JPH02189664A - Method for processing fault in channel control device - Google Patents

Abstract

PURPOSE:To surely execute the processing of fault recovery by extracting the states of hardwares constituting a channel control device, the state of control information to be used at the time of controlling channel devices and the execution trace of a microprogram for controlling the hardwares constituting the channel devices and analyzing the cause and range of a fault. CONSTITUTION:At the time of detecting a fault in the channel control device 4 in a condition that the device 4 for controlling plural channel devices 5 to 7 with time division is not connected to any of the channel devices 5 to 7, the state of the hardwares constituting the device 4, the state of the control information to be used at the time of controlling the channel devices 5 to 7 and the execution trace of the microprogram for controlling the hardwares of the channel devices 5 to 7 are extracted. The cause and range of the generated fault are analyzed, an optimum channel number to be used for a fault report to a CPU 1 is selected and the fault report to the CPU 1 is executed by using the channel number. Thus, the channel device generating the fault can be specified and the processing of fault recovery can be surely executed.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a method for handling failures detected in a channel control device that controls a plurality of channel devices in a time-sharing manner.

[Conventional technology]

In recent years, channel devices that control data transfer between central processing units and input/output devices have been trending toward miniaturization and function substitution for central processing units. It has begun to take on the character of an output processing device. On the other hand, the external device called the channel control device cannot be recognized from the central processing unit side.
Failures due to abnormality detection in the channel control device shall be reported as channel device failures. However, depending on the type of abnormality in the channel control device, it may not be possible to immediately identify the faulty channel among all channels controlled by the channel control device. For example, if a failure occurs in the channel control unit while the channel control unit is searching for a control request interrupt from a channel or a channel activation instruction from the central processing unit, the channel number that caused the failure is immediately retrieved. may not be able to be identified. Therefore, in the past, the following methods were used: (1) Reporting to the central processing unit using an "external device abnormality interrupt," (2) Using a predetermined channel number, for example, the lowest number among the channel devices controlled by oneself. Either the channel control device reports to the central processing unit using an "input/output interrupt" using the number of the last channel it controlled, or ■When a failure is detected, the channel control device temporarily suspends reporting to the central processing unit and , when a control request interrupt from a channel, or a channel activation request from the central processing unit occurs, the channel number that made these requests is used to report to the central processing unit by "input/output interrupt". The method was flawed.

[Problem to be solved by the invention]

However, method (2) has the disadvantage that all channels controlled by the channel control device must undergo failure recovery processing. In other words, "interrupt due to external device error"
Subsequent failure recovery processing must be performed on a plurality of channel devices, which impairs system processing speed and availability. On the other hand, in the case of method (2), since the channel determined to be abnormal is not suitable for actual operation, there is a risk that recovery processing will not be performed in the program. Furthermore, although recovery processing is performed with a fairly high probability in method (2), there is a drawback that there is no guarantee that 100% recovery processing will be performed in all cases.

[Means to solve the problem]

A fault processing method for a channel control device according to the present invention is a method for processing a fault detected in a channel control device that controls a plurality of channel devices in a time-sharing manner, the channel control device controlling any particular channel device. If a failure is detected in the channel control unit while searching for a control request from the channel device or a startup instruction for the channel device from the central processing unit, the failure is reported to the central processing unit. The state of the hardware that is included in the channel control device that makes up the channel control device, the state of the control information used to control the channel device, and the hardware that makes up the channel device. By extracting the execution trajectory of the microprogram to be controlled and analyzing the cause and scope of the failure that has occurred, the optimal channel number to be used when reporting the failure to the central processing unit is selected, and the channel number is This is used to report failures to the central processing unit.

[For production]

Therefore, if the channel control device is not controlling any specific channel device and is searching for a control request from the channel device or a startup instruction for the channel device from the central processing unit, the channel control device When a fault is detected in the system, the channel device in which the fault has occurred can be identified, and fault recovery processing can be performed reliably.

〔Example〕

Next, embodiments of the present invention will be described with reference to the drawings.

FIG. 1 is a connection configuration diagram of input/output channels.

The input/output processing device 2 includes a plurality of channels 5 to 7 and a channel control section 4 that controls the channels 5 to 7 in a time-sharing manner, and is connected to the central processing device 1 via a human output interface 9. Each channel 5 to 7 is connected to the input/output device 3 via an I10 interface 11, and further connected to the channel control section 4 via a channel interface 10. The channel control unit 4 has a control storage (CS) a, and controls channels 5 to 7 based on a microprogram stored therein.

Next, the operation of this embodiment will be described using as an example the processing of a C3 continuous output failure in the channel control unit 4.

FIGS. 2 and 3 are block diagrams of the processing flow of the microprogram. The processing flow of the microprogram can be classified into three routines.

The first routine is the idle cortin, which waits for control requests from channels 5-7 or channel activation instructions from central processing unit 1. Usually, this routine executes microinstructions for self-diagnosis. The second routine is a "channel control" routine, and when a control request, etc. from channels 5 to 7 occurs, the microprogram is forcibly branched to the start address of the routine (called Blnibreak-in), and processing is performed. Start. The third routine is the "Fault Handling J Routine" which is started due to a read error in C58.The forced branch operation to the "Fault Handling J Routine" is distinguished from the BI operation, and the 5TL
(steal) action.

FIG. 2 is a flowchart of processing when a C88 read error is detected while the microprogram is executing the "channel control" routine. A "channel control" request occurs at points A and 0, and the microprogram transitions from the "idle" routine to the "channel control" routine. At point B, the processing of the "channel control" routine is completed,
Back to the "idle" routine again. The behavior at this time is B
The I operation is called a BO (break out) operation.

In the example of FIG. 2, no C58 read failure occurs between ANBs, and a "channel control" request occurs again at point 0, causing a Bl operation. When a reading failure occurs in the C58 at point 0 in the request, the microprogram starts "failure processing" by STL operation. After performing failure recovery processing such as rewriting the C58, it reports to the central processing unit 1 that a failure has occurred in the channel that was performing "channel control", and completes the "failure processing". In this way, a failure in the channel control unit 4 detected during “channel control” can be reported as a failure in the channel that was under control of the channel control unit 4. On the other hand, when the channel control section 4 is not controlling any channel, that is, when it is "idle", it cannot be determined that a fault in the channel control section detected is a fault in the other channel.

FIG. 3 is a flowchart of failure processing according to this embodiment. When a read failure of CS8 is detected at point A of the "idle" routine, a "failure handling" routine is initiated by STL operation. At this time, the operation of the "fault handling" routine will be explained using FIG. FIG. 4 is an embodiment of the present invention, and shows an example of the configuration of a portion of the channel control unit 4 that is related to the present invention. The microinstruction read from CS8 is set in C3 read register 12, and its operation section becomes an input to decode circuit 16 via signal line 20. The decoded output of this decode circuit 16 is given to each part of the device, thereby executing the microinstruction.

On the other hand, normally the address part of the C8 read register 12 is set in the CS address register 21 via the signal line 19, and this is given to the CS8 as the address of the next microinstruction. When a control request from a channel arrives through a signal line 24 and is accepted by the request selection circuit 14, a control request acceptance signal line 25 is excited. The surface water latch circuit 15 during channel control is activated by this control request acceptance signal line 25.
is set, and furthermore, the BI address signal line 30 is set in the CS address register 21 through the signal line 23. That is, the microprogram forcibly branches to the start address of the "channel control" routine, and processing of the "channel control" routine is started. It is assumed that the surface water latch circuit 15 during channel control is reset through its reset signal line 27 when the decode circuit 16 decodes the end of channel control. Therefore, when the channel control surface water latch circuit 15 is in the reset state ("idle" surface water), the channel control display signal line 2
6 allows the C88 to know that the channel control unit 4 is in the "idle" state, and processes the "self-diagnosis" routine by sending the "self-diagnosis" routine microprogram to the C5 read register 12. are doing. Now, if there is a read failure of C38 during "channel control" or "idle", it will be detected by the check circuit 13, and the check output signal line 22 will be excited so that a new signal will arrive through the signal line 24 and the request selection will be made. circuit 1
4, the control request instructed from the control request reception signal line 25 is forcibly suppressed, the STL address signal line 31 is set to the CS address 21 through the signal line 23, and a forced branch is made to the "fault handling" routine. do. At this time, upon learning from the channel control indication signal line 26 that the channel control unit 4 was in the "idle" state, the C58 "sends a microinstruction for handling the failure during idle j to the C3 read register 2. At this time, as fault processing during idle, the state of the hardware constituting the channel control unit 4,
Control unit 111 used to control channels 5 to 7 (information for controlling the hardware and microprograms that configure channels 5 to 7, specifically memory tables, data, and registers for firmware in C5) content, etc.) and the execution trajectory of the microprogram that controls the hardware constituting the channel devices 5 to 7, and analyzes the cause and scope of the failure that has occurred,
The optimum channel number to be used for reporting a failure to the central processing unit 1 is selected, and further processing for reporting the failure to the central processing unit 1 is performed using the channel number.

〔Effect of the invention〕

As explained above, the present invention provides a method for configuring a channel control device when a fault is detected in the channel control device in a state where the channel control device that controls a plurality of channel devices in a time-sharing manner is not connected to any channel device. Analyze the cause and scope of the failure by extracting the hardware status, the status of the control information used to control the channel device, and the execution trajectory of the microprogram that controls the hardware that makes up the channel device. Then, by selecting the optimal channel number to be used for reporting the fault to the central processing unit, and then reporting the fault to the central processing unit using the channel number, the faulty channel device can be identified and the fault can be resolved. This has the effect of ensuring that the recovery process is carried out reliably.

[Brief explanation of the drawing]

FIG. 1 is a connection configuration diagram of the human output channel targeted by the present invention, FIGS. 2 and 3 are flowcharts of the processing of the microprogram in the channel control unit 4, and FIG. It is a block diagram. DESCRIPTION OF SYMBOLS 1... Central processing unit, 2... Input/output processing unit, 3... Input/output device, 4... Channel control unit, 5.6.7-... Channel, 8... Control storage , 9... Input/output interface, 10-... Channel interface, 11... I10 interface, 12... CS read register, 13... Check circuit, 14... Request selection circuit, 15... Channel control surface water latch circuit, 16-・
・Decode circuit, 21...CS address register, 20.23.24...signal line, 22...check output signal line, 25...control request reception signal line, 26...channel control display Signal line. 27--Reset signal line, 30--Bl address signal line, 31-STL address signal line.

Claims (1)

[Claims] 1. A method for handling a failure detected by a channel control device that controls a plurality of channel devices in a time-sharing manner, wherein the channel control device does not control any specific channel device;
When a failure is detected in the channel control device while searching for a control request from the channel device or a startup instruction for the channel device from the central processing unit,
Temporarily suspends reporting the failure to the central processing unit,
The state of the hardware included in the channel control device that makes up the channel control device, the state of the control information used to control the channel device, and the execution trajectory of the microprogram that controls the hardware that makes up the channel device. By analyzing the cause of the failure and the scope of the failure, the optimum channel number to be used when reporting the failure to the central processing unit is selected, and the channel number is used to report the failure to the central processing unit. A fault handling method for a channel control device that reports faults.