JPH04337849A - Memory dump method - Google Patents

Abstract

PURPOSE:To eliminate a futile dump processing so as to save a memory dump sheet by inputting a command designating the type of a memory to be dumped. CONSTITUTION:When CPU 2 is started, it is judged whether the memory dump sheet is installed on an FDD installation part 7 or not (S1). Usually, it is not installed and therefore OS is loaded from HDD (S2). An operator inputs the command by K/B 10 (S3). OS executes a program instructed by the command (S4). When a fault occurs while S4 is operated, the operator installs the memory dump sheet, starts CPU 2, loads the memory dump program on a standard memory 3 and executes a processing. (S5). The operator inputs the type of the memory becoming the object of dump from K/B 10 (S6). The memory dump program judges whether all the memories are the standard memory 3 or a two ports memory (S7) and dumps only the necessary memories.

Description

[Detailed description of the invention]

0001

[Industrial Application Field] When a failure occurs in a control device using a CPU, the present invention provides a memory dump system that saves the contents of memory at the time of failure for the purpose of investigating the cause of the failure. It is about the method.

0002

2. Description of the Related Art FIG. 2 is a block diagram showing the hardware of a computer, which is the object of the present invention.
is the system board on which the main parts of the computer are mounted, 2 is the CPU in system board 1, 3 is the standard memory (capacity is e.g. 640KB), 4 is the expansion memory (capacity is e.g. 0.5 to 10MB), and 5 is the A disk control section, 6 a hard disk (abbreviated as HDD) mounting section, 7 a flexible disk (abbreviated as FDD) mounting section, 8 a display and K/B control section for controlling a display device and a keyboard, 9
10 is a display device (abbreviated as CRT), and 10 is a keyboard (
(abbreviated as K/B), 11 is a 2-port memory control unit.

Further, 12 is a communication control device, 13 is a CPU in the communication control device 12, 14 is a local memory, 15 is a communication line, 16 is a communication control unit, and 17 is a 2-port memory (
The capacity is, for example, 256 KB), and the display, K/B control unit 8, CRT 9, and K/B 10 constitute interactive input means.

[0004] When this computer is used as a terminal by connecting it to a host computer (not shown) via a communication network, when an operator inputs a terminal emulator startup command from the K/B 10, the command is stored in the standard memory 3. operating system (abbreviated as OS)
loads the terminal emulator program from the HDD 6 to the standard memory 3 and starts it.

[0005] Then, when the activated terminal emulator program writes transmission data and a transmission request to the two-port memory 17, the CPU 13 transmits the transmission data to the communication line 15 via the communication control section 16. Further, when receiving data from the communication line 15, the CPU 13 writes the received data and a reception request to the 2-port memory 17, and
PU2 reads and processes the received data.

In addition, the CPU 13 writes into the 2-port memory 17 communication execution traces of commands/responses used for communication control based on the communication protocol, and statistical information on line errors and communication protocol errors.

[0007] When using this computer for office automation (abbreviated as OA) work, the operator terminates the execution of the terminal emulator program,
Key-in the OA program start command. The OA program is loaded into the standard memory 3 and started. If the OA program requires a large memory capacity for its work, the additional memory 4 is used.

FIG. 3 is a diagram showing memory addresses seen from the CPU 2 in FIG. 2, and the addresses are expressed in hexadecimal. For example, standard memory 3 of 640KB is allocated from address 0 to address 9FFFF, and has a size of 0.5 to 10MB.
The additional memory 4 of is allocated from address 100000,
This area is used for OA program work. 256
KB 2-port memory 17 is from address F80000 to FB
Allocated to address 0000.

FIG. 4 is a flowchart showing the operation of the conventional method. In the figure, S1 to S5 and S
9 indicates each step. When the CPU 2 is started, it is determined whether a memory dump sheet, which is an FDD storing a memory dump program, is attached to the FDD attachment section 7 (S1). In normal operation, the memory dump sheet is not installed, so S2
, and load the OS from the HDD.

[0010] When the OS is started, it waits for command input from the K/B 10 (S2). The operator can use the K/B10 to launch a terminal emulator program or use the OA
Enter the command to start the program (S3)
. The OS reads the program specified by the input command from the HDD 6, inputs it to the standard memory 3, and executes it (S4). Once this execution is complete, the program will wait for command input again.

[0011] Although the storage area in the standard memory 3 can be used as the working memory for executing the terminal emulator program, a wide memory area is required for the working memory for executing the OA program. In some cases, an additional memory 4 is used in addition to the standard memory 3.

[0012] If a failure occurs during the operation of step S4, the operator attaches the memory dump sheet to the FDD mounting section 7 and restarts the CPU 2. At this restart, care is taken to prevent the contents of the standard memory 3, additional memory 4, and 2-port memory 17 from evaporating. Moving from step S1 to S5, a memory dump program is loaded from the memory dump sheet into the standard memory 3 and executed. By executing the memory dump program, standard memory 3, expansion memory 4, 2-port memory 17
The contents of all memories are dumped onto a memory dump sheet (S9) and used for analyzing the fault that has occurred.

[0013]

[Problems to be Solved by the Invention] The problem to be solved is that in the conventional memory dump method as described above, all memory is dumped in step S9, but while the terminal emulator program is running, the additional memory is 4 is not used and there is no need to dump it, but even if a failure occurs while the terminal emulator program is being executed, the entire memory will be dumped, which is wasteful. In other words, if the capacity of the additional memory 4 is 3MB, the capacity of the memory dump sheet is usually 1MB.
Since it is about B, just dumping the additional memory 4 will save 3
This method requires multiple memory dump sheets, which leads to problems such as wasting a lot of time.

The present invention has been made to solve the above problems, and an object of the present invention is to provide a memory dump method that dumps only the memory contents necessary for failure analysis.

[0015]

[Means for solving the problems] A memory device according to the present invention
The dump method is to load the memory dump program from the memory dump sheet and start its execution.
The main feature is that a command indicating which memory contents should be dumped is input using an interactive input device. Prior art related to memory dumps include Japanese Patent Application Laid-Open No. 64-25255 and Japanese Patent Application Laid-Open No. 2-28.
There are methods disclosed in Japanese Patent Application Laid-open No. 2838 and Japanese Patent Application Laid-Open No. 2-280239, but these prior arts do not have a method for inputting a command from an operator each time a memory dump program is executed, and the method of the present invention does not have a method. However, the problem remains that they are unable to adapt well to changes in the situation.

[0016]

[Operation] In the present invention, the operator reads the memory dump program from the memory dump sheet, loads the program into a specific area of standard memory, and then starts the memory dump program. By specifying the type, you can dump only the memory contents necessary for investigating the cause of the failure.

[0017]

[Embodiment] FIG. 1 is a flowchart showing an embodiment of the present invention. In the figure, the same reference numerals as in FIG. 4 indicate the same or corresponding steps, and S6, S7, and S8 are added in this embodiment. The following steps are shown below. Note that in the same steps as in FIG. 4, the same or equivalent operations as those in the conventional memory dump method are executed, so the explanation thereof will be omitted here.

When step S5 is executed, a display is displayed on the CRT 9 to wait for input of the type of memory to be dumped. The operator inputs the memory type to be dumped from the K/B 10 (S6). Memory dump program is K/
Determine whether the memory type input from B10 is all memory, standard memory or 2-port memory (S7)
. Then, the process moves to step S8 or step S9 according to this determination result.

Although only the contents of the standard memory 3 and the two-port memory 17 are dumped in step S8, the dump of the additional memory 4 is no longer necessary to analyze a failure that occurs during execution of the terminal emulator program.

Note that memory dump programs may be created for each type of memory to be dumped, and which program is to be executed may be specified in step S6.

[0021]

[Effects of the Invention] As explained above, the memory dump method of the present invention allows the operator to specify and select the memory area to be dumped in the memory dump program, so that trouble analysis is possible. This method has advantages such as being able to dump only the memory contents necessary for the process, eliminating unnecessary dump processing, and saving memory dump sheets.

[Brief explanation of the drawing]

FIG. 1 is a flowchart showing one embodiment of the present invention.

FIG. 2 is a block diagram showing the hardware of a computer that is the subject of the present invention.

FIG. 3 is a diagram showing memory addresses seen from the CPU in FIG. 2;

FIG. 4 is a flowchart showing operations in a conventional method.

[Explanation of symbols]

3 Standard memory 4. Expansion memory 17 2-port memory

Claims (2)

[Claims] [Claim 1] When transmitting and receiving data via a communication control device as a terminal emulator, a CPU uses a standard memory and a 2-port memory serving as an interface with the communication control device to control office automation. In the memory dump method when the above CPU uses the above standard memory and additional memory, if the need for memory dump is recognized, a memory dump sheet, which is a non-volatile storage device for memory dump, is installed. a step in which the CPU loads the memory dump program from the memory dump sheet into the standard memory; a step in which the CPU starts executing the memory dump program; a step of instructing a memory range to be dumped to the memory dump sheet through an interactive input means between the CPU and the operator; A memory dump method comprising the step of dumping to a dump sheet. [Claim 2] Multiple types of memory dump programs are created according to the memory range to be dumped and stored in a memory dump sheet, and when the CPU is restarted, all of the multiple types of memory dump programs are transferred to the standard memory. 2. A memory dump method according to claim 1, wherein the memory dump program is loaded into a memory dump program and an operator instructs which memory dump program to execute through interactive input means.