Multi-device test management method, device, computer readable storage medium

Abstract

The application discloses a multi-device test management method, device and computer readable storage medium. The method comprises the following steps: obtaining a plurality of to-be-tested devices; in response to a first selection instruction of a user, determining a plurality of target to-be-tested devices; in response to a second selection instruction of the user, determining a test item; and performing parallel testing on the plurality of target to-be-tested devices by using a pseudo-random number generation algorithm according to the test item. In this way, the problem of insufficient one-to-one test efficiency is solved, and the test efficiency of the to-be-tested devices is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of device testing, in particular to a multi-device testing management method, device and computer readable storage medium. BACKGROUND

[0002] At present, in the testing process of the semiconductor industry, multi-device testing technology is used in the process of automatic testing / sorting, and the testing efficiency and accuracy / consistency are improved through the automatic testing system of software.

[0003] However, the traditional testing software communicates with the device one by one, and the deficiency is that the efficiency can only serve one testing device, or manual intervention is required to manage multiple software, which causes work redundancy and is prone to human negligence. SUMMARY

[0004] The technical problem solved by the present application is to provide a multi-device testing management method, device and computer readable storage medium, to solve the problem of insufficient one-to-one testing efficiency and improve the testing efficiency of the testing device.

[0005] One technical solution adopted by the present application is to provide a multi-device testing management method, which comprises: obtaining a plurality of testing devices; in response to a first selection instruction of a user, determining a plurality of target testing devices; in response to a second selection instruction of the user, determining a testing item; and performing parallel testing on the plurality of target testing devices by using a pseudo-random number generation algorithm according to the testing item.

[0006] The parallel testing on the plurality of target testing devices by using the pseudo-random number generation algorithm comprises: the formula of the pseudo-random number generation algorithm is as follows:

[0007]

[0008] wherein X k represents the state after the kth iteration; g represents the multiplication factor; c represents the addition constant; m represents the cycle length; k_(j) represents the jth bit in the binary representation of k; mod2^m represents dividing a number by 2 raised to the power of m and then taking the remainder.

[0009] In the parallel testing process, the parallel program is written in MPI standard.

[0010] In the program written in MPI standard, each process has a corresponding memory space and can run independently, and the processes communicate and cooperate through message passing.

[0011] The MPI parallel program is debugged by using the method of embedding based on the LLVM compilation framework intermediate representation.

[0012] The method for debugging the MPI parallel program embedded based on the LLVM compilation framework intermediate representation comprises the following steps: obtaining a program to be tested; converting the program to be tested into a feature intermediate representation embedding through LLVM intermediate representation vector embedding; obtaining a feature subset according to the feature intermediate representation embedding; inputting the feature subset into a decision tree model to obtain a prediction result of the program to be tested; and the prediction result comprises a correct result or an error result.

[0013] The method for debugging the MPI parallel program embedded based on the LLVM compilation framework intermediate representation comprises the following steps: obtaining a program to be tested; converting the program to be tested into a feature intermediate representation embedding through LLVM intermediate representation vector embedding; obtaining a feature subset according to the feature intermediate representation embedding; inputting the feature subset into a decision tree model to obtain a prediction result of the program to be tested; and the prediction result comprises a correct result or an error result.

[0014] The method for debugging the MPI parallel program embedded based on the LLVM compilation framework intermediate representation comprises the following steps: obtaining a program to be tested; converting the program to be tested into a feature intermediate representation embedding through LLVM intermediate representation vector embedding; obtaining a feature subset according to the feature intermediate representation embedding; inputting the feature subset into a decision tree model to obtain a prediction result of the program to be tested; and the prediction result comprises a correct result or an error result.

[0015] Another technical solution adopted by the present application is to provide a multi-device test management device, which comprises a processor and a memory connected to the processor; the memory is used for storing a computer program, and the computer program is used for implementing the method provided by the above technical solution when executed by the processor.

[0016] Another technical solution adopted by the present application is to provide a computer readable storage medium for storing a computer program, and the computer program is used for implementing the method provided by the above technical solution when executed by the processor.

[0017] The beneficial effects of the present application are as follows: Different from the prior art, the multi-device test management method, device and computer readable storage medium provided by the present application. The method obtains a plurality of test devices to be tested; in response to a first selection instruction of a user, a plurality of target test devices to be tested are determined; in response to a second selection instruction of the user, a test item is determined; and the plurality of target test devices to be tested are tested in parallel by using a pseudo-random number generation algorithm according to the test item, so as to solve the problem of insufficient one-to-one test efficiency and improve the test efficiency of the test device to be tested. BRIEF DESCRIPTION OF DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments description. Obviously, the drawings in the following description only constitute some embodiments of the present application. For those skilled in the art, other drawings can also be obtained from these drawings without any creative effort.

[0019] Figure 1 is a flowchart of an embodiment of the multi-device test management method provided by the present application;

[0020] Figure 2 is a flowchart of another embodiment of the multi-device test management method provided by the present application;

[0021] Figure 3 is a flowchart of an embodiment of step 14 provided by the present application;

[0022] Figure 4 is a flowchart of an embodiment of step 142 provided by the present application;

[0023] Figure 5 is a structural diagram of an embodiment of the multi-device test management device provided by the present application;

[0024] Figure 6 is a structural diagram of an embodiment of the computer readable storage medium provided by the present application. DETAILED DESCRIPTION

[0025] The technical solutions in the embodiments of the present application will be described clearly and completely with reference to the drawings in the embodiments of the present application. It can be understood that the specific embodiments described herein are only used to explain the present application, but not to limit the present application. In addition, it should be noted that, for the convenience of description, only the parts related to the present application are shown in the drawings, but not all the structures. Based on the embodiments in the present application, all the other embodiments obtained by those skilled in the art without any creative effort are within the scope of protection of the present application.

[0026] In this document, reference to“an embodiment” means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the application. The appearances of the phrase“in an embodiment” in various places in the specification are not necessarily all referring to the same embodiment, nor are they necessarily mutually exclusive of one another. Those skilled in the art will understand that the embodiments described herein are merely examples of the application and are not the only way in which the application can be practiced.

[0027] At present, in the test link of the semiconductor industry, the multi-device test technology is needed in the process of using the automatic test / Sorting, and the test efficiency and accuracy / consistency are improved by the automatic test system of the software.

[0028] But the traditional test software is one-to-one communication with the device, the deficiency is that the efficiency can only serve one test device, or manual intervention in the case of management of multiple software, resulting in redundant work while prone to human oversight.

[0029] Based on this, the present application proposes to obtain a plurality of devices to be tested; in response to the first selection instruction of the user, determine a plurality of target devices to be tested; in response to the second selection instruction of the user, determine the test items; according to the test items, use the pseudo-random number algorithm to generate a plurality of target devices to be tested in parallel testing mode, so as to solve the problem of one-to-one test efficiency, and improve the test efficiency of the device to be tested. For specific technical solutions, please refer to any of the following embodiments.

[0030] Reference Figure 1 , Figure 1 is a flowchart of an embodiment of the multi-device test management method provided by the present application. The control method comprises:

[0031] Step 11: obtaining a plurality of devices to be tested.

[0032] In some embodiments, the plurality of devices to be tested can be connected to the computer (multi-device test management device) through the connection line. The multi-device test management device can display the selection items and the related information of the devices to be tested through the display interface.

[0033] Step 12: in response to the first selection instruction of the user, determine a plurality of target devices to be tested.

[0034] In some embodiments, the user can select on the display interface of the multi-device test management device to determine the target device to be tested. For example, the number of tests is selected, and a plurality of target devices to be tested are determined according to the number of tests.

[0035] In some embodiments, the display interface of the multi-device test management device displays the device identifiers of the plurality of devices to be tested. The user can select on the display interface of the multi-device test management device to determine the target device to be tested. For example, the device identifier is selected, and a plurality of target devices to be tested are determined.

[0036] Step 13: in response to the second selection instruction of the user, determine the test items.

[0037] In some embodiments, the display interface of the multi-device test management device displays the item identifiers of the plurality of test items. The user can select on the display interface of the multi-device test management device to determine the target item identifier. For example, the item identifier is selected, and a plurality of test items are determined.

[0038] In an application scenario, steps 11 to 13 can be the following process: obtaining the connected plurality of to-be-tested devices through the corresponding computer, and then selecting the number of devices according to the user's selection. The specific steps are: clicking the "DUT" drop-down selection box to select the device mode and number; checking whether the selected device mode and number meet the requirements. Then, the test project is selected according to the user's selection. The specific steps are: clicking the "TestFlow" drop-down selection box to select the test project.

[0039] Step 14: According to the test project, a plurality of target to-be-tested devices are tested in parallel by using the pseudo-random number generation algorithm.

[0040] Among them, the plurality of target to-be-tested devices are tested in parallel by using the pseudo-random number generation algorithm, including: the formula of the pseudo-random number generation algorithm is as follows:

[0041]

[0042] Among them, X k represents the state after the kth iteration; g represents the multiplication factor; c represents the addition constant; m represents the cycle length; k_(j) represents the jth bit in the binary representation of k; mod2^m: represents dividing a number by 2 raised to the power of m, and then taking the remainder.

[0043] The working principle of the above pseudo-random number generation algorithm is as follows:

[0044] 1. Decompose the binary representation of k into m bits.

[0045] 2. Multiply each bit by the corresponding power of 2.

[0046] 3. Multiply these products and take modulo 2^m.

[0047] 4. Multiply c by the sum of 1 to g^(k-1) and take modulo 2^m.

[0048] 5. Add the results of steps 3 and 4 and take modulo 2^m.

[0049] Assuming that the parameters of the pseudo-random number generation algorithm are: g = 1103515245; c = 12345; m = 32. Then, the calculation result when k = 10 is:

[0050] X_10 = (g^0)^1 * (g^2)^0 * (g^4)^1 * (g^8)^1 * (g^16)^0 * (g^32)^0 + c * (1 + g +... + g^9) = 1140712840.

[0051] The advantages of the pseudo-random number generation algorithm are: the LCG iteration of any number can be quickly calculated, and the LCG state with a guaranteed offset relative to another state can be generated, which reduces the probability of repetition of the pseudo-random number sequence.

[0052] In the embodiment, a plurality of to-be-tested devices are acquired, a plurality of target to-be-tested devices are determined in response to a first selection instruction of a user, a test item is determined in response to a second selection instruction of the user, and the plurality of target to-be-tested devices are tested in parallel by using the pseudo-random number generation algorithm according to the test item, so as to solve the problem of low one-to-one test efficiency and improve the test efficiency of the to-be-tested devices.

[0053] Further, the parallel program in the parallel test process is written by using the MPI standard.

[0054] In the program written by using the MPI standard, each process has a corresponding memory space and can be independently run, and the processes communicate and cooperate through message passing.

[0055] The advantages of the MPI program are:

[0056] Portability: The MPI standard is universal and can be ported to different platforms.

[0057] Ease of use: MPI provides a simple interface that is easy to learn and use.

[0058] Extensibility: MPI can support large-scale parallel computing.

[0059] In the present application, the applicant finds that debugging the MPI parallel program is still a major challenge, and therefore adopts a method of embedding based on the LLVM (Low Level Virtual Machine) compilation framework intermediate representation to solve the error identification problem in the MPI program. That is, the MPI parallel program is debugged by using the method of embedding based on the LLVM compilation framework intermediate representation.

[0060] Referring to Figure 2 , the MPI parallel program can be debugged by using the method of embedding based on the LLVM compilation framework intermediate representation, and the process can be as follows:

[0061] Step 21: Acquire the to-be-tested program.

[0062] Step 22: Convert the to-be-tested program into a feature intermediate representation embedding by using the LLVM intermediate representation vector embedding.

[0063] Step 23: Obtain a feature subset according to the feature intermediate representation embedding.

[0064] Step 24: Input the feature subset into a decision tree model to obtain a prediction result of the to-be-tested program.

[0065] The prediction result includes a correct result or an incorrect result. The incorrect result indicates that there is an error recognition problem in the MPI program. The correct result indicates that the MPI program is normal.

[0066] LLVM (Low Level Virtual Machine) is a compiler infrastructure project consisting of a series of modular compiler components and toolchains. Based on the SSA (Static Single Assignment) compilation strategy, it can support static and dynamic compilation of any programming language.

[0067] LLVM intermediate representation is an important part of the LLVM architecture. It serves as the hub connecting the compiler front-end and back-end, allowing compilers for different source languages to generate LLVM intermediate representations, which are then optimized by the LLVM back-end to generate executable programs. In simple terms, the LLVM intermediate representation replaces the role of C language in modern compiler implementations. We can compile the source code of our own language into LLVM intermediate representation, and then optimize this intermediate code by the LLVM back-end and compile it into binary programs suitable for different platforms.

[0068] Flexibility: LLVM intermediate representation can serve as an intermediate layer to connect different compiler components, enabling flexible compilation processes.

[0069] Optimization: By optimizing on the LLVM intermediate representation, more efficient target code can be generated.

[0070] Cross-platform: LLVM intermediate representation can be converted and linked to machine-dependent assembly language code for specific target platforms.

[0071] The goal of LLVM intermediate representation vector embedding is to represent programs as embeddings in continuous space to better handle their syntax and semantics. By combining representation learning methods and flow information, language- and machine-independent embeddings can be generated. Specifically, LLVM intermediate representation vector embedding models entities in the intermediate representation of source code as relationships and learns their representation form to form a seed embedding vocabulary. Then, LLVM intermediate representation vector embedding proposes two incremental encoding methods: symbolic encoding and flow-aware encoding. Symbolic encoding is obtained from the seed embedding vocabulary, while flow-aware encoding is obtained by combining symbolic encoding with flow information. Using this method can achieve good results in multi-process optimization tasks.

[0072] Training phase:

[0073] Training program: First, provide a set of preprocessed training programs that have been processed and converted into feature intermediate representation embeddings by LLVM intermediate representation vector embedding.

[0074] Feature selection: Next, feature selection is performed on these features to obtain a feature subset.

[0075] Label: Used to indicate whether these programs are correct or contain known MPI errors.

[0076] Learning algorithm: Taking these feature subsets and labels as input, we obtain a decision tree model (DT model) through a learning algorithm.

[0077] Validation phase:

[0078] New program: Now, a new MPI program to be tested is provided, which also goes through similar steps: vector embedding processing through LLVM intermediate representation, and then feature selection to obtain another feature subset.

[0079] DT model validation: Inputting this feature subset into the previously trained DT model, a prediction is obtained, labeled as “correct” or “incorrect”.

[0080] Furthermore, step 14 above can be as shown in Figure 3 ,

[0081] Step 141: Obtain a plurality of test programs according to a test item.

[0082] Step 142: Determine the concurrency order of the plurality of test programs using a concurrency optimizer.

[0083] Referring to Figure 4 , step 142 can be the following flow:

[0084] Step 41: Call the worker program through the MPI interface.

[0085] Step 42: Determine the initial running time using the worker program.

[0086] Step 43: Iterate the initial running time to obtain the minimum simulation running time.

[0087] Step 44: Optimize the concurrency order of the plurality of test programs based on the minimum simulation running time.

[0088] Step 143: Perform parallel testing on a plurality of target devices to be tested based on the concurrency order using a pseudo-random number generation algorithm.

[0089] Further, the concurrent processes corresponding to the plurality of test procedures are acquired, and the query request and the query request serial number are inquired through the MPI interface. Then, the corresponding working procedure is inquired by the concurrent optimizer, and the corresponding query result and serial number and the running time and serial number are obtained. The initial running time is iterated to obtain the minimum simulation running time, and then the simulation running time of the working procedure with the same serial number is updated. The concurrent order of the plurality of concurrent processes is optimized through the minimum simulation running time, and the overall test efficiency is improved.

[0090] In an application scenario, a plurality of to-be-tested devices connected to a PC are acquired. The to-be-tested devices will be fixed as a serial number after being connected to the PC. In this embodiment, the serial port names of the plurality of devices connected to the PC are acquired. The steps are specifically as follows: the devices are connected to the PC through hardware. After the user installs the software, the serial number list of the devices can be seen in the device connection window.

[0091] Then, the selection of the number of devices is performed according to the user's selection. In order to support different numbers of devices and meet different user needs, in this embodiment, the actual running device and mode are selected through the number of test devices selected by the user. The specific operation is as follows: the device mode and number are selected by clicking the “DUT” drop-down selection box. The mode is divided into synchronous and parallel. Synchronous means testing and management of a plurality of test devices at the same time in the literal sense. Parallel means testing and management in a mode of 5 groups. Then, whether the selection meets the requirements is checked. In this embodiment, the parameters selected by the user in the “DUT” are compared with the actual number of available test devices, and the selection is confirmed to be valid before proceeding. For example, in an environment where 3 test devices are connected, assuming that the user selects synchronous 5 in the “DUT” option, the software compares and then issues a prompt that it is not available and needs to be reselected to meet the actual number of options.

[0092] Then, the selection of the test project is performed according to the user's selection. In this embodiment, the required test projects are integrated into a software list on the software, and the user can select the test project to be performed. The specific operation is as follows: the test project is selected by clicking the “TestFlow” drop-down selection box.

[0093] Then, the test project is started. In this embodiment, the setting of the test device has been completed, and the user can start the test at this stage. Once the test is started, the test data will be updated on the software in real time.

[0094] Then, the test project is stopped and ended. In this embodiment, the software system has started the test, and the user can stop the test at this stage or until the test is ended. In the test process, the test method of any one of the above embodiments is combined.

[0095] Further, an automated test system is provided. The system comprises a test equipment connection module 100, a test equipment mode and quantity selection module 200, a test item selection module 300, a test control module 400, and a machine automation test system module 500 for interacting with the operation of the equipment. The test equipment connection module 100 is configured to acquire information of a plurality of equipment and establish a communication connection; the test equipment mode and quantity selection module 200 is configured to select an operation mode and a quantity of the equipment; the test item selection module 300 is configured to select a test item; the test control module 400 is configured to control a test process, such as starting and stopping; and the machine automation test system module 500 is configured to interact with the operation of the machine equipment, so as to realize mechanical and automated operation. In this embodiment, the plurality of equipment chips to be tested are classified according to the test system operation result, and the wafer is processed on the machine. The automated test system can implement the method of any one of the above embodiments.

[0096] Referring to Figure 5 , Figure 5 An embodiment of a multi-equipment test management device provided by the present application is shown in the structural schematic diagram. The multi-equipment test management device 50 comprises a processor 51 and a memory 52 connected to the processor 51; the memory 52 is configured to store a computer program, and the computer program, when executed by the processor 51, is configured to implement the following method:

[0097] acquire a plurality of equipment to be tested; in response to a first selection instruction of a user, determine a plurality of target equipment to be tested; in response to a second selection instruction of the user, determine a test item; and perform parallel testing on the plurality of target equipment to be tested according to the test item by using a pseudo-random number generation algorithm.

[0098] It can be understood that the computer program, when executed by the processor 51, is further configured to implement the method provided in any one of the above embodiments.

[0099] Referring to Figure 6 , Figure 6 An embodiment of a computer readable storage medium provided by the present application is shown in the structural schematic diagram. The computer readable storage medium 60 is configured to store a computer program 61, and the computer program 61, when executed by a processor, is configured to implement the following method:

[0100] acquire a plurality of equipment to be tested; in response to a first selection instruction of a user, determine a plurality of target equipment to be tested; in response to a second selection instruction of the user, determine a test item; and perform parallel testing on the plurality of target equipment to be tested according to the test item by using a pseudo-random number generation algorithm.

[0101] It can be understood that the computer program 61, when executed by the processor, is further configured to implement the method provided in any one of the above embodiments.

[0102] In summary, the multi-device test management method, device and computer readable storage medium provided by the application. The method acquires a plurality of to-be-tested devices; in response to a first selection instruction of a user, determines a plurality of target to-be-tested devices; in response to a second selection instruction of the user, determines a test item; and in accordance with the test item, uses a pseudo-random number generation algorithm to perform parallel testing on the plurality of target to-be-tested devices, thereby solving the problem of insufficient one-to-one test efficiency and improving the test efficiency of the to-be-tested devices.

[0103] In several embodiments provided in the present application, it should be understood that the disclosed methods and devices can be implemented in other ways. For example, the device embodiments described above are only illustrative, and for example, the division of the modules or units is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed.

[0104] The units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, i.e., can be located in one place or distributed on a plurality of network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the present embodiment scheme.

[0105] In addition, the functional units in each embodiment of the present application can be integrated in one processing unit, or each unit can be physically present alone, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0106] The integrated units in the above other embodiments, if realized in the form of a software functional unit and sold or used as an independent product, can be stored in a storage medium. Based on such understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to execute all or part of the steps of the methods described in the embodiments of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0107] The above merely describes the embodiments of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation, or direct or indirect application in other related technical fields, which is made by using the content of the present application specification and drawings, is also included in the patent protection scope of the present application.

Claims

1. A multi-device test management method, characterized in that, The method includes: Acquire multiple devices to be tested; In response to the user's first selection command, several target devices to be tested are identified; In response to the user's second selection instruction, determine the test items; According to the test items, the pseudo-random number generation algorithm is used to perform parallel testing on the several target devices under test; The parallel testing of the plurality of target devices under test using a pseudo-random number generation algorithm includes: The formula for the pseudo-random number generation algorithm is as follows: Among them, X k The state after the k-th iteration is represented by: g represents the multiplication factor; c represents the addition constant; m represents the loop length; k_(j) represents the j-th bit in the binary representation of k; mod2^m: means dividing a number by 2 to the power of m and taking the remainder. The parallel program in the parallel testing process is written using the MPI standard; In the program written using the MPI standard, each process has its own memory space and can run independently, and the processes communicate and cooperate with each other through message passing. Debugging MPI parallel programs is done using an intermediate representation embedding method based on the LLVM compilation framework; The method for debugging MPI parallel programs using intermediate representation embedding based on the LLVM compilation framework includes: Obtain the program to be tested; The test program is converted into a feature intermediate representation embedding by LLVM intermediate representation vector embedding; A feature subset is obtained by embedding the intermediate representation of the features; The feature subset is input into the decision tree model to obtain the prediction result of the program under test; the prediction result includes correct result or incorrect result; The step of performing parallel testing on the plurality of target devices under test using a pseudo-random number generation algorithm according to the test items includes: Obtain multiple test programs based on the test project; Use a concurrency optimizer to determine the concurrency order of multiple test programs; Parallel testing of the target devices under test is performed using a pseudo-random number generation algorithm based on concurrent order. The method of using a concurrency optimizer to determine the concurrency order of multiple test programs includes: Call the working program via the MPI interface; The initial running time is determined using the aforementioned working procedure; Iterate through the initial running time to obtain the minimum simulation running time; Optimize the concurrent order of multiple test programs based on the minimum simulation runtime.

2. A multi-device test management device, characterized in that, The multi-device test management device includes a processor and a memory connected to the processor; the memory is used to store a computer program, which, when executed by the processor, is used to implement the method as described in claim 1.

3. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store a computer program, which, when executed by a processor, is used to implement the method as described in claim 1.

Images