JPS63310040A - Data sampling system - Google Patents

Abstract

PURPOSE:To sample all data on a main storage device at the production of an error by outputting information including hardware information of a system control area lost by the application of IPL of an error analysis program by means of the error analysis program. CONSTITUTION:A data in a software operating area (SUA) 102 is outputted to an external device such as a magnetic tape device. Then a diagnostic instruction is used to transfer a data in a system control area (SCA) 103 to the SUA 102. The data in the SCA 103 transferred to the SUA 102 is outputted to an external storage device. Then the data outputted to an external storage device is edited and the result is outputted to a line printer or the like.

Description

[Detailed Description of the Invention] [Field of Industrial Application] The present invention relates to a data collection method in a computer system that uses a part of the memory area of a main storage device as a system control area dedicated to hardware, and in particular, The present invention relates to a data collection method for collecting hardware information in a system control area when a failure occurs.

[Conventional technology]

In recent computer systems, hardware dedicates a part of the memory area of the main storage device to a dedicated system control area for the purpose of system control.
It has come to be used as an area. for example,
In computer systems equipped with the extended Lee Yarnel system, unit control word (UCW), logical control unit word (LCUW)
Hardware information for input/output control, such as the following, is stored in a predetermined area of the system control area of the main storage device.

In such computer systems, the system control area of the main memory is an area dedicated to hardware, and software controls this system control area.
The contents of the area are not directly referenced. Even if software attempts to access this system control area, it will be treated as an access exception and will not be visible to the software. In other words, the software can refer to the memory area of the software-used area other than the system control area of the main memory, and although the system control area is part of the main memory, it is not visible to the software. It is like that.

By the way, when a serious failure occurs in a computer system that causes the system to go down, it is common to collect memory information at the time of the system failure (memory dump) in order to analyze the cause of the failure. It will be done.

In computer systems that use part of the memory area of the main storage device as a system control area dedicated to hardware, hardware information is stored in the system control area, making it easy to analyze the cause of abnormal operation. This requires a memory dump, including the information stored in this system control area.

However, the contents of this system control area are normally not read by software, so for example,
As described in the publication, only at the time of memory dump,
Opening the system control area to software so that the contents of the main memory including the contents of the system control area can be read,
Perform a memory dump and collect data for analyzing the cause of abnormal operation.

[Problem that the invention seeks to solve]

By the way, a memory dump for collecting data for analyzing the cause of an abnormal operation is performed by executing a failure analysis program (memory dump program). usually,
The failure analysis program does not reside in main storage;
In order to run the failure analysis program, first perform the initial program load (IPL) of the failure analysis program.
) must be loaded onto main memory.

In a computer system that uses part of the memory area of the main storage device as a system control area dedicated to hardware, various information necessary to control the computer system is stored inside the system control area. However, some of this information is initialized by the hardware when an IPL operation is executed.

Therefore, in computer systems that use part of the memory area of the main storage device as a system control area dedicated to hardware, the system control area is opened and the system control area where hardware information is stored is - There was a problem in that simply by making it possible to read the area information, a failure analysis program that performs a memory dump could not collect the current hardware information of the system control area at the time of failure.

The present invention has been made to solve the above problems.

An object of the present invention is to eliminate the current state of the system control area that is lost due to IPL of a failure analysis program in a computer system that uses a part of the memory area of the main storage device as a system control area dedicated to hardware. The goal is to output information, including hardware information, using a failure analysis program.

The above and other objects and novel features of the present invention will become apparent from the description of this specification and the accompanying drawings.

[Means for solving problems]

In order to achieve the above object, in the present invention, a part of the memory area of the main storage device is dedicated to a hardware system.
In a computer system used as a control area, a buffer area is set in a memory area that is not initialized by a reset operation, and data in a predetermined area of the system control area that is initialized by a reset operation is transferred to the buffer area. After the data is transferred to the system control area, the data in the buffer area is read by a failure analysis program, and hardware information data in the system control area is output.

[Effect]

According to the above means, in a computer system that uses a part of the memory area of the main storage device as a hardware-dedicated system control area, a buffer area is set in a memory area that is not initialized by a reset operation. After data in a predetermined area of the system control area initialized by a reset operation (IPL operation) is transferred to the buffer area, the data in the buffer area is read by a failure analysis program, and the system control area is read out by a failure analysis program. Outputs area hardware information data.

When a failure occurs in a computer system, the failure analysis programmer uses IPLL to perform a memory dump and restore the system.
Outputs memory data, including hardware information data in the control area, but here, first,
Among the information inside the system control area, information in a predetermined area initialized by a reset operation (IPL operation) is saved to a buffer area. Thereafter, the failure analysis program reads the data in the buffer area and outputs the hardware information data in the system control area. Therefore, among the information inside the system control area, hardware information that is normally lost during the IPL operation of the failure analysis program can be collected by the failure analysis program.

〔Example〕

Hereinafter, one embodiment of the present invention will be specifically described using the drawings.

FIG. 1 shows the configuration of a memory area in a main storage device of a computer system in which the present invention is implemented. Main storage device I (MS) 101 in this computer system
, the memory area is the operating system (O8
) and other software used by software, and the system control area (SUA) 102 used by hardware for hardware control such as input/output control.
It is divided into two areas (SCA) 103 and used.

This system control area 103 is an area dedicated to hardware, and is normally not directly referenced by software.

The system control area 103 stores input/output control information, which is hardware information. Due to the hardware configuration, a part of the predetermined area on the system control area 103 where this input/output control information is stored is initialized when a reset operation is executed by the initial programmer 110-code (IPL), etc. It is configured so that In addition, this system control area 10
3, a buffer area 105 is provided in an area that is not initialized by the reset operation. This buffer area 105 is secured at the time of system startup, and is used as the system control area 10.
The start address 106 is registered in the directory 104 of No. 3.

Although not shown here, this computer system is equipped with a central processing unit, an input/output processing unit, an input/output device, etc., as well as a service processor that has a console function, a function to collect hardware information about each part of the system, etc. ing.

When a failure occurs in a computer system and a memory dump is taken, store status processing is executed by console operation of the service processor prior to IPLing the failure analysis program. This store status processing transfers the information in the area initialized by IPL among the hardware information in the system control area to the buffer area.

FIG. 2 shows a processing sequence when store status processing is executed in a computer system. When execution of store status processing is instructed by the service processor (SVP) console operation, the service processor (SVP) executes the central processing unit (CPU).
) to start store status processing. When the CPU receives activation of the store status process, the CPU executes the operation of the store status process under the control of a microprogram stored in the control storage device (O8) of the CPU. In this store status processing, as shown in Figure 2, the central processing unit (CPU) stores hardware control information such as general-purpose registers and control registers in the CPU into a predetermined area on the software usage area (SUA). and then I
The service processor (
The process ends by reporting the completion to SVP).

By performing store status processing through such service processor console operations, the system control initialized by the reset operation is
Data in a predetermined area of the area is transferred to the buffer area.

FIG. 3 is an explanatory diagram showing in detail an example of a process in which information in the system control area is saved to a buffer area as a part of the store status process.

In Figure 3, the system control area (S
cA) Among the information stored in 306, the information stored in area A, territory area, and area C shall be initialized when IPL is executed.

The microprogram of the central processing bag (CPU) first reads the start address 302 of area A and the start address of the buffer area from the system control area directory 301 stored in a predetermined area on the system control area. 305 and transfers the information stored in area A to buffer area 310. Similarly, the start address 303 of area B and the start address 304 of area C are obtained, respectively, and the information stored in each area is transferred to the buffer area.

In this way, the information in the predetermined area of the system control area that is initialized by the IPL operation (reset operation) is saved to the buffer area. It is possible to output information in the system control area, including information saved in the buffer area.

Next, a method for the failure analysis program to output information inside the system control area including evacuation information will be explained.

FIG. 4 shows an example of a diagnostic command used to read data inside the system control area in a fault analysis program. Reference numeral 401 represents the instruction format, and the first operand address obtained by adding the contents of the pace register specified by B1 and displacement D1 gives the start address of the first control table 402. The first control table is 8
It consists of bytes, L (bits 24 to 31) is data specifying the transfer byte, and L+1 is the number of transfer bytes. MS address 1 (bits 33 to 63) provides the start address of the second control table 403. The second control table consists of 8 bytes and consists of MS address 2 (bits 1 to 2).
31) indicates the start address of the area where data read from the system control area is stored, and SCA
The inner address (bits 33 to 63) indicates the start address of the area within the system control area from which data is read.

By using such diagnostic instructions in a fault analysis program, up to 256 bytes of data at a time can be transferred from the System Control Area (SCA) to the Software Usage Area (SUA).

FIG. S is a flowchart outlining the processing procedure when the failure analysis program outputs a memory dump.

The failure analysis program uses the diagnostic instructions as explained in FIG. 4 to output a memory dump through the following processing steps.

This memory dump output processing will be explained with reference to FIG. That is, in step S1, data in the software usage area (SUA) is first output to an external storage device such as a magnetic tape device.
At step 2, data in the system control area (SCA) is transferred to the SUA using diagnostic instructions such as those described in FIG. Then, in the first step S3, the data in the SCA transferred to the SUA is output to an external storage device. Thereafter, in step S4, the data output to the external storage device is edited and output to a line printer or the like. By executing such processing steps 81 to S4, the failure analysis program outputs a memory dump.

As described above, according to this embodiment, in a computer system that uses a part of the memory area of the main storage device as a hardware-dedicated system control area, a buffer is stored in a memory area that is not initialized by a reset operation. After setting the area and transferring data in a predetermined area of the system control area initialized by a reset operation to the buffer area, the failure analysis program reads the data in the buffer area and transfers the data to the buffer area. - Since the area's hardware information data is output, it can be used at the time a failure occurs, including information on the current system control area before it is initialized at IPL, without adding any special hardware. This has the advantage that all the data on the main memory can be output by the failure analysis program.

The present invention has been specifically explained above based on examples, but
It goes without saying that the present invention is not limited to the embodiments described above, and can be modified in various ways without departing from the spirit thereof.

〔Effect of the invention〕

As described above, according to the present invention, in a computer system that uses a part of the memory area of the main storage device as a hardware-dedicated system control area, the failure analysis program is lost by IPL. Information including current hardware information in the system control area can be output by the failure analysis program. Therefore, the IPL information within the system control area is usually
It is possible to collect data including hardware information that would otherwise be lost during operation, and all information (data) on the main storage device at the time of a failure can be collected, making it easy to analyze the cause of a failure. That's quite an advantage.

[Brief explanation of the drawing]

Figure 1 is an explanatory diagram of the configuration of the memory area in the main storage of a computer system. Figure 2 is an explanatory diagram showing the processing sequence when store status processing is executed in the computer system. Figure 3 is an explanatory diagram of the system・An explanatory diagram showing an example of the process of saving information in the control area to the buffer area. FIG. 4 explains an example of the diagnostic command used by the failure analysis program to read data inside the system control area. The explanatory diagram, FIG. 5, is a flowchart showing an outline of the processing procedure of the failure analysis program. In the figure, 101...main storage device (MS), 102...
Software usage area (SUA), 103 . . . system control area (SCA).

Claims (1)

[Claims] 1. In a computer system that uses part of the memory area of the main storage device as a hardware-dedicated system control area, a buffer area is set in a memory area that is not initialized by a reset operation, and is initialized by a reset operation. After transferring data in a predetermined area of the system control area to the buffer area, the data in the buffer area is read by a failure analysis program, and hardware information data in the system control area is output. data collection method.