JP2010237743A - System, method and program for detecting problematic point of data conflict - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To detect a point to be corrected from a log at the time of data conflict. <P>SOLUTION: In the system for detecting a problematic point of data conflict of a program to be inspected, an information processor includes: a first log storage means which stores a log at the time of data conflict caused by the program to be inspected which executes parallel processing of processes or threads, as a first log; an individual execution means which individually and successively executes the processes or threads of the program to be inspected; a second log storage means which stores a log of the result of the individual execution means, as a second log; and a difference extraction means which compares the first log with the second log to extract a difference between the logs. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  TECHNICAL FIELD The present invention relates to a data conflict problem location detection system, a problem location detection method, and a problem location detection program, and in particular, a problem location detection system, a problem location detection method, and a problem location detection method capable of detecting a location to be corrected from a log when a problem occurs. The present invention relates to a problem location detection program.

  In an information processing system that operates a plurality of processes and threads in parallel or in parallel, a problem of data contention occurs unless exclusive control is performed correctly.

  When a data race failure is discovered during software testing, simulation, or operation, the problem location of a program is generally identified by manually reading a log. Here, the problem part of the program is a part in the target program (target source code) where a problem is caused by a description different from the developer's expectation, and the problem does not occur by correcting the problem part. However, in the case of problems that occur due to the influence of multiple processes and threads, the logs of each process and thread are output together, so knowledge and experience are required to analyze the logs and identify the problem location. It took a long time to analyze.

  An example of a related program verification system is described in Patent Document 1. As shown in FIG. 13, the program verification system described in Patent Literature 1 includes a change detection unit, a source code classification unit, a model generation unit, and a verification unit. The system having such a configuration operates as follows. In other words, the function that has been changed and the function that uses the shared variable that is affected by the change are detected, and the part that is affected by the change, the control structure part, and other parts are detected, and based on the detection results. To generate a model. Thereafter, the verification target program after the change is verified by comparing the model of the verification target program before the change with the model of the verification target program after the change.

  However, in the verification system described in Patent Document 1, when a problem is found in the verification, the execution order in which the problem occurs can be output, but the work of reading the log and identifying the actual problem part is not It is left to humans. This required knowledge and experience and took time to analyze.

JP 2007-11690 A

  In the verification system described in Patent Document 1, it is necessary for an operator having a lot of experience and knowledge to grasp the operation of a plurality of process threads and search for a problem location from a large number of logs.

  In the case of a data race defect caused by the influence of a plurality of processes and threads, the logs of the respective processes and threads are output in a mixed manner, so that the log cannot be easily analyzed. That is, as a problem, regarding log analysis of data competition in parallel computing, it takes a great deal of knowledge and experience to specify a problem part, and a lot of man-hours are required for the analysis.

  An object of the present invention is to provide a problem location detection system, a problem location detection method, and a problem location detection program for detecting a problem location of a data conflict to be corrected using a data conflict log.

  A system for detecting a problem location of data conflict in a program to be inspected according to the present invention includes a first log that stores, as a first log, a log when a data conflict occurs by executing a program to be inspected that executes processes or threads in parallel. A storage unit, an individual execution unit that sequentially and sequentially executes a process or a thread of the inspection target program, a second log storage unit that stores a log of an execution result of the individual execution unit as a second log, A difference extraction unit that compares the first log with the second log and extracts a difference between the logs is provided.

  According to another aspect of the present invention, there is provided a method for detecting a problem location of a data conflict in a test target program, wherein the information processing apparatus stores a test target program that is a data conflict when executing processes or threads in parallel. A log at the time of contention is stored as a first log, the process or thread of the program to be inspected is individually and sequentially executed, a log of an execution result is stored as a second log, and an information processing apparatus is used. Then, the first log and the second log are compared to extract a log difference, and the difference part of the extracted log is presented to the operator so that the difference can be recognized.

  Further, the data conflict problem location detection program of the inspection target program of the present invention is a log when data conflict occurs by executing the inspection target program for executing the process or thread in parallel with the control unit of the information processing apparatus, First log storage means for storing as a first log, individual execution means for individually and sequentially executing processes or threads of the inspection target program, and a log of execution results of the individual execution means as a second log It functions as a second log storing means for storing, and a difference extracting means for comparing the first log and the second log to extract a log difference.

  According to the present invention, it is possible to provide a problem location detection system, a problem location detection method, and a problem location detection program for detecting a problem location of data contention to be corrected from a data conflict log.

It is a block diagram which shows the structure of the 1st Embodiment of this invention. It is a flowchart which shows the operation | movement of the 1st Embodiment of this invention. It is a block diagram which shows the structure of the 2nd Embodiment of this invention. It is a flowchart which shows the operation | movement of the 2nd Embodiment of this invention. It is a block diagram which shows the structure of the 3rd Embodiment of this invention. It is a flowchart which shows the operation | movement of the 3rd Embodiment of this invention. It is the block diagram which showed the structure of the problem location detection system in an Example. It is explanatory drawing which illustrates the control | contrast program in an Example. It is explanatory drawing which illustrates the log at the time of the malfunction occurrence in an Example. It is explanatory drawing which illustrates the sequential execution result of the thread | sled in an Example. It is explanatory drawing which illustrates the sequential execution result of the thread | sled in an Example. It is explanatory drawing which illustrates the result of having extracted the location different from a 2nd log from a 1st log. It is a block diagram which shows the program verification system described in patent document 1.

  Embodiments of the present invention will be described in detail with reference to FIGS.

  Referring to FIG. 1, the first embodiment of the present invention includes a target program storage means 110 for storing a target source code, and a program individual execution means 120 capable of individually executing a plurality of process threads. A first log storage means 130 for storing a first log when process threads are executed in parallel; a second log storage means 140 for storing a second log when processes are individually executed; , And a difference extracting means 150 for extracting a difference by comparing the first log and the second log.

Each of these means generally operates as follows.
The program individual execution means 120 acquires the source code to be inspected from the target program storage means 110, and individually decomposes (replaces and supplements) the process threads described in the source code, Run threads sequentially. The order in which each is sequentially executed may be determined in advance, or may be specified by an operator based on experience or the like. The log obtained at the time of execution is stored in the second log storage unit 140 as a log of the second log.

The difference extraction means 150 compares the first log and the second log to extract and store a difference portion. The stored difference portion is presented to the operator so as to be recognized as a candidate for a data conflict problem portion assumed from the difference portion. The difference portion may be presented by the difference extraction means 150 or by other presentation means.
The difference extraction means 150 includes, as necessary, a process or thread corresponding to the extracted difference portion, a program specifying means for specifying a corresponding position of the process or thread corresponding to the extracted difference portion, and an extracted difference portion. Source code specifying means for specifying the corresponding source code and the corresponding position on the source code corresponding to the extracted difference portion may be provided.

  The content presented to the worker corresponds to the log of the difference part where the progress and result differ between the first log and the second log, the process corresponding to the log of the difference part, the corresponding position of the process, and the log of the difference part Threads and corresponding positions of threads, processes or threads of differences, or corresponding positions of source codes and source codes corresponding to logs or logs can be displayed in a desired display format, and one or more results can be switched sequentially Present.

Next, the overall operation of the present embodiment will be described in detail with reference to the flowcharts of FIGS.
It is assumed that the target source code is stored in advance in the target program storage unit 110, and the log when a failure occurs when processes are executed in parallel (in parallel) is stored in the first log storage unit. . Here, the collected log includes the process ID, the line number of the source code corresponding to the part being executed, the value of the variable, and the like. In addition, when the same source code is operated a plurality of times in the same execution order, the same log is output. For this reason, information such as an indeterminate intermediate state and time is not included in the log, or is excluded from the difference extraction by the difference extraction means 150.

First, the program individual execution means 120 acquires the target source code from the target program storage means 110 (step A1 in FIG. 2).
Next, the program individual execution unit 120 individually executes the process and the thread, and stores the obtained log in the second log storage unit 140 (step A2).
Next, the difference extraction means 150 reads the first log and the second log (step A3). Next, the difference extraction means 150 compares the first log and the second log and extracts a difference (step A4).
Finally, the difference extraction means 150 presents the candidate of the problem part assumed from the obtained difference part to an operator (step A5).
The presentation content at this time is related to the source code of the process corresponding to the difference part, whether it is the difference part of either the first log or the second log, or the process corresponding to the difference part. It may be information.

  Next, the effect of this embodiment will be described. In the present embodiment, the first log obtained when the processes are executed in parallel (in parallel) and the second log obtained when the processes are executed individually are read to extract a difference, which can be used for finding a problem location. Since the information is presented, the log can be confirmed centering on the location, and even a worker with little knowledge or experience can effectively find the problematic location.

  Next, a second embodiment of the present invention will be described in detail with reference to the drawings.

  The second embodiment of the present invention is different from the first embodiment described above in that it includes a data flow analysis means for analyzing the flow of information by analyzing the target program. Specifically, referring to FIG. 3, the second embodiment of the present invention is similar to the first embodiment in the target program storage means 110, the program individual execution means 120, and the first log storage. Means 130, second log storage means 140, difference extraction means 150, and data flow analysis means 160 for analyzing the flow of information by analyzing the source code data flow.

Each of these means generally operates as follows.
It is assumed that the target source code is stored in advance in the target program storage unit 110, and the log when a failure occurs by executing a process thread in parallel is stored in the first log storage unit. The program individual execution means 120 acquires the source code from the target program storage means 110 and individually executes the process thread described in the source code. The log obtained at the time of execution is stored in the second log storage means 140.

  The data flow analyzing unit 160 acquires the inspection target program (source code) from the target program storage unit 110, and whether or not the variable / function (information) included in the inspection target program affects the final result of executing the process / thread. Is checked by data flow analysis of the source code.

  The difference extraction unit 150 compares the first log and the second log to extract the difference, selects a candidate problem part from the obtained difference part, and based on the result of the data flow analysis Excludes candidates for variables that do not affect the final results that were executed, and presents candidate candidates for problems to the operator.

Next, the overall operation of the present embodiment will be described in detail with reference to the flowcharts of FIGS.
First, the individual program execution unit 120 acquires the target source code from the target program storage unit 110 (step B1 in FIG. 4), executes the program individually, and stores the obtained log in the second log storage unit 140. Store (step B2).
The data flow analysis means 160 acquires the target source code from the target program storage means 110, and checks whether or not a predetermined variable affects the final result executed by the data flow analysis of the source code (step B3).
Next, the difference extraction means 150 reads the first log, the second log, and the data flow analysis result (step B4), compares the first log and the second log, and extracts the difference ( Step B5). At this time, using the result of the data flow analysis, a variable that does not affect the final result executed is not recognized as a difference.

  Finally, the difference extraction means 150 presents the candidate of the problem part assumed from the obtained difference part to an operator (step B6).

  Next, the effect of this embodiment will be described. In the present embodiment, in addition to the effects of the first embodiment, erroneous detection of a problem location can be reduced by using the result of data flow analysis for presenting a problem location candidate. Even if this does not affect the final result, the first log and the second log may take different values temporarily during the calculation. In this embodiment, such a difference is detected from the detection target. This is because it is excluded.

  Next, a third embodiment of the present invention will be described in detail with reference to the drawings. The third embodiment of the present invention is different from the first embodiment described above in that it includes an execution order determining unit 170 that determines an execution order for individually executing process threads. Specifically, referring to FIG. 5, the third embodiment is similar to the first embodiment in target program storage means 110, individual program execution means 120, first log storage means 130, 2 log storage means 140, difference extraction means 150, and execution order determination means 170 for determining the execution order for individually executing processes.

Each of these means generally operates as follows.
The execution order determination unit 170 acquires the source code from the target program storage unit 110 and determines all execution orders that can be executed by the process thread.
As an example, the execution order determination unit 170 identifies all the process threads that can be executed and passes them to the program individual execution unit 120. For example, when there are three processes X, Y, and Z, “X → Y → Z” “X → Z → Y” “Y → X → Z” “Y → Z → X” “ All the six types of process execution order of “Z → X → Y” and “Z → Y → X” are recognized and generated as information related to the execution order as information that can be notified to the individual program execution means 120.
In another example, the operator inputs a predetermined work time (analysis time), and determines the execution order of as many combinations as possible that can execute all the processes including the processes described later within that time. Information regarding the execution order may be used.

The individual program execution unit 120 acquires the source code from the target program storage unit 110, acquires the execution order (information on the execution order) determined from the execution order determination unit 170, and is described in the source code according to the extracted execution order. Process threads are executed individually and serially. The log obtained at the time of execution is stored in the second log storage means 140.
The difference extraction means 150 compares the first log and the second log, extracts the difference, and displays a candidate for a problem location assumed from the obtained difference location.

Next, the overall operation of the present embodiment will be described in detail with reference to the flowcharts of FIGS.
First, the execution order determination unit 170 acquires the source code from the target program storage unit 110 and determines the execution order of process threads (step C1 in FIG. 6).

The individual program execution means 120 acquires the target source code from the target program storage means 110, and is described in the source code until all of the instructed execution orders are completed according to the execution order received from the execution order determination means 170. The obtained process is individually executed, and the obtained log is stored in the second log storage means 140 (step C2).
Next, the difference extraction means 150 reads the first log and the second log (step C3), compares the first log and the second log, compares them, and extracts the difference (step C4).

  Finally, the difference extraction unit 150 displays a candidate for a problem portion assumed from the obtained difference portion (step C5).

  Next, the effect of this embodiment will be described. In the present embodiment, in addition to the effects of the first embodiment, in the present embodiment, the execution order is determined by the execution order determination unit 170 and the process threads are sequentially executed individually. False detection can be reduced. This is because the intermediate state may differ depending on the context of execution of the process, and both such intermediate states are considered to be normal values. Further, by extracting only the log different from the result (second log) executed in any execution order from the first log as a candidate for a problem location, a highly accurate result can be obtained.

Next, the first embodiment will be described using a specific example.
In the embodiment, as shown in FIG. 7, a system that detects a problem portion of a program related to data conflict can be realized using a general information processing apparatus (computer) 200. In the problem location detection system shown in FIG. 7, various programs stored in the auxiliary storage device 210 are expanded in the RAM and read into the control unit, whereby the target program 214 stored in the auxiliary storage device 210 is read. Operates as a system to detect problem areas. The control unit functions as a program individual execution unit, a difference extraction unit, a program identification unit, a source code identification unit, and the like based on various programs read into the RAM. The auxiliary storage device 210 and the RAM function as target program storage means, first log storage means, and second log storage means. The program may be configured so that the control unit functions as a data flow analysis unit and an execution order determination unit.

  Here, the target program 214 is a program in which a process or thread operates in parallel or in parallel, in which a data race failure is discovered by an operator during software testing, simulation, or operation. It is a program that requires correction. The target program 214 has many child source codes under the parent source code, and based on each source code, the control unit of the information processing apparatus generates processes and threads and performs arithmetic processing.

The information processing apparatus 200 that operates as the problem location detection system stores the first log and the second log based on the operator's instruction (selection of the target program 214, etc.) input via the input unit or the network interface. Store and extract log differences and output to the display unit information identified from the log differences that would be important to the operator.
In other words, the information processing apparatus 200 stores the difference analysis program 211 recorded in the auxiliary storage device 210 to operate the control unit as the difference extraction unit, the program identification unit, and the source code identification unit, and the log difference obtained by extracting the control unit By operating based on the result display program 212 that operates as a presentation means for displaying the location on the display unit to the operator, and the log collection program 213 that operates the control unit as a program individual execution means, from among the data-competed logs, It is possible to detect the problem part of the data conflict that should be corrected.

Next, a processing example of the information processing apparatus 200 will be shown to describe the present invention.
FIG. 8 shows an example of the target source code. The target program (source code) includes a thread A and a thread B, and uses variables X, Y, and Z. In the operation of the source code, a problem relating to data conflict has occurred, and the log when the problem occurs is as shown in FIG.

Next, an operation at the time of analysis of the information processing apparatus 200 will be described.
The control unit expands and reads various programs recorded in the auxiliary storage device 210 into the RAM, causes the program individual execution means to function, and acquires the target source code from the target program (corresponding to step A1 in FIG. 2). .

  Next, the control unit causes the program individual execution means to function, and sequentially executes processes and threads individually, and stores and holds the obtained log in the auxiliary storage device as the second (corresponding to step A2). Here, when execution is performed in the order of thread A → thread B, a log as shown in FIG. 10 is obtained. Further, when the processing is executed in the order of thread B → thread A, a log as shown in FIG. 11 is obtained. In this example, the log name is a thread name on the left side of “:” and a variable value in “()”.

  Next, the control unit causes the difference extraction means to function, and reads a log (second log) when a process or thread is individually executed and a log (first log) when a failure occurs (corresponding to step A3). ) Compare two types of logs, extract the difference and store it (corresponding to step A4).

  When the log of FIG. 9 is compared with the logs of FIG. 10 and FIG. 11, the value of the variable when the X value is acquired for the second time is 100 or − in the logs of FIG. 10 and FIG. However, it is 30 when the problem occurs, and the value of the variable when the X value is updated for the second time is 70, but it is -30 when the problem occurs.

  The control unit uses the stored difference portion and displays a possible problem portion candidate on the display unit (corresponding to step A5).

FIG. 12 is an explanatory diagram illustrating the result of extracting a portion different from the second log from the first log.
In the information presented to the worker, as shown in FIG. 12, a thread and a variable that are differences are acquired and displayed from a log in which a problem has occurred. As shown in FIG. 12, it can be seen that by extracting the difference, a portion having a high possibility of a problem can be extracted from the portions where the variable is updated.
12 will be described in detail. From the result shown in FIG. 12A, there is a high possibility that there is a problem in the location where the thread B acquires / updates the variable X in the first log where the problem has occurred. From the result shown in FIG. 12 (b), it is understood that there is a high possibility that there is a problem in the location where the thread A acquires / updates the variable X in the first log where the problem has occurred. . Thus, by performing the exclusive control correctly and preventing the acquisition / update of the variable X from being executed at the same time, the problem of data contention occurring at the time of acquiring the first log can be solved. It can also be seen that the variables Y and Z do not need to be changed.

  In the present embodiment, the first embodiment has been described. However, the second and third embodiments can be similarly realized using a general information processing apparatus. In this case, the data flow analysis means and the execution order determination means may be incorporated as part of the log collection program, or may be prepared as separate programs. Further, the above-described program may be integrally created as a program for detecting a problem location of data conflict, and the creator may arbitrarily divide and combine the programs.

  As described above, according to the present invention, there is a high possibility that the difference between the log when the process or thread is executed in parallel and the log when the process is executed individually is a problem location. By checking the log, the problem part can be found at an early stage, and the trouble part can be found effectively even for an operator having little knowledge and experience.

  That is, with the above-described configuration and operation, it is possible to provide a problem location detection program and a problem location detection system for detecting a problem location of data conflict to be corrected from a log, and to realize a problem location detection method.

  In the above description, the data conflict detection method is described using the coefficient (function). However, the data conflict is not limited to the coefficient (function). For example, the present invention can be applied even when data contention is contention for a memory resource, a shared memory file, or the like.

  In addition, this invention is not limited to the said embodiment and Example. The present invention achieves the same effect even if the configuration or operation is changed without departing from the gist of the invention.

  INDUSTRIAL APPLICABILITY According to the present invention, when a data conflict problem occurs, the present invention can be applied to the use of finding and correcting a problem location. Further, the present invention can be applied to an application such as a test result display function of a tool that detects a data conflict problem.

110 Target Program Storage Means 120 Program Individual Execution Means 130 First Log Storage Means 140 Second Log Storage Means 150 Difference Extraction Means 160 Data Flow Analysis Means 170 Execution Order Determination Means 200 Information Processing Device 210 Auxiliary Storage Device 211 Difference Analysis Program (Difference extraction means)
212 Result display program (presentation means)
213 Log collection program (individual execution means)
214 Target program (source code)

Claims (18)

A first log storage unit that stores a log when a data conflict occurs by executing a test target program that executes processes or threads in parallel;
Individual execution means for individually and sequentially executing processes or threads of the inspection target program;
Second log storage means for storing a log of the execution result of the individual execution means as a second log;
A system for detecting a problem portion of a data conflict of a program to be inspected, comprising: difference extraction means for extracting a difference between logs by comparing the first log and the second log.   The said difference extraction means has a program specification means which specifies the process or thread | sled corresponding to the extracted difference location, and the applicable position of the said process or thread | sled corresponding to the extracted difference location. Problem detection system.   The said difference extraction means has a source code specific | specification means which specifies the source code corresponding to the extracted difference location, and the applicable position on the source code corresponding to the extracted difference location. The problem location detection system described in 1. Data flow analysis is performed on the program to be inspected, and further includes data flow analysis means for analyzing a variable information passing relationship when a process or thread is executed,
When inspecting the problem part of the inspection target program,
Analyzing whether or not it affects the final result of executing a process or thread for each variable information of the program to be inspected,
The problem point detection system according to any one of claims 1 to 3, wherein a difference relating to variable information that does not affect the final result is excluded from the extracted difference based on the analysis result. Execution order determining means for deriving a combination of a plurality of orders for executing processes or threads;
The process or thread described in the program to be inspected is sequentially executed based on a plurality of execution orders derived by the execution order determination unit, and the execution result is stored in the second log storage unit. The problem point detection system according to any one of claims 1 to 4. It has a presentation means to present the difference part of the extracted log to the worker so that it can be recognized,
The difference part of the log includes a log having a different process and result, a process corresponding to the log and a corresponding position of the process, a thread corresponding to the log and a corresponding position of the thread, the process or the thread or a source corresponding to the log. The problem location detection system according to any one of claims 1 to 5, wherein the system is one or a combination of corresponding positions of a code and a source code. In the information processing apparatus, a test target program that is a data conflicting program when executing processes or threads in parallel is stored,
A log when data conflict of the inspection target program is stored as a first log,
The process or thread of the program to be inspected is sequentially executed separately, and a log of the execution result is stored as a second log,
Using the information processing apparatus, the first log and the second log are compared to extract a log difference,
A method for detecting a problem portion of data conflict of a program to be inspected, characterized in that a difference portion of an extracted log is presented to a worker so as to be recognized. From the log difference extraction results, identify the process or thread corresponding to the extracted difference location, and the corresponding position of the process or thread corresponding to the extracted difference location,
The problem location detection method according to claim 7, wherein the identified result is presented to an operator in a recognizable manner. From the extraction result of the log difference, identify the source code corresponding to the extracted difference location, and the corresponding position on the source code corresponding to the extracted difference location,
The problem location detection method according to claim 7 or 8, wherein the identified result is presented to an operator in a recognizable manner. Analyzing data flow of the program to be inspected using an information processing device, analyzing the passing relationship of variable information when executing a process or thread,
Analyzing whether or not it affects the final result of executing a process or thread for each variable information of the program to be inspected,
10. The problem location detection method according to claim 7, wherein differences relating to variable information that does not act on the final result are excluded from the extracted differences based on the analysis results. Deriving a combination of multiple sequences of executing processes or threads,
The process or thread described in the inspection target program is sequentially executed based on the plurality of derived execution orders, and the execution result is stored as the second log. The method for detecting a problem location according to any one of the above. The log difference portion is presented to the worker by a log having a different process and result, a process corresponding to the log and a corresponding position of the process, a thread corresponding to the log and a corresponding position of the thread, the process or the thread or the log. 12. The problem location detection method according to claim 7, wherein the source code corresponds to any one of or a combination of the corresponding positions of the source code. The control unit of the information processing device
A first log storage unit that stores a log when a data conflict occurs by executing a test target program that executes processes or threads in parallel;
Individual execution means for individually and sequentially executing processes or threads of the inspection target program;
Second log storage means for storing a log of the execution result of the individual execution means as a second log;
A program for detecting a problem location of a data conflict in a program to be inspected, which functions as a difference extraction unit that compares the first log and the second log and extracts a difference between the logs.   The said difference extraction means has a program specification means which specifies the process or thread | sled corresponding to the extracted difference part, and the applicable position of the said process or thread | sled corresponding to the extracted difference part. Problem location detection program.   The said difference extraction means has a source code specific | specification means which specifies the applicable position on the source code corresponding to the extracted difference location, and the source code corresponding to the extracted difference location, characterized by the above-mentioned. The problem location detection program described in 1. The control unit performs a data flow analysis of the program to be inspected and functions as a data flow analysis unit that analyzes a variable information passing relationship when a process or thread is executed,
When inspecting the problem part of the inspection target program,
Analyzing whether or not it affects the final result of executing a process or thread for each variable information of the program to be inspected,
16. The problem location detection program according to claim 13, wherein differences relating to variable information that does not act on the final result are excluded from the extracted differences based on the analysis result. Causing the control unit to function as an execution order determination unit that derives a combination of a plurality of orders for executing processes or threads;
The process or thread described in the program to be inspected is sequentially executed based on a plurality of execution orders derived by the execution order determination unit, and the execution result is stored in the second log storage unit. The problem part detection program according to any one of claims 13 to 16. The control unit functions as a presentation means for presenting the difference portion of the extracted log so as to be recognizable to the worker,
The difference part of the log includes a log having a different process and result, a process corresponding to the log and a corresponding position of the process, a thread corresponding to the log and a corresponding position of the thread, the process or the thread or a source corresponding to the log. The problem location detection program according to any one of claims 13 to 17, wherein the program is a corresponding position or a combination of a code and a source code.

Images