JPS59221753A - Subroutine tracer - Google Patents

Abstract

PURPOSE:To shorten the time needed for debugging by collecting the parameters which are delivered when the control is shifted to a slave subroutine from a master subroutine. CONSTITUTION:A subroutine name (block A1) which called out a trace routine and its master subroutine name (block A2) are collected together with the contents (block A3) of the parameter which is shifted to the slave subroutine from the master subroutine. Then the subroutine name collected by the block A1 is collated with the trace subroutine name in trace management table TMT to decide a trace designated subroutine. If the coincidence is obtained from said collation, the subroutine name collected by the block A1 is collated with the break subroutine name in the table TMT. If a break subroutine name is decided, it is decided whether the contents of the parameters collected by the block A3 are coincident with the break conditions set at the TMT. Then the processing is forcibly discontinued when the coincidence is obtained.

Description

[Detailed Description of the Invention] [Field of Application of the Invention] The present invention relates to a tray length for collecting information that allows easy understanding of the program operation sequence of a data processing system. This invention relates to a subroutine tracer suitable for the system.

[Background of the invention]

It is known that logical errors in programs are propagated through data (parameters) passed between subroutines and common areas (commons) between subroutines. Therefore, in order to identify the problem when a program malfunctions, it is a good idea to insert logic (for example, a WRITE statement) to print this data in each subroutine when creating the program. This is the quickest route. However, this method has problems such as requiring a great deal of labor and increasing the amount of printing when the scale of the program increases.

Therefore, in the past, a method was often used to insert temporary logic in order to investigate the cause of a malfunction in a program after it occurred, which was one of the causes of reduced debugging and maintenance efficiency. .

[Purpose of the invention]

SUMMARY OF THE INVENTION An object of the present invention is to provide a subroutine tracer that collects TFF information that can grasp the operating sequence of a program in a data processing system, and facilitates finding the cause of a program malfunction.

[Summary of the invention]

The main point of the present invention is to provide a trace function for each subroutine using a simple procedure, and to provide flexibility that allows all functions such as specifying a subroutine to collect information, specifying a collected information output device, and setting break subroutines and break conditions. It's in the subroutine tracer.

[Embodiments of the invention]

An embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 shows a configuration diagram of a flow chart automatic drawing system according to an embodiment of the present invention. This system 2 is composed of two input analyzers, a logic analyzer 2B, and a layout drawing device 2C, and automatically converts a source program 1 into a flownayat 3 as input data.

FIG. 2 shows an aspect of the program built into each device 2A, 2B, and 2C. In this figure, the main program M
This shows that the subroutine S1 is called within the subroutine S1, and the subroutines S2 and S3 are called from within the subroutine S1. Now, if there is a logic error in the program latent at point P in subroutine SL, it will be propagated to subroutine 82.83 in a different form, and the defect will be exposed at point Q in M of the main program, for example. I will do it. In general, the further apart the potential location of the defect is from the exposed location, the more difficult it becomes to trace the cause. Therefore, the present invention attempts to isolate the problem in the program by collecting all data (parameters) that are passed when control is transferred from a parent subroutine to a child subroutine.

FIG. 3 shows how to use the subroutine tracer and outlines the processing of the trace routine. Does this mean that by simply calling the trace routine at the entrance of each leave routine (no parameters required), it is possible to collect information useful for understanding the program's operating sequence and running environment? Show - [
There is.

In the trace routine, the name of the subroutine that called the trace routine (block AI) and the name of its parent subroutine (block A2) =, (collected, and the contents of parameters passed from the parent to the child subroutine (block A3) ) to collect.

That is, it collects which subroutine (parent) to which subroutine (child) control was transferred, and what values the parameters were at that time. Note that this data can be collected by using addresses in the register save area used by the monitor during program execution, similar to the method used by traceback utilities for large computers, and using known techniques. It is.

The subroutine tracer has the following characteristics,
TMT (TraCe Management) shown in FIG.
-ment Ta1)Ie), perform the following processing in FIG.

(1) Subroutine name collected in block A1,
Check the trace subroutine name in the TMT table and determine whether it is a trace specified subroutine. (Block Bl) (2) As a result of (1) above, if it is a trace specification subroutine, edit the E-J& collected in blocks A1 to A3 into a visually easy-to-read format, and use the output device set in the TMT table. Output to. (Block B2) (3) Next, the subroutine name e collected in block A1 is matched with the break subroutine name in the TMT table to determine whether it is a break specified subroutine or not.

(Block CI) (4) As a result of (3) above, if it is a break specification subroutine, it is further determined whether or not the parameter contents collected in block A3 match the break conditions set in the TMT table.

(Block C2) (5) As a result of (3) and (4) above, if all break conditions are met, the process is forcibly stopped. (Block C3
) Here, information can be added to the TMT table arbitrarily before program execution, so depending on the progress of debugging,
If you change the output device to a line printer or disk,
You can perform flexible debugging by increasing or decreasing the number of traced subroutines, or by setting break conditions to immediately stop processing when a presumed problem occurs.

Further, after the program is completed, by making the trace routine a dummy, it is possible to almost completely eliminate any impact on the execution time.

An example of an actual subroutine trace output and a time chart are shown in FIG.

According to an embodiment of the present invention, conventionally, WRITE is used to collect data for investigating the cause of a problem after it occurs.
From reactive debugging such as inserting statements, to proactive debugging such as constantly storing the program's running trajectory and running environment on disk using simple procedures, and extracting and analyzing the data if a problem occurs. Since it can be converted, it greatly contributes to improving the efficiency of program development and maintenance.

〔Effect of the invention〕

According to the present invention, it is possible to collect information that allows easy understanding of the complex operation sequence of a program in a data processing system and its running environment, which is effective in finding the cause when a program malfunction occurs.

[Brief explanation of the drawing]

Fig. 1 is a configuration diagram of an automatic flowchart drawing system according to an embodiment of the present invention, Fig. 2 is a diagram of the program built in each device in Fig. 1, and Fig. 3 is a procedure for using the subroutine tracer of the present invention. 4 is a content diagram of the TMT table used in the subroutine tracer, and FIG. 5 is an example of the subroutine tracer output (a) and a time chart (b). 1... Source program, 2... Flowchart automatic glue 1 Kuchikurei 2 fig. 4 Ku Me 50 (0-) Ax I CALLl: DBY (M) 38 CAL
LSr(E, t2o, o, -to)ax' 2 C/
ILLE()BY(Sl 74 CALL
δZ(/,!;,o)*-2CALLEE)BY(
Sl) E2 CALL δ3 Ku-
4, 2, 3) (Jn (-4・'・5);King's second turn -1

Claims (1)

[Claims] 1. A data processing system that organically combines many subroutines with a single function to execute a target process, characterized in that each of the subroutines has a tracing function. Subroutine tracer.