JPH0244437A - Fault detecting method - Google Patents

Abstract

PURPOSE:To improve debugging efficiency by finding out and outputting a fault rapidly by providing a loop counter to monitor an execution program and a timer to judge a normal operation by an instruction outputted from the execution program. CONSTITUTION:In the execution program, the loop counter 4 to monitor the execution program, and the timer to judge the normal operation by a timer reset instruction outputted from the execution program and to compare a timer value with a prescribed value are provided, and a tracer 3 to detect a loop detecting point 6 is provided, and the value of the loop counter 4 is compared with the prescribed value. When the timer value is larger than the prescribed value, or when a loop counter valued is larger than the prescribed value, it is decided as the generation of the fault, then, trace information in a processor 1 is outputted by an output device 5. In such a manner, it is possible to find out the fault rapidly in the debugging of the execution program, and to improve the debugging efficiency.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to debugging of an execution program and a fault detection method for a processing device.

[Conventional technology]

Conventional debugging of execution programs
As described in Japanese Patent No. 767, a breakpoint is set at an arbitrary location in a program, and the state within the processing device is checked when the breakpoint I. is reached.

[Problem to be solved by the invention]

In conventional technology, even if breakpoints are set in the execution program, if a failure such as a runaway of the processing unit occurs,
The breakpoint you set will never be reached, so you end up wasting your time.

An object of the present invention is to provide a fault detection method that can detect hardware and software faults in order to debug an execution program in a relatively short time.

[Means to solve the problem]

The above purpose is to quickly detect and detect failures by providing a loop counter that monitors the execution program and a timer that determines normal operation based on instructions output from the execution program.
This is achieved by outputting.

[Effect]

In the execution program, a loop counter that monitors the inside of the execution program and a timer that determines normal operation based on a timer reset instruction output from the execution program are provided. At the same time, means for comparing the timer value and a predetermined value;
A means for detecting a loop detection point and a means for comparing the loop counter value with a predetermined value are provided, and if the timer value is greater than the predetermined value or the loop counter value is greater than the predetermined value, it is treated as a failure occurrence. Output trace information in the device using an output device. This allows faults to be quickly identified during debugging of the executed program, thereby improving debugging efficiency.

[Example] Hereinafter, an example of the present invention will be described with reference to FIGS. 1 and 2.

In this example, by using two fault detection methods,
The purpose is to quickly discover faults.

First, a method for detecting failures when loop detection points are not used will be explained. In FIG. 1, a timer reset instruction is inserted at the beginning of the execution program, and timer 2 is reset when the execution program starts operating. The timer 2 starts operating when the execution program is started, and ends when the execution program ends. The timer 2 measures the execution program operation time, and compares and judges the program operation end time set in advance for each processing device operation with the value of the timer 2. If the value of the timer 2 is within a predetermined value, it is normal, and as the execution program starts and ends, the tracer 3 that starts and ends acquires trace information such as the address of the processing bag @1. If the timer value is larger than a predetermined value, it is regarded as an abnormality such as an abnormal loop in the execution program, a downtime, or a runaway of the processing device 1, and is treated as a failure.

Next, we will explain how to find faults when using loop detection points. The presence or absence of a loop detection point 6 set in advance at an arbitrary location such as processing is detected. When the loop detection point 6 is detected, the loop counter 4 is counted and amplified, and the value of the loop counter 4 is compared with a predetermined number of loops set in advance.

The loop counter 4 is reset each time a new loop detection point is detected by a means for resetting the loop counter 4 in the tracer 3. If the value of the loop counter 4 is greater than the predetermined number of loops, it is assumed that an abnormal loop is occurring in the execution program and a failure occurs.

When a failure is detected, an alarm sounds and trace information at the time of the failure is outputted by an output device 5 such as a display printer.

The flowchart in FIG. 2 shows the fault discovery process in the fault detection method. First, reset timer 2,
6. Compare the value of timer 2 with a predetermined value.If the value of timer 2 is larger than the predetermined value, it is determined as a failure, but if it is within the predetermined value, trace information is acquired by tracer 4. Next, detect the loop detection point 6, and if it is not detected, try again.
The value of timer 2 is compared with a predetermined value. If detected.

It is compared whether it is the same as the previously detected loop detection point, and if it is the same, the loop counter 4 is counted up, and if it is not the same, the loop counter 4 is reset and then counted up. Furthermore, the value of the loop counter 4 is compared with a predetermined value.

If the value of the loop counter 4 is larger than a predetermined value, it is determined as a failure, but if it is within the predetermined value, the value of the timer 2 and the predetermined value are compared again. When a failure is detected, the output device 5 outputs an output.

〔Effect of the invention〕

According to the present invention, in debugging an execution program,
Since hardware and software failures can be quickly discovered and output, there is an effect of improving debugging efficiency.

[Brief explanation of the drawing]

FIG. 1 shows a block diagram of an embodiment of the invention, and FIG. 2 shows a flowchart of the invention. 1... Processing device, 2... Timer, 3... Tracer, 4... Loop counter, 5... Output device, 6...
・Loop detection point.

Claims (1)

[Claims] 1. In a system that performs predetermined processing using an execution program, a timer that is reset by a reset command inserted at the beginning of the execution program, a tracer that traces addresses, etc. in a processing device, and an optional a loop detection point for detecting a program loop set in the tracer, and means for detecting the loop detection point provided in the tracer;
means for determining whether the loop detection point is the same as the previously detected loop detection point; and means for resetting a loop counter if the loop detection point is a new loop detection point; a loop counter that counts up each time the loop detection point is found within a predetermined time; and failure detection. 1. A failure detection method characterized in that a device for outputting trace information is provided.