JP2008242999A - Information processing apparatus and memory dump method - Google Patents

Abstract

A program can be restarted quickly while dumping memory.
An information processing apparatus acquires an access request to a memory during a program activation process, area specifying information for specifying a storage area to be accessed included in the acquired access request, and an access request acquisition order Is stored, and after the start processing of the program is finished, the execution of the program being executed is stopped, and after the execution of the program is stopped, the area designation information stored in the access pattern storage unit and the acquisition order are read and the acquisition In order, the contents of the storage area indicated by each area designation information are dumped, and the program is started during the dump.
[Selection] Figure 1

Description

  The present invention relates to an information processing apparatus and a memory dump method.

In an information processing apparatus, the contents of a memory are dumped when a program failure occurs. In the system described in Patent Literature 1, when a failure occurs in the operating system, after the memory area necessary for loading the operating system is dumped, the operating system is reloaded and the memory area to be used is already dumped. If not, it dumps before use, and dumps the memory area that has not been dumped in order to shorten the time until the program is restarted.
JP-A-10-333944

  In the system described in Patent Document 1, the operating system startup process and the memory dump process are executed in parallel. However, in the system described in Patent Document 1, the memory area accessed during the operating system startup process is not always dumped. When an area that has not been dumped is accessed during startup processing, dump processing is performed each time access is performed, and memory is accessed after the dump processing, and the startup processing of the operating system takes time. End up.

  The present invention has been made in view of such a background, and an object of the present invention is to provide an information processing apparatus and a memory dump method that can quickly restart a program while dumping a memory.

  A main invention of the present invention for solving the above-described problem is an information processing apparatus, wherein a CPU, a memory, and an access to the memory from the application program during a startup process of the application program executed by the CPU An access request acquisition unit that acquires a plurality of requests; an area designation information that is information that specifies an access target storage area included in the access request; and an access pattern storage unit that stores an acquisition order of the access requests; After the start processing of the application program, a stop processing unit that stops the execution of the application program that is being executed, and after the execution of the application program is stopped, the area designation information and the acquisition order are read from the access pattern storage unit, Specify each area in the order of acquisition And dump processor to dump the contents of the storage area specified by the broadcast, during the dump by the dump processing unit, and be provided with a restart processing unit that activates the application program.

  According to the present invention, it is possible to quickly restart a program while dumping a memory.

  FIG. 1 is a hardware configuration diagram of an information processing apparatus 10 according to an embodiment of the present invention. The information processing apparatus 10 according to the present embodiment is assumed to be a general computer such as a personal computer or a workstation. As illustrated in FIG. 1, the information processing apparatus 10 includes a physical CPU 11, a physical memory 12, a storage device 13, an input device 14, and an output device 15. The storage device 13 is, for example, a hard disk drive, a CD-ROM drive, or a flash memory that stores various programs and data. The physical CPU 11 performs various arithmetic processes and controls the information processing apparatus 10. The physical CPU 11 implements various functions by reading the program stored in the storage device 13 into the physical memory 12 and executing it. The input device 14 is a keyboard, mouse, touch panel, microphone, or the like that accepts input of information, and the output device 15 is a display, printer, speaker, or the like that outputs information.

  In the information processing apparatus 10 of the present embodiment, various resources such as the physical CPU 11, the physical memory 12, and the storage device 13 are logically divided, and different resources are assigned to each logical partition to realize a virtual computer. Server virtualization (hereinafter simply referred to as virtualization) is performed. In the following description, the physical CPU 11 virtualized in each logical partition is referred to as a virtual CPU, and the virtualized physical memory 12 is referred to as a virtual memory. On the virtualized computer, an operating system (called a guest operating system, hereinafter abbreviated as a guest OS) operates.

  FIG. 2 shows a logical configuration of the information processing apparatus 10 that has been virtualized. As shown in the figure, in the information processing apparatus 10 of this embodiment, a guest OS 20 and a hypervisor 30 are executed. In the physical memory 12, a hypervisor area 121 that is a storage area allocated to the hypervisor 30 and a guest OS area 122 that is a storage area allocated to the guest OS 20 are secured.

  Arithmetic commands and virtual memory access requests from the guest OS 20 (hereinafter referred to as memory access requests) are converted by the hypervisor 30 into arithmetic commands to the physical CPU 11 and physical memory 12 access requests. By such conversion processing by the hypervisor 30, virtualization in the information processing apparatus 10 is realized. In the following description, only processing related to the memory access request will be described.

  FIG. 3 is a functional block diagram of the information processing apparatus 10 according to the present embodiment. As shown in the figure, the information processing apparatus 10 includes a guest OS 20 and a hypervisor 30, and the guest OS 20 includes a memory access request transmission unit 21 and a failure notification unit 22. The hypervisor 30 includes an access pattern table 31, a dump Output table 32, size management table 33, access pattern table creation unit 41, dump output table creation unit 42, size management table registration unit 43, memory access request reception unit 44, memory access processing unit 45, failure detection unit 46, guest OS A stop processing unit 47, a guest OS activation processing unit 48, and a dump processing unit 49 are provided.

  The memory access request transmission unit 21 transmits a memory access request designating a storage area of the virtual memory to be accessed to the hypervisor 30. In this embodiment, the storage areas of the physical memory 12 and the virtual memory are addresses on the physical memory 12 (hereinafter referred to as physical addresses) or addresses on the virtual memory (hereinafter referred to as virtual addresses), and the number of blocks. It shall be represented by.

  When a failure occurs in the guest OS 20, the failure notification unit 22 notifies the hypervisor 30 to that effect. The failure notification unit 22 notifies the failure by, for example, an interrupt to the virtual CPU.

The size management table 33 manages setting values for various data sizes. FIG. 4 shows a configuration example of the size management table 33. As shown in the figure, the size management table 33 manages an access pattern size 331, a core size 332, and a block size 333. The access pattern size 331 indicates the size of the access pattern table 31 described later. The core size 332 indicates the size of a storage area required for starting the guest OS 20. The core size 332 may be, for example, a data size of a loader program for reading an application program (hereinafter referred to as an OS program) for realizing the guest OS 20 on the physical memory 12 or a data size of the OS program itself. it can. The block size 333 is the size of the blocks constituting the physical memory 12 and the virtual memory. In this embodiment, it is assumed that the physical memory 12 and the virtual memory are managed in units of blocks.
The size management table registration unit 43 receives input of the access pattern size 331, core size 332, and block size 333 from the user via the input device 14, and registers the received values in the size management table 33.

  The access pattern table 31 stores the area of the physical memory 12 accessed from the guest OS 20 and the access order (hereinafter referred to as an access pattern) during the startup process of the guest OS 20. A configuration example of the access pattern table 31 is shown in FIG. In the example of FIG. 5, the access pattern table 31 stores the physical addresses of the accessed blocks in the order of access. In the present embodiment, after starting the guest OS 20, every time a memory access request is received from the guest OS 20, a physical address indicating a block to be accessed is added to the end of the access pattern table 31, and the same. When a plurality of blocks are accessed based on the memory access request, each physical address for each block is stored in the access pattern table 31.

  In the present embodiment, the period during which the guest OS 20 is in the startup process refers to a predetermined number of blocks (obtained based on the access pattern size 331 of the size management table 33) after the guest OS 20 starts up. The period until access is made. As a period during the activation process, for example, a period from when the guest OS 20 is activated until the number of operation instructions to the virtual CPU transmitted from the guest OS 20 exceeds a predetermined number, or after the guest OS 20 is activated. It is also possible to employ a period until a process for accepting user login is started.

  The access pattern table creation unit 41 creates the access pattern table 31. Details of the process of creating the access pattern table 31 will be described later.

The dump output table 32 manages the storage area in the guest OS area 122 that has been dumped. In the present embodiment, the dump output table 32 manages whether or not each block is dumped. A configuration example of the dump output table 32 is shown in FIG. As shown in the figure, the dump output table 32 stores an output flag 322 indicating whether or not the contents of the block have been dumped in association with the physical address 321 indicating the head of the block. The output flag 322 is set to either “complete” indicating that the contents of the block have already been dumped or “incomplete” indicating that the block has not been dumped yet.
The dump output table creation unit 42 creates the dump output table 32 at the start of dump processing related to the guest OS area 122. Details of the dump output table creation process will be described later.

The memory access request reception unit 44 receives the access request transmitted by the memory access request transmission unit 21 of the guest OS 20.
The memory access processing unit 45 performs data input / output processing for the physical memory 12 in response to the received memory access request. The flow of input / output processing for the physical memory 12 by the memory access processing unit 45 is shown in FIG. FIG. 7 shows the flow of input / output processing when dump processing of the guest OS area 122 is not performed, and input / output processing to the physical memory 12 during dump processing will be described later.

  As shown in FIG. 7, the memory access processing unit 45 acquires the virtual address and the number of blocks included in the memory access request (S501), and converts the acquired virtual address into a physical address (S502). The conversion from the virtual address to the physical address is obtained, for example, by adding the virtual address to the top physical address of the guest OS area 122. The memory access processing unit 45 initializes the variable N to 0 (S503), and ends the process when N becomes the number of blocks (S504: YES).

  If N is not the number of blocks (S504: NO), the memory access processing unit 45 adds a value obtained by multiplying N by the block size 333 of the size management table 33 to the physical address to obtain a block address. (S505). The memory access processing unit 45 inputs / outputs data to / from the block indicated by the block address (S506), and increments N (S507).

  As described above, the memory access processing unit 45 inputs / outputs data to / from the physical memory 12 in response to a memory access request.

  The failure detection unit 46 detects a failure notification from the failure notification unit 22 of the guest OS 20. Note that the failure detection unit 46 may monitor the operation of the guest OS 20 and detect a failure of the guest OS 20 by, for example, processing timeout.

  The guest OS stop processing unit 47 stops the execution of the guest OS 20 when the failure detection unit 46 detects a failure of the guest OS 20. For example, the guest OS stop processing unit 47 may transmit a signal instructing to end the program to the guest OS 20, or the hypervisor 30 may not process a calculation instruction from the guest OS 20 to the virtual CPU. Thus, the guest OS 20 may be prevented from proceeding by making the guest OS 20 appear to have stopped the virtual CPU.

  The guest OS activation processing unit 48 activates the guest OS 20 after execution of the guest OS 20 is stopped by the guest OS termination processing unit 47. That is, the guest OS 20 is restarted by the guest OS stop processing unit 47 and the guest OS startup processing unit 48.

  The dump processing unit 49 dumps the contents of the guest OS area 122 corresponding to the stopped guest OS 20 after the guest OS stop processing unit 47 stops the execution of the guest OS 20. In the present embodiment, dumping the contents of the storage area means writing the data stored in the storage area to the storage device 13 or outputting it to the output device 15. The data output by the dump is used, for example, for analyzing a program defect at the time of a program failure.

Hereinafter, the details of the dump process when the guest OS 20 fails will be described.
FIG. 8 is a diagram for explaining the flow of dump processing of the guest OS area 122 in the information processing apparatus 10 of this embodiment.

First, the hypervisor 30 activates the guest OS 20 (S601), and creates the access pattern table 31 (S602). FIG. 9 shows a flow of processing for creating the access pattern table 31.
The access pattern table creation unit 41 creates an empty access pattern table 31 (S621). Next, the access pattern table creation unit 41 multiplies the number of records registered in the access pattern table 31 by the block size 333 of the size management table 33 to obtain a registered size (S622). If the registered size is equal to or larger than the access pattern size 331 of the size management table 33 (S623: YES), the access pattern table creation unit 41 ends the process.
When the registered size is smaller than the access pattern size 331 (S623: NO), when the memory access request receiving unit 44 receives the memory access request (S624), the memory access processing unit 45 is included in the memory access request. The virtual address is converted into a physical address (S625). The access pattern table creation unit 41 determines whether or not the converted physical address is registered in the access pattern table 31 (S626). If the physical address is not registered (S626: NO), the physical address is used as the access pattern. It is added to the end of the table 31 (S627), and the processing from step S622 is repeated.
The access pattern table 31 is created as described above.

Next, the hypervisor 30 creates a dump output table 32 (S603). The flow of the creation process of the dump output table 32 is shown in FIG.
The dump output table creation unit 42 sets the quotient obtained by dividing the size of the guest OS area 122 by the block size 333 of the size management table 33 as the number of blocks (S641), and sets the head address of the guest OS area 122 as the block address ( S642). The dump output table creation unit 42 counts the number of records registered in the dump output table 32, and ends the process if the counted number of records is equal to or greater than the number of blocks (S643: YES).
If the number of records in the dump output table 32 is smaller than the number of blocks (S643: NO), the dump output table creation unit 42 additionally registers a block address in the dump output table 32 (S644), and an output flag 322 corresponding to the block address. Is set to “incomplete” (S645). The dump output table creation unit 42 adds the block size 333 to the block address (S646), and repeats the processing from step S643.
The dump output table 32 is created as described above.

  After the guest OS 20 is started and the access pattern table 31 and the dump output table 32 are created as described above, when the failure detection unit 46 detects a failure notification (S604), the guest OS stop processing unit 47 displays the guest OS 20 The virtual CPU provided to the server is stopped (S605), and the execution of the guest OS 20 is stopped.

  When the execution of the guest OS 20 is stopped, the hypervisor 30 performs a dump process on a storage area (hereinafter referred to as an access pattern area) in the guest OS area 122 accessed during the activation of the guest OS 20 activated in step S601. (S606). The flow of dump processing for the access pattern area is shown in FIG.

The dump processing unit 49 sets the activation flag to “false” (S661), sets the dump size to “0” (S662), and performs the following processing for each physical address stored in the access pattern table 31. .
When the activation flag is “false” and the dump size is equal to or larger than the core size 332 of the size management table 33 (S663: YES), the guest OS activation processing unit 48 activates the guest OS (S664), The dump processing unit 49 sets the activation flag to “true” (S665).
The dump processor 49 reads the output flag 322 corresponding to the physical address from the dump output table 32 (S666), and if the read output flag 322 is “incomplete” (S667: YES), the block corresponding to the physical address. Dump the contents of. Specifically, the dump processing unit 49 dumps data for the block size 333 of the size management table 33 from the physical address (S668).
After dumping the contents of the block, the dump processing unit 49 sets “complete” to the output flag 322 of the dump output table 32 corresponding to the physical address (S669).

  By repeating the above processing, the dump processing unit 49 dumps a block corresponding to each physical address stored in the access pattern table 31. When the total amount of data dumped by the dump processing unit 49 exceeds the core size 332, the guest OS activation processing unit 48 activates the guest OS 20.

  As described above, the hypervisor 30 of the present embodiment can dump the contents of the block according to the access pattern of the block stored in the access pattern table 31. Therefore, the hypervisor 30 can preferentially dump the block accessed when the guest OS 20 is first started when the guest OS 20 is restarted. Since the block accessed when the guest OS 20 is started is highly likely to be accessed when the guest OS 20 is restarted, the contents of the block that is highly likely to be accessed when the guest OS 20 is restarted are processed by the above processing. A dump can be made prior to the startup of the OS 20. Therefore, the dump process and the boot process of the guest OS 20 can be performed in parallel.

  However, when the guest OS 20 accesses a block different from that at the time of initial startup, or when the startup processing of the guest OS 20 is executed faster than the dump processing, the dump processing is finished. An access request from the guest OS 20 may be transmitted to a block that does not exist.

  In this case, the hypervisor 30 of this embodiment is configured to perform input / output processing after dumping the block that is the target of the access request if it has not been dumped yet. The flow of input / output processing to the memory during dumping is shown in FIG.

  The process of FIG. 12 is obtained by adding the processes of steps S621 to S623 to step S506 of the process shown in FIG. That is, the memory access processing unit 45 calculates a block address by adding a value obtained by multiplying N and the block size 333 of the size management table 33 to the physical address (S505), and then outputs an output flag corresponding to the block address. 322 is read from the dump output table 32, and it is determined whether or not the read output flag 322 is “incomplete” (S681). When the output flag 322 is “incomplete” (S681: YES), the dump processing unit 49 dumps the contents of the block indicated by the block address (S682), and sets the output flag 322 corresponding to the block address to “completed”. The setting is made (S683).

  In this way, when the hypervisor 30 receives an access request for a block that has not been dumped during dump processing, the hypervisor 30 dumps the contents of the block and then accesses the physical memory 12 according to the access request. Processing can be performed. Therefore, even if the dump process and the startup process of the guest OS 20 are executed in parallel, the contents of the guest OS area 122 at the time of the previous stop of the guest OS 20 can be reliably dumped.

  When the access pattern area dump process is completed as described above, the hypervisor 30 next dumps the storage area of the guest OS area 122 that has not yet been dumped (hereinafter referred to as an undumped area). (S607). FIG. 13 is a diagram showing the flow of dump processing for an undumped area. For each record stored in the dump output table 32, the dump processing unit 49 dumps the contents of the block corresponding to the physical address if the record output flag 322 is “incomplete” (S701: YES) ( S702), “complete” is set in the output flag 322 (S703). In this way, the dump processing unit 49 performs a dump process for an undumped area.

  As described above, according to the information processing apparatus 10 of this embodiment, the access pattern of the block (storage area) accessed when the guest OS 20 is first started is stored, and when the guest OS 20 is restarted, the access pattern is accessed. The guest OS area 122 can be dumped according to the pattern. Generally, the access pattern to the virtual memory is often almost the same in the startup process of the operating system and other application programs. Therefore, by performing dumping according to the access pattern at startup, it can be accessed when the guest OS 20 is restarted. It is possible to preferentially dump high-performance blocks. Therefore, the restart of the guest OS 20 can be started more quickly than in the case where the guest OS 20 is not started until all of the dump processing is completed as in the prior art. As a result, the downtime of the service provided by the guest OS 20 can be shortened. In particular, when the guest OS area 122 allocated to the guest OS 20 has a large capacity as in recent years, it is effective because it takes a long time to dump all the contents of the guest OS area 122. .

  In the information processing apparatus 10 according to the present embodiment, when a block that has not been dumped is accessed while the guest OS area 122 is being dumped, the block access process is performed after dumping the contents of the block. Therefore, when the guest OS 20 is restarted, the guest OS area 122 can be reliably dumped even when an access is made to a block different from that at the first startup.

  On the other hand, in the above processing, when the dump of the block to be accessed is incomplete, the contents of the block are dumped, and data input / output is performed after the dump for the block is completed. Therefore, the information processing apparatus 10 according to this embodiment can preferentially dump blocks that are likely to be accessed when the guest OS 20 is restarted as described above. The possibility of accessing an incomplete block can be reduced. Therefore, it is possible to reduce overhead related to the dump processing of the guest OS area 122 and suppress the processing load of the information processing apparatus 10. In addition, since the guest OS 20 can also be activated quickly, the downtime of the service provided by the guest OS 20 can be shortened.

  In the information processing apparatus 10 of the present embodiment, only one guest OS 20 is executed. However, the present invention is not limited to this, and a plurality of guest OSs 20 may be executed. In this case, the hypervisor 30 is provided with an access pattern table 31, a dump output table 32, a size management table 33, and the like for each guest OS 20.

  In the information processing apparatus 10 according to the present embodiment, the physical address pattern is stored in the access pattern table 31, but a virtual address pattern may be stored. In this case, it is possible to cope with a case where the area on the physical memory 12 allocated to the guest OS area 122 is changed.

  In the present embodiment, the physical address is stored for each block in the access pattern table 31, but the physical address and the number of blocks may be stored in association with each other. In this case, the virtual address included in the memory access request is converted into a physical address, and the physical address and the number of blocks included in the memory access request are associated with each other and registered in the access pattern table 31.

  In the present embodiment, the physical address of the block is stored in the access pattern table 31 in the order of access from the guest OS 20, but the access order and the accessed area may be stored. For example, the access pattern table 31 may store a number indicating the access order and a physical address in association with each other. In this case, the access pattern table 31 can be realized as a table in a relational database management system that does not manage the storage order of records.

  In the present embodiment, the dump output table 32 manages the output flag 322 in association with the physical address. However, the output flag 322 may be managed in association with the virtual address.

  In this embodiment, the access pattern is stored in units of blocks. However, the present invention is not limited to this, and may be stored in units of variable-length storage areas. In this case, for example, the physical address and the data length are registered in the access pattern table 31 in association with each other.

  Although the present embodiment has been described above, the above embodiment is intended to facilitate understanding of the present invention and is not intended to limit the present invention. The present invention can be changed and improved without departing from the gist thereof, and the present invention includes equivalents thereof.

It is a hardware block diagram of the information processing apparatus 10 which concerns on one Embodiment of this invention. It is a figure explaining the logical structure of the information processing apparatus 10 virtualized. It is a functional block diagram of information processor 10 concerning this embodiment. 4 is a diagram illustrating a configuration example of a size management table 33. FIG. 4 is a diagram illustrating a configuration example of an access pattern table 31. FIG. It is a figure which shows the structural example of the dump output table. FIG. 4 is a diagram showing a flow of input / output processing for a physical memory 12. FIG. 6 is a diagram for explaining a flow of dump processing in a guest OS area 122. It is a figure which shows the flow of a creation process of the access pattern table. It is a figure which shows the flow of a creation process of the dump output table. It is a figure which shows the flow of the dump process about an access pattern area | region. It is a figure which shows the flow of the input-output process to the memory in dumping. It is a figure which shows the flow of the dump process about an undump area | region.

Explanation of symbols

10 Information processing apparatus 11 Physical CPU
12 Physical memory 13 Storage device 14 Input device 15 Output device 20 Guest OS
DESCRIPTION OF SYMBOLS 21 Memory access request transmission part 22 Failure notification part 30 Hypervisor 31 Access pattern table 32 Dump output table 321 Physical address 322 Output flag 33 Size management table 331 Access pattern size 332 Core size 333 Block size 41 Access pattern table creation part 42 Dump output table Creation unit 43 Size management table registration unit 44 Memory access request reception unit 45 Memory access processing unit 46 Fault detection unit 47 Guest OS stop processing unit 48 Guest OS activation processing unit 49 Dump processing unit 121 Hypervisor area 122 Guest OS area

Claims (6)

CPU, memory,
An access request acquisition unit that acquires a plurality of access requests to the memory from the application program during the startup process of the application program executed by the CPU;
Area specification information that is information for specifying a storage area to be accessed, included in the access request, and an access pattern storage unit that stores the acquisition order of the access requests;
A stop processing unit for stopping execution of the application program being executed after the start processing of the application program;
After the execution of the application program is stopped, a dump processing unit that reads out the area designation information and the acquisition order from the access pattern storage unit, and dumps the contents of the storage area designated by the area designation information in the acquisition order; ,
During a dump by the dump processing unit, a restart processing unit that starts the application program,
An information processing apparatus comprising: The information processing apparatus according to claim 1,
A core size storage unit that stores a core size that is a size of a storage area necessary for starting the application program;
Each time the dump processing unit dumps the contents of the storage area, it calculates a dump size that is a cumulative value of the size of the dumped storage area,
The restart processing unit starts the application program when the cumulative value is equal to or larger than the core size;
An information processing apparatus characterized by the above. The information processing apparatus according to claim 1,
An access processing unit for inputting / outputting data to / from the storage area specified by the area specifying information included in the access request;
A dump output table that stores a flag value indicating whether or not the contents of the storage area are dumped in association with the area designation information;
With
When the dump processing unit dumps the contents of the storage area, the dump processing unit registers the flag value indicating that the dump is associated with the area designation information in the dump output table,
The access processing unit obtains the flag value of the dump output table corresponding to the area specifying information included in the access request from the dump output table, and the contents of the storage area are dumped by the obtained flag value. Waiting for the contents of the storage area to be dumped if the contents of the storage area have not been dumped,
An information processing apparatus characterized by the above. The information processing apparatus according to claim 1,
A failure detection unit that detects that a failure has occurred in the application program;
The stop processing unit stops the execution of the application program triggered by the detection of the failure;
An information processing apparatus characterized by the above. CPU, memory,
An access request acquisition unit that acquires a plurality of access requests to the memory from the operating system during startup processing of the operating system executed on the virtual machine;
Area specification information that is information for specifying a storage area to be accessed included in the access request, and an access pattern storage unit that stores the access request acquisition order;
A stop processing unit that stops execution of the operating system that is being executed after the start-up process of the operating system;
After the execution of the operating system is stopped, the area designation information and the acquisition order stored in the access pattern storage unit are read, and the contents of the storage areas designated by the area designation information are dumped in the acquisition order. A dump processing unit;
During a dump by the dump processing unit, a restart processing unit for starting the operating system,
An information processing apparatus comprising: A method for dumping memory contents,
A computer including a CPU and a memory
During the startup process of the application program executed by the CPU, a plurality of access requests to the memory from the application program are acquired,
Storing area designation information, which is information for designating a storage area to be accessed, included in the access request, and the access request acquisition order;
After the startup process of the application program is finished, stop the execution of the application program being executed,
After stopping execution of the application program, the area designation information and the acquisition order are read from the access pattern storage unit, and the contents of the storage areas specified by the area designation information are dumped in the read acquisition order,
Launching the application program during dump processing;
A memory dump method characterized by the above.

Images