JPS59229658A - Information processor - Google Patents

Abstract

PURPOSE:To investigate easily the cause of an incorrect exception by stopping the operation of a status history storage means when the incorrect exception is generated, and reading out information stored in the state history storage means. CONSTITUTION:When an exception detecting part 31 detects an exception during the execution of a program by an operation control part 3, the detected signal is sent to a state history storage control part 34 through an exception detecting part 32 and the operation of the status history storage part 33 is stopped. Subsequently, information stored in the status history storage part 33 is read out and stored in an exception information saving area in a storage part 1. After the completion of reading-out from the state history storage part 33, a diagnosis control part 21 specifies the start of the operation of the state history storage part 33.

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of Industrial Application) The present invention relates to exception processing in an information processing device, and particularly to an exception information processing method using state history storage means.

(Prior Art) Conventionally, in this type of information processing device, when an exception is detected, a program routine corresponding to the exception number is activated, and thereafter all processing is performed by the program. Exceptions are mainly caused by program errors. However, sometimes a hardware failure causes the device to malfunction, resulting in an exception.

Therefore, if an exception occurs due to a program error, the cause of the error becomes clear through the processing of the program started by the exception. but,
If the exception occurred due to a hardware failure, the processing of the program started by the exception alone will not solve the problem.
There was a drawback that it was difficult to investigate the cause of the error.

Therefore, a method has been proposed in which the state of the hardware when an exception occurs is logged in a maintenance diagnostic device. but,
When paging processing is performed using virtual memory, many functional exceptions occur, so if the maintenance/diagnosis device were to operate every time an exception occurred, performance would deteriorate.

In addition, when an exception occurs due to a hardware failure, information at that point is important, but information about the progress that led to the occurrence of the exception is usually even more important, and with conventional equipment, progress information cannot be obtained. The drawback was that it could not be done.

(Object of the Invention) The object of the present invention is to stop the operation of the state history storage means when an illegal exception occurs, read out and hold the information stored in the state history storage means, and read and hold the information stored in the state history storage means, To provide an information processing device that solves the above drawbacks by transferring the information to an area where it can be processed by a program, and is configured so that it is extremely easy to investigate the cause of an incorrect exception. It is in.

(Structure of the Invention) An information processing device according to the present invention is composed of a storage section, a system control section, an arithmetic control section, an input/output control section, and a maintenance/diagnosis device. history means;
It includes first and second exception detection means and means for controlling the state history storage means.

The first exception detection means is for detecting an exception, and the second exception detection means is for detecting an exception such that the exception detected by the first exception detection means is a predetermined range-0 exception. This is to detect that.

On the other hand, each means for controlling the state history storage means includes a stopping means, a reading means, a holding means, a starting means, and a transferring means.

The stopping means is for stopping the operation of the state history storage means based on the output of the second exception detection means, and the reading means is for reading information from the state history storage means whose operation has stopped. It is. The holding means is for holding the read information, and the starting means is for starting the operation of the state history storage means after the information has been read from the state history storage means. The transfer means is for transferring the information held by the holding means to the designated area of the storage section in response to instructions or commands.

(Example) Next, the present invention will be explained in detail with reference to the drawings.

In FIG. 1 showing the first embodiment of the present invention, a 1Fi storage section, a 2Fi system control section, a 3i/'i calculation control section,
4Fi input/output control unit, 5Fi maintenance diagnosis device, 21 is a diagnosis control unit, 22 is a storage access control unit, 31 is a first exception detection unit, 32 is a second exception detection unit, 33Fi state history storage unit, 34Fi Status register M storage control unit, al''-84 is an address signal line, ω1 to ω4 are write data signal lines, r
l %i4 Fi read data signal line, el, e2
*jl, t2, t4 are control signal lines, respectively.
t3 indicates a state history storage unit information read signal line. FIG. 2 is a diagram showing in detail the configuration of the storage unit 1 shown in FIG. 1. It is roughly divided into two areas 11 and 12 by the storage unit IFi boundary 13. Area 12 is an area where programs or data are normally stored, and can be accessed by the program. Area 11 is an area that stores microprograms and control tables for the information processing device, and cannot be accessed directly from the program. Exception information save area 14 is placed in the Fi area 11 and is roughly divided into two areas 141 and 142.

When the arithmetic control section 3 detects an exception by the first exception detection section 31 during program execution, a detection signal is sent to the second exception detection section 32 through the control signal line el.

The second exception detection section 32 investigates whether the exception detected by the first exception detection section 31 is an invalid exception or a functional exception. Finally, if a specific group of exception numbers is found when examining the exception numbers, it is detected that an unauthorized exception has occurred. The detected signal passes through the control signal line eg to the state history storage control unit 3.
Sent on 4th. At this time, a microprogram that executes illegal exception handling is activated. This microprogram reads the contents of the internal registers and scratchpad memory of the arithmetic control unit 3, stores them in the exception information save area 142 of the storage unit 1, and then instructs the exception handling program to stop the operation of the state history storage unit 33. is carried out via the control signal line t4, and the diagnosis control unit 2 indicates that the operation of the state history storage unit 33 has stopped via the control signal line t1.
Notify 1. Upon learning that the operation of the state history storage section 33 has stopped due to an invalid exception, the diagnostic control section 21 reads out the information stored in the state history storage section 33 via the state history storage section information readout signal line ts, and stores the information. Section 10 stores the exception information in the save area 141. After the diagnostic control unit 21 finishes reading from the state history storage unit 33, the diagnostic control unit 21 connects the control signal line t.
2 to start the operation of the state history storage section 33.

When a program started by an exception examines the exception status and determines that it is impossible to continue the program that generated the exception, the program executes a teardown process. If the program can continue, the exception handling program executes the necessary processing and then starts the exception generating program. On the other hand, when it is determined that it is impossible to continue operating the system due to the occurrence of an exception, system crash processing is executed. Processing processes such as program aborts and system crashes are also caused by exceptions that occur due to program errors, and are also caused by exceptions that occur due to hardware failures. Usually, hardware failures are detected by a failure detection circuit, but sometimes they go undetected and result in data errors or exceptions. Therefore, a read command for the exception information save area 14 is issued by a routine such as a program abort or a system crash, and the exception information in area 11 is read out to any area 15 or 16 in area 12. The file is saved to a magnetic disk along with the program environment information, and the file is output to a printer. Since the area 11 cannot be accessed by the program, a dedicated instruction is provided for reading the contents of the exception information save area 14. According to this command, any area within the area 14 can be specified. Therefore, the arithmetic control unit 3 activates a microprogram routine to execute this instruction, and the microprogram reads out the information in the fixed area 14 and writes it to the designated areas 15 and 16, thereby executing the instruction function. It has been realized.

Whether or not the designated area 15 exists within the area 12 is constantly checked. If area 15 does not exist in area 12, it is detected as an exception, and state history storage section 33 continues to record predetermined information at predetermined sample timings. Normally, when a fault is detected by the arithmetic control section 3, the state history storage section 33 stops operating, and the information is continuously outputted to the maintenance diagnostic device 5 along with other fault information via the diagnostic control section 21.

In this embodiment, the storage unit 1 is used as a storage means for the information read out from the state history storage unit 33, but the diagnostic control unit 2
1 or the maintenance/diagnosis device 5.

(Effects of the Invention) As explained above, the present invention has the following advantages: When an illegal exception occurs, the operation of the state history storage device is stopped without deteriorating the performance, and the stored information is read and retained. By configuring this to be transferred to an area where siprogram processing is possible according to instructions from the program, it is possible to investigate the cause when a siprogram abort or system crash occurs due to an invalid exception. This has the effect of being able to be carried out efficiently.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing an embodiment of an information processing apparatus according to the present invention. FIG. 2 is a block diagram showing details of the storage section shown in FIG. 1. 1-...Storage unit 2...Fist system control unit 3-Life/arithmetic control unit 4...Input/output control unit 5...Maintenance diagnosis device 11-16,141,142,151,152゜161
．． 162.Memory area 13.@ψ inter-area boundary 21...Diagnosis control unit 22...Storage access control unit 31.32...Exception detection unit 33...State history storage unit 34... State history storage control unit al ′a4, ω1 ”-114, rl ”=
r4, el, e2. 11~t4...e--signal line

Claims (1)

[Claims] A storage unit, a system control unit, and an arithmetic control unit. In an information processing device comprising an input/output control unit and a maintenance diagnosis device and equipped with a state history storage means, a first exception detection means for detecting an exception; a second exception detection means for detecting that the exception falls within a predetermined range; and an output of the second exception detection means to stop the operation of the front state history storage means. a reading means for reading information from the state history storage means whose operation has stopped; a holding means for holding the read information; and a continuation of information from the state history storage means. a starting means for causing the state history storage means to start operating the state history storage means; and a start means for causing the state history storage means to start operating the state history storage means, and transferring the stored information from the storage means to a designated area of the storage unit in response to an instruction or a command. An information processing device characterized by comprising: a transfer means for