JPS5960656A - Data processing system - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION (a1 Techniques of the invention 1, 1, 1 minute IF) The present invention provides a method for recording data processing systems such as computer computers, especially malfunctioning states of data processing systems.
This paper relates to a tracer that can be used to reproduce this phenomenon and investigate the causes of °ζ malfunctions.

(Background of BL technology In order to process data with a single electronic meter, in addition to bar IS software, microprograms and software are required to operate it. Also, malfunctions in data processing systems are caused by hardware There are various causes, such as cases where the cause is solely due to the hardware, such as failure of a computer or design error, cases where the cause is due to the design of the microprogram or software, and cases where the cause is caused by a complex interplay of hardware, microprograms, and software. .

Generally, when such a failure occurs, it is difficult to find out the cause and it takes a long time.

Moreover, in the case of 2 hardware, malfunctions occur intermittently, and the occurrence rate is extremely low, occurring once every few months or once a year. In addition, it often recovers naturally, and normal operation can be resumed by retrying.

To find out the cause of malfunctions that occur in this way,
The state before and after the malfunction occurred in some form (:11j-
Currently, it has been put into practical use as a so-called l-laser.

(C1 Prior art and its problems Figure 1 is a block diagram showing a conventional 1-laser.M
is a data processing system that includes a main storage device 1, a central processing unit 2, a display control unit 13, a printer control unit 14, a disk control unit 5, and a power supply unit 6.
It is equipped with A display device 7 and a keyboard 8 are connected to the display control unit 3, a printer 9 is connected to the printer control unit 4, and a magnetic disk device 10 is connected to the disk if1'I tall unit 5. has been done.

In order to trace the status of a data processing system MO) having a similar configuration, a router equipped with an oscilloscope 11 (MO) is connected to the central processing unit 2. The tracer unit 112 mainly consists of a C memory, and is connected to the central processing unit 2 via signal lines 13, which have the same number of channels as the signals to be traced, and can simultaneously handle several channels to more than ten channels. The signal can be taken out and written to the memory of the tracer unit 12 as 1 and 0.

And note 1. ☆) and the data on the oscilloscope 1
1, the result will be as shown in FIG. 2, for example. That is, on the horizontal axis or time T, the signals Sl, corresponding to each signal line 13...
Changes in the waveforms of S2, S3, etc. are displayed.

In this way, by reproducing the state of each signal stored in the tracer unit 1-12 and checking whether the signal 53 is stationed, we can determine which signal is causing the malfunction and what the malfunction state is. can be confirmed. There is also an rbl device that has a dedicated printer and can print loose results.

In this way, the conventional tracer has a tracer unit 1-12.
and an oscilloscope 11, as well as a printer dedicated to printing race results, maintenance personnel transport these devices to the user's site, connect them to the data processing system M for six months to one year, and keep them in operation. Allows recording of operating status.

However, from the user's perspective, Rina Okawa's device is always connected to the data processing system M, which gives the user a strange feeling and raises suspicions, reducing trust in the data processing system. There is a risk that it will go down a notch. Moreover, it is a large and expensive device, and must be installed for a long period of time until a malfunction signal is generated, so in reality, it is only possible to install it at a few locations nationwide in the case of nationwide work systems. . Therefore] (Il
Once you have traced the location and investigated the cause of the malfunction, you may have to move it to the next location, or it may be too large to transport. There are problems such as.

(d1 Purpose of the invention 4 - Invention 41, to solve such problems with conventional 1 - lasers, to reduce the size of the laser so that the user does not notice it, to make it easy to transport, and to effectively investigate the cause of malfunctions. rules
The purpose is to realize the following.

[Each configuration of the ie1 invention] In order to achieve the object, the present invention provides a central processing unit,
In a data processing system having a master 1,0 device and a peripheral device such as a display device, an l-L--server unit is equipped with a dedicated memory for tracing and recording system malfunctions, and a writing/reading circuit for the memory. is incorporated into the data processing system, the data bus and address bus for reading the memory of the l-laser unit are shared with the data bus and address bus of the main storage device, and the writing section to the tracer unit is configured to prevent malfunctions of the data processing system. The configuration is such that it is connected to the signal extraction section.

(f1 Embodiment of the Invention Next, an embodiment will be used to explain how the data processing system according to the invention is actually implemented.

FIG. 3 is a block diagram showing an embodiment of the data processing system J according to the present invention, and parts corresponding to the configuration of the data processing system M shown in FIG. 1 are given the same reference numerals.

Quick start 1.1 Device 1, central processing unit 2, display/
C control Unison 1-3, printer control "'-2'-nol
4, disk control unit 5, and power source 1-
6, the display control unit 1-3 is connected to the display device 7 and the keyboard 18, the printer control unit 4 is connected to the printer 9, and the display control unit 1 and 5 are connected to the magnetic screen. Devices 10 are connected to each other.

14 is a tracer unit according to the present invention, physically:
It is installed using a part of the 1e device 1 described above. The 1-Ray Uni 71-14 is configured, for example, on a plurality of print boards or on a single printed circuit board 15 as shown in FIG. This printer 1-board 15 can be replaced with one memory printer I-board in the mounting area of the 1a device 1, or the mounting space for the tracer unit 1 can be set at 1$ (Ii:f) in advance. When it becomes necessary to install the Lury unit node 14, this space may be utilized for mounting.

i.Rayli Unino L configured on Print 1 board 15
] 4 Lt, memorize the trace results as shown in Figure 4'
an address circuit 17 for updating the address of this memory 1G, an address counter 18, and a write circuit 19 for writing trace results into the memory 16.
, a read circuit 20 for reading data recorded in the memory 16 to the outside, a clear circuit 21 for clearing the memory 16 before starting tracing, and a record for stopping recording to the memory 16 based on a malfunction detection signal. Stop circuit 22
, a timing circuit 23 that sets the write timing to the memory 16.

The write circuit 19 is connected to the medium heat processing device 2 through signal lines 24 having the required number of channels, and traces the signals obtained by the central processing device 2 and records them in the memory 16. In addition, in order to output the trace data recorded in the memory 16 to the outside, the take iJt extraction circuit 20 is connected to the data bus DB, and the address circuit 17 is connected to the address bus AB.
Control circuits (not shown) are respectively connected to the control lot CB. And these lotus D 113, Al
1. C13- is shared with each lotus of the main offender 1 and 0 device 1, respectively. In this way, the tracer unit 14 is connected to the main fa device 1 on the common lotus, so from the rough 1-wear, the dirt 1j! It can be handled in the same way as some data in the device. Therefore, the I recorded in the memory 16
- The race results can be output and displayed on the display device 7 or printed on the printed circuit board 15 using the printer 9 in exactly the same way as some other data.

Therefore, unlike in the past, a special oscilloscope was installed,
There is no need to install a dedicated printer.

Next, the trace operation to the memory 16 and the read operation of I/race results will be explained. First, trace the I-laser operator panel 25 on line 18 26.
14, the clear circuit 21ζ memory 16 is cleared by operating the operator panel 25. Then, a signal requiring one race is taken in from the central processing unit 2 which is currently in operation through the signal line 24, and the write circuit I
9, it is written to the memory 16. At this time, the address circuit 17 and the address counter 18 sequentially advance the addresses in the memory 16 to record the take, or
Since the capacity is limited to 41, once the last 71-res is written, it returns to the original state and rewrites again, recording endlessly. In this way, the data inputted at a constant writing cycle P shown in FIG. 2 is recorded in the memory 16°ζ
go. The write cycle P at that time may differ depending on the type of malfunction signal to be detected, but by resetting the timing circuit 23 by operating the operator panel 25, the desired write cycle P can be obtained.

When the recording stop circuit 22 receives a signal indicating that a malfunction has occurred from the malfunction detection section, the recording stop circuit 22 performs a1. The address counter 18 is stopped and writing is stopped.

By the time a malfunction occurs, the state of the signal that is the cause of the malfunction has already been input and written into the memory 6 from the signal line 24, so there is no problem even if the writing is stopped once a malfunction occurs. do not have. If the state after the occurrence of a malfunction is also set to 1, the writing may be stopped from "ζ" after a predetermined time delay.

Once the signal that causes the malfunction is traced and recorded in the memory 16, the content is displayed on the display 7 or printed on the printer 9-r to investigate the cause of the malfunction. In other words, since the memory 16 of the Rayli Uniso I/j4 is also connected on the same bus as the main storage device 1, the data processing system M's own program fully utilizes peripheral devices such as display devices and printers to output data. Since each block is shared with the main memory 1, the data recorded in the memory 16 can be handled in the same way as some other data, so the processing function of the data processing system itself can be You can use it to process and output it in hexadecimal or decimal format, and if you want to obtain malfunction data regarding the hardware operation sequence, you can output it in time chart or binary format, and in the case of Alres emotion, Data in a format suitable for the purpose, such as output -J- in hexadecimal format, can be easily obtained.In the case of an intermittent malfunction, the data processing and the system itself must be kept under pressure even after the malfunction. Therefore, it is easy to process and output using the functions of the data processing system itself as described above.

The blade operator paddle 25 is used for setting the Zumblink cycle P of the malfunction signal, selecting a trace signal, and clearing the memory 16 when recording starts.

(g1 Effects of the Invention As described above, according to the present invention, a data processing system having a peripheral device such as a display device is equipped with a memory dedicated to tracing malfunctions and its 1. loading and reading circuit. In addition to incorporating the tracer unit, the tracer unit bus and atrus bus are shared with the data bus of the main storage device and the data bus of the main memory, so that output can be made using the tracer unit processing system's own display device and printer. Therefore, there is no need to bring in the conventional L-lace equipment and install it at the user's location for a long period of time, and it is possible to
The disk is installed as an option in the - section of the main memory,
Since it is sufficient to connect the operator pipes and pipes, there is no risk of giving a suspicious window to the user or lowering the reliability of the system. In addition, the tracer unit can be configured with a printed circuit board of about 1 pc, and the trace results can be output using the user's data processing system's own display device or printer, making Tracer UniSoto low-cost and suitable for use in nationwide network data processing systems. However, by placing a large number of them in various locations, we can quickly respond to any malfunction.
It can greatly contribute to improving the reliability of data processing systems.

] - Since the laser unit has a compact configuration, it can be transported to remote locations using parcels or other means of transportation.
Even in areas with inconvenient transportation, we can deliver quickly and inexpensively. Moreover, by effectively utilizing the processing functions of the data processing system itself, it is possible to process and output data in the most advantageous form for investigating the cause of malfunctions, making it easier and faster to investigate the causes of malfunctions. It becomes possible to do so.

[Brief explanation of drawings]

Figure 1 is a block diagram showing the 1-laser from (jL);
FIG. 3 is a diagram showing the output waveform of the recording signal, FIG. 3 is a knock diagram of the data processing system which also functions as a router according to the present invention, and fourth is a block diagram of the laser unit 1-. In the figure, M is a data processing system, ■ is a main memory, 2 is a central processing unit, 7 is a display device, 9 is a printer, 14 is a tracer unit, 16 is a memory, 17 is an address circuit, and 18 is an address counter. , 19 is a write circuit, 20 is a readout circuit, 22 is a recording stop circuit, D
B represents a data bus, AB represents an Arres bus, and C10 represents a control bus. Patent Applicant Fujitsu Limited Agent Patent Attorney Minoru Aoyagi Procedural Amendment 1. Indication of case: Japanese Patent Application No. 57-1714962, Title of invention: Name of data processing system (522) Fujitsu Ltd. 5, Date of amendment order: Voluntary action 6, Subject of amendment: “Detailed description of the invention” column 7, Amendment As shown in the attached sheet 1, this will be explained using an example on page 6, line 15 of the specification. ” shall be amended as follows. An example will be described in which a malfunction of the r CPU is investigated. ”2
, page 13, lines 13 to 14, the statement ``Malfunction may occur.'' has been corrected as follows. "Even if this tracer function is added midway through, it will not affect the data processing system at all, so it will be possible to investigate the cause of malfunctions more easily, quickly and accurately."

Claims (1)

[Claims] In a data processing system having peripheral devices such as a central processing unit, a main memory device, and a display device, a dedicated memory for tracing and recording malfunctions of the system and a writing/reading circuit for the memory are provided. In addition to linking the tracer unit to the data processing system, the data bus and the Arres bus for reading the memory of the Laser Uni/1 are made common to the data bus and Arres bus of the main storage device, and the writing unit to the tracer unit is connected to the data processing system. The data processing system is connected to a malfunction signal extraction section of the data processing system.