JPH04169941A - Debugging information output processing system - Google Patents

Abstract

PURPOSE:To execute the processing at a high speed by generating a call instruction to an interrupt function processing part for debugging in the source level before object information of each statement in a source program. CONSTITUTION:A C language processing program 202 consists of an input part 204 which inputs source program information from an input file 201, a syntax analyzing part 205 which analyzes source program information, and an object output part 206 which generates and outputs an object program of the analyzed source program. Source program information is inputted from the input file 201 and is analyzed to generate object information, and object information including debugging information is outputted to a prescribed object file, and the call instruction to a prescribed debugging processing part is added before object information to an interrupt function in the source program included in object information. Thus, the debugging operation is simplified to shorten the operation time for execution.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a debug information output processing method, and particularly relates to a debug information output processing method for a language processing program when performing source level debugging.

[Conventional technology]

There are two conventional debug information output processing methods for language processing programs. One is to output information that is directly associated with the source program, such as line number information, and the other is to output information that is directly associated with the source program, such as a call instruction that calls a routine that checks whether the stack is available. , is a method of outputting information generated as machine language instructions at the beginning of a function.

Information that is different from machine language, such as the former line number information that is information for source level debugging, is used by the debugger when performing source level debugging, such as when step execution is performed line by line. used for.

The information generated as machine language instructions, such as the latter call instruction that calls a routine that checks whether the stack is available at the beginning of a function, is useful when an error may occur at runtime or when an error occurs. It is used to call special processing routines, output messages, handle errors, etc.

Below, a method for debugging an interrupt function among conventional source-level debugging will be explained.

Interrupt functions are used regardless of the flow of the source program.
This relationship is called when a hardware interrupt or software interrupt occurs. For this reason, in step execution, which is one of the normal methods of debugging functions,
Since the timing is different, debugging is not possible, so follow the steps below to debug.

(1) Set breakpoints in the entry sections of all interrupt functions.

(2) An interrupt occurs and execution is interrupted at the beginning of the interrupt function.

(3) Perform step execution of the interrupt function and debug it.

Currently, the number of interrupt sources has increased due to the enhancement of Tepais' functionality, so the number of interrupt functions has also increased accordingly. In addition, due to the increase in the amount of communication with the in-circuit emulator for setting breakpoints in the entry sections of all interrupt functions, the response time for commands for Tebacker users has increased, and the response time for commands has become faster. There is a demand for change.

[Problem to be solved by the invention]

In the conventional language processing program debug information output processing method described above, when debugging an interrupt function, breakpoints must be set at the entry portions of all interrupt functions. Furthermore, since the flow of program operations in interrupt functions is not continuous, branch destinations cannot be predicted from the source program, and the flow of program operations cannot be followed. It is also possible that the program will end at the branch point, and you will have to start debugging from the beginning again.

Therefore, the conventional debugging information output processing method for language processing programs requires extra time for debugging.
The drawback is that it takes a lot of time to respond during execution.

[Means to solve the problem]

A debug information output processing method of the present invention is a debug information output processing method for a language processing program, which includes means for inputting source program information from a predetermined file, and generating object information by analyzing the input source program information. means for outputting the object information including debug information to a predetermined object file; and a means for outputting the object information including debug information to a predetermined object file; and means for adding.

〔Example〕

Next, the present invention will be explained with reference to the drawings.

FIG. 1 is a flowchart of an object output section of a C language processing program in an embodiment of the present invention, and FIG. 2 is a system configuration diagram of the embodiment.

In FIG. 2, the C-language processing program 202 includes an input unit 204 that inputs source program information from an input file 201, a parsing unit 205 that parses the input source block information, and a parsed source program. It is composed of an object output unit 206 that generates and outputs an object program.

FIG. 3 shows the contents of the input foil 201. The numbers on the left are line numbers in the source program for explanation purposes.

FIG. 4 shows an object file 2 according to an embodiment of the present invention.
03 is an information explanatory diagram showing the contents (object program).

The overall processing flow will be explained with reference to FIG. Input section 20
4 inputs program information from the input file 201. Next, a syntax analysis unit 205 analyzes the source program. Finally, the object output unit 206 generates an object program and outputs it to the output file 203.

The flowchart shown in FIG. 1 will now be explained.

However, the explanation will focus on functions and statements that are objects of object output of a C language processing program.

(1) One line of the input file 201 is analyzed by source program analysis processing (step 101).

(2) As a result of the processing in (1), it is determined whether the object is to be output. If one line of the source program is the start of a declaration or block, it is not an object to be output, so the process returns to step 101 and the next ' line is analyzed (steps 1 and 2).

(3) In step 103, it is determined whether the source program has ended or not. If the source program has ended, the process moves to end processing (step 104) and ends. If the source program has not ended, the process moves to the next step 105.

(4) In step 105, as a result of the processing in step 101, if it is the beginning of an interrupt function, the processing starts with the call instruction output processing of the interrupt open number processing section (step 10).
Move on to 6). In the interrupt function processing section call instruction output process (step 106), a call instruction for the interrupt function processing section is output.

(5) After outputting the call instruction for the interrupt open number processing unit, the process moves to machine language instruction output processing for the statement within the function (step 107). In the machine language instruction output process for a statement within a function (step 1o7), a machine language instruction for a statement within a function is output.

Next, a process in which the C language processing program 202 of FIG. 2 inputs the source program shown in FIG. 3 and outputs the object program shown in FIG. 4 will be explained according to the flowchart of FIG. 1. In the object program shown in FIG. 4, only the overall configuration is described for the sake of brevity. Also,
The entire process from the input of source program information to the output of the object program is the same as the description of the process flow above, so the explanation will be omitted.

(1) Source program analysis processing (step 101)
The first line of FIG. 3 is analyzed by: Since no object is generated for this line in the object output determination process (step 102), the process moves to the source program analysis process (step 101) and analyzes the second line.

(2) The second line is object output determination processing (step 1
02>, it is determined that the object is to be output, and furthermore, in the end determination process (step 103), it is determined that the source program has not ended, so the process moves to the interrupt function determination process (step 105).

(3) In the function determination process (step 105), it is determined that the second line is not an interrupt function, so the process moves to machine language output processing for the statement within the function (step 1.7).

(4) Machine language output processing for statements in functions (step 1
07), the machine language for lines 4 to 6 in FIG. 3 is output. These are a function preprocessing section 402, a machine language instruction 403 for a statement within the function, a function postprocessing section 404, and a return instruction 405 in FIG.

(5> Processing is source program analysis processing (step 1)
01) and analyze the 7th line in FIG. Since no object is generated for this line in the object output determination process (step 102), the process moves to the source program analysis process (step 101>) and analyzes the 8th line.

(6) The 8th line is the object appearance determination process (step 1
02), it is determined that the object is to be output, and furthermore, in the end determination process (step 103), it is determined that the source program has not ended, so the process moves to the processing relationship determination process (step 1o5).

(7) In the function determination process (step 1o5), the 8th line is determined to be the beginning of the interrupt function, so the call instruction output process (step 106) of the interrupt function processing unit
Outputs a call instruction for the interrupt function processing section. This is the call instruction 401 of the interrupt-related processing section shown in FIG.

(8) Next, the machine language pressure processing for the sentence within the function (step 107) is performed to output the machine language for the sentence. This is the function preprocessing unit 402 in FIG. 4. Machine language instructions 403 for statements within the function. These are a function post-processing unit 404 and a return instruction 405.

In this way, for the source program in Figure 3,
6. The object shown in FIG. 4 is generated Next, a method for debugging a source level interrupt function will be explained using the output object program shown in FIG.

With a source debugger, you can set a breakpoint in the interrupt function processing section and run the program to be debugged.
When an interrupt occurs due to a hardware interrupt or a software interrupt and the execution of the program to be debugged is transferred to the interrupt function processing section, the operation of the program to be debugged is stopped by the source debugger. As a result, even if an interrupt function is called while the debugged program is being debugged, the execution of the debugged program will stop at the beginning of the called function, making it impossible to follow the flow of the debugged program's operations. It would be a waste if the execution of the debugged program ended in a called function and the debugging had to be restarted from the beginning.

〔Effect of the invention〕

As explained above, the present invention generates a call instruction for the interrupt function processing section for source-level debugging before the object information for each statement in the source program, so that the entry section of all interrupt functions is This not only eliminates the hassle of setting and executing breakpoints, but also enables faster execution compared to setting breakpoints at the entry sections of all interrupt functions. effective.

[Brief explanation of drawings]

FIG. 1 is a flowchart of object output processing of a C language program in an embodiment of the present invention, FIG. 2 is a system configuration diagram of the execution example, FIG. 3 is a diagram showing the contents of an input file, and FIG. It is an information explanatory diagram showing the contents of an object file in an example. In the figure, 101... Source program analysis, 1
02...Object output determination processing, 103...End determination processing, 104...End processing, 105...Relationship determination processing, 1'06...Read command output processing of interrupt function processing unit, 107. ...Machine language instruction output processing for sentences, 201...Input file, 202...C language processing, 203...Object file, 204...
Input section, 205...Syntax analysis section, 206...Object output section, 401 Interruption function processing section read instruction, 402...Function preprocessing section, 403...Machine language instruction for the statement within the function , 404...Function post-processing unit, 40
5...Return command.

Claims (1)

[Claims] In a debug information output processing method for a language processing program, means for inputting source program information from a predetermined file, means for analyzing the input source program information to generate object information, and the object including the debug information. The present invention is characterized by comprising means for outputting information to a predetermined object file, and means for adding a call instruction for a predetermined debug processing unit before the object information for the interrupt function in the source program included in the object information. Debug information output processing method.