Method for testing server hardware signals and electronic device

Abstract

The application discloses a kind of server hardware signal test method and electronic equipment, it is related to server hardware signal test technical field, including by pre-set fixed parameter standard value, with the standard value as comparison benchmark, avoid different determination to test parameter due to individual experience difference;With the monotonicity trend of waveform edge as the basis for judging waveform quality, combined with image recognition technology, the waveform chart derived by oscilloscope is standardized analyzed, and the subjective interpretation deviation when manually observing waveform is excluded.For the problem of high manpower consumption and low efficiency, this method does not require engineers to examine waveform charts one by one, manually record parameters, and can automatically obtain test files containing waveform charts and test parameters, automatically compare parameters with standard values, image recognition of waveform charts and result determination, and finally generate a test report according to the test file and the determination result, reduce the repeated operation of manual operation, improve the test efficiency, and reduce the labor cost.

Description

Technical Field

[0001] This application relates to the field of server hardware signal testing technology, and in particular to a method and electronic device for testing server hardware signals. Background Technology

[0002] In the field of server hardware signal testing technology, board-level signal quality testing is a crucial step in ensuring server hardware performance. The quality of the signal directly determines the reliability and lifespan of the server. Currently, in practice, server hardware signal testing still relies heavily on manual intervention by hardware test engineers. Specifically, test engineers must manually select test points and execute tests. After testing, they must also manually classify and assess the test waveforms, as well as compile and output test reports. The entire process lacks fast, effective, and intelligent processing methods.

[0003] The aforementioned manually-led testing scheme has significant drawbacks. Firstly, when multiple engineers participate in the testing, the judgment of waveform quality is easily influenced by individual subjective factors, leading to inconsistent judgment standards among different engineers. This not only undermines the consistency of test results but may also negatively impact the consistency of server hardware design and the quality of the final product. Secondly, tasks such as waveform recognition, classification, and report integration all rely on manual labor, requiring substantial manpower and resulting in low overall testing efficiency, failing to meet the demands for large-scale and efficient server hardware testing. Summary of the Invention

[0004] This application provides a method and electronic device for testing server hardware signals, which at least solves the problem that in related technologies, server board-level signal quality testing is performed manually by engineers, resulting in inconsistent judgment standards among multiple engineers, high manpower consumption, and low efficiency.

[0005] This application provides a method for testing server hardware signals, comprising: acquiring a test file for the server hardware signals, the test file including a test waveform diagram and test parameters; the test waveform diagram being a waveform diagram of the server hardware signals exported from an oscilloscope; comparing the test parameters with preset standard values; if the test parameters meet the preset standard values, performing image recognition on the test waveform diagram to obtain an image recognition result; determining whether the test waveform of the server hardware signals meets the monotonicity trend of the waveform edges based on the image recognition result to obtain a test result; the test result being used to reflect whether the quality of the test waveform of the server hardware signals meets the standard; and generating a test report based on the test file and the test result.

[0006] This application also provides a testing device for server hardware signals, comprising:

[0007] The acquisition module is used to acquire test files for server hardware signals. The test files include test waveforms and test parameters. The test waveforms are waveforms of server hardware signals exported from an oscilloscope.

[0008] The comparison module is used to compare the test parameters with preset standard values;

[0009] The image recognition module is used to perform image recognition on the test waveform to obtain the image recognition result when the test parameters meet the preset standard values.

[0010] The judgment module is used to determine whether the test waveform of the server hardware signal meets the monotonicity trend of the waveform edge based on the image recognition result, so as to obtain the test result; the test result is used to reflect whether the quality of the test waveform of the server hardware signal meets the standard.

[0011] The report generation module is used to generate test reports based on test files and test results.

[0012] This application also provides an electronic device, including: a memory for storing a computer program; and a processor for executing the computer program to implement the steps of the above-described test method for server hardware signals.

[0013] This application also provides a computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the steps of the above-described test method for server hardware signals.

[0014] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the above-described server hardware signal testing method.

[0015] This application uses preset, fixed parameter standard values ​​as a comparison benchmark to avoid engineers making different judgments on test parameters due to differences in personal experience. It uses the monotonicity trend of waveform edges as the basis for judging waveform quality, and combines image recognition technology to standardize the analysis of waveforms exported from the oscilloscope, eliminating subjective interpretation biases from manual waveform observation, ensuring that different engineers follow the same judgment logic when performing tests. Addressing the problems of high manpower consumption and low efficiency, this method automates the entire process of core testing: engineers no longer need to examine waveforms one by one or manually record parameters; it automatically obtains test files containing waveforms and test parameters, automatically compares parameters with standard values, performs image recognition of waveforms, and judges results. Finally, it generates a test report based on the test files and judgment results, reducing repetitive manual operations, improving testing efficiency, and lowering labor costs. Attached Figure Description

[0016] To more clearly illustrate the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 A schematic diagram of the specific hardware architecture on which the server hardware signal testing method provided in this application depends;

[0018] Figure 2 A flowchart illustrating a method for testing server hardware signals provided in an embodiment of this application;

[0019] Figure 3A A test waveform diagram provided for an embodiment of this application;

[0020] Figure 3B Another test waveform diagram provided for an embodiment of this application;

[0021] Figure 4 An image recognition result provided in an embodiment of this application;

[0022] Figure 5 A schematic diagram of a server hardware signal testing device provided in an embodiment of this application;

[0023] Figure 6 This application provides a schematic diagram of the structure of an electronic device. Detailed Implementation

[0024] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the protection scope of this application.

[0025] It should be noted that, in the description of this application, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. The terms "first," "second," etc., in this application are used to distinguish similar objects and are not used to describe a specific order or sequence.

[0026] To more clearly illustrate the embodiments of this application, the technical terms used in the embodiments will be briefly introduced below:

[0027] Comma-Separated Values ​​(CSV) is a comma-separated value file format that stores tabular data in plain text. It converts the required binary data into ASCII characters and saves it as ASCII data, which can be opened with Excel, MATLAB, or text editors.

[0028] In the field of server hardware signal testing technology, board-level signal quality testing is a crucial step in ensuring server hardware performance. The quality of the signal directly determines the reliability and lifespan of the server, and this testing stage primarily relies on oscilloscopes as the core testing equipment. Currently, Tektronix is ​​a leading oscilloscope manufacturer, and its oscilloscope products are equipped with open data interfaces, supporting users to transmit and process test data, providing basic hardware support and data processing convenience for signal quality testing.

[0029] However, in current server hardware signal testing practices, the entire process still relies heavily on manual intervention by hardware test engineers. Specifically, test engineers must manually select test points and execute tests. After the tests are completed, they must also manually classify the test waveforms, assess their quality, and integrate and output test reports. The entire process lacks fast, effective, and intelligent processing methods.

[0030] The aforementioned manually-led testing scheme has significant drawbacks. Firstly, when multiple engineers participate in the testing, the judgment of waveform quality is easily influenced by individual subjective factors, leading to inconsistent judgment standards among different engineers. This not only undermines the consistency of test results but may also negatively impact the consistency of server hardware design and the quality of the final product. Secondly, tasks such as waveform recognition, classification, and report integration all rely on manual labor, requiring substantial manpower and resulting in low overall testing efficiency, failing to meet the demands for large-scale and efficient server hardware testing.

[0031] To address all or part of the aforementioned technical problems, this application provides a method for testing server hardware signals. This method replaces subjective human judgment with a standardized process and reduces human input through automated processing, fundamentally solving the pain points of purely manual testing.

[0032] Specifically, firstly, the test parameters exported from the oscilloscope are mechanically compared with preset standard values, replacing the subjective judgment of engineers based on experience regarding whether the parameters meet the standards. This avoids inconsistencies in standards caused by different engineers' understanding, ensuring that the parameter-level judgment results are unique and traceable. Secondly, based on the premise that the parameters meet the standards, discrete sampling point data of the target sampling time period in the test waveform is automatically extracted through image recognition. Then, monotonicity is judged according to preset algorithms (e.g., rising edges require "positive time difference and positive voltage difference", falling edges require "positive time difference and negative voltage difference"), completely abandoning the fuzzy judgment mode of manually observing waveform trends. This solidifies the edge quality judgment logic into executable code rules, eliminating deviations caused by human factors. From test file acquisition, parameter comparison, image recognition, monotonicity judgment to the final generation of a structured report containing waveform diagrams, parameter comparisons, and test results, the entire process is completed automatically by the system, eliminating the need for engineers to analyze each waveform individually, manually record, or format reports. This not only reduces repetitive work (such as manually calculating differences and organizing data), but also shortens the single waveform test cycle from several minutes manually to seconds automatically processed by the system, improving testing efficiency, especially suitable for batch testing scenarios of server board-level signals. It solves the problem of inconsistent judgment standards among multiple engineers, reduces labor costs, improves testing efficiency, and achieves standardization and high efficiency in server board-level signal quality testing.

[0033] To enable those skilled in the art to better understand the present application, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0034] The specific application environment architecture or specific hardware architecture on which the execution of a test method for server hardware signals depends is described herein.

[0035] like Figure 1 The diagram illustrates the specific hardware architecture upon which the server hardware signal testing method relies. The execution of this method depends on a specific hardware architecture comprising a test execution layer, a data transmission layer, and a data processing layer, with each layer working together to support fully automated and standardized testing.

[0036] The core hardware of the test execution layer includes a server board under test for providing general-purpose input / output (GPIO) and timing-waiting test hardware signals; an oscilloscope for real-time acquisition of voltage-time changes of signals, generating test waveforms and calculating test parameters such as maximum / minimum voltage values ​​and time differences; and a hardware automation test device that can replace some manual operations, enabling batch and efficient signal acquisition and preliminary testing. It can work in conjunction with an oscilloscope to complete automated triggering and data capture, and can be used in conjunction with manual testing to assist verification in special scenarios.

[0037] The data transmission layer needs to have data transmission interfaces and links, including wired transmission interfaces (such as Ethernet interfaces) between the oscilloscope and the host computer, to stably transmit the test waveforms, quantization parameters and raw sampled data in CSV format generated by the oscilloscope to the host computer, ensuring that the data is not lost or distorted, and providing a complete basis for subsequent processing.

[0038] The core hardware of the data processing layer is the host computer (usually a high-performance computer), which needs to have sufficient computing power and storage resources to support the operation of image recognition algorithms (such as discrete sampling point extraction and difference calculation), automated comparison of test parameters and preset standard values, waveform classification and organization, and generation of Excel format test reports. At the same time, it needs to be compatible with oscilloscope data transmission protocols and test file formats to ensure the smoothness of data parsing and processing, and ultimately achieve full-process hardware support from data acquisition to result output.

[0039] The embodiments of this application provide a method for testing server hardware signals. The method is described in detail below with reference to the execution flow of the server hardware signal testing method.

[0040] like Figure 2 As shown, the method includes the following steps:

[0041] S201, Test file for acquiring server hardware signals.

[0042] The test file includes test waveforms and test parameters. The test waveforms are waveforms exported from an oscilloscope after using automated hardware testing equipment or manually testing server hardware signals, for example... Figure 3A As shown in Figure 3B, the voltage and time variation characteristics of the signal are presented intuitively. Hardware automation devices can achieve batch, efficient signal detection, while manual testing is used for special scenarios or to verify the results of automation.

[0043] Test parameters are the set test data, such as the maximum and minimum voltage values, time difference, etc.

[0044] In terms of file management, test files can be named according to the signal names of server hardware signals. Test files can be categorized by test type, including but not limited to timing, GPIO, monotonicity, etc. Each test type contains subcategories, such as positive and negative. For example, "GPIO signal - positive rising edge test waveform.csv". This hierarchical classification method enables rapid data retrieval and management, meeting the needs for traceability and analysis of test data in different scenarios.

[0045] S202. Compare the test parameters with the preset standard values.

[0046] Among them, the preset standard (spec) value is a pre-set standard value or standard range used to determine whether the test parameter is qualified.

[0047] Comparing the actual test parameters with preset standard values ​​allows for a preliminary determination of whether the measured parameters meet the specifications. For each waveform, the measured parameters are compared with the preset standard values. If all measured parameters of the waveform meet the specifications, the process continues to the next waveform. If any test parameter does not meet the preset standard value, the server hardware signal is determined to be non-compliant with quality standards, and the waveform test result is judged as FAIL (Failure to Meet Standards). The subsequent processing of the current waveform is terminated, and the process moves to the next waveform to be processed. For example, when there is a mismatch between the measured parameters of a waveform (such as maximum / minimum voltage, time difference, edge monotonicity, etc.) and the preset standard value, it will be judged as "FAIL". This means that the performance of the hardware signal does not meet the design requirements, which may affect the functional stability of the server, requiring further investigation into hardware faults, signal interference, or problems with the testing process.

[0048] S203. When the test parameters meet the preset standard values, perform image recognition on the test waveform to obtain the image recognition result.

[0049] In some embodiments, when the test parameters meet preset standard values, more specifically when all test parameters meet preset standard values, the test waveform is sampled to obtain discrete sampling points. Each discrete sampling point Pi contains a sampling time Ti and a corresponding voltage value Vi, denoted as Pi(Ti, Vi), i = 1, 2, 3…n. The low-level and high-level voltage values ​​of the test waveform are determined, and then the first sampling time corresponding to the low-level voltage value and the second sampling time corresponding to the high-level voltage value are found from the discrete sampling points. The period from the start of the first sampling time to the end of the second sampling time is marked as the target sampling time period. For any adjacent sampling points within the target sampling time period, the sampling time difference and voltage difference between them are calculated. At this point, the image recognition result includes the sampling time difference and the voltage difference.

[0050] The low-level voltage value VL is a reference used to determine the low level of server hardware signals. The high-level voltage value Vh is a reference used to determine the high level of server hardware signals.

[0051] The first sampling time TL corresponding to the low-level voltage value is used to define the time range during which the server hardware signal is at a low level, thus defining the initial time boundary for subsequent waveform analysis. The second sampling time TH corresponding to the high-level voltage value is used to define the time range during which the server hardware signal reaches a high level, ensuring the accuracy of waveform analysis in the high-level region.

[0052] The above embodiments first confirm that all test parameters meet preset standard values, eliminating fundamental issues such as substandard core quantitative indicators like voltage extremes and time differences. This ensures that the object entering in-depth waveform analysis has passed preliminary compliance verification, avoiding invalid analysis and improving overall testing efficiency. Secondly, by determining the high and low voltage values ​​and finding the corresponding first and second sampling times, the target sampling time period (i.e., the core interval where the waveform transitions from low to high level) is accurately marked. This interval focuses on the most critical change stage of the waveform, eliminating redundant stable level data before and after, achieving precise focus of the analysis range, and reducing interference from irrelevant data in subsequent judgments. Finally, the sampling time difference and voltage difference are calculated for adjacent sampling points within the target interval and incorporated into the image recognition results, transforming the continuous waveform into discrete, quantifiable feature data. These data directly reflect the change trend of the waveform in the core transition stage, providing a standardized and calculable basis for subsequent judgment of deep quality indicators such as waveform edge monotonicity, jitter-free and distortion-free characteristics. This extends waveform quality judgment from parameter compliance to process quality, further ensuring the stability and reliability of server hardware signals.

[0053] Optionally, after finding the first sampling time corresponding to the low-level voltage value, if there are multiple first sampling times, the earliest first sampling time is selected; and / or, after finding the second sampling time corresponding to the high level, if there are multiple second sampling times, the earliest second sampling time is selected.

[0054] If multiple TH or TL exist, select the earliest TL (the point that reaches the low level earliest) and the latest TH (the point that reaches the high level latest) to ensure that the effective data width of the server hardware signal can be captured accurately and completely, thereby ensuring the accuracy and reliability of waveform analysis.

[0055] Specifically, for low-level voltage values ​​VL, selecting the group with the smallest TL (Transmission Time) clarifies the earliest time point when the server hardware signal enters the low-level state, defining the starting boundary of the low-level region for subsequent analysis. For high-level voltage values ​​Vh, selecting the group with the largest TH (Transmission Time) determines the latest time point when the server hardware signal enters the high-level state, defining the ending boundary of the high-level region. This selection method maximizes the coverage of the complete time period from low to high level, avoiding misjudgments of the effective waveform width due to improper data point selection. It provides accurate and complete time and voltage range benchmarks for subsequent analysis steps such as waveform monotonicity judgment and voltage fluctuation verification, ensuring the rigor and effectiveness of the entire waveform analysis process.

[0056] S204. Based on the image recognition results, determine whether the test waveform of the server hardware signal satisfies the waveform edge monotonicity trend to obtain the test results.

[0057] The waveform edge monotonicity trend includes a monotonically increasing rising edge and a monotonically decreasing falling edge. If the test waveform of the server hardware signal satisfies the waveform edge monotonicity trend, from the perspective of signal quality, it indicates that the waveform edge integrity is good and there is no interference or distortion; from the perspective of testing, it indicates that the sampled data is accurate and reliable, and there are no abnormal sampling points affecting the judgment.

[0058] In some embodiments, the image recognition result includes the sampling time difference and voltage difference between any adjacent sampling points within a target sampling time period from the first sampling time to the second sampling time. Based on the image recognition result, and combined with the rising edge in the test waveform, it is determined whether the sampling time difference and voltage difference are positive. If both the sampling time difference and voltage difference are positive, it is determined that any adjacent sampling points within the target sampling time period satisfy the waveform edge monotonicity trend, indicating that the waveform of the server hardware signal is monotonically increasing within the target time period.

[0059] Within the target sampling time period, both the time difference and voltage difference between any two adjacent sampling points are positive, indicating that the first reciprocal of the function between these two sampling points is positive. A consistently positive voltage difference at the rising edge signifies that the voltage value continuously increases over time without any instantaneous drop.

[0060] like Figure 4 As shown, the high-level voltage value corresponds to sampling point Pn, and the low-level voltage value corresponds to sampling point P1. For any adjacent sampling points P3 and P4, calculate the sampling time difference ΔT34 and the voltage difference ΔV34 between them.

[0061] Specifically, the target sampling time period from the first sampling time to the second sampling time is locked from the image recognition results. This sampling time period is aligned with the effective analysis range of the rising edge in the test waveform diagram to ensure coverage of the complete transition process of the rising edge from low level to high level. At the same time, the sampling time difference (i.e., the time difference between the next sampling point and the previous sampling point) and voltage difference (i.e., the voltage difference between the next sampling point and the previous sampling point) of all adjacent sampling points in this interval are extracted from the image recognition results to form the original data pair to be verified.

[0062] Secondly, for each set of adjacent sampling points' time difference and voltage difference, a double positive judgment is performed. First, it is verified whether the sampling time difference is positive to ensure that the sampling process proceeds continuously in chronological order; then, it is verified whether the corresponding voltage difference is positive to ensure that the voltage value of the later sampling point is higher than that of the previous sampling point, which conforms to the basic characteristics of voltage increase as the rising edge progresses. This process needs to cover all adjacent sampling points within the target sampling time period, without omitting any data pair, to avoid misjudgment due to local data anomalies.

[0063] Finally, the waveform monotonicity conclusion is determined based on the matching results. If the time difference and voltage difference of all adjacent sampling points within the target sampling time period satisfy the double positive condition, it indicates that the voltage steadily increases with time throughout the entire rising edge transition process, without reverse fluctuations or abnormal sudden changes. It can be determined that the sampling points in this interval satisfy the waveform edge monotonicity trend, and thus it can be determined that the server hardware signal waveform is monotonically increasing within the target time period. If any set of sampling points does not satisfy the double positive condition (such as a negative time difference, a negative or zero voltage difference), further investigation of sampling anomalies or waveform interference is required, and it is not determined to meet the monotonicity requirements for the time being.

[0064] For example, such as Figure 3A The rising edge waveform of the server hardware signal shown starts from a stable low level, rises rapidly, and then enters a stable high level phase. The time difference and voltage difference between adjacent sampling points during the rising process are both positive, which meets the monotonicity requirement of a monotonically increasing rising edge. Furthermore, there is no significant fluctuation exceeding (Vh±1%) in the high-level region, and the low-level region is also stable near VL. This rising edge waveform can be judged as passing (PASS), indicating that the rising edge waveform quality of the server hardware signal meets the standard.

[0065] In other embodiments, based on the image recognition results and the falling edge in the test waveform, it is determined whether the sampling time difference and voltage difference are negative. If the sampling time difference is positive but the voltage difference is negative, it indicates that the waveform of the server hardware signal is monotonically decreasing within the target time period. Therefore, it is determined that any adjacent sampling points within the target sampling time period satisfy the monotonicity trend of the waveform edges. Specifically, a consistently negative voltage difference at the falling edge means that the voltage value continuously decreases over time without any instantaneous rebound.

[0066] Specifically, firstly, the target sampling time period from the first sampling time to the second sampling time is determined from the image recognition results. This target sampling time period needs to be precisely aligned with the effective analysis range of the falling edge in the test waveform diagram to ensure complete coverage of the entire process from the start of the falling edge at a high level to the stabilization of the low level. At the same time, the sampling time difference (i.e., the time difference between the next sampling point and the previous sampling point) and voltage difference (i.e., the voltage difference between the next sampling point and the previous sampling point) of all adjacent sampling points within this interval are extracted from the recognition results and organized into one-to-one corresponding data pairs to be verified, providing the original basis for subsequent judgment.

[0067] Secondly, for each pair of adjacent sampling points extracted, a dual verification is performed to ensure both a positive time difference and a negative voltage difference. The first step verifies whether the sampling time difference is positive, which is fundamental to ensuring the sampling process proceeds in the correct time sequence, eliminating anomalies such as time backtracking and sampling misalignment, and guaranteeing the validity of the data's temporal dimension. The second step verifies whether the corresponding voltage difference is negative. This verification directly corresponds to the core characteristic of voltage decrease as the falling edge progresses, ensuring that the voltage of the subsequent sampling point is always lower than the previous sampling point, conforming to the essential law of the falling edge transitioning from high to low level. The entire verification process must traverse all adjacent sampling points within the target time period without skipping any data set, avoiding misjudgments of the overall monotonicity of the waveform due to missing local data.

[0068] Finally, the waveform monotonicity is determined based on the verification results. If all adjacent sampling points within the target sampling time period meet the conditions of a positive sampling time difference and a negative voltage difference, it indicates that the voltage consistently decreases steadily over time throughout the entire falling edge transition process, without any abnormal fluctuations such as reverse rise or voltage stagnation. This confirms that the adjacent sampling points within this interval satisfy the waveform edge monotonicity trend, and thus determines that the waveform of the server hardware signal is in a monotonically decreasing state within the target time period. If any set of sampling points does not meet this condition (e.g., a negative sampling time difference, a positive or zero voltage difference), it indicates that the waveform is abnormal, and further investigation is needed to identify problems such as sampling equipment failure or signal interference. In this case, it is not considered to meet the falling edge monotonicity requirement.

[0069] For example, such as Figure 3B The falling edge waveform of the server hardware signal shown drops rapidly from a high level, but there is significant voltage fluctuation after the drop (the waveform has obvious undulations). The voltage at some sampling points may exceed the allowable range of (Vl±1%). If such fluctuations still exist after considering the monotonicity of positive sampling time difference and negative voltage difference, the falling edge waveform will be judged as FAIL, indicating that there is interference or distortion in the falling edge of the signal, and problems in hardware interfaces, signal transmission paths, etc. need to be investigated.

[0070] Analyze all sampling points within the time interval TL to TH, and calculate the sign of the first derivative by comparing the time difference (ΔT) and voltage difference (ΔV) between every two adjacent points. For the rising edge, ΔV must be positive when ΔT is positive, meaning the voltage value is higher as the time progresses; for the falling edge, ΔV must be negative when ΔT is positive, meaning the voltage value is lower as the time progresses.

[0071] For example, for a simulated waveform of a rising edge signal, first capture the high-level voltage value Vh and the low-level voltage value VL. Then, find the corresponding time TH for Vh and the corresponding time TL for VL in the test waveform graph. Next, for the data within the time interval from TL to TH, compare the time and voltage values ​​of adjacent sampling points sequentially. If the time difference between adjacent sampling points is positive and the voltage difference is also positive, that is, the first derivative is positive, then the monotonicity of the waveform is determined to meet the requirements. If all data meet the requirements, the waveform is determined to be PASS.

[0072] For simulating waveforms of falling edge signals, the process is similar to that of rising edge signals but with some differences. First, capture the high-level voltage value Vh and the low-level voltage value VL, and find the corresponding time TL for VL and the corresponding time TH for Vh. Then, for the data within the time interval from TH to TL, compare the time and voltage values ​​of adjacent sampling points sequentially. It is necessary to ensure that the time difference between adjacent sampling points is positive and the voltage difference is negative, i.e., the first derivative is negative. This is used to determine whether the waveform monotonicity meets the requirements. If all data meets the requirements, the waveform is judged as PASS.

[0073] In some embodiments, if any adjacent sampling points within the target sampling time period do not satisfy the monotonicity trend of the waveform edges, an intermediate sampling time is first calculated based on the first and second sampling times. Then, for discrete sampling points before the intermediate sampling time, it is determined whether their voltage values ​​are within the range of the first voltage value. For discrete sampling points after the intermediate sampling time, it is determined whether their voltage values ​​are within the range of the second voltage value.

[0074] The intermediate sampling time TM is the same as the interval between the first sampling time TL and the second sampling time TH. The first voltage value range is related to the low-level voltage value, for example, VL ± 1%; the second voltage value range is related to the high-level voltage value, for example, Vh ± 1%. This 1% fluctuation range is defined as normal to avoid misjudging small voltage value changes in the low-level or high-level regions as abnormal.

[0075] If any adjacent sampling points within the target sampling time period do not satisfy the monotonically increasing trend of the waveform edges, the voltage value of discrete sampling points before the intermediate sampling time TM, which is close to the low-level end, is determined to be within the range of VL±1% to avoid misjudgment caused by sudden changes in voltage value in the low-level region; and the voltage value of discrete sampling points after the intermediate sampling time TM, which is close to the high-level end, is determined to be within the range of Vh±1% to avoid misjudgment caused by sudden changes in voltage value in the high-level region.

[0076] For example, for the simulated waveform of the rising edge signal, if there are sampling points that do not conform to the waveform monotonicity, further verification is needed. The voltage value at the time point before the intermediate sampling time TM is compared with VL±1%, and the voltage value at the time point after TM is compared with Vh±1%. Fluctuations within 1% are ignored. If all data meet the requirements, the waveform is judged as PASS, indicating that the rising edge transition of the server hardware signal is stable and without distortion, and the signal quality meets the standard; if there are still non-compliant data points, it is judged as FAIL.

[0077] The above embodiment first calculates an intermediate sampling time TM with equal intervals using the first sampling time TL and the second sampling time TH. This precisely divides the target sampling time period into two sub-intervals: "before TM (low-level to intermediate transition phase)" and "after TM (intermediate to high-level transition phase)," providing clear boundaries for scenario-specific verification and avoiding errors caused by uniform judgment across the entire domain. Secondly, differentiated voltage judgment criteria are set for the two sub-intervals: a first voltage range (e.g., VL ± 1%) associated with the low-level voltage value VL before TM, and a second voltage range (e.g., Vh ± 1%) associated with the high-level voltage value Vh after TM. This aligns with the waveform's transition logic from low to high and clarifies the normal voltage fluctuation threshold for each stage. Finally, this design effectively distinguishes between true anomalies and acceptable fluctuations. If the monotonicity is not satisfied due to minor voltage fluctuations in the low / high level region, it is judged as a normal error, avoiding misjudgment as FAIL. If the fluctuation exceeds the corresponding range, it is confirmed as a true anomaly such as waveform distortion or interference, ensuring that the judgment results are both rigorous and flexible, guaranteeing the accuracy of server hardware signal testing.

[0078] Similarly, if any adjacent sampling points within the target sampling time period do not satisfy the monotonically decreasing trend of the waveform edge, the voltage value of discrete sampling points before the intermediate sampling time TM and close to the high-level end is determined to be within the range of Vh±1%; and the voltage value of discrete sampling points after the intermediate sampling time TM and close to the low-level end is determined to be within the range of VL±1%.

[0079] For example, for the simulated waveform of the falling edge signal, if there are any non-compliant sampling points, further verification is performed by comparing the voltage value at the time point before TM with Vh±1%, and the voltage value at the time point after TM with VL±1%. Fluctuations within 1% are ignored. If all data meet the requirements, the waveform is judged as PASS, indicating that the falling edge waveform quality of the server hardware signal meets the standard and no abnormalities affecting the effectiveness of signal transmission have occurred; if there are still non-compliant data points, it is judged as FAIL, indicating that the waveform has obvious signal distortion, sampling failure, or external interference.

[0080] In the above scenario, firstly, the interval is divided using the intermediate sampling time TM as the boundary to accurately match the core logic of the falling edge transitioning from high to low level. Before TM, the focus is on sampling points closer to the high-level end, and after TM, the focus is on sampling points closer to the low-level end. This ensures that voltage verification is consistent with the natural transition law of the waveform, avoiding judgment deviations caused by a uniform standard across the entire domain. Secondly, differentiated voltage judgment thresholds are set: before TM, the threshold is associated with ±1% of the high-level voltage value Vh, and after TM, it is associated with ±1% of the low-level voltage value VL. This clarifies the normal voltage fluctuation boundaries of each stage and conforms to the level characteristics of different intervals of the falling edge. For example, small fluctuations at the high-level end are allowed but do not exceed the reasonable range of Vh, and the same applies to the low-level end. Finally, this design effectively filters out non-core anomalies: if the monotonicity is not satisfied due to minor fluctuations in the high / low level region (not exceeding the corresponding ±1% range), it is judged as a normal error, avoiding misjudging qualified waveforms as FAIL; if the fluctuation exceeds the corresponding range, it is confirmed as a true anomaly such as waveform distortion or signal interference, ensuring that the judgment result is both fault-tolerant and does not relax the requirements for the quality of core signals, providing reliable support for the accurate testing of falling edge signals of server hardware.

[0081] S205. Generate a test report based on the test documents and test results.

[0082] Based on the different types of test files, obtain the name and test parameters of the test waveforms. Insert the test waveforms into the specified positions of their names in the test report, and insert the test parameters into the specified cells of the test report; store the test results corresponding to the test waveforms.

[0083] The naming convention for the test waveforms corresponds to the previously defined signal names and classifications, allowing users to locate specific sections within the test report based on the waveform's name. Adjusting the waveform image size and inserting it into the corresponding cell enables precise archiving of the visualized data.

[0084] In some embodiments, the test report includes, but is not limited to: basic test information (such as test object, time, environment, and device model), test item classification (by major categories such as timing, GPIO, and monotonicity, and minor categories such as positive / negative), waveform data area, parameter comparison area, and result judgment. First, signal names and classification labels are extracted from the stored test files and filled into the report header according to a preset format. For each waveform, the resized and positioned test waveform is included in the visualization area under the corresponding category. Simultaneously, the measured parameters of each waveform (such as maximum and minimum voltage values, time difference, etc.) are filled into the measured value column of the parameter comparison area, and preset standard values ​​are automatically associated with the standard value column, forming a one-to-one parameter comparison table. For each waveform, based on the previous test results (PASS / FAIL), it is automatically marked in the result judgment area. If it is FAIL, specific abnormal items need to be associated in the remarks column (such as "voltage difference is negative, not meeting rising edge monotonicity" or "high-level fluctuation exceeds Vh+1%)" to achieve accurate problem tracing. The system summarizes test results from all categories, calculates the number of pass / fail results and pass rate, and generates a report summary. After manual or automated script verification for format compliance (e.g., image placement, parameter alignment, consistency of decision logic), the report is output in archiveable formats such as Excel and PDF, supporting subsequent review, auditing, or hardware problem troubleshooting. This ensures the accuracy, standardization, and traceability of the test report, providing an intuitive and reliable basis for server hardware signal quality assessment.

[0085] In summary, this application provides a method for testing server hardware signals. This method acquires signal waveforms and parameters, and combines a file management system based on signal naming and classification to achieve structured storage and rapid retrieval of test data. This lays an orderly data foundation for subsequent analysis and report generation, improving data traceability efficiency. By determining the low-level voltage value (VL), high-level voltage value (Vh), and corresponding sampling time (TL / TH), the target sampling time period (rising edge from TL to TH, falling edge from TH to TL) is locked. Redundant stable-level data is eliminated, focusing on the core transition phase of the waveform from low to high (or high to low), reducing irrelevant data interference and improving the relevance of the analysis. By comparing measured parameters (voltage extremes, time differences, etc.) with spec values, waveforms that clearly fail to meet standards are quickly screened out, eliminating fundamental performance issues and improving testing efficiency. The time and voltage differences between adjacent sampling points within the target interval are judged using either "double positive (rising edge)" or "positive time, negative voltage (falling edge)" criteria to accurately identify abnormal trends in the waveform transition phase, ensuring signal changes conform to design logic. When monotonicity is not satisfied, threshold checks of VL±1% (before rising edge / after falling edge) and Vh±1% (after rising edge / before falling edge) are applied to sampling points before and after TM, using the intermediate time (TM) as the boundary, to distinguish between acceptable minor fluctuations and true anomalies, avoiding misjudgments and balancing the rigor and fault tolerance of the judgment. The system automatically organizes waveform images, fills in measured parameters and spec values, and links PASS / FAIL results and anomaly causes to generate a structured report, ensuring consistency between data and conclusions. It also supports subsequent hardware problem investigation and auditing, improving the standardization and reliability of the testing process.

[0086] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method.

[0087] like Figure 5 As shown, embodiments of this application also provide a testing apparatus for server hardware signals, the apparatus comprising:

[0088] The acquisition module 501 is used to acquire test files of server hardware signals. The test files include test waveforms and test parameters. The test waveforms are waveforms of server hardware signals exported from an oscilloscope.

[0089] The comparison module 502 is used to compare the test parameters with preset standard values;

[0090] The image recognition module 503 is used to perform image recognition on the test waveform to obtain the image recognition result when the test parameters meet the preset standard values.

[0091] The judgment module 504 is used to judge whether the test waveform of the server hardware signal meets the waveform edge monotonicity trend based on the image recognition result, so as to obtain the test result; the test result is used to reflect whether the quality of the test waveform of the server hardware signal meets the standard.

[0092] The report generation module 505 is used to generate a test report based on the test files and test results.

[0093] As an optional implementation provided in this application, the image recognition module 503 is specifically used for: sampling the test waveform to obtain discrete sampling points when the test parameters meet preset standard values; the discrete sampling points include sampling time and corresponding voltage values; determining the low-level voltage value and high-level voltage value of the test waveform; finding the first sampling time corresponding to the low-level voltage value and the second sampling time corresponding to the high-level voltage value from the discrete sampling points; calculating the sampling time difference and voltage difference for any adjacent sampling points within the target sampling time period; wherein, the target sampling time period is the sampling time period from the first sampling time to the second sampling time, and the image recognition result includes the sampling time difference and voltage difference.

[0094] As an optional implementation provided in this application, the image recognition module 503, after finding the first sampling time corresponding to the low-level voltage value and the second sampling time corresponding to the high-level voltage value from the discrete sampling points, and before calculating the sampling time difference and voltage difference for any adjacent sampling points within the target sampling time period, is further configured to: select the earliest first sampling time when there are multiple first sampling times; and / or, select the latest second sampling time when there are multiple second sampling times.

[0095] As an optional implementation provided in this application, the judgment module 504 is specifically used to: determine whether the sampling time difference and voltage difference are positive numbers for the rising edge waveform in the test waveform diagram; if the sampling time difference and voltage difference are both positive numbers, then determine that any adjacent sampling points within the target sampling time period satisfy the monotonicity trend of the waveform edge.

[0096] As an optional implementation provided in this application, the judgment module 504 is specifically used to: determine whether the sampling time difference and voltage difference are negative for the falling edge waveform in the test waveform diagram; if the sampling time difference is positive but the voltage difference is negative, then determine that any adjacent sampling points satisfy the monotonicity trend of the waveform edge.

[0097] As an optional implementation provided in this application, the judgment module 504 is further configured to: calculate the intermediate sampling time based on the first sampling time and the second sampling time when any adjacent sampling points within the target sampling time period do not satisfy the waveform edge monotonicity trend; determine whether the voltage value of the discrete sampling points before the intermediate sampling time is within the range of the first voltage value; wherein the range of the first voltage value is related to the low-level voltage value; if so, determine that the discrete sampling points before the intermediate sampling time satisfy the waveform edge monotonicity trend.

[0098] As an optional implementation provided in this application, the judgment module 504 is further configured to: determine whether the voltage value of discrete sampling points after the intermediate sampling time is within the range of a second voltage value; wherein, the range of the second voltage value is related to the high-level voltage value;

[0099] If so, then the discrete sampling points after the intermediate sampling time are determined to satisfy the monotonicity trend of the waveform edges.

[0100] As an optional implementation provided in this application, the report generation module 505 is specifically used for: inserting the test waveform into the specified position corresponding to the name in the test report according to the name of the test waveform; inserting the test parameters into the specified cell of the test report; and storing the test results corresponding to the test waveform.

[0101] As an optional implementation provided in this application, the comparison module 502 is further used to: determine that the quality of the test waveform of the server hardware signal is substandard when the test parameters do not meet the preset standard values.

[0102] For a description of the features in the embodiment corresponding to the server hardware signal testing device, please refer to the relevant description in the embodiment corresponding to the server hardware signal testing method, which will not be repeated here.

[0103] like Figure 6 As shown, embodiments of this application also provide an electronic device, including a memory 601 and a processor 602. The memory 601 stores a computer program, and the processor 602 is configured to run the computer program to perform the steps in any of the above-described server hardware signal testing method embodiments.

[0104] Embodiments of this application also provide a computer-readable storage medium storing a computer program, wherein the computer program is configured to execute the steps in any of the above-described server hardware signal testing method embodiments when running.

[0105] In one exemplary embodiment, the aforementioned computer-readable storage medium may include, but is not limited to, various media capable of storing computer programs, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard disk, magnetic disk, or optical disk.

[0106] Embodiments of this application also provide a computer program product, which includes a computer program that, when executed by a processor, implements the steps in any of the above-described server hardware signal testing method embodiments.

[0107] Embodiments of this application also provide another computer program product, including a non-volatile computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps in any of the above-described server hardware signal testing method embodiments.

[0108] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0109] The foregoing has provided a detailed description of a server hardware signal testing method and electronic device provided in this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only intended to help understand the method and core ideas of this application. It should be noted that those skilled in the art can make various improvements and modifications to this application without departing from its principles, and these improvements and modifications also fall within the protection scope of the claims of this application.

Claims

1. A method for testing server hardware signals, characterized in that, include: Test files for acquiring server hardware signals, including test waveforms and test parameters; The test waveform is a waveform diagram of the server hardware signals exported by an oscilloscope; The test parameters are compared with preset standard values; When the test parameters meet the preset standard values, image recognition is performed on the test waveform to obtain image recognition results; the image recognition results include sampling time difference and voltage difference. Based on the sampling time difference and the voltage difference, determine whether the test waveform of the server hardware signal satisfies the waveform edge monotonicity trend, so as to obtain the test result; The test results are used to reflect whether the quality of the test waveform of the server hardware signal meets the standard; The monotonicity trend of the waveform edges includes a monotonically increasing trend at the rising edge and a monotonically decreasing trend at the falling edge of the test waveform. A test report is generated based on the test file and the test results; Wherein, the step of performing image recognition on the test waveform to obtain the image recognition result when the test parameters meet the preset standard value includes: When the test parameters meet the preset standard values, the test waveform is sampled to obtain discrete sampling points; the discrete sampling points include the sampling time and the corresponding voltage value. Determine the low-level voltage value and high-level voltage value of the test waveform; From the discrete sampling points, find the first sampling time corresponding to the low-level voltage value and the second sampling time corresponding to the high-level voltage value; For any adjacent sampling points within the target sampling time period, calculate the sampling time difference and the voltage difference; wherein, the target sampling time period is the sampling time period from the first sampling time to the second sampling time; The method further includes: if any adjacent sampling points within the target sampling time period do not satisfy the monotonicity trend of the waveform edge, calculating an intermediate sampling time based on the first sampling time and the second sampling time; the intermediate sampling time is the same as the interval between the first sampling time and the second sampling time; For discrete sampling points before the intermediate sampling time, if their voltage values ​​are determined to be within a first voltage value range, then the discrete sampling points before the intermediate sampling time are determined to satisfy the waveform edge monotonicity trend; wherein, the first voltage value range is related to the low-level voltage value; For discrete sampling points after the intermediate sampling time, if their voltage values ​​are determined to be within the second voltage value range, then the discrete sampling points after the intermediate sampling time are determined to satisfy the waveform edge monotonicity trend; wherein, the second voltage value range is related to the high-level voltage value.

2. The method according to claim 1, characterized in that, After finding the first sampling time corresponding to the low-level voltage value and the second sampling time corresponding to the high-level voltage value from the discrete sampling points, and before calculating the sampling time difference and voltage difference for any adjacent sampling points within the target sampling time period, the method further includes: If there are multiple first sampling times, the earliest first sampling time is selected; And / or, If there are multiple second sampling times, the latest second sampling time shall be selected.

3. The method according to claim 1, characterized in that, The step of determining whether the test waveform of the server hardware signal satisfies the waveform edge monotonicity trend based on the sampling time difference and the voltage difference to obtain the test result includes: For the rising edge waveform in the test waveform diagram, determine whether the sampling time difference and the voltage difference are positive numbers; If both the sampling time difference and the voltage difference are positive, then any adjacent sampling points within the target sampling time period are determined to satisfy the waveform edge monotonicity trend.

4. The method according to claim 1, characterized in that, The step of determining whether the test waveform of the server hardware signal satisfies the waveform edge monotonicity trend based on the sampling time difference and the voltage difference to obtain the test result includes: For the falling edge waveform in the test waveform diagram, determine whether the sampling time difference and the voltage difference are negative; If the sampling time difference is positive but the voltage difference is negative, then any adjacent sampling points within the target sampling time period satisfy the waveform edge monotonicity trend.

5. The method according to claim 1, characterized in that, The step of generating a test report based on the test file and the test results includes: According to the name of the test waveform, insert the test waveform into the specified position corresponding to the name in the test report; Insert the test parameters into the specified cell of the test report; The test results are stored in correspondence with the test waveform diagram.

6. The method according to claim 1, characterized in that, After comparing the test parameters with preset standard values, the method further includes: If the test parameters do not meet the preset standard values, it is determined that the quality of the test waveform of the server hardware signal is substandard.

7. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor, configured to implement the steps of the test method for server hardware signals as described in any one of claims 1 to 6 when executing the computer program.

Images