JPH03154131A - Collecting and editing system for fault processing data - Google Patents

Abstract

PURPOSE:To curtail the hardware quantity and to execute the processing at high speed by extracting the data of various registers required for a fault processing from the hardware information of an arithmetic processor and editing them to a prescribed logical format. CONSTITUTION:When a fault is generated in an arithmetic processor 1, the diagnostic control part 6 of a diagnostic controller 2 brings hardware informa tion of the arithmetic processor 1 to scan-out to a hardware information storage part in the diagnostic controller 2. Thereafter, in accordance with each position information held in a fault processing data editing information storage part 5, the data of each register are extracted from in the hardware signal storage part 3 and stored in a work register 8, and when the fault processing data of a format conforming to a prescribed logical format is completed on the working register 8, it is transferred to a fault processing data storage part 4. In such a manner, the data of various registers required for a fault processing are collected and can be edited to the prescribed logical format at the high speed with a small hardware quantity.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to a diagnostic control device that handles failure processing of an arithmetic processing unit, and in particular, the present invention relates to a diagnostic control device that handles failure processing of an arithmetic processing unit. The present invention relates to a failure processing data collection and editing method for collecting register data from an arithmetic processing unit and editing it into a predetermined logical format.

[Conventional technology]

When a failure occurs in the arithmetic processing unit, the diagnostic control device connected to the arithmetic processing unit executes predetermined failure handling to cause the arithmetic processing unit to retry instructions, etc. In order to do this, the data of various predetermined registers in the arithmetic processing unit that has stopped operating due to the occurrence of a fault is collected in the diagnostic control device, and the data is collected so that it can be recognized by the diagnostic control program that handles the fault processing. , a process of arranging the collected data in a predetermined order, that is, a process of editing it into one piece of data having a predetermined logical format is required.

In order to collect and edit such fault processing data, conventionally, a dedicated interface is provided between the arithmetic processing unit and the diagnostic control unit to individually read the data in multiple registers necessary for fault processing, and the diagnostic control unit performs the diagnosis. This was achieved by executing a control program to sequentially read the data in each register via its dedicated interface, and performing editing processing such as connecting the data.

[Problem to be solved by the invention]

Although it is possible to collect and edit data for fault handling using the conventional method described above, the problem is that the amount of hardware increases because a dedicated interface is required to read the registers required for fault handling. there were. Furthermore, it is necessary to execute several software instructions to edit the read data of each register into a predetermined logical format, making high-speed editing difficult.

The present invention solves these conventional problems, and its purpose is to collect data from various registers necessary for failure handling and edit it into a predetermined logical format with a small amount of hardware and at high speed. The purpose is to do so.

[Means to solve the problem]

In order to achieve the above object, the present invention collects data held in a plurality of registers in the arithmetic processing unit necessary for handling the failure when a failure occurs in the arithmetic processing unit, and edits it into a predetermined logical format. In a diagnostic control device, a hardware information storage unit holds hardware information, a fault processing data editing information storage unit stores position information indicating a position to be extracted from this hardware information storage unit, and a a failure processing data storage unit that retains failure processing data; and a failure processing data storage unit that scans out hardware information including data of the plurality of registers from the arithmetic processing unit via a scan path when a failure occurs in the arithmetic processing unit. After storing data in the hardware information storage unit, in response to a register editing command, extracting data of the plurality of registers from the hardware information storage unit according to the position information held in the failure handling data editing information storage unit. and a control unit that edits the error processing data and stores it in the failure processing data storage unit.

[Operation] Generally, in an information processing device that adopts a scan path configuration,
When a fault occurs in a processing unit, the hardware information in the processing unit is scanned out via a scan path that can be controlled by the diagnostic control unit. Also included is data held in multiple registers within the processing unit. Therefore, in the present invention, the scanned out hardware information is temporarily held in the hardware information storage section, and then the individual position information of the data of the plurality of registers held in the hardware information storage section is stored. The control unit extracts the data of each register from the hardware information storage unit in response to a register editing command issued from the diagnostic control program while referring to the contents of the fault processing data editing information storage unit holding the fault processing data editing information storage unit. By editing and storing the data in the fault processing data storage unit, it is possible to edit one collection of fault processing data.

〔Example〕

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

FIG. 1 is a block diagram of essential parts of an embodiment of the present invention.

The arithmetic processing device l that constitutes the information processing device and the diagnostic control device 2 that controls its diagnostic control are connected by a scan path 7 that can be controlled by the diagnostic control device 2, and if a failure occurs in the arithmetic processing device l, The configuration is such that hardware information in the arithmetic processing device 1 is collected by the diagnostic control device 2 via the scan path 7. Here, what kind of data is scanned out via the scan path 7 and in what order is determined in advance. Contains data for various necessary registers.

The diagnostic control device 2 includes a hardware information storage section 3. Failure processing data storage unit 4. Failure processing data editing information storage unit 59 diagnosis control unit 6. A work register 8 and a program storage section 9 for storing a diagnostic control program IO are provided.

The hardware information storage unit 3 is a storage unit that holds hardware information scanned out via the scan path 7 from the arithmetic processing device 1 in which a failure has occurred, and is written and read by the diagnostic control unit 6.

The fault processing data storage unit 4 is a storage unit that holds fault processing data that has been edited into a predetermined logical format, and is written and read by the diagnostic control unit 6.

The failure processing data editing information storage unit 5 determines the location of the hardware information held in the hardware information storage unit 3.
This is a storage unit that stores in advance position information indicating whether data of individual registers necessary for troubleshooting is stored in a predetermined order, and is read out by the diagnostic control unit 6. Here, as shown in the figure, the location information regarding one register that is necessary for failure handling includes the relative address of the hardware information storage unit 3 that contains the data of that register (the beginning of the hardware information storage unit 3). It consists of a (relative value from the address) a, a bit position (first bit position) b indicating in which part of the address the data is stored, and the number of bits C.

The work register 8 is a register that is temporarily used when the diagnostic control unit 6 extracts and collects register data necessary for trouble handling, and is written and read from the diagnostic control unit 6 .

When a failure occurs in the arithmetic processing unit 1, the diagnostic control unit 6 reads hardware information from the arithmetic processing unit 1 via the scan path 7 by executing a diagnostic control program 10 stored in the program storage unit 9. While sequentially storing the information in the hardware information storage unit 3 from the beginning,
The register editing command C subsequently issued from the diagnostic control program 10 and the parameter 11 given to the command
In response to this, the data of each register is extracted from the hardware information storage unit 3 according to each position information held in the failure processing data editing information storage unit 5, and stored in the work register 8 sequentially from the beginning to the left. When all necessary register data is extracted and fault processing data in a format that matches a predetermined logical format is completed on the work register 8, it has a function to transfer it to the fault processing data storage section 4. There is. Here, the parameter 11 is composed of the start address PI of the hardware information storage section 3, the address P2 of the wl damage processing data storage section 4, and the address P3 of the fault processing data editing information storage section 5. FIG. 2 shows the flow of processing that the diagnostic control unit 6 executes in response to the above register editing command. These processes are realized by the hardware, firmware, etc. in the diagnostic control unit 6. .

The operation of this embodiment configured as described above will be explained below.

When a failure occurs in the arithmetic processing device 1, the diagnostic control unit 6 of the diagnostic control device 2 stores the hardware information of the arithmetic processing device 1 in the diagnostic control device 2 by executing the diagnostic control programmer J, 10. Scan out to Part 3.

When the hardware information of the processing unit l is scanned out to the hardware information storage unit 3, the start address pt of the hardware information storage unit 3 of the processing unit l, the address P2 of the fault processing data storage unit 4, and the fault processing A register editing command C to which the address P3 of the data editing information storage section 5 is given as a parameter 11 is executed by the diagnostic control section 6.

Referring to FIG. 2, the diagnostic control unit 6 first stores the data of the first register necessary for failure handling from the address P3 of the failure handling data editing information storage unit 5 in the given parameters 11. Read the relative address a of the hardware information storage unit 3 that is located (Sl), and then add the start address P1 of the hardware information storage unit 3 in the parameter 11 to this relative address a to find the address of the data to be extracted. (S2). Next, the bit position and bit number C of the data to be extracted are read from the failure processing data editing information storage unit 5 (S3).

Then, the data corresponding to the bit number 0 from the bit position in the address obtained in the hardware information storage unit 3 is read and stored in the work register 8 in a left-justified manner (S
4).

When the extraction of the data in the first register is completed, the diagnostic control unit 6 returns to the process of step Sl, refers to the next position information in the failure processing data editing storage unit 5, and performs the same operation as described above. Such operations are repeated until a predetermined condition is satisfied, such as when data corresponding to the number of bits in the work register 8 has been extracted. When all the necessary register data is collected on the work register 8, the data on the work register 8 is transferred to the fault processing data storage section 4 specified by the address P2 of the fault processing data storage section 4 in the parameter ll. The register editing command is stored at the address (S5), and the execution of the register editing command is ended. The fault processing data stored in the fault processing data storage section 4 is then analyzed by the diagnostic control program IO, and fault processing of the arithmetic processing unit, such as instruction retry, is performed.

〔Effect of the invention〕

As explained above, the present invention extracts the data of various registers necessary for failure processing from the hardware information of the arithmetic processing unit scanned out via the scan path and edits it into a predetermined logical format. This eliminates the need for a dedicated interface to read the registers required for fault handling as in the past, making it possible to reduce the amount of hardware. In addition, the process of extracting and editing only the data of the target register from the hardware information collected in the hardware information storage section in the diagnostic control bag W is performed by one instruction called the register collection instruction of the diagnostic control software. This process is performed by executing the following steps, so high-speed processing is possible.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of a main part of an embodiment of the present invention, and FIG. 2 is a diagram showing the flow of processing when the diagnostic control unit 6 executes a register editing command. In the figure, 1...Arithmetic processing unit 2...Diagnostic control device 3...Hardware information storage section 4...Fault processing data storage section 5...Fault processing data editing information storage section 6... Diagnostic control unit 7...Scan path 8...Work register 9...Program storage unit 10...Diagnostic control program 11...Parameter C...Register editing command

Claims (1)

[Scope of Claims] A diagnostic control device that collects data held in a plurality of registers in the arithmetic processing unit necessary for troubleshooting when a failure occurs in the arithmetic processing unit and edits it into a predetermined logical format, A hardware information storage unit that holds hardware information; a failure handling data editing information storage unit that stores positional information indicating a position to be extracted from the hardware information storage unit; and a failure handling data editing information storage unit that stores edited failure handling data. a processing data storage unit; and when a failure occurs in the arithmetic processing unit, hardware information including data in the plurality of registers is scanned out from the arithmetic processing unit via a scan path and stored in the hardware information storage unit. After storing, in response to a register editing command, data of the plurality of registers is extracted from the hardware information storage section and edited according to the position information held in the fault processing data editing information storage section, and the fault processing is performed. 1. A collection and editing method for failure processing data, comprising: a control section for storing data in a data storage section.