JPS62241040A - Hardware action tracing device - Google Patents

Abstract

PURPOSE:To secure the effective application of a store area for action state information and to hold the information of high accuracy for a long period of time, by adding a tracing time point and the importance degree of information to the action state information of hardware as the control information. CONSTITUTION:An action state information collecting mechanism 112 is started by an event detecting mechanism 111 and collects the trace information on a device 100 to be traced. A control information adding mechanism 114 adds the importance degree stored in an action state information importance degree memory mechanism 113 and the collecting time point information on the trace information clocked by a clock mechanism 115 to the trace information of the device 100. Then a store area control mechanism 116 checks the idle state of the store area of an external memory device 120 and refers to the time point and the importance degree added to the action state information if no idle area is detected and selects the oldest action state information having a low degree of importance. Then the new action state information is stored (overlayed) into the store area of the selected action state information. Thus it is possible to use effectively the store area of the action state information and to hold the information of high accuracy for a long period of time.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a vulware operation tracing device that traces the operating status of various types of hardware in a data processing system.

[Conventional technology]

Conventionally, when trying to collect operating status information with a hardware operation tracing device, the operating status of the hardware (hereinafter referred to as the traced device) is collected for each event, and this collected operating status information (hereinafter referred to as trace information) is collected for each event. ) is stored in an external storage device or an internal storage area within the hardware operation tracing device.

Steps 1 and 2: If this storage area becomes full with trace information, either stop the tracing of the traced device at that point, or cyclically overlay the trace information in the storage area from the oldest information. , continue tracing.

[Problem that the invention seeks to solve]

As mentioned above, in conventional hardware operation tracing devices, when the trace information storage area runs out,
This is handled using the fil trace interruption method or (2) overlay method. By the way, in the trace collection process, when a man-machine interface or a host device exists.

In order to restart a trace that has been interrupted, trace information is generally saved using a nodeware interface or the like. Therefore, when using this trace relay method, a man-machine interface or a software interface is involved, making the trace collection process complicated.

Moreover, there is a problem in that it lacks real-time performance.

In order to solve these problems, please! - Race information storage area It is necessary to set the race information storage area sufficiently large in advance, but if the storage area becomes full, tracing will be interrupted.

On the other hand, with the overlay method, tracing is not interrupted, but since the oldest information is overlaid, information necessary to understand the operating status of the hardware may be erased. There is a problem in that a sufficiently large storage area must be set in order to preserve the data.

[Means for solving problems]

The hardware operation tracing device of the present invention includes an event detection mechanism that detects the timing to collect trace information necessary for grasping the operation status of the traced device, and a timing detected by the event detection mechanism that an operation status information collection mechanism that collects trace information; an operation status information importance storage mechanism that stores the importance of the trace information; and a clock mechanism that manages time information.

A management information addition mechanism that adds management information consisting of the above importance level and time information notified from the clock mechanism to trace information, and an operation status importance level that stores trace information with this management information added in a storage area. The present invention is characterized by having a storage area management mechanism that searches for low trace information by referring to management information and uses the storage area of the searched trace information as a storage area of the latest trace information.

〔Example〕

Next, the present invention will be explained with reference to the drawings.

First, referring to FIG. 1, a traced device 100 and an external storage device 120 are connected to a hardware operation tracing device 110 via an interface (not shown).

The hardware operation tracing device 110 includes an event detection mechanism 111 that detects the timing of collecting trace information (operation status information) of the traced device 100, and a trace information of the traced device 1oo at the timing detected by the event detection mechanism 111. an operating status information collection mechanism 112 that collects trace information; an operating status information importance storage mechanism 113 that stores the importance of trace information; and a clock mechanism 115 that manages time information; Remembered importance.

and a management information addition mechanism 114 that adds time information notified from the clock mechanism 115 (this importance level and time information together are referred to as management information) to the trace information collected by the operating status information collection mechanism 112; The operating status information storage mechanism 117 stores the trace information with added information in the external storage device 120, and the free space management of the trace information storage area is performed. A storage area management mechanism 11 that searches for low trace information by referring to management information and uses this trace information storage area as a storage area for the latest trace information.
6.

The event detection mechanism 111 detects the timing of collecting trace information necessary to understand the operating status of the traced device 100. There are two types of timing detection in the event detection mechanism 111: (a) a traced device-driven method and a) a hardware operation trace device-driven method.

In the case of the traced device-initiated method, the timing of trace information collection is notified from the traced device 100 to the event detection mechanism 111. - In the case of the hardware operation tracing device-driven method, the event detection mechanism 111 constantly checks the state of the traced device 100, and when a change in state occurs.

In order to collect trace information, the operating status information collection mechanism 1
12 is activated.

The operation status information collection mechanism 112 activated by the event detection mechanism 111 collects information about the traced device 10 at the time of activation.
Collect trace information (operation status information) for 0. In addition,
What operating status information is collected?

Although it differs depending on the traced device 100, it is specified in advance. For example, hardware! status.

If data transfer is involved, the length of the transfer data, etc.

The clock mechanism 115 is a timekeeping mechanism in a computer system, and there are two types: (a) an uninterruptible timekeeping method and (b) an internal timer method.

In the case of the uninterruptible timing system, time is continuously measured regardless of whether the hardware operation trace device 110 is powered on or off. On the other hand, in the internal timer method, the timing mechanism is activated after the hardware operation tracing device 110 is powered on. Therefore, it is necessary to synchronize the time with the traced device 100.

The trace information collected by the operating status information collection mechanism 112 includes the importance level stored in the operating status information importance level storage mechanism 113 by the management information addition mechanism 114, and the collection of trace information timed by the clock mechanism 115. Time information is added.

This level of importance is specified in advance.

Importance and time information storage device 1 by management information addition mechanism 114
20. Note that this storage location is determined by the storage area management mechanism 116. That is, the storage area management mechanism 116 examines the external storage device 120, determines the storage location of the above-mentioned collected trace information, and notifies the operating status information storage mechanism 117.

Here, the operation of the storage area management mechanism 116 will be explained with reference to FIG.

The storage area management mechanism 116 checks the free status of the storage area of the external storage device 120 (step 300). If a free area exists, the starting position of the free area is determined as shown in FIG. 5 (step 31O). Then, this leading position, that is, the storage position, is notified to the operating status information storage mechanism 117 (step 320).

On the other hand, if the result of checking the free space status of the storage area is that there is no free space, a plurality of pieces of trace information (including control information) are extracted from the storage area in order of the longest stored trace information. That is.

A plurality of pieces of trace information are taken out in order from the oldest trace information in the storage area of the external storage device 120 (step 301). The number of pieces of trace information to be extracted is set in advance. For example, set the number of trace information to be extracted to 5.
As shown in FIG. 4, storage area management mechanism 1
16 refers to the trace time information and selects A2 + in order of oldest
A3 r AS yA 9 t A 13 e
Five pieces of trace information are extracted (time information is numbered in ascending order of 1, 2, 3, etc., with l representing the oldest one). The least important trace information is found from the 95 extracted trace information (step 302).
). 2. The highest importance level is 5. the lowest t
-1. Therefore * A2 + A3 e
Among A6 r A9 + A13, the least important trace information A3 is obtained. Then, the storage location of this trace information A3 is determined (step 303), and this storage location is notified to the operating status information storage mechanism 117 (step 320). The operating status information storage mechanism 117 overlays the latest trace information on the above-mentioned trace information storage area. Note that if there is a plurality of pieces of trace information having the same degree of importance, the piece of trace information with the oldest time information is selected.

Note that FIG. 2 shows the hardware operation tracing device 11 without using an external storage device as a storage area for trace information.
An example using an internal storage area 118 prepared for 0 is shown (
The operation of the hardware operation tracing device iio is the same as in the embodiment described above).

〔Effect of the invention〕

As explained above, in the present invention, the time at the time of tracing and the importance of the information are added to the hardware operating status information as control information and stored in a storage area such as an external storage device. If there is no free space, all the oldest and least important operating status information is selected by referring to the next time and importance added to the operating status information, and new operating status information is stored in the storage area for that information (overlay). ), the effect is that the storage area for operating status information can be used efficiently.

[Brief explanation of drawings]

1 and 2 are configuration diagrams of a hardware operation tracing device according to the present invention, and FIG. 3 is a flowchart illustrating the operation of a storage area management mechanism. FIGS. 4 and 5 are diagrams specifically explaining the processing of the storage area management mechanism. 100... Traced i! , 110... Bartware operation tracing device, 111... Event detection mechanism. 112... Operating status information collection mechanism, 113... Operating status information importance storage mechanism, 114... Management information adding machine groove, 115... Clock mechanism, 116... Storage area management mechanism, 117. ...Operating status information storage mechanism, 11
8... Internal storage area, 12o... External storage device. Figure 1 Figure 2 Figure 3 Cause Figure 4 Figure 5

Claims (1)

[Claims] 1. An event detection mechanism that is used to collect information necessary to understand the operating status of the hardware that makes up the data processing system, and detects the timing of collecting the operating status information of the hardware to be traced. and,
an operating status information collection mechanism that collects operating status information of the hardware at the timing detected by the event detection mechanism; an operating status information importance storage mechanism that stores the importance level of the operating status information; a clock mechanism for managing information; a management information adding mechanism for adding management information consisting of the importance level and time information notified from the clock mechanism to the operating status information; and operating status information to which the management information has been added. an operating status information storage mechanism that stores operating status information in a storage area, and manages free space in the storage area, and if the free area does not exist, the operating status information that has already been stored and is of low importance is referred to the management information. and a storage area management mechanism that searches for and sets the storage area of the searched operating status information as a storage area of the latest operating status information, thereby shortening the residence period of operating status information of low importance in the storage area. A hardware operation tracing device characterized in that: