to search for faulty computer hardware increases the utilization rate of the computer.

ten

Claims (2)

1. Japanese Patent 521831, cl. O 06 F 11/00, published 1978 2. Author's certificate of the USSR I 556441, cl. U 06 F 11/00, 1970. l g R l 2 . J l .J