JP2015022667A - Trace information collection device, trace information collection method, and computer program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To collect information to be used for investigating the factor of stall occurrence in a plurality of programs whose processing operates in association with each other.SOLUTION: A trace information collection device includes: a collection part 10 for collecting the trace information of a plurality of programs; a first information storage part 20 including a plurality of buffers (22 to 24) capable of storing the trace information of a first program and management information 21 for storing information necessary for access to the buffer in which the current trace information is collected among those buffers; and a second information storage part 30 including a plurality of buffers (32 to 34) capable of storing the trace information of a second program and management information 31 for storing information necessary for access to the buffer in which the current trace information is collected among those buffers. When a stall is detected, a management part 40 simultaneously switches the collection destination of the trace information from the buffer in which the current trace information is collected to the buffer different from the buffer.

Description

  The present invention relates to a trace information collection device, a trace information collection method, and a computer program.

  If a stall occurs in the middle of the program being executed, if there is detailed trace information before the stall occurs, the process leading to the stall can be confirmed, and the cause of the stall Is likely to be found. Here, the stall refers to a stop or abnormal delay state in which the program processing does not proceed at all, such as deadlock or a CPU (Central Processing Unit) not being assigned.

  If this detailed trace information is always collected, the amount of trace information collected is enormous. For this reason, there are many cases in which the shortage of disk capacity is prevented by cyclically outputting trace information. To output cyclically, after storing trace information to the end in the address space of the storage area where the trace information is stored, update the trace information already stored from the beginning of the storage area by overwriting. Thus, the storage of trace information is continued.

  In a program such as a multi-thread program that cooperates between threads and that operates in association with a plurality of processes, the program may stall due to a processing delay of other programs or an unlock leak. Therefore, in such a program, not only the trace information of the stalled process but also detailed trace information regarding other related processes are required for investigation of the cause.

  However, when the investigation is performed with reference to the trace information, the trace information at the time when the stall occurs may not be left by cyclically outputting the trace information.

  For example, even if it is determined that there is an abnormality due to a response waiting timeout occurring in another thread due to a certain thread being stopped, the stopped thread may resume its operation and then operate normally.

  Even if another thread keeps waiting for exclusive acquisition due to omission of exclusion of a certain thread, the thread that has lost omission of exclusion continues to operate normally.

  In addition, even when another thread keeps waiting for a response because a certain thread does not return a response, the thread that does not return a response continues to operate normally.

  Or, if a thread performs multiple notifications asynchronously to another thread and one of them is an illegal notification, the thread that made the illegal notification is Will continue to work normally.

  In such cases, the operator checks the trace information when the thread stops, the trace information when the exclusion release is omitted, the trace information when the response is forgotten to be returned, and the trace information when the incorrect notification is made. There is a need to. However, when an abnormality such as a response waiting timeout is detected, there is a problem that the trace information is cyclically output, so that it is already overwritten and the necessary trace information may not be obtained.

JP 2002-007172 A JP 2000-181748 A JP 2000-057013 A Japanese Patent Application Laid-Open No. 11-184736 JP 5093175 A Japanese Patent No. 4114879 Japanese Patent Laid-Open No. 08-044595

  Here, as related technologies existing prior to the present application, there are, for example, the following patent documents.

  By using the technique proposed in Patent Document 1, it is possible to determine that a stall has occurred based on the update state of the time stamp.

  By using the technique proposed in Patent Document 2, it is possible to collect trace information of each space of a program operating in a multi-memory space.

  By using the technique proposed in Patent Document 3, there are multiple buffers for storing trace information, and when the buffer for storing trace information is full, switch to the next buffer and use multiple buffers repeatedly. can do.

  By using the technique proposed in Patent Document 4, data can be collected when a stall is detected.

  By using the technique proposed in Patent Document 5, it is possible to collect trace information when an event for which trace information is to be collected occurs.

  By using the technique proposed in Patent Document 6, by comparing the operation stop time ratios of the trace collection target programs, it is possible to collect trace information when it is determined that the program is operating in a different operation. Can do.

  By using the technique proposed in Patent Document 7, in a process in which a trace collection target program runs, trace information is stored in a buffer and input to the file from the buffer in a separate process. The load required for collecting traces can be reduced.

  However, even when the techniques proposed in Patent Documents 1 to 7 are used, the collection of traces in a program in which a plurality of processes operate in relation is not considered, and the above-described problems have not been solved.

  The present invention provides a trace information collection device and the like that can collect information used for investigating the cause of a stall occurrence in a program that operates in association with a plurality of processes, such as a multi-thread program that cooperates between threads. Is the main purpose.

  In order to achieve the above object, a trace information collecting apparatus according to the present invention has the following configuration.

That is, the trace information collecting apparatus according to the present invention is
A collecting means for collecting trace information of the programs in an environment in which a plurality of programs operate in parallel or pseudo-concurrency;
Among the plurality of programs, a plurality of buffers capable of storing the trace information of the first program, and management information for storing information necessary for accessing the buffer that is currently collecting the trace information among the buffers, A first information storage unit including:
Among the plurality of programs, a plurality of buffers capable of storing the trace information of the second program, and management information for storing information necessary for accessing a buffer that is currently collecting the trace information among the buffers, A second information storage unit including:
When a stall is detected by determining a predetermined stall determination condition, from the buffer that is currently collecting the trace information among the plurality of buffers with respect to the first and second information storage units And a management unit for controlling the collection of the trace information to be simultaneously switched to a buffer different from the buffer and storing the trace information.

The trace information collection method according to the present invention that achieves the same object,
In an environment where a plurality of programs operate in parallel or pseudo-parallel, the information processing apparatus
Among the plurality of programs, a plurality of buffers capable of storing the trace information of the first program, and management information for storing information necessary for accessing the buffer that is currently collecting the trace information among the buffers, Collecting trace information of the first program in the buffer based on the management information for the first information storage unit including
Among the plurality of programs, a plurality of buffers capable of storing the trace information of the second program, and management information for storing information necessary for accessing a buffer that is currently collecting the trace information among the buffers, Collecting trace information of the second program in the buffer based on the management information for the second information storage unit including
When a stall is detected by determining a predetermined stall determination condition, from the buffer that is currently collecting the trace information among the plurality of buffers with respect to the first and second information storage units The collection of the trace information is simultaneously switched to a buffer different from the buffer, and control is performed to save the trace information.

  Further, the object is achieved by a computer program for realizing the trace information collecting method and the trace information collecting apparatus having the above-described configuration by a computer, and a computer-readable storage medium storing the computer program. Is done.

  According to the present invention described above, it is possible to collect information used for investigating the cause of the occurrence of a stall in a program that operates in association with a plurality of processes, such as a multi-thread program that cooperates between threads.

It is a block diagram which shows the structure of the trace information collection apparatus which concerns on the 1st Embodiment of this invention. It is a block diagram which shows the structure of the trace information collection apparatus which concerns on the 2nd Embodiment of this invention. It is a figure explaining the management information which concerns on the 2nd Embodiment of this invention. It is a figure explaining the buffer information which concerns on the 2nd Embodiment of this invention. It is a flowchart which shows the process which stores the trace information in the shared memory in the trace information collection part which concerns on the 2nd Embodiment of this invention. It is a flowchart which shows the stall monitoring process in the buffer management part which concerns on the 2nd Embodiment of this invention. It is a flowchart which shows the state initialization process in the buffer management part which concerns on the 2nd Embodiment of this invention. It is a flowchart which shows the process regarding the file writing in the buffer management part which concerns on the 2nd Embodiment of this invention. It is a flowchart which shows the file writing process in the file output part which concerns on the 2nd Embodiment of this invention. It is a block diagram which shows the structure which deform | transformed the trace information collection apparatus which concerns on the 2nd Embodiment of this invention. It is a block diagram which shows the structure of the trace information collection apparatus which concerns on the 3rd Embodiment of this invention. It is a block diagram which shows the structure of the trace information collection apparatus which concerns on the 4th Embodiment of this invention. It is a figure which illustrates illustartively the hardware constitutions of the computer (information processing apparatus) which can implement | achieve the 1st thru | or 4th embodiment of this invention.

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

<First Embodiment>
FIG. 1 is a block diagram showing a configuration of a trace information collecting apparatus according to the first embodiment of the present invention.

  The trace information collection device includes a collection unit 10, an information storage unit 20, an information storage unit 30, and a management unit 40.

  The collection unit 10 collects trace information from the process 60 and stores the collected trace information in the information storage unit 20. Further, the collection unit 10 collects trace information from the processing 70 and stores the collected trace information in the information storage unit 30.

  The information storage unit 20 includes management information 21, a buffer 22, a buffer 23, and a buffer 24.

  The management information 21 is information representing a buffer that is the storage destination of the current trace information among these three buffers (details will be described later with reference to FIG. 3 in the second embodiment).

  As shown in FIG. 1, the information storage unit 30 has the same configuration as the information storage unit 20, and includes management information 31, a buffer 32, a buffer 33, and a buffer 34.

  The management unit 40 detects the occurrence of a stall and switches the buffers that are the storage destinations of the trace information all at once in order to secure the trace information. That is, the management unit 40 changes the management information 21 and 31 in the information storage unit 20 and the information storage unit 30 to information on a buffer that stores trace information next.

  According to the present embodiment, there is an effect that information used for investigating the cause of the occurrence of a stall can be collected in a program that operates in association with a plurality of processes.

  The reason is that each process stores the trace information in a separate shared memory and prepares a plurality of buffers in each shared memory as a storage destination of the trace information. As a result, when the possibility of stall in a certain process is detected, the trace information of other processes up to that point can be left by changing and saving the buffer for storing the trace information all at once.

<Second Embodiment>
FIG. 2 is a block diagram showing the configuration of the trace information collecting apparatus according to the second embodiment of the present invention.

  The trace information collection apparatus according to this embodiment includes a process 100, a trace information collection unit 200, shared memories 301 and 302, attributes 361 and 362, a buffer management unit 400, a file output unit 500, and a file 600. Have

  The process 100 includes a thread 101 and a thread 102.

  An application program (not shown) operating in the thread 101 stores trace information in the shared memory 301 by the trace information collection unit 200.

  Similarly, an application program (not shown) that operates in the thread 102 stores trace information in the shared memory 302 by the trace information collection unit 200.

  The shared memory 301 includes management information 311 and three buffers (buffer A321, buffer B331, and buffer C341) that are storage destinations of trace information.

  The buffer management unit 400 includes buffer information 401.

  FIG. 3 is a diagram for explaining the management information 311 according to the second embodiment of the present invention.

  The management information 311 is information that is referred to in order to obtain information indicating the storage location of the trace information when the trace information collection unit 200 stores the trace information. That is, as shown in FIG. 3, the management information 311 includes a storage position 811 that is a position for storing trace information, a buffer start position 812 that is a start position of a buffer that is currently storing trace information, For each buffer end position 813 that is the buffer end position, an address 814 that is information representing each address is included.

  The management information 311 also includes buffer information 815 that is information indicating the addresses of the start position and end position of each of the three buffers (buffer A321, buffer B331, and buffer C341).

  The buffer information 815 is information that is referred to when the storage position 811, the buffer start position 812, and the buffer end position 813 are updated in order to switch the buffer where the trace information is stored.

  The trace information collection unit 200 stores the trace information in the buffer represented by the management information 311 among the three buffers of the buffer A321, the buffer B331, and the buffer C341. The trace information collection unit 200 uses the buffer B331 after the buffer A321 and the buffer C341 after the buffer B331 as a storage destination of the trace information. Further, the trace information collection unit 200 uses the buffer A321 after the buffer C341.

  The attribute 361 is information representing the attribute of the shared memory 301 provided by the operating system, for example. The attribute 361 includes, for example, the creation time and creator of the shared memory 301 and the latest update time of the shared memory 301. When writing to the shared memory 301, the operating system sets the time of writing to the update time included in the attribute 361.

  The shared memory 302 has the same configuration as the shared memory 301. Management information 312, buffer A 322, buffer B 332, buffer C 342, and attribute 362 of the shared memory 302 correspond to management information 311, buffer A 321, buffer B 331, buffer C 341, and attribute 361 of the shared memory 301, respectively.

  The file output unit 500 writes the trace information stored in the designated buffer to the file 600 according to an instruction from the buffer management unit 400.

  The buffer management unit 400 manages the shared memory 301 and all the buffers in the shared memory 302.

  The buffer management unit 400 monitors whether the time since the last update of the shared memory 301 or 302 has exceeded a predetermined time. When the buffer management unit 400 detects that the update has not been performed for a certain time, the buffer management unit 400 changes the buffer in which the trace information collection unit 200 stores the trace information from the currently stored buffer to the next buffer. As shown, the management information 311 and 312 are updated.

  The buffer management unit 400 has buffer information 401.

  FIG. 4 is a diagram for explaining buffer information 401 according to the second embodiment of the present invention.

  The buffer management unit 400 has buffer information 401 for managing the buffers A 321 and 322, the buffer B 331 and 332, and the buffers C 341 and 342.

  The buffer information 401 includes a buffer identification name 821, a status 822, and a final time 823. In other words, the buffer identification name 821, the status 822, and the final time 823 are associated in the buffer information 401 as shown in the table conceptually shown in FIG.

  The buffer identification name 821 is information for identifying a buffer.

  The status 822 is information representing four states, for example, “stored”, “free”, “writing”, and “temporarily saved”. The buffer management unit 400 sets and changes the status 822.

  The status 822 “stored” indicates that the trace information collection unit 200 is storing trace information in the buffer.

  The status 822 “free” indicates that the buffer is not used.

  The status “822” indicates that the file output unit 500 is writing the trace information in the buffer to the file 600.

  The status 822 “temporarily saved” indicates that the trace information collection unit 200 is not storing the trace information but is holding the trace information as it is without being overwritten.

  The last time 823 is information indicating the time when the buffer for storing the trace information is changed.

  When the buffer management unit 400 changes the buffer for storing the trace information and the buffer is no longer in the “storage” state, the time is set to the final time 823.

  For example, when the trace information collection unit 200 stores the trace information in the buffers A321 and 322, the status 822 of the buffer A is “in storage”. When the buffer management unit 400 changes the trace storage destination of the trace information collection unit 200 from the buffer A (321 and 322) to the buffer B (331 and 332), the status 822 of the buffer A is set to “temporary storage”. The status 822 of the buffer B is set to “stored”.

  Further, the buffer management unit 400 sets the time when the buffer to be stored is changed to the last time 823 of the buffer B.

  Further, the buffer management unit 400 stores the start address of the buffer B331 in the address 814 column corresponding to the storage location 811 and the buffer start location 812 of the management information 311 and the buffer B331 in the address 814 column corresponding to the buffer end location 813. Set the end address.

  In addition, the buffer management unit 400 performs similar settings for the management information 312. That is, the buffer management unit 400 stores the start address of the buffer B 332 in the address 814 column corresponding to the storage position 811 and the buffer start position 812 of the management information 312, and the buffer B 332 in the address 814 column corresponding to the buffer end position 813. Set the end address.

  With these processes, the trace information collection unit 200 stores the trace information in the buffers B331 and 332.

  Next, the operation of the second exemplary embodiment of the present invention will be described in detail with reference to the flowcharts of FIGS.

  FIG. 5 is a flowchart showing a process of storing the trace information in the shared memory 301 in the trace information collection unit 200 according to the second embodiment of the present invention.

  First, the trace information collection unit 200 receives a trace information storage request from the thread 101 and refers to the management information 311 (step S11).

  The trace information collection unit 200 adds the size of the trace information that has received the storage request to the address of the storage location 811. The trace information collection unit 200 checks whether or not the position obtained by the addition exceeds the address of the buffer end position 813 (step S12).

  When the address of the buffer end position 813 is exceeded (“Yes” in step S12), the trace information collection unit 200 stores the trace information at the head of the buffer currently being stored. That is, the trace information collection unit 200 stores the trace information at the address of the buffer start position 812 (step S13). The trace information collection unit 200 changes the address of the storage location 811 to a value obtained by adding the size of the stored trace information to the address of the buffer start location 812 (step S14).

  When the address of the buffer end position 813 is not exceeded (“No” in step S12), the trace information collection unit 200 stores the trace information at the address of the storage position 811 (step S15). The trace information collection unit 200 changes the address of the storage location 811 to a value obtained by adding the size of the stored trace information (step S16).

  FIG. 6 is a flowchart showing a stall monitoring process in the buffer management unit 400 according to the second embodiment of the present invention.

  The stall monitoring process is executed periodically, for example.

  The buffer management unit 400 acquires the update times of the shared memories 301 and 302 in the attributes 361 and 362 (step S21). The buffer management unit 400 can obtain the attributes 361 and 362 using, for example, a function provided by the operating system.

  If there is a shared memory that has not been updated for a certain time with reference to the update time (“Yes” in step S22), the buffer management unit 400 stores the trace information before the update time of the shared memory. It is determined whether the time is later than the last time 823 of the buffer that has been stored (step S23). Here, the fixed time is a time determined in advance, and there is no particular limitation on the value of the time. For example, a time is set so that the buffer is not filled with the trace information storage within the fixed time.

  If the update time of the shared memory is later (“Yes” in step S23), the buffer management unit 400 determines that the possibility of a new stall has been detected.

  On the other hand, if “No” in step S23, the buffer management unit 400 does not detect the possibility of a new stall, but determines that the state where the possibility of a stall has already been detected continues. To do.

  When the possibility of a stall is newly detected, the buffer management unit 400 checks whether or not the status 822 of the buffer that stores the trace information is “empty” (step S24).

  If the next buffer is “free”, the buffer management unit 400 sets the current time at the last time 823 corresponding to the buffer currently being “stored”, and changes the status 822 to “temporarily saved”. The buffer management unit 400 changes the status 822 corresponding to the next buffer to “stored” (step S25). Further, in the management information 311 and 312, the addresses 814 corresponding to the storage position 811, the buffer start position 812, and the buffer end position 813 are updated with values indicating the next buffer, respectively (step S 26).

  FIG. 7 is a flowchart showing a state initialization process in the buffer management unit 400 according to the second embodiment of the present invention.

  The buffer management unit 400 compares the current time with the final time 823 of the buffer whose status 822 is “temporary storage” (step S31).

  As a result of the comparison, if the predetermined time has passed (“Yes” in step S32), the buffer status 822 is changed to “free” (step S33). Note that the fixed time here is, for example, assumed that the trace information acquisition work (write processing to the file by the file output unit 500) is performed after an abnormality such as a timeout occurs, and is set to “temporary storage”. It takes longer than the time required to acquire the trace information.

  FIG. 8 is a flowchart showing processing related to file writing in the buffer management unit 400 according to the second embodiment of the present invention.

  The buffer management unit 400 checks whether a notification is received from the file output unit 500 (step S41).

  If there is a notification from the file output unit 500 (“Yes” in step S42), the buffer management unit 400 checks whether there is a buffer whose status 822 is “temporarily saved” (step S43).

  When there is a “temporary save” buffer (“Yes” in step S43), the buffer management unit 400 instructs the file output unit 500 to write the contents of the buffer to the file for the “temporary save” buffer. (Step S44).

  If there is no “temporary save” buffer (“No” in step S43), the buffer management unit 400 notifies the file output unit 500 that there is no “temporary save” buffer (step S48).

  Next, the buffer management unit 400 changes the status 822 of the buffer instructed to write the file to the file output unit 500 from “temporary storage” to “writing” (step S45).

  The buffer management unit 400 waits for a file write completion notification from the file output unit 500 (step S46).

  Upon receiving the notification, the buffer management unit 400 changes the status 822 of the buffer to “free” (step S47).

  FIG. 9 is a flowchart showing a file writing process in the file output unit 500 according to the second embodiment of the present invention.

  First, the file output unit 500 notifies the buffer management unit 400 that it will write to a file (step S51).

  The file output unit 500 waits for a response from the buffer management unit 400 (step S52).

  If the response from the buffer management unit 400 indicates that there is no “temporary storage” buffer, the file output unit 500 ends the process (“No” in step S53).

  When the buffer management unit 400 receives a notification of a buffer whose status 822 is “temporarily stored” (“Yes” in step S53), the file output unit 500 stores the trace information stored in the designated buffer in the file 600. (Step S54). For example, when “buffer B” is notified from the buffer management unit 400 as a “temporary storage” buffer, the file output unit 500 stores the trace information in the buffer B (331 and 332) in each shared memory. Write to file 600.

  After completing the writing, the file output unit 500 notifies the buffer management unit 400 of the completion of the file writing (step S55).

  In the present embodiment, three buffers in one shared memory have been described. However, the number of buffers is not limited to this, and may be three or more.

  FIG. 10 is a block diagram showing a modified configuration of the trace information collecting apparatus according to the second embodiment of the present invention.

  In FIG. 10, the shared memory 301 includes four buffers: a buffer A321, a buffer B331, a buffer C341, and a buffer D351. Similarly, the shared memory 302 includes four buffers: a buffer A 322, a buffer B 332, a buffer C 342, and a buffer D 352.

  As shown in FIG. 10, even when four buffers are targeted, the processing flow is not changed and the description thereof is omitted.

  For convenience, the number of buffers included in the shared memories 301 and 302 is described as the same number, but the number of buffers is not limited to the same number as long as the other can be specified.

  As described above, the second embodiment has an effect that information used for investigating the cause of the occurrence of a stall can be collected in a multi-thread program that cooperates between threads.

  The reason is that each thread stores the trace information in a separate shared memory and prepares a plurality of buffers in each shared memory as a storage destination of the trace information. As a result, when the possibility of stall in a certain process is detected, the trace information of other processes up to that point can be left by changing and saving the buffer for storing the trace information all at once.

<Third Embodiment>
FIG. 11 is a block diagram showing a configuration of a trace information collecting apparatus according to the third embodiment of the present invention.

  The change from FIG. 2 to FIG. 11 is that the process 100 is changed to a process 103 and a process 104. That is, the second embodiment of the present invention is configured to store trace information from a plurality of threads in separate shared memories, but the third embodiment of the present invention has a plurality of processes (processes 103 and 104). ) To store the trace information in separate shared memories.

  As described above, the third embodiment has an effect that it is possible to collect information used for investigating the cause of stall occurrence in a program linked between processes.

The reason is that each process stores trace information in a separate shared memory and prepares a plurality of buffers in each shared memory as a storage destination of the trace information. As a result, when the possibility of stall in a certain process is detected, the trace information of other processes up to that point can be left by changing and saving the buffer for storing the trace information all at once. <Fourth Embodiment>
FIG. 12 is a block diagram showing a configuration of a trace information collecting apparatus according to the fourth embodiment of the present invention.

  In the present embodiment, as shown in FIG. 12, stall determination units 701 and 702 are provided for each shared memory.

  Stall determination units 701 and 702 each have a stall determination condition, and the possibility of a stall can be determined based on the stall determination condition.

  As described above, the fourth embodiment has an effect that the definition of a stall can be defined in advance and information used for investigating the cause of the occurrence of the stall can be collected in a multi-thread program linked between threads. is there.

  The reason is that each thread stores the trace information in a separate shared memory and prepares a plurality of buffers in each shared memory as a storage destination of the trace information. In addition, the stall determination condition is set in advance. As a result, when the possibility of a stall is detected in a certain process based on the set stall determination condition, the buffer for storing the trace information is changed and saved all at once, so that Trace information of the process can also be left.

(Hardware configuration)
In the embodiment described above, each unit shown in FIGS. 1, 2, 10, 11 and 12 may be practiced by a dedicated device, but is regarded as a function (processing) unit (software module) of a software program. be able to. However, the division of each part shown in these drawings is a configuration for convenience of explanation, and various configurations can be assumed for mounting. An example of the hardware environment in such a case will be described with reference to FIG.

  FIG. 13 is a diagram for exemplifying the hardware configuration of a computer (information processing apparatus) capable of realizing the first to fourth embodiments of the present invention. That is, it is a configuration of a computer (information processing apparatus) such as a server that can realize the whole or a part of the trace information collection apparatus shown in FIG. 1, FIG. 2, FIG. 10, FIG. The hardware environment which can implement | achieve each function in a form is represented.

  An information processing apparatus 9000 shown in FIG. 13 includes a CPU (Central Processing Unit) 9001, a display 9002, a communication interface (I / F) 9003, a ROM (Read Only Memory) 9004, a RAM (Random Access Memory) 9005, a hard disk device ( HD) 9006, and these are connected via a bus 9007. The hard disk device (HD) 9006 stores a program group 9006A and various kinds of stored information 9006B. The program group 9006A is, for example, a computer program for realizing a function corresponding to each block (each unit) shown in FIG. 1, FIG. 2, FIG. 10, FIG. Various kinds of stored information 9006B are, for example, the information storage unit 20 and the information storage unit 30 shown in FIG. A communication interface 9003 is a general communication unit that realizes communication with an external device via the network 9100.

  The present invention described by taking the above-described embodiment as an example is a block configuration diagram referred to in the description of the embodiment, or after supplying a computer program capable of realizing the function of the flowchart, This is achieved by reading the data to the CPU 9001 of the hardware and using the hardware resources of the information processing apparatus 9000 shown in FIG. Specifically, when the collection unit 10 or the trace information collection unit 200 is realized by the information processing device 9000, the information processing device 9000 may realize a computer program corresponding to the flowchart shown in FIG. When the management unit 40 or the buffer management unit 400 is realized by the information processing device 9000, the information processing device 9000 may realize a computer program corresponding to the flowcharts shown in FIGS. When the file output unit 500 is realized by the information processing device 9000, the information processing device 9000 may realize a computer program corresponding to the flowchart shown in FIG. The computer program supplied to the information processing device 9000 may be stored in a readable / writable temporary storage memory (9005) or a non-volatile storage device (storage medium) such as the hard disk device 9006.

  In the above case, the computer program can be supplied to the apparatus from the outside through various storage media such as a CD-ROM or from the outside via a communication line such as the Internet. Currently, a general procedure can be adopted, such as a downloading method. In such a case, the present invention can be understood to be constituted by a code constituting the computer program or a computer-readable storage medium in which the code is recorded.

DESCRIPTION OF SYMBOLS 10 Collection part 20 Information storage part 21 Management information 22 Buffer 23 Buffer 24 Buffer 30 Information storage part 31 Management information 32 Buffer 33 Buffer 34 Buffer 40 Management part 100 Process 101 Thread 102 Thread 103 Process 104 Process 200 Trace information collection part 301 Shared memory 302 Shared memory 311 Management information 312 Management information 321 Buffer A
322 Buffer A
331 Buffer B
332 Buffer B
341 Buffer C
342 Buffer C
351 Buffer D
352 Buffer D
361 attribute 362 attribute 400 buffer management unit 401 buffer information 500 file output unit 600 file 701 stall determination unit 702 stall determination unit 811 storage position 812 buffer start position 813 buffer end position 814 address 815 buffer information 821 buffer identification name 822 status 823 final time 9000 Information processing device (computer)
9001 CPU
9002 Display 9003 Communication interface (I / F)
9004 ROM
9005 RAM
9006 Hard disk device (HD)
9006A Program group 9006B Various stored information 9007 Bus 9100 Network

Claims (8)

A collecting means for collecting trace information of the programs in an environment in which a plurality of programs operate in parallel or pseudo-concurrency;
Among the plurality of programs, a plurality of buffers capable of storing the trace information of the first program, and management information for storing information necessary for accessing the buffer that is currently collecting the trace information among the buffers, A first information storage unit including:
Among the plurality of programs, a plurality of buffers capable of storing the trace information of the second program, and management information for storing information necessary for accessing a buffer that is currently collecting the trace information among the buffers, A second information storage unit including:
When a stall is detected by determining a predetermined stall determination condition, from the buffer that is currently collecting the trace information among the plurality of buffers with respect to the first and second information storage units A trace information collecting apparatus comprising: a management unit that controls the collection of the trace information to a buffer different from the buffer at a time and controls the trace information to be saved. 2. The trace information collecting apparatus according to claim 1, wherein the stall determination condition is determined based on a result of referring to a time when the first and second information storage units were most recently updated. The plurality of buffers included in the second information storage unit can store the trace information, and correspond to the plurality of buffers included in the first information storage unit on a one-to-one basis. The trace information collection device according to claim 1 or 2. The trace information collection apparatus according to claim 1, wherein an environment in which the plurality of programs operate in parallel or pseudo-parallel is a multi-thread environment. In an environment where a plurality of programs operate in parallel or pseudo-parallel, the information processing apparatus
Among the plurality of programs, a plurality of buffers capable of storing the trace information of the first program, and management information for storing information necessary for accessing the buffer that is currently collecting the trace information among the buffers, Collecting trace information of the first program in the buffer based on the management information for the first information storage unit including
Among the plurality of programs, a plurality of buffers capable of storing the trace information of the second program, and management information for storing information necessary for accessing a buffer that is currently collecting the trace information among the buffers, Collecting trace information of the second program in the buffer based on the management information for the second information storage unit including
When a stall is detected by determining a predetermined stall determination condition, from the buffer that is currently collecting the trace information among the plurality of buffers with respect to the first and second information storage units A trace information collecting method comprising: controlling the collection of the trace information to a buffer different from the buffer at once, and saving the trace information. 6. The trace information collecting method according to claim 5, wherein the stall determination condition is determined based on a result obtained by referring to a time when the first and second information storage units were most recently updated. The plurality of buffers included in the second information storage unit can store the trace information, and correspond to the plurality of buffers included in the first information storage unit on a one-to-one basis. The trace information collecting method according to claim 5 or 6. In an environment in which a plurality of programs operate in parallel, a plurality of buffers capable of storing the trace information of the first program among the plurality of programs and the current trace information among the buffers are collected. Collecting the trace information of the first program in the buffer based on the management information, with respect to the first information storage unit including management information for storing information necessary for accessing the current buffer;
Among the plurality of programs, a plurality of buffers capable of storing the trace information of the second program, and management information for storing information necessary for accessing a buffer currently collecting the trace information among the buffers. And collecting the trace information of the second program in the buffer based on the management information for the second information storage unit including
When a stall is detected by determining a predetermined stall condition, the buffer that currently collects trace information among the plurality of buffers is detected for the first and second information storage units. A computer program for causing a computer to realize a control function for switching the collection destination of trace information to a buffer different from the buffer at a time and saving the trace information.

Images