JP3491379B2 - Information processing apparatus and trace information storage method - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an information processing apparatus and a trace information storage method, and more particularly to a trace information storage method for storing a plurality of types of trace information that are asynchronously generated in a processing unit, and the trace information storage method. The present invention relates to an information processing device to which is applicable.

[0002]

2. Description of the Related Art In an information processing apparatus, for example, communication with an external device or a specific process such as access to a storage medium such as a magnetic disk is performed, or a slight error occurs. Each time it occurs, information (trace information) representing the content and result of the processing, the content of error, etc. is stored in a trace information storage area provided in a memory or the like. According to the above
By reading the trace information stored in the trace information storage area when a failure occurs, it is possible to easily determine the history of processing that was performed before the failure occurred and the state of the information processing device. It is possible to easily analyze the cause of failure.

By the way, in an information processing apparatus having a plurality of processors operating independently of each other, a trace information storage area is generally provided for each processor, and each processor stores the trace information in the trace information storage area. Is performed asynchronously and irregularly. However, even in such an information processing apparatus, in order to facilitate the analysis of the cause of failure, the trace information stored in each trace information storage area is stored in the correct order (for example, stored in the trace information storage area). Order) is required.

For this reason, Japanese Patent Laid-Open No. 5-40658 discloses a system in which a trace information storage area is provided for each processor and a counter is provided on a shared memory accessible by each processor. It is described that when each processor stores trace information, the counter value is added to the trace information and stored in the trace information storage area, and the counter value on the shared memory is updated. According to the above, by referring to the value of the counter added to each trace information, each trace information can be correctly arranged in the order of storage in the trace information storage area.

However, in the above, since the Torres information is distributed and stored in the trace information storage area provided corresponding to each processor, it is necessary to transfer the trace information between the processors when reading the trace information. Further, if any one of the plurality of processors stops its operation due to a failure or the like, it is possible to read the trace information from the trace information storage area provided on the memory of the processor that stopped the operation. There was an inconvenience that it would not be possible.

Further, in order to unify management of trace information, Japanese Patent Laid-Open No. 5-54010 discloses that fault information (log data) generated in a slave processor is transferred to a master processor, and the master processor transfers the failure information. A fault logging method for storing fault information has been proposed. However, in the above method, if the failure information is stored immediately every time the failure information occurs, the failure information is overwritten and the data becomes inconsistent due to the overwriting of the failure information.

For this reason, in Japanese Patent Laid-Open No. 5-54010, if another failure information is stored when trying to store the failure information, the failure is stored until the storage of the other failure information is completed. The so-called exclusive control is performed to temporarily stop the storage of information. However, the exclusive control is a complicated control, and in order to realize the exclusive control, it is necessary to provide a buffer for temporarily holding the failure information, so that there is a problem that the configuration becomes complicated. It should be noted that the loss or inconsistency of the above-mentioned data is caused by multiple processes (tasks) in a single processor.
Even in an information processing apparatus that executes processing in so-called multitasking, which is executed in parallel by time division or the like, it may occur if the program that stores the trace information is a reentrant program (reentrant program).

The present invention has been made in consideration of the above facts, and an information processing apparatus and a trace capable of storing a plurality of types of trace information that occur asynchronously with each other in a correct order, can be stored easily and reliably. The purpose is to obtain an information storage method.

[0009]

In order to achieve the above-mentioned object, an information processing apparatus according to the invention of claim 1 has a storage area group composed of a large number of storage areas each capable of storing trace information. Any of a plurality of storage areas of the storage area group, and a plurality of trace information storage means provided corresponding to a plurality of types of trace information that are asynchronously generated by a plurality of processes executed in parallel A holding means for holding the storage position information indicating whether to store the trace information in the area, and each time the trace information is generated,
Monotonic increase to be added to the trace information as identification information
An additional number is generated for each of the multiple types of trace information.
Each time the trace number is generated and the incrementing number generation means ,
Of the many storage areas of the storage area group corresponding to the generated trace information of the trace information storage means, the trace in which the monotonically increasing number is added to the storage area represented by the storage position information held by the holding means Trace information storage control means for storing information and updating the storage position information held in the holding means.

According to a second aspect of the present invention, there is provided a trace information storing method, wherein a processing unit for executing a plurality of processes in parallel and a storage area group constituted by a large number of storage areas each capable of storing trace information are the processing. The trace information is stored in a plurality of trace information storage means corresponding to a plurality of types of trace information that are asynchronously generated by the execution of the processing by the unit, and in any one of the storage areas of the storage area group. An information processing device including a holding unit that holds storage position information indicating whether or not each time trace information is generated, the trace information is added as identification information to the trace information.
The monotonically increasing number for the above-mentioned multiple types of trace information
The monotonically increasing number in the storage area represented by the storage position information held in the holding means, out of a large number of storage areas of the storage area group corresponding to the generated trace information in the trace information storage means . stores the trace information obtained by adding, is characterized by updating the storage location information held in said holding means.

[0011]

According to the present invention, a plurality of processes are executed in parallel in the processing section, and a plurality of types of trace information are generated asynchronously with each other. In addition, the processing unit has a plurality of CPs.
The configuration may be such that each CPU includes U and performs processing independently of each other, or a configuration that includes a single CPU and performs a plurality of processings in parallel by time division or the like. In addition, the type of trace information is such that the processing unit has multiple CPUs.
When the processing unit has a single CPU, the trace information generated by different CPUs may be distinguished as the trace information of different types. Trace information different in communication or access to a storage medium) may be distinguished as different types of trace information.

According to the first aspect of the present invention, a storage area group composed of a large number of storage areas each capable of storing trace information has a plurality of storage areas that are asynchronously generated by a plurality of processes being executed in parallel by the processing unit. A plurality of trace information storage means are provided corresponding to the types of trace information, and the monotonically increasing number generation means generates the trace information.
Each time, for adding as identification information to the trace information
Generate a monotonically increasing number for each type of trace information,
In the trace information storage control means, every time the trace information is generated, among the many storage areas of the storage area group corresponding to the generated trace information of the trace information storage means, the storage position information held in the holding means is stored. The trace information to which the monotonically increasing number is added is stored in the storage area shown, and the storage position information held in the holding means is updated.

In the above, the determination of the storage area in which the trace information is stored among the many storage areas of the storage area group is made based on the storage position information held by the holding means. A plurality of types of trace information that are generated are provided, and the generated trace information is stored in the storage area of the corresponding storage area group. Therefore, for example, a plurality of different types of trace information are generated at the same time or almost at the same time. However, even if the trace information storage control means stores the generated trace information completely in parallel, the trace information will not be lost by being overwritten with each other, and complicated exclusive control will be performed. It is not necessary to provide a buffer or the like for performing the exclusive control.

As described above, according to the first aspect of the present invention, a plurality of types of trace information that occur asynchronously with each other can be stored easily and reliably. Further, as described above, since the storage position information held in the holding means is commonly used when storing different types of trace information, the storage position of the trace information in each of the plurality of storage area groups is It also changes every time the trace information is stored in the information storage means. Therefore, it is possible to determine the generation order of the different types of trace information stored in the different storage area groups based on the storage position in each storage area group (the position of the storage area in which the trace information is stored). Therefore, a plurality of types of trace information that occur asynchronously with each other can be stored so that the correct order can be determined.

[0015] In the present invention of claim 2, wherein, each time the trace information of a predetermined type by the processing unit occurs, the generated trace information
Multiple monotonically increasing numbers to be added to the information as identification information
Raises for each type of trace information, among a plurality of storage areas of said generated corresponding storage area group in the trace information in the trace information storage means, the storage area indicated by the storage position information held in the holding means, monotonous Add increment number
Since the trace information is stored and the storage position information held in the holding means is updated, a plurality of types of trace information that occur asynchronously with each other can be stored in the correct order, as in the first aspect of the invention. Can be stored and can be stored easily and surely.

[0016]

Embodiments of the present invention will now be described in detail with reference to the drawings. FIG. 1 shows an information processing apparatus 1 according to this embodiment.
0 is shown. The information processing device 10 includes a first processor 12 and a second processor 14 as processing units.

The first processor 12 is a CPU 16, RO
An M18 and a RAM 20 are provided, and these are connected to each other via a local bus 22. The local bus 22 is connected to the main bus 24 of the information processing device 10. Also, the second processor 14 is the CPU 2
8, a ROM 30, and a RAM 32, which are connected to each other via a local bus 34. The local bus 34 is also connected to the main bus 24. Therefore, the first processor 12 and the second processor 1
4 can exchange various data with each other via the main bus 24.

A RAM 36 is connected to the main bus 24. The RAM 36 can be accessed by the first processor 12 and the second processor 14.
The storage area of the RAM 36 is provided with a storage position information holding area 38 as a holding means of the present invention and a trace information storage area 40 as a trace information storage means of the present invention. As shown in FIG. 2, the trace information storage area 40 has a storage area group 4 for the first processor 12.
0A and a storage area group 40B for the second processor 14
And are provided. These storage area groups 40A, 4
0B has a plurality of storage areas (single storage areas are indicated by □ in FIG. 2) each capable of storing one unit of trace information (referred to as unit trace information) (n in FIG. 2).
It is provided.

An address for identifying each storage area in the storage area group is assigned in advance to each storage area of the storage area groups 40A and 40B (from "1" in FIG. 2).
The storage location information holding area 38 stores the address of a predetermined storage area as storage location information. In this embodiment, the trace information generated by the first processor 12 and the trace information generated by the second processor 14 are different types of trace information. These trace information are
As the processes (tasks) are executed in the first processor 12 and the second processor 14, they occur asynchronously with each other.

Next, the operation of this embodiment will be described. In FIG. 2, among various functions of the CPU 16 of the first processor 12 and the CPU 28 of the second processor 14,
Functions related to storage and output of trace information are shown in blocks. With the execution of the task in the first processor 12 (shown as “task A” in FIG. 2),
When an event in which the trace information is to be stored occurs (for example, a process such as communication in which a process history should be left is executed), the first processor 12 causes the monotonically increasing number generation unit 44,
The trace information storage unit 46, the storage position information acquisition unit 48, and the next storage position calculation unit 50 are activated, and the trace information storage process (described later) is executed. Of these , monotonically increasing numbers
The raw part 44 corresponds to the monotonically increasing number generating means of the present invention.
The others correspond to the trace information storage control means of the present invention.

Similarly, when an event in which the trace information is to be stored occurs due to the execution of the task in the second processor 14 (shown as "task B" in FIG. 2), the monotonically increasing number generating section 54, the trace information. The storage unit 56, the storage position information acquisition unit 58, and the next storage position calculation unit 60 are activated, and the trace information storage process (described later) is executed. Among these, the monotonically increasing number generating unit 54 is the monotonically increasing number of the present invention.
It corresponds to the number generation means, and the other corresponds to the trace information storage control means of the present invention. FIG. 3 shows the trace information storage processing executed by the first processor 12.
This will be described with reference to the flowchart in FIG.

In step 100, a number to be added to the trace information is acquired. The acquisition of this number is performed by activating the monotonically increasing number generation unit 44. The monotonically increasing number generation unit 44 holds the number output in the past, and every time it is activated, 1 is added to the held number to output a new number. In step 100, the number output from the monotonically increasing number generation unit 44 is acquired as a number to be added to the trace information. In step 102, the acquired number is added as identification information to the trace information to be stored.

At the next step 104, the storage position information held in the storage position information holding area 38 is acquired.
The storage position information is acquired by the storage position information acquisition unit 48, and the acquired storage position information is held in the trace information storage unit 46. In the next step 106, the trace information is transferred to the RAM 36, and the trace information is stored in the storage area group 40A of the trace information storage area 40 for the first processor which corresponds to the acquired storage location information. This processing is performed by the trace information storage unit 46.

In step 108, the next storage position is calculated. The storage position information acquisition unit 4 calculates the next storage position.
This is performed by the next storage position calculation unit 50 according to the instruction from 8. In the next storage position calculation unit 50, when the address of the storage area at the end of the storage area group (the storage area to which the symbol “n” is added in FIG. 2) is set as the current storage position information, As the storage location information, the address of the top storage area (the storage area to which the reference numeral “1” is attached in FIG. 2) of the storage area group is obtained, and as the current storage location information, the storage areas other than the end of the storage area group When the address is set, the address of the next storage area is obtained as the next storage position information. Then, in step 110, the calculated next storage position information is set in the storage position information holding area 38, and the process ends.

Next, the trace information storage processing executed by the second processor 14 will be described with reference to the flowchart of FIG. This processing is almost the same as the processing shown in FIG. 3, and in step 120, the monotonically increasing number generation unit 54
To acquire a number to be added to the trace information, the acquired number is added to the trace information to be stored as identification information in step 122, and the storage position information holding area 38 is stored in step 124. The stored storage position information is acquired by the storage position information acquisition unit 58.

In step 126, the trace information storage unit 5
6, the trace information is transferred to the RAM 36, and the trace information is stored in the storage area group 40B for the second processor of the trace information storage area 40, which corresponds to the acquired storage position information. Step 128
Then, in accordance with an instruction from the storage position information acquisition unit 58, the next storage position calculation unit 60 calculates the next storage position in the same manner as described above. Then, in step 130, the calculated next storage position information is set in the storage position information holding area 38, and the process ends.

By the above processing, as shown in FIG. 5A as an example, the trace information stored in the storage area group 40A of the trace information storage area 40 from the first processor 12 includes the first processor 12 in the trace information. A monotonically increasing number that increases every time trace information is generated is added as identification information (in FIG. 5A, the identification information of the monotonically increasing number N is
N>), and the trace information stored in the storage area group 40B of the trace information storage area 40 from the second processor 14 has a monotonic value that increases every time the trace information is generated in the second processor 14. The increase number is added as identification information.

The first processor 1 is shown in FIG.
2 and the example of the sequence of the trace information storage by the second processor 14, FIG. 5B shows the trace information stored in the storage area groups 40A and 40B of the trace information storage area 40 in the sequence of FIG. 5A. (The correspondence between the trace information of FIG. 5B and the trace information storage sequence of FIG. 5A is indicated by circled numbers),
As is clear from FIG. 5, the storage position information holding area 3
The storage position information held in 8 is updated every time the first processor 12 or the second processor 14 stores the trace information, and the trace information generated in the first processor 12 and the second processor 14 is stored. Based on the storage position information held in the position information holding area 38, it is stored in a predetermined storage area of a corresponding storage area group among a plurality of storage area groups provided for each processor.

As a result, the trace information is generated in the first processor 12 and the second processor 14 substantially at the same time (for example, in FIG. 5A, the trace information and the identification information <24> to which the identification information <30> is added). Even if the other trace information is stored in parallel based on the storage position information before update before updating the storage position information according to the storage of one trace information). ,
The trace information will not be lost due to overwriting. Therefore, it is possible to easily and surely store a plurality of types of trace information that occur asynchronously with each other without performing complicated exclusive control or providing a buffer or the like for realizing exclusive control.

Further, in the above, the storage position information holding area 3
Since the storage position information held in 8 is commonly used for storing the trace information of the first processor 12 and the trace information of the second processor 14, the storage position information is stored in each of the storage area groups 40A and 40B. Storage location of trace information (location of storage area for storing trace information)
Similarly changes each time the trace information is stored in the first processor 12 or the second processor 14. Therefore, as described below, when the trace information is edited and output, the generation order of the trace information of the first processor 12 and the second processor 14 stored in the storage area groups 40A and 40B will be described later. In addition, the determination can be made based on the storage position in each storage area group (the position of the storage area storing the trace information).

That is, when a failure or the like occurs in the information processing apparatus 10 and an instruction to edit / output the trace information is given to analyze the cause of the failure, the trace information edit / output unit 62 shown in FIG. 2 is activated. Then, the trace information edit output process is executed. The trace information edit output unit 62 may be provided in either the first processor 12 or the second processor 14. The trace information edit output process executed by the trace information edit output unit 62 will be described with reference to the flowchart of FIG.

In step 140, the first processor 12
As the number of the pre-trace information of the first processor 1 and the number of the pre-trace information of the second processor 14 respectively.
The latest numbers are generated by the monotonically increasing number generator 44 of No. 2 and the monotonically increasing number generator 54 of the second processor 14 and added to the trace information as identification information. In step 142, as the read storage area address, the address of the storage area in which the latest trace information is stored (the storage area immediately before the storage area whose address is stored in the storage location information storage area 38 as storage location information) Address).

In step 144, it is determined whether or not all the storage areas of each storage area group have been searched. If the determination is negative, in step 146 the first processor 12
The trace information is read from the storage area corresponding to the previously set read storage area address in the storage area group 40A for use with. In step 148, it is determined whether or not the number added as the identification information to the read trace information is two or more older (the value is smaller) than the number of the previous trace information of the first processor 12 set previously.

If the determination is affirmative, the read trace information is set to "invalid" in step 150, and the process proceeds to step 154. If the determination is negative, the read trace information can be determined to be the newest trace information that has not been output yet among the trace information of the first processor 12 stored in the trace information storage area 40. The process proceeds to step 152, the number of the identification information of the read trace information is set as the number of the previous trace information of the first processor 12, and the process proceeds to step 154.

In step 154, the previous step 146
Similarly to the storage area group 40B for the second processor 14,
Of the above, the trace information is read from the storage area corresponding to the read storage area address. In step 156, it is determined whether or not the number added as the identification information to the read trace information is two or more older (smaller in value) than the previously set number of the previous trace information of the second processor 14.

If the determination is affirmative, the read trace information is set to "invalid" in step 158, and the process proceeds to step 162. When the determination is negative, it can be determined that the read trace information is the latest non-output trace information among the trace information of the second processor 14 stored in the trace information storage area 40. In step 160, the number of the identification information of the read trace information is set as the number of the previous trace information of the second processor 14, and the process proceeds to step 162.

In step 162, of the trace information read out as described above, only the trace information which is not set as "invalid" in step 150 or step 158 is arranged in synchronization trace information (the trace information of each of the processors 12 and 14 is arranged in the order of occurrence). Output).

In step 164, it is determined whether the address currently set as the read storage area address is the address of the last storage area of the storage area groups 40A and 40B. If the determination is positive, step 16
Storage area group 4 as the address of the read storage area at 6
Set the address of the first storage area of 0A and 40B,
When the determination is negative, the address of the next storage area is set as the address of the read storage area in step 168, and then the process returns to step 144. After steps 144 to 168 are repeated a predetermined number of times, step 1
If the determination at 44 is affirmative, the trace information edit output process ends.

From the above, the storage area groups 40A, 40B
The trace information of the first processor 12 and the trace information of the second processor 14, which are respectively stored in, can be accurately output in the order in which they are stored in the trace information storage area 40, facilitating the analysis of the cause of a failure or the like. Can be done. Further, since the trace information storage area 40 is provided in the RAM 36, even if the operation of the first processor 12 or the second processor 14 is stopped due to a failure or the like, the trace information of the stopped processor can be read. It is possible.

When storing the trace information in the trace information storage area 40 and reading the trace information from the trace information storage area 40, the first processor 12
Since it is not necessary to transfer the trace information between the second processor 14 and the second processor 14, a large load is applied to both processors at the same time in order to transfer the trace information, and the processing capability of the information processing device 10 is temporarily reduced. It doesn't get lost.

The first processor 12 and the second processor 14 are configured to include a plurality of processors (CPUs) that cooperate to execute one process (for example, to execute a lower layer process in communication or the like). A configuration including a sub processor or the like ) may be used.

Further, in the above description, two processors (first processor 12 and second processor 14) are provided as processing units according to the present invention, and storage area groups 40A and 40B are provided corresponding to the respective processors. As explained in the example,
The number of processors and the number of storage area groups are not limited to this, and a larger number of processors and storage area groups may be provided, or a plurality of processes (tasks) may be executed in parallel by time division or the like. Alternatively, the processing unit may be configured by a single processor, and the number of storage area groups may be provided according to the number of types of trace information generated by the processing unit.

[0043]

As described above, according to the present invention, every time trace information is generated , identification information is added to the generated trace information.
Tray of multiple types with a monotonically increasing number to be added as
Of the plurality of storage areas corresponding to the generated trace information in the trace information storage means, the storage area indicated by the storage position information held in the holding means indicates a monotonically increasing number. Since the trace information added with is stored and the storage position information held in the holding means is updated, the correct order can be determined easily and reliably for multiple types of trace information that occur asynchronously with each other. It has an excellent effect that it can be stored.

[Brief description of drawings]

FIG. 1 is a block diagram showing a schematic configuration of an information processing apparatus according to an embodiment.

FIG. 2 is a functional block diagram showing the functions related to storage and output of trace information among the various functions of the CPUs of the first processor and the second processor of the information processing apparatus, divided into blocks.

FIG. 3 is a flowchart illustrating a trace information storage process executed by the CPU of the first processor.

FIG. 4 is a flowchart illustrating a trace information storage process executed by the CPU of the second processor.

5A is a sequence diagram showing an example of a flow of trace information storage by trace information storage processing, and FIG. 5B is an image diagram showing a trace information storage area in which trace information is stored according to the sequence of FIG. is there.

FIG. 6 is a flowchart illustrating a trace information edit output process.

[Explanation of symbols]

10 Information processing equipment 12 First Processor 14 Second processor 36 RAM 38 Storage location information holding area 40 Trace information storage area

Front page continuation (58) Fields surveyed (Int.Cl. ⁷ , DB name) G06F 11/28-11/34 G06F 9/46 G06F 15/16-15/177 JSTPLUS file (JOIS)

Claims (2)

(57) [Claims] 1. A storage area group composed of a large number of storage areas each capable of storing trace information is converted into a plurality of types of trace information that are asynchronously generated by a plurality of processes being executed in parallel in a processing unit. Trace information storage means provided correspondingly, holding means for holding storage position information indicating in which storage area of a large number of storage areas of the storage area the trace information is stored, and trace information is generated. Each time the trace information is
A monotonically increasing number to be added as a report
A monotonically increasing number generating means for generating each trace information, and each time the trace information is generated, held in the holding means among a large number of storage areas of the storage area group corresponding to the generated trace information in the trace information storage means The monotonically increasing number is stored in the storage area indicated by the stored location information.
An information processing device comprising: trace information storage control means for storing the added trace information and updating the storage position information held in the holding means. 2. A processing unit for executing a plurality of processes in parallel,
A plurality of storage area groups each of which is capable of storing trace information, and a plurality of trace information storage means are provided corresponding to a plurality of types of trace information that are generated asynchronously with each other by the execution of the processing by the processing unit; In an information processing device including: a storage unit that stores storage position information indicating which storage area of the plurality of storage areas to store the trace information, each time trace information is generated, Identification information in the trace information
A monotonically increasing number to be added as a report
Generated for each trace information, among the many storage areas of the storage area group corresponding to the generated trace information of the trace information storage means, in the storage area represented by the storage position information held in the holding means, the monotonic With increasing number
A method for storing trace information, characterized by storing the added trace information and updating the storage position information held in the holding means.