JPH05233481A - Program tracing stop device of debugging device - Google Patents

Abstract

PURPOSE:To automatically stop program tracing synchronized with the fault occurrence of a host device. CONSTITUTION:The program tracing stop device of the debugging device 6 that has a program tracing part 5 which reads out a program being executed by the host device of an information processing system wherein plural work stations 3 are connected to the host device 1 through a work station channel 2 and polling from the work station channel 2 to the work stations 3 is performed for control and a buffer part 5 which updates and stores the read program trace in order is equipped with a polling response part 7 which is connected to the host device 1 through the work station channel 2 and responds to the polling similarly to the work stations 3 and a timer part 8 which measures intervals of the polling to the polling response part and stops the new update of the program trace to the buffer part when the interval of the polling exceeds a specific value.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a program trace stopping device in a debug device, and more particularly, by connecting a plurality of workstations to a host device through workstation channels and polling each workstation from the host device. The present invention relates to a program trace stopping device in a debug device having a buffer unit for sequentially updating and storing a program being executed by a host device of an information processing system for controlling.

[0002]

2. Description of the Related Art As an information processing system, as shown in FIG.
In addition, a display device, a printer, etc. are connected via a workstation channel 24 (hereinafter simply referred to as channel 24), and a host device 21 and a workstation 22 are connected.
There is a method for controlling the transmission of data to and from the channel 24 by polling the channel 24. In the figure, reference numeral 25 is a central processing unit (CPU) provided in the host device 21, 26.
Indicates memory.

As described above, the workstations controlled by polling are assigned machine numbers, and as shown in FIG. 5, normally, the polling (P) is automatically sent from the channel to the workstation of each machine number. In response to this polling, each workstation responds to the channel with "no interrupt factor response" when there is no interrupt factor.
(N) is sent, and when there is an interrupt factor such as data transfer, “response with interrupt factor” is sent to the channel (R)
Is sent.

When this "response with interrupt factor" (R) is output, the channel stops polling (P) and raises an interrupt to the host device. By this processing, the host device performs processing (D) such as transmission / reception of data requested by the workstation, and after exchanging a signal of “data transmission / reception completed” (A) between the workstation and the channel, The polling (P) is instructed to restart.

The example shown in FIG. 5 is an example in which three work stations are connected to the host device, and the work station 1 responds to the second polling from the channel by "response with interrupt factor" (R). Is output, data transmission (D) is received from the central processing unit (CPU) of the host device, and "data transmission / reception completed" (A) is output. In this way, the information processing system is controlled.

By the way, during the development of such an information processing system, a debug device is used when performing debugging by tracing and removing an error in a program for controlling the system or other software. In order to collect the program trace which is the history information of the execution process of the program processed by the device 21, as shown in FIG. 4, a program trace mechanism unit 28 connected to the CPU 25 of the host device and tracing the history information of the program, And a buffer unit 29 for sequentially updating and writing program history information. Here, since the storage capacity of the program in the buffer unit 29 is limited, the programs written in the buffer unit 29 are changed over one after another, and even if the program is written once, after a certain period of time elapses. Is a new program that is overwritten and erased.

In order to debug the information processing system using such a debugging device, the debugging device 27 is connected to the host device 21 so that the host device may suffer a system failure, for example, stoppage of polling. The conditions are input, the program trace is stopped when a failure occurs in the system, and the program stored in the buffer unit 29 is analyzed.

[0008]

By the way, the above-mentioned processing for stopping the program trace must be performed at the timing of occurrence of a failure in the information processing system.
Unable to get valid program trace information.
When the timing for stopping the program trace is delayed, a new program is overwritten and necessary information is erased.

However, in the above-mentioned conventional debugging device, the operator stops the program trace by monitoring the information processing system and confirming that the system has failed, so that the failure occurs. There is a problem that it is difficult to obtain a program trace that matches the timing. SUMMARY OF THE INVENTION It is therefore an object of the present invention to provide a program trace stop device for a debug device that can automatically stop the program trace according to the occurrence of a failure in the host device.

[0010]

In the present invention, as a means for solving the above problems, as shown in FIG. 1, a plurality of workstations 3 are connected to a host device 1 via a workstation channel 2. , A program trace function unit 4 for reading a program being executed by a host device 1 of an information processing system for controlling by polling each workstation 3 from a workstation channel 2, and sequentially updating the read program trace. Debugging device 6 having buffer section 5 for storing
, The polling response unit 7 connected to the host device 1 via the workstation channel 2 and responding to polling similarly to the workstation 3, and the polling interval input to the polling response unit 7 is measured. A program trace stop device in a debug device, further comprising a timer mechanism unit 8 for stopping new update of the program trace to the buffer unit 5 when the polling interval exceeds a predetermined value.

[0011]

According to the present invention, the polling response unit 7 of the debug device connected to the host device 1 is similar to a normal workstation when a normal poll is generated from the workstation channel 2 of the host device 1. A non-interrupt request response (N) is sent to. On the other hand, when polling is stopped due to an abnormality in the host device 1 or has not occurred for a long time, and when the polling interval measured by the timer mechanism unit 8 exceeds a predetermined value, the buffer unit 5 of the program trace function unit 4 is entered. Stop the program trace of and stop the new update of the program trace. Therefore, the buffer unit 5 stores the program trace up to the time when the host device 1 becomes abnormal and polling does not occur for a certain period of time or more, and the operating condition of the host device 1 can be reliably investigated.

[0012]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of a program trace stopping device in a debug device according to the present invention will be described below with reference to the drawings. 2 and 3 show an embodiment of the program trace stop device in the debug device according to the present invention.

In the present embodiment, as shown in FIG. 2, the debug device includes a program trace function unit 14 for reading a program being executed by a host device, a buffer unit 15 for sequentially updating and storing the read program trace,
It is connected to the host device via the workstation channel and outputs "No interrupt request response" (N) to polling as with other workstations.
At the same time as this output, the polling response unit 17 that interrupts the timer mechanism unit 18 and the interval of the interrupt from this polling response unit 17 are measured, and when the interval of the interrupt exceeds a predetermined value, the program trace to the buffer unit 15 is performed. And a timer mechanism section 18 for stopping the new update of.

In the present embodiment, the program trace function unit 14 has a flag 11 as a switch for starting and stopping the program trace function of the program trace function unit 14. By turning this flag on or off, the trace function can be started or stopped, and this flag can be turned on or off manually from the outside or turned off by a signal from the timer mechanism section 18. It is supposed to be possible.

Further, the timer mechanism section 18 is provided with a timer 12 for resetting and starting the interruption by the polling response section 17 and turning off the flag when the timer value reaches a preset value as a polling abnormality. ing. Next, the operation of the program trace stopping device in the debug device according to the present embodiment will be described. FIG. 3 is a flowchart showing the operation of the program trace stop device in the debug device according to the present embodiment.

First, when the first interrupt from the polling response unit occurs, the timer 12 resets the timer value to "0" (ST1) and adds the timer values until the interrupt from the polling response unit 17 occurs. (ST2, ST
4. When the timer value reaches the set value (ST3), the flag 11 of the program trace function unit 14 is turned off (ST5), and the program trace function unit 14 stops the program trace.

Therefore, according to this embodiment, the bus of the program trace function unit is connected to the host device, and further,
Connect the polling response unit to the channel of the host device,
The program trace is started to cause the host device to operate so as to cause a failure. Then, the program trace is stopped when a failure occurs in the host device and the occurrence of polling becomes abnormal, and the program trace of the host device when the failure occurs can be easily collected.

[0018]

As described above, according to the present invention, the program trace stop device in the debug device stops the program trace when the host device fails and polling becomes abnormal. Therefore, it is possible to easily collect the program trace of the host device at the time of occurrence of a failure, and it is possible to shorten the time required for investigating the cause of the failure.

[Brief description of drawings]

FIG. 1 is a diagram showing the principle of the present invention.

FIG. 2 is a block diagram showing an embodiment of a program trace stop device in a debug device according to the present invention.

FIG. 3 is a flowchart showing an operation of a program trace stop device in the debug device shown in FIG.

FIG. 4 is a diagram showing an information processing system in which a debug device according to the present invention is used, together with a conventional debug device.

FIG. 5 is a diagram showing a polling operation of the information processing system.

[Explanation of symbols]

 1 host device 2 workstation channel 3 workstation 4 program trace function unit 5 buffer unit 6 debug device 7 polling response unit 8 timer mechanism unit

Claims (1)

[Claims] 1. A plurality of workstations (3) to a host device (1) via workstation channels (2).
And a program trace function unit (4) for reading a program being executed by a host device (1) of an information processing system that controls by connecting each workstation (3) from a workstation channel (2). , A buffer device (5) for sequentially updating and storing the read program trace, a debug device (6) connected to a host device (1) via a workstation channel (2), and a workstation (3). Polling response unit (7) that responds to polling similarly to
And a timer mechanism section for measuring a polling interval input to the polling response section (7) and stopping new update of the program trace to the buffer section (5) when the polling interval exceeds a predetermined value. (8) A program trace stop device in a debug device, comprising: