JPH0152774B2 - - Google Patents

Description

[Detailed Description of the Invention] [Summary] When an abnormality in line operation is detected when diagnosing a fault in a data communication system, the central processing unit sets the first diagnostic mode instruction in the command register, thereby preventing an interrupt. The processing unit is notified, and the processing unit causes the corresponding communication line control unit to form a transmission/reception return path based on the first diagnosis mode command, and then transmits predetermined diagnostic data from the communication control unit to pass through the return path. Diagnose the failure of the communication line control unit by receiving the
If no fault is detected by the diagnosis, the processing section notifies the central processing unit, and the central processing unit sets the second diagnosis mode command in the command register, so that the processing section controls the communication line control section. This document describes how to perform failure diagnosis of the line and terminal device downstream from the communication line control section by restoring the transmission/reception return path and then transmitting and receiving predetermined diagnostic data from the communication control section to and from the terminal device.

[Industrial application field]

The present invention relates to a diagnostic method for data processing systems. The diagnostic method according to the present invention is used for diagnosing faults in data communication systems.

[Prior art and problems to be solved by the invention]

Computer systems such as the central processing unit (CPU), internal and external memory, and communication control unit (CMC),
A data communication system (data processing In the past, when an abnormality occurred in the LA provided for each line, the line, or the terminal equipment connected to that line, the operator manually
Fault diagnosis was then performed by operating the line blocking switch to form a return circuit and return to the normal state. However, such operations that require operator intervention are likely to cause operational errors, which may cause the entire system to go down.
Additionally, there is a problem in that the time required for fault diagnosis becomes longer.

SUMMARY OF THE INVENTION In view of the above-mentioned problems of the prior art, an object of the present invention is to provide a method that can appropriately diagnose a fault without requiring operator intervention and shorten the time required for fault diagnosis. be.

[Means for solving problems]

In the present invention, a central processing unit, a communication control unit connected to the central processing unit, a communication line control unit corresponding to each line connected to the communication control unit via a bus, and a communication line control unit corresponding to each line connected to the communication control unit via a bus. A diagnostic method for a data processing system that uses a terminal device connected to a communication line control unit to diagnose a line abnormality by forming a return path for transmission and reception corresponding to the line in which the abnormality has occurred, the communication control unit A processing section provided in the section;
Using a command register that receives a diagnostic mode command sent from the central processing unit, and an encoder that is activated upon reception of the diagnostic mode command and notifies the processing unit of the cause of the interrupt, when an abnormality in line operation is detected, the By setting the first diagnostic mode command in the command register by the central processing unit, an interrupt is notified to the processing unit, and the processing unit performs transmission/reception in the corresponding communication line control unit based on the first diagnostic mode command. diagnosing a fault in the communication line control unit by forming a return path, and then transmitting predetermined diagnostic data from the communication control unit and receiving it via the return path;
If no failure is detected by the diagnosis, the processing unit notifies the central processing unit, and the central processing unit sets a second diagnosis mode command in the command register, so that the processing unit The transmission/reception loop path of the communication line control unit is restored and then the communication control unit transmits/receives predetermined diagnostic data to/from the terminal device, thereby diagnosing a fault in the line and terminal device downstream from the communication line control unit. A diagnostic method for a data processing system is provided.

〔Example〕

The present invention will be explained in detail below using the drawings.

FIG. 1 is an overall schematic diagram of a data processing system according to an embodiment of the present invention. In the figure, 10 is a central processing unit (CPU), 12 is a main memory, and 1 is a central processing unit (CPU).
4 indicates a communication control unit (CMC). CMC 14 and other channels not shown are connected to CPU 10 and memory 1 via bus 16.
Connected to 2. Communication line control units (line adapters, LA) 20-0, 20-1, ..., 20 corresponding to each line are connected to the CMC 14 via a bus 18.
-n is connected, and beyond that each communication line 2
Terminal devices (I/O) 24-0, 24-1, ..., 24-m via 2-0, 22-1, ..., 22-n
is connected.

FIG. 2 is a block diagram of a part of the CMC 14 shown in FIG. In the figure, 16 and 18 are the CPU side bus and LA side bus respectively shown in FIG. 30 is a random access memory (RAM) having a program area for storing the program of the MPU 28, a transmission/reception buffer area, a table area for storing diagnostic mode flags, etc., 32 is an interrupt for the MPU 28 A control encoder, 34 is a memory that stores the start addresses of the transmitting buffer and receiving buffer and the transmitting buffer length in the RAM 30 for direct memory access (DMA), and 36 is a LA20-0
Address decoders 20-n through 20-n are shown, respectively.

FIG. 3 is a block diagram of a part of the LA shown in FIG. 1. In the figure, 18 is the aforementioned LA side bus, and 38 is an LSI element for high-level transmission control procedure (HDLC). This HDLC・LSI
A switch circuit 40 for return control (blocking control) is provided at the line side output of the element 38.
This switch circuit 40 includes an HDLC/LSI element 38
When the DTR terminal becomes a logical "1" output, switches 40a and 40b are opened, and switches 40c and 40d are closed, setting the loopback mode in both directions. Conversely, when the DTR terminal becomes a logic "0" output, switches 40a and 40b are closed, and switch 40
c and 40d are opened to enter normal communication mode (through mode). In Figure 3, 42
is an address compare circuit that detects that the address of this LA has been accessed.

Next, the operation of this embodiment will be explained. When the CPU 10 recognizes that an abnormality has occurred in a specific line, its LA, or its terminal device, the CPU 10
The first diagnostic mode pattern is written into the command register 26 of the controller. This pattern represents the loop test (LTST) mode for diagnosing the line by blocking the line, and one of the LAs that seems to have an abnormality (hereinafter referred to as LA20-0)
Contains the number. When the final address of the command register 26 is written, bit 26a is set indicating that this register 26 is full, and the interrupt control encoder 32 is activated.
An interrupt request is made to the MPU 28. MPU28
This accesses the encoder 32, learns the cause of the interrupt, takes in the information in the command register 26, and analyzes the pattern. This allows the MPU to know that it is in LTST mode.
28 sets a flag in the bit 30a for LA 20-0 at a predetermined position in the table area of the RAM 30 to indicate that the LA 20-0 is in the diagnostic mode. moreover
The MPU 28 connects the corresponding LA 20-0 via an address decoder 36 and an address compare circuit 42.
, turn on the chip select (CS) terminal of that HDLC/LSI element 38, and
A flag indicating the LTST mode is set in a specific bit 38a of a register in the HDLC/LSI element 38. As a result, the terminal DTR of the HDLC/LSI element 38 becomes a logical "1" output, and the switch circuit 40 operates as described above to form a return path. Furthermore, the MPU 28 forms the HDLC diagnostic data for LTST in the transmission buffer 30b of the RAM 30, writes its start address and buffer length to the memory 34, and writes the data to the reception buffer 30.
c is prepared and its start address is written into the memory 34. Next, DMA startup takes place, and CMC14 and
After predetermined exchanges such as a DMA transfer request (REQ) and its acknowledgment (ACK) are performed with the LA 20-0, the diagnostic data in the transmission buffer 30b is transmitted by DMA and returned by the switch circuit 40. The signal is then received again by the DMA and stored in the reception buffer 30c. After that, MPU28
are the contents of the sending buffer 30b and the receiving buffer 30.
It is determined whether or not there is a fault in the circuit up to the switch circuit 40. LTST of more than
When the MPU 28 recognizes that there is no abnormality in the mode, the MPU 28 notifies the CPU 10 that there is no abnormality, and the CPU 10 receives this and writes the second diagnostic mode pattern into the command register 26.
This pattern represents the transfer test (TTST) mode for unblocking the line and performing diagnosis in the normal through mode. The operation of CMC14 in this TTST mode is almost the same as in LTST mode, but
The specific bit 38a of the HDLC/LSI element 38 of the LA 20-0 is not flagged, and the content of the HDLC diagnostic data formed in the transmission buffer 30b is
The difference is that it is different from the case of LTST mode. Therefore, in the TTST mode, the DTR terminal of the HDLC/LSI element 38 becomes logic "0", the switch circuit 40 is restored, and a return path is not formed and becomes a through path. Therefore, the diagnostic data in the transmission buffer 30b is sent to the line 22-0 via the LA 20-0 by DMA, and is either returned as is at the terminal device, or some response based on the procedure is returned, or else no response is made. Dont return. Accordingly
Receive buffer 3 after DMA transmission/reception
By examining the contents of 0c by the MPU 28, it is possible to determine whether or not there is a failure in the line or terminal device downstream from the switch circuit 40.

A specific example of automatic failure diagnosis in the apparatus shown in FIG. 1 is shown in a flowchart as shown in FIG. 4, for example. An example of forming a diagnostic mode instruction used in the flowchart of FIG. 4 is also shown in FIGS. 5a and 5b.

In FIG. 4, preparations for transmitting normal data are first made, and it is determined whether all data has been transmitted. If all data has not been transmitted, a request is made to the communication control unit to transmit normal data, and a determination is made as to whether or not a transmission abnormality has been detected. When a transmission abnormality is detected, a diagnosis mode is instructed to the communication control section, for example, using the instruction format shown in FIG. 5a. Next, it is determined whether or not a diagnostic abnormality has been detected. If YES, the communication control section branches to the abnormality processing section, and if YES, the return data from the communication control section is checked. After checking the returned data from the communication control unit, it is determined whether the reception is normal or not. If it is a yes, a branch is made to the abnormality processing unit of the communication control unit. For example, the diagnosis mode is instructed to the terminal device using the instruction format shown in FIG. 5B, and the return data from the terminal device is checked.

After checking the return data from the terminal device, it is determined whether the reception is normal or not. If it is YES, a branch is made to the error processing section of the terminal device, and if YES, the terminal device An instruction to return to normal mode is given to the device and the communication control unit.

[Effects of the Invention] In the present invention, when an abnormality in line operation is detected, the central processing unit sends a first diagnosis mode command to the communication control unit, so that the corresponding communication line control unit A transmission/reception loop path is formed, and then predetermined diagnostic data is sent and received from the communication control unit to diagnose a failure in the communication line control unit.If no failure is detected, the central processing unit A second diagnostic mode command is sent to the communication control section to release the transmission/reception loop path of the communication line control section, and then predetermined diagnostic data is sent and received from the communication control section to control the communication line. Fault diagnosis is performed for lines and terminal devices downstream from the main unit, and there is no need for an operator to intervene in fault diagnosis. Therefore, there is no risk of an operator's erroneous operation, and no system down or failure will occur due to this. Furthermore, since fault diagnosis is automatically performed online by the program, the time required for fault diagnosis can be shortened.

[Brief explanation of drawings]

Fig. 1 is an overall schematic diagram of an embodiment of the present invention, Figs. 2 and 3 are detailed block diagrams of a part of Fig. 1, and Fig. 4 is a flowchart of a specific example of automatic fault diagnosis. The chart diagrams in FIGS. 5a and 5b are diagrams showing examples of diagnosis mode instruction formats. 10...Central processing unit, 12...Main memory, 14...Communication control unit, 16, 18...
Bus, 20-0 to 20-n...Communication line control unit, 22-0 to 22-n...Communication line, 2
4-0 to 24-m...Terminal device, 26...Command register, 28...Microprocessor, 3
0...Random access memory, 32...Interrupt control encoder, 38...HDLC LSI element,
40...Switch circuit.

Claims (1)

[Claims] 1. A central processing unit, a communication control unit connected to the central processing unit, a communication line control unit corresponding to each line connected to the communication control unit via a bus, and a communication control unit connected to the communication control unit via a bus. A data processing system diagnostic method for diagnosing a line abnormality by forming a return path for transmission and reception corresponding to the line where the abnormality has occurred, using a terminal device connected to each communication line control unit, A communication control unit includes a processing unit, a command register that receives a diagnostic mode command sent from the central processing unit, and an encoder that is activated upon reception of the diagnostic mode command and notifies the processing unit of the cause of the interrupt. When a line operation abnormality is detected, the central processing unit sets a first diagnostic mode command in the command register, thereby notifying the processing unit of an interrupt, and the processing unit performs the first diagnosis. A transmission/reception return path is formed in the corresponding communication line control unit based on the mode command, and predetermined diagnostic data is then transmitted from the communication control unit and received via the return path, thereby preventing the failure of the communication line control unit. Diagnosis is performed, and when no failure is detected by the diagnosis, the processing unit notifies the central processing unit, and the central processing unit sets a second diagnosis mode command in the command register, thereby causing the The processing unit releases the transmission/reception return path of the communication line control unit, and then transmits and receives predetermined diagnostic data from the communication control unit to and from the terminal device, thereby controlling the line and terminal device downstream from the communication line control unit. A diagnostic method for a data processing system characterized by performing fault diagnosis.