JPH039435A - Method for testing system abnormality - Google Patents

Abstract

PURPOSE:To detect an abnormal state having low reproducing frequency by repeating operation for sending collected test result information to an object to be tested as information to be sent for a test until a previously determined test stopping condition is satisfied and then collating test expected value information with final test result information. CONSTITUTION:A test pattern is formed by a testing carrier information forming means 300 and stored in a testing sending information storing memory device 102. The data are divided into data blocks, respective data blocks are trans ferred to a circuit to be tested, after ending prescribed data transfer, the trans ferred data are successively read out from the circuit to be tested and stored again in the corresponding addresses of the memory 102. The operation for sending the collected test result information to the circuit to be tested as the information to be sent for the test is repeated until the previously determined test stopping condition is satisfied and then the test expected value information is collated with the final test result information. Even when a device to be tested is expanded, an abnormal state having low reproducing frequency can be efficiently detected.

Description

[Detailed Description of the Invention] [Field of Industrial Application] The present invention relates to a system abnormality testing method, and in particular to a system abnormality test (hereinafter also simply referred to as "test j") that can efficiently detect system abnormalities that rarely reproduce. Regarding the method.

[Conventional technology]

In recent years, in information processing systems and communication systems with advanced functions, even if the frequency of system abnormalities is low,
The impact when it occurs is often significant. Conventionally, a repeat test method has been used to detect such system abnormalities.

For example, in the conventional repetitive test method for memory devices, etc., write data (test transmission information) is stored in a test transmission information storage means, and this information is written to the memory to be tested. This was then read out and collected and compared with the test transmission data described above.

By repeating the above procedure for all test areas, it is possible to detect abnormalities that occur less frequently. This will be explained in more detail below.

FIG. 4 is a conceptual diagram of a conventionally used test method. In the figure, 100 is a test execution unit that executes and manages the test, and 101 is a test target unit that is tested under the control of the test execution unit 100. Further, +02 indicates a memory device that stores the above-mentioned test transmission information, and +03 indicates a device to be tested (for example, a large-capacity magnetic disk storage device).

Considering the configuration described above, in the conventional test execution method, first, the above-mentioned test transmission information is retrieved from the memory device +02 and sent to the device under test 103, and then,
In order to verify whether this information has been written normally, after writing one byte or multiple bytes, the written information is read out and read into the information storage area 104, and then the test sending information stored in the memory device 102 is continued. The method was to use the matching means +05 to match each time.

FIG. 5 is a time chart when the above-mentioned test method is adopted. Considering the case where the test transmission information is sent to the test target device +03 in units of blocks, the time required for the above-mentioned test method is shown as shown in this diagram. That is,
Now, in Fig. 5, assuming that the average sending time of 1 byte of data is 2 μ seconds and considering that the sending time is proportional to the amount of data (number of bytes), the time required to send out the first 1 MB of information is W1. , 2 seconds. Furthermore, if the conditions for reading this are also the same as above, the time R1 required to read the IM byte will also be 2 seconds. Further, assuming that the time for comparing this information is Iμ seconds per byte, the comparison time C1 for comparing 1 Mbyte of information will require approximately 1 second.

Therefore, it will take about 5 seconds to complete the verification after sending 1 Mbyte of information, and more than 40 minutes will be required for 500 Mbytes of information. If a test were to be performed in which this was repeated 100 times per unit element (for example, 1 byte), approximately 70 hours (approximately 3 days) would be required. Incidentally, FIG. 6 shows an operation flowchart.

In addition, regarding this, for example, H6Y, Zhang et al.
Translated by Ukai et al. [Fault diagnosis of digital systems J
t Diagnosis of Digital
Systems) Chapter 2 (first edition: published in 1972)
The description is helpful.

[Problem to be solved by the invention]

If the conventional test method is adopted, test interruptions are relatively small in conventional small-scale systems with little test transmission information, but in large-scale systems such as large-capacity magnetic disk storage devices that are becoming highly integrated, When testing capacity storage means using such a test method, there is a problem in that the test time becomes long. Also, in an attempt to shorten the exam time,
When the number of repetitions per unit element is reduced, there is a problem in that it becomes difficult to detect abnormalities that are rarely reproduced.

The present invention has been made in view of the above-mentioned circumstances, and its purpose is to solve the above-mentioned problems in the conventional technology, and to solve abnormal conditions that are not frequently reproduced even when the scale of the test target device is increased. The object of the present invention is to provide a test method that can efficiently detect

[Means for Solving the Problems] The above object of the present invention is to provide a means for providing test expected value information for comparing with test result information corresponding to test sending information, In a system abnormality test method that includes means for transmitting information to a test subject and means for collecting the test result information from the test subject, and collates the test expected value information and test result information, the collected test results Providing means for transmitting information to the test object as test transmission information, and repeating the operation of transmitting the collected test result information to the test object as test transmission information until a predetermined test stop condition is satisfied. This is achieved by a system abnormality testing method characterized in that the test expected value information is then compared with the final test result information.

[Effect]

In conventional test methods, after sending test sending information to a device under test such as a large-capacity magnetic disk storage device, it is necessary to check each byte or multiple bytes against an expected value. It was necessary to perform the sending procedure, the procedure for reading the test result information, and the procedure for collating them for all areas to be tested. For this reason, as the scale of the test object increases, the test execution time increases.

On the other hand, in the test method according to the present invention, (1) after the test transmission information generated at the start of the test is transmitted to the test target device, the test result information is used as the test transmission information.

(2) The test is repeatedly executed when a separately specified test execution time or number of test executions is reached, or until the number of test executions specified by the tester is reached (during the test execution period).

(3) During the above test execution period, no comparison with expected information (expected value) will be performed, and even if test result information different from the expected information is collected, the result information will be used as the test transmission information for the next test. .

(4) When the test execution unit ends, the test result information is compared with the test transmission information at the start of the test.

(5) By detecting a mismatch between the test transmission information, which is the expected information, and the read result information, it is known that an abnormality has occurred during the test execution period. If a discrepancy is detected even at - degrees, repeated tests are conducted intensively at or near the discrepancy.

By adopting this procedure, abnormalities detected during the test execution period are different from conventional methods in that the longer the test execution period is set, the more concentrated the abnormalities are.

As a result, it is possible to reduce the verification operation that has conventionally been performed each time test transmission information is written, thereby making it possible to shorten the test execution time for a large-scale test target device.

[Example] Hereinafter, an example of the present invention will be described in detail based on the drawings.

FIG. 1 is a conceptual diagram of a test method showing an embodiment of the present invention. In the figure, 100a is a test execution unit that executes and manages the test, +01 is the test execution unit +0
The unit under test is tested under the control of 0a,
Reference numeral 102 indicates the test transmission information storage memory device, 103 indicates the device under test, and +05 indicates a verification circuit. Also,
Reference numeral 300 denotes test transmission information generation means, which may be generated with @ or may be generated each time a test is started.

In this embodiment, the test execution unit 1ooa may be a general-purpose processing device, and in that case, the test transmission information generating means 300 may be a program that can generate a certain data pattern or a pattern of the same type that is stored in advance in the memory. In the test method of this embodiment, which is configured as described above, when the test person instructs to start a test, first, the test transmission information generating means 300 creates a data pattern, and It is stored in the transmission information storage memory device 102. If the data was created by Ryu, the information may be transferred to the test transmission information storage memory device 102. Next, this data is divided into several data blocks and transferred to the circuit under test for each data block, and then, when the specified data transfer is completed, the transferred data is transferred from the circuit under test that is the transfer destination. The information is read out sequentially and stored again at the corresponding address on the sending information storage memory +02 for test vII.

If you repeat the above operating steps, in the meantime,
When writing errors or reading errors occur, the erroneous information is accumulated one after another. Here, the location where the erroneous data occurred is the test transmission information storage memory 102
This can be easily pointed out by analyzing the address information above.

Since this embodiment has the above-mentioned system characteristics, the fewer times the test transmission information is compared with the read information from the circuit under test, the more erroneous read information will be accumulated. It will be possible. Focusing on this point, as stated in claim 2 of the claims,
If the data is collated all at once after the test is completed, the cumulative test execution time can be shortened.

Furthermore, as the scale of the circuit under test increases, the number of accesses from a certain part of the circuit under test becomes extremely small. Therefore, if the frequency of occurrence of data errors is low, the chance of encountering an error in that location next time is extremely small. In such a case, as stated in claim 3 of the claims, by conducting the test in two stages,
It is possible to improve reproducibility.

FIG. 2 is a time chart when the test method of this embodiment is applied. According to this embodiment, the information read in R1 is overwritten in the write source area 2 written in Wl, and the information is written again in R2. Even if a writing or reading error occurs during this procedure, the conditions for stopping the test execution, such as when the number of test executions exceeds the specified number of times, when the test execution time exceeds the specified time, and when sending If an error such as a parity check error is detected during reading, the above-mentioned test transmission information is sent or read repeatedly until the situation is corrected, and when the test stop conditions are met, the data block matching procedure is performed during that time. , it is possible to significantly reduce the time required for matching, which was conventionally performed on a data block basis. Incidentally, FIG. 3 shows an operational flowchart of the main parts of this embodiment.

It goes without saying that the above embodiment is shown as an example of the present invention, and that the present invention is not limited thereto.

[Effects of the Invention] As described in detail above, according to the present invention, there is provided a means for providing test expected value information for comparing with test result information corresponding to test sending information, and a means for providing test expected value information for comparing with test result information corresponding to test sending information. In a system abnormality test method that includes a means for transmitting to a test subject as test transmission information and a means for collecting the test result information from the test subject, the system abnormality test method collates the test expected value information and the test result information. Providing means for transmitting the test result information to the test subject as test transmission information,
After repeating the operation of sending the collected test result information to the test subject as test sending information until a predetermined test stop condition is met, the test expected value information is compared with the final test result information. This has the remarkable effect of realizing a test method that can efficiently detect abnormal states that are rarely reproduced even when the device to be tested is increased in scale.

[Brief explanation of the drawing]

Fig. 1 is a conceptual diagram of a test method showing an embodiment of the present invention, Fig. 2 is a time chart when the test method of the embodiment is applied, Fig. 3 is an operation flowchart of the embodiment, and Fig. 4 is a conventional method. Figure 5 is a conceptual diagram of the test method used since
FIG. 6 is a time chart when the test method shown in the figure is adopted, and FIG. 6 is a flowchart of the same operation. 100a: Test execution unit, +01: Test target unit, 102: Test transmission information storage memory device, +03
Test target device, 105: Verification circuit, 300: Test transmission information generation means, Wl: Transmission time of the first data block to the test target, R1: Reading time of the first data block from the test target, C: First verification time after sending and reading multiple data blocks.

Claims (3)

[Claims] (1) A means for providing test expected value information for comparison with test result information corresponding to the test transmission information, a means for transmitting the test expected value information to the test subject as the test transmission information, and a test subject. In a system abnormality testing method that includes means for collecting the test result information from a computer, and collating the test expected value information with the test result information, the collected test result information is sent to the test subject as test transmission information. After repeating the operation of transmitting the collected test result information to the test subject as test transmission information until a predetermined test stop condition is satisfied, the test expected value information and the final test result information are transmitted. A system abnormality test method that is characterized by checking against (2) While repeating the operation of sending the collected test result information to the test subject as test sending information until a predetermined test stop condition is met, the expected test value information and the test result information are The system abnormality test method according to claim 1, characterized in that no verification is performed. (3) Initially, as a result of conducting a test targeting a predetermined first wide range, if the test result information and the test expected value information do not match, a second narrow range including the relevant test area is Claim 1 characterized in that the test is conducted on a subject.
Or the system abnormality test method described in 2.