JPH08171520A - Faulty package detection system - Google Patents

Abstract

PURPOSE: To provide a faulty package detection system which can detects the number of a faulty package immediately from an address where a fault occurs in case of fault occurrence by providing a subroutine of an application program with a fault index table wherein addresses indicating memories or I/O devices and the package numbers of the memories or I/O devices are made to correspond to each other. CONSTITUTION: A common subroutine (step S3) stores parameter information including a memory or I/O device, or its address passed from the application program (step A1) in the fault index table (step A8), and actuates a function for accessing the memory or I/O device to access the corresponding I/O device or memory. If the accessed I/O device or memory becomes faulty to cause a non-maskable interrupt, the corresponding package number stored in the fault index table (step A8) is detected in the fault index table (step A8) and passed to the application program.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a failure package detection method, and more particularly to NMI (Non Maska) by Xack Time Out.
The present invention relates to a fault package detection method for identifying a fault package by using a firmware program when a ble interrupt occurs.

[0002]

2. Description of the Related Art FIG. 3 is a flowchart showing a conventional fault package detection method. A software program monitors a failure state for a unit consisting of a control package and its subordinate packages via a firmware program or directly periodically. If there is a failure, the software program is notified via the firmware program or directly. The maintenance person performs decoding based on the content of the failure and identifies the faulty package.

As described above, in the conventional fault package detection method, the control package for controlling the fault package is
In a unit of a unit including a fault package, a control package, and a package other than the fault package, the fact that the unit is out of order and the details of the fault are notified to the software program by the firmware program.

[0004]

In the above-mentioned conventional failure package detection method, the control package notifies the software program by the firmware program that the unit is in failure and the failure content is in units of the unit. Therefore, it is difficult to identify the faulty package, and there is no choice but to guess the faulty package from the fault content.

In view of the above problems, the present invention provides a fault index table in which addresses and package numbers are associated with each other, and when a fault occurs, a faulty package number can be immediately detected from the faulty address. The purpose is to provide a scheme.

[0006]

According to the fault package detection method of the present invention, when an application program driven by a CPU accesses a package having a package number corresponding to the given address based on the given address, When an application program accesses a memory or an I / O device that is not mounted or is in a failed state and the ready signal does not return, the ready signal is forcibly generated, and non-maskable interrupt, that is, N
A failure package detection method when an Exakk Time Out that causes MI (Non Maskable Interrupt), that is, an Xack Time Out occurs, wherein the application program includes a function for accessing the memory or the I / O device. , A common subroutine having a failure index table capable of storing each package number for a given I / O or memory and the address assigned to each package number in association with each other,
The application program is a memory or I / O
When accessing the device, the application program passes the memory or I / O device to be accessed and its address to the common subroutine as parameter information, and the common subroutine stores the passed parameter information in the fault index table and executes the function. When a non-maskable interrupt occurs, the corresponding package number is detected from the fault index table and passed to the application program.

[0007]

The common subroutine stores the parameter information delivered from the application program in the fault index table, activates the access function, and accesses the corresponding I / O device or memory. When a failure occurs in the accessed I / O device or memory and a non-maskable interrupt occurs, the corresponding package number stored in the failure index table is detected from the failure index table and passed to the application program. . This delivery can be executed immediately because it can be executed only by searching the fault index table.

[0008]

Embodiments of the present invention will now be described with reference to the drawings. FIG. 1 is a flow chart showing an embodiment of the fault package detection method of the present invention, and FIG. 2 is a fault in which each package number of an I / O device or memory and an address corresponding to each package number are stored in a common subroutine. It is a figure which shows the memory content of an index table.

Control package ROM (Read Only Memo)
The application program stored in ry) is started (step A1) and indicates whether it is an I / O device or a memory, or which address of the package number corresponds to the I / O device or the memory. The parameters are prepared (step A2) and passed to the common subroutine (step A3). The common subroutine stores the delivered parameters in the fault index table and activates a function for accessing the I / O device and the memory (step A4). Function is the corresponding I / O
If an access is made to the device or the memory (step A5), but there is a failure in the accessed package, Xack Time Out that causes NMI (Non Maskable Interrupt) occurs (step A6).

When the Xack Time Out occurs, the common subroutine detects it (step A7), searches the fault index table held in the common subroutine (step A8), and detects when the Xack Time Out occurs. The corresponding package number is detected from the address of the access destination, the detected package number is called as a common subroutine as parameter information (step A9), and is delivered to the application program.

[0011]

As described above, according to the present invention, by using the common subroutine, the parameter information is passed from the application program to the common subroutine for storage, and the address and package number of the memory or I / O device are stored. A fault index table that associates with the above is prepared, and the package number can be immediately detected from the address at the time of the fault by the function that accesses the address of the I / O device or the memory from the common subroutine. There is an effect.

[Brief description of drawings]

FIG. 1 is a flowchart showing an embodiment of a fault package detection method of the present invention.

FIG. 2 is a diagram showing stored contents of a failure index table stored in a common subroutine in the embodiment of FIG.

FIG. 3 is a flowchart showing a conventional fault package detection method.

[Explanation of symbols]

 A1-A9 steps

Claims (3)

[Claims] 1. When an application program driven by a CPU accesses a package having a package number corresponding to the given address on the basis of a given address, the application program is not mounted or has a faulty memory or I / O. Access the O device,
In a failure package detection method when an exaac time-out occurs, in which a ready signal is forcibly generated when a ready signal is not returned and a non-maskable interrupt is generated, the application program includes the memory or I / O function and the I / O provided
There is provided a common subroutine having a package index number for O or a memory and a failure index table capable of storing the address assigned to the package number in association with each other, and the application program stores the memory or the I / O.
When accessing the device, the application program passes the memory or I / O device to be accessed and its address to the common subroutine as parameter information, and the common subroutine stores the passed parameter information in the fault index table and executes the function. Is started, the corresponding I / O or memory is accessed, and when a non-maskable interrupt occurs,
A failure package detection method that detects the corresponding package number from the failure index table and passes it to the application program. 2. The fault package detection method according to claim 1, wherein the application program is a firmware program. 3. The fault index table comprises memory and I / O.
The fault package detection method according to claim 1, wherein the parameter information relating to the O device can be stored separately.