JPS5827537B2 - History information storage processing method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention provides a history information storage processing system, particularly a data processing system having a history information storage means for storing history information in a plurality of history stacks. The present invention relates to a history information storage processing method in which time information is stored together and timing relationships between pieces of information stored in respective history stacks can be determined.

Conventionally, in data processing systems, for example, machine instruction address information, microinstruction address information, and other important control signals such as interfaces with equipment and equipment are stored in a history stack to prevent bird failures and software failures. Measures will be taken to deal with obstacles and harm.

Such history information is stored each time it is updated, or when a branch occurs in the address information or the like.

Since the above-mentioned history information is stored in chronological order, it is easy to know, for example, the time relationship between pieces of information stored in one Hiss I-'J stack.

However, when a plurality of history stacks exist, it is not easy to know the time relationships among the stacks.

For this purpose, for example, in a data processing system, machine instruction address information in the event of a bang condition,
It is not easy to know the time relationship between microinstruction address information and other control signals.

Furthermore, in a multi-data processing system, it is not easy to know the time relationship of the husband's information when a deadlock situation occurs.

In such cases, in the past, at best, a software website expert and/or a bird expert would spend a great deal of effort and time to learn about the time-1 relationship by chance, so to speak, and discover the bug.

The present invention aims to solve the above-mentioned problems, and the history information storage processing method of the present invention includes a history information storage means for sequentially storing information corresponding to the processing progress in a data processing system on a history stack. In a data processing system with It is characterized in that it is stored in each of the above-mentioned history stacks together with the information stored in the history stack.

This will be explained below with reference to the drawings.

The figure shows an embodiment of the present invention, in which 1 is a data processing system, 2-0.2-1 is a data processing device, and 3-0 is a data processing system.
．． 3-1 are machine instruction address counters, 4-0
．． 4-1 are microinstruction address counters, 5-
0.5-1 are various control signals (or register contents)
, 6-0.6-1 are data processing devices 2-0.2-1, respectively.
Real time clock (or real time information) provided by 7-0
．． Reference numerals 7-1.8-0.8 and 1.9-0.9-1 each represent a history stack, and 10-0 to 12-1 each represent an information coupling circuit.

Each data processing device 2-0.2-1 independently executes processing in a well-known manner.

and each machine instruction address counter 3-0.3-
Microinstruction address counters 4-0 and 4 operate to access a main memory (not shown) based on the contents of microinstruction address counters 4-0 and 4-0, and to access a control memory (not shown) in accordance with machine instructions read from the main memory. The control memory is sequentially accessed according to the contents of -1, and processing is executed based on the microinstructions read from the control memory.

During this time, the machine instruction address counter 3-0.
Contents of 3-1, microinstruction address counter 4-0
．． The contents of 4-1 and other control signals 5-0.5-1 change sequentially.

Each of these pieces of information is paired with the time information of the real time clock 6-0.6-1 by each of the coupling circuits 10-0 to 12-1, respectively, and stored in the corresponding history stack 7-0.7.
-1.8-0.8-1.9-0.9-1.

It goes without saying that the real time clocks 6-0 and 6-1 are synchronized with each other.

The respective history information extracted as described above can be arranged on a time axis based on real-time information, and bugs can be prevented to deal with the bang state and deadlock state as described at the beginning of this specification. Analysis processing for discovery becomes extremely easy.

In the above description, it has been shown that the real time information from the real time clocks 6-0 and 6-11 is stacked together with their respective history information.

However, the time information is not necessarily limited to real time information, and may be anything that ultimately serves as a reference for arranging each history information on the time axis, such as an interval timer, a CPU timer, etc. , or the contents of a counter.

Furthermore, regarding the above-mentioned time information, it is not necessary to store all the bits constituting the time information, and it is sufficient that the time of occurrence of each piece of history information can be identified and the necessary minimum precision is sufficient.

[Brief explanation of the drawing]

The figure shows an embodiment of the invention. In the figure, 1 is a data processing system, 2-0.2-1 are data processing units, 3-0.3-1 are machine instruction address counters, and 4-0 and 4-1 are microinstruction address counters. , 5-0゜5-1 are various control signals,
6-0 and 6-1 represent husband and wife time information, and 7-0 to 9-1 each represent a history stack.

Claims (1)

[Scope of Claims] 1. A data processing system having a history information storage means for sequentially storing information corresponding to the processing progress in the data processing system on the history IJ stack (herein, information corresponding to the processing progress in the data processing system is In addition to providing a plurality of history stacks to be stored, the information to be stored in each history stack σ (this information, which is a common time reference, is stored in each of the above history stacks) A history information storage processing method characterized in that the information is stored in an IJ stack. 2. The information stored in each of the plurality of history stacks is information of a plurality of locations in a single data processing device. A history information storage processing method according to claim 1, characterized by A history information storage processing method according to claim 1, characterized in that: 4. A history information storage processing method according to claim 1, wherein the time information serving as the common time reference is real time information. The history information storage processing method according to any one of Items 1 to 3.