Linux real-time development platform

Abstract

The invention provides a Linux real-time development platform, and relates to the field of intelligent development, and the Linux real-time development platform comprises the steps: firstly obtaining the characteristic description of a hardware platform, determining an adaptive Linux system configuration option based on the characteristic description of the hardware platform, then installing the adaptive Linux system on the hardware platform, and finally carrying out the real-time performance test on the hardware platform provided with the adaptive Linux system. And obtaining a performance test result. In this way, the Linux system can be subjected to configuration optimization according to the characteristics of the hardware platform, and then the reliability and stability of the Linux system are improved.

Description

Technical Field

[0001] The present application relates to the field of intelligent development, and more specifically, to a Linux real-time development platform. Background Art

[0002] With the rapid development of the Internet of Things (IoT), industrial automation, and driverless driving, embedded systems have increasingly higher requirements for real-time performance. Applications in these fields usually require the system to respond to external events within strict time limits to ensure the reliability and security of the system. As a flexible and powerful operating system, Linux has been widely used in embedded systems and other fields due to its open source, flexibility, and scalability.

[0003] However, as multi-core processors become mainstream, although they provide more powerful computing capabilities, they also bring new challenges to the real-time and deterministic nature of the system. Multi-core processors allow multiple tasks to be executed concurrently, thereby improving the overall throughput of the system. However, this also increases the complexity of scheduling and synchronization, which may result in the inability to guarantee the response time of real-time tasks. Specifically, task scheduling in a multi-core environment must not only consider the priority and execution time of tasks, but also deal with data dependencies and resource sharing issues between tasks. In addition, interrupt processing and context switching occur frequently, which increases the system overhead, making the execution time and response time of real-time tasks more unpredictable, thus affecting the reliability and stability of the system.

[0004] Therefore, a Linux real-time development platform is desired. Summary of the invention

[0005] In order to solve the above technical problems, this application is proposed. The embodiment of this application provides a Linux real-time development platform.

[0006] According to one aspect of the present application, a Linux real-time development platform is provided, which includes: a hardware platform feature description data acquisition module, used to obtain a feature description of the hardware platform; a Linux system configuration option determination module, used to determine, based on the feature description of the hardware platform, configuration options of the Linux system adapted to the hardware platform; a Linux system installation module, used to install the Linux system on the hardware platform based on the configuration options of the Linux system to obtain a hardware platform with the Linux system installed; and a performance test result generation module, used to perform real-time performance testing on the hardware platform with the Linux system installed to obtain performance test results.

[0007] In the above-mentioned Linux real-time development platform, the Linux system installation module includes: a Linux system download patch transplantation unit, which is used to download the Linux system adapted to the hardware platform on the hardware platform and transplant the real-time patch to the Linux system; a Linux system configuration unit, which is used to configure the Linux system after the real-time patch is transplanted based on the configuration options of the Linux system using the makemenuconfig or make xconfig tool, and select the real-time scheduler and set the interrupt priority according to the hardware characteristics of the hardware platform; a Linux system compilation unit, which is used to compile the configured Linux system and generate an image file; an image file deployment unit, which is used to deploy the image file on the hardware platform; and a Linux system inspection unit, which is used to perform an inspection after starting the system to ensure that the Linux system can run normally and load the real-time patch.

[0008] In the above Linux real-time development platform, the Linux system configuration option determination module includes: a hardware platform feature description word embedding encoding unit, which is used to perform word segmentation processing on the feature description of the hardware platform and then perform word embedding encoding to obtain a sequence of platform feature description word embedding encoding vectors; a hardware platform feature feature descriptive significant enhancement unit, which is used to perform platform feature feature descriptive significant enhancement on the sequence of platform feature description word embedding encoding vectors to obtain a sequence of enhanced expression platform feature description word embedding encoding vectors, wherein the hardware platform feature feature descriptive significant enhancement unit includes: a platform feature significant description factor calculation first-level subunit, which is used to calculate each platform feature in the sequence of platform feature description word embedding encoding vectors The invention relates to a method for generating a configuration option recommendation result, wherein the method comprises: a first-level sub-unit for enhancing the platform characteristic descriptor word embedding coding vector, a first-level sub-unit for enhancing the platform characteristic descriptor word embedding coding vector, and a second-level sub-unit for enhancing the platform characteristic descriptor word embedding coding vector, based on the sequence of platform characteristic significance descriptor factors, to perform hierarchical modulation enhancement on the sequence of platform characteristic descriptor word embedding coding vectors to obtain the sequence of enhanced expression platform characteristic descriptor word embedding coding vectors; and a configuration option recommendation result generating unit, wherein the unit is used for generating a configuration option recommendation result based on the enhanced expression hardware platform characteristic descriptor word granularity contextual semantic association features obtained by context encoding the sequence of enhanced expression platform characteristic descriptor word embedding coding vectors, and the configuration option recommendation result is used to represent the kernel version recommendation type label of the Linux system.

[0009] In the above-mentioned Linux real-time development platform, the hardware platform feature description word embedding encoding unit is used to: perform word segmentation processing on the feature description of the hardware platform to obtain the word granularity distribution of the platform feature description; use the feature description word embedding matrix to perform word embedding encoding on the word granularity distribution of the platform feature description to obtain a sequence of the platform feature description word embedding encoding vectors.

[0010] In the above-mentioned Linux real-time development platform, the first-level sub-unit for calculating the platform characteristic significance description factor includes: a second-level sub-unit for mapping the characteristic description word embedding coding vector hyperbolic space, which is used to map each platform characteristic description word embedding coding vector in the sequence of the platform characteristic description word embedding coding vector to a hyperbolic space to obtain a sequence of platform characteristic description word embedding coding mapping vectors; a second-level sub-unit for generating the platform characteristic significance description factor, which is used to calculate the characteristic significance description factor of each platform characteristic description word embedding coding mapping vector in the sequence of the platform characteristic description word embedding coding mapping vector to obtain a sequence of the platform characteristic significance description factors; wherein the second-level sub-unit for mapping the characteristic description word embedding coding vector hyperbolic space is used to: after matrix-vector multiplication of the first weight matrix and the platform characteristic description word embedding coding vector, vector-matrix multiplication of the obtained platform characteristic description word embedding coding weight vector and the second weight matrix to obtain the platform characteristic description word embedding coding mapping vector.

[0011] In the above-mentioned Linux real-time development platform, the platform characteristic significance description factor generation secondary subunit is used to: calculate the square of the norm of the platform characteristic descriptor embedding coding mapping vector to obtain the platform characteristic descriptor embedding coding mapping norm value; add and subtract the constant one from the platform characteristic descriptor embedding coding mapping norm value to obtain a first platform characteristic descriptor embedding coding mapping norm modulation value and a second platform characteristic descriptor embedding coding mapping norm modulation value, and divide the first platform characteristic descriptor embedding coding mapping norm modulation value by the second platform characteristic descriptor embedding coding mapping norm modulation value to obtain the platform characteristic descriptor embedding coding mapping norm modulation value; calculate the logarithmic function value of the platform characteristic descriptor embedding coding mapping norm modulation value with base two as the logarithm to obtain the platform characteristic significance description factor.

[0012] In the above-mentioned Linux real-time development platform, the platform characteristic description word embedding coding vector enhancement first-level subunit includes: a platform characteristic significant modulation weight calculation second-level subunit, which is used to perform hierarchical mask modulation on the sequence of the platform characteristic significant description factors to obtain a sequence of platform characteristic significant modulation weights; a platform characteristic description word embedding coding vector weighted enhancement second-level subunit, which is used to use each platform characteristic significant modulation weight in the sequence of the platform characteristic significant modulation weights as a weight, and respectively perform weighted enhancement on the sequence of the platform characteristic description word embedding coding vectors to obtain the sequence of the enhanced expression platform characteristic description word embedding coding vectors.

[0013] In the above Linux real-time development platform, the secondary subunit for calculating the platform characteristic significant modulation weight includes: a tertiary subunit for normalizing the platform characteristic significant description factor, which is used to normalize the sequence of the platform characteristic significant description factor to obtain a sequence of normalized platform characteristic significant description factors; a tertiary subunit for generating the platform characteristic significant modulation weight, which is used to hierarchically modulate the sequence of the normalized platform characteristic significant description factor to obtain a sequence of the platform characteristic significant modulation weight, wherein the tertiary subunit for generating the platform characteristic significant modulation weight is used to: compare each normalized platform characteristic significant description factor in the sequence of the normalized platform characteristic significant description factor with a preset threshold value. Comparison is performed to obtain a sequence of significant modulation weights of the platform characteristics; wherein, in response to the normalized platform characteristic significance description factor being greater than a second preset threshold, the product of the normalized platform characteristic significance description factor and the first feature scaling factor is used as the significant modulation weight of the platform characteristics; in response to the normalized platform characteristic significance description factor being greater than the first preset threshold and less than or equal to the second preset threshold, the normalized platform characteristic significance description factor is used as the significant modulation weight of the platform characteristics; in response to the normalized platform characteristic significance description factor being less than or equal to the first preset threshold, the product of the normalized platform characteristic significance description factor and the second feature scaling factor is used as the significant modulation weight of the platform characteristics.

[0014] In the above-mentioned Linux real-time development platform, the configuration option recommendation result generating unit is used to: pass the sequence of the enhanced expression platform feature descriptor embedding coding vectors through a converter-based context encoder to obtain an enhanced expression hardware platform feature descriptor granularity context semantic association feature vector as the enhanced expression hardware platform feature descriptor granularity context semantic association feature; pass the enhanced expression hardware platform feature descriptor granularity context semantic association feature vector through a classifier-based configuration option recommender to obtain the configuration option recommendation result, and the configuration option recommendation result is used to represent the kernel version recommendation type label of the Linux system.

[0015] In the above-mentioned Linux real-time development platform, the performance test result generation module is used to: use latencytop or rt-tests to perform real-time performance testing on the hardware platform installed with the Linux system to obtain the performance test results, and the performance test results include system response time, jitter data and throughput.

[0016] Compared with the prior art, the Linux real-time development platform provided by the present application performs word segmentation and word embedding coding on the feature description of the hardware platform by adopting the natural language processing model and technology of artificial intelligence, and then describes the embedding coding features of each platform feature description word after embedding, and significantly enhances it, so as to intelligently recommend the kernel version type of the Linux system according to the contextual semantic association of each enhanced expression platform feature description word embedding coding feature. In this way, the feature description of the hardware platform can be automatically analyzed, and the configuration options of the Linux kernel version that best suits the hardware platform can be intelligently recommended. This significantly reduces the time and expertise required for manual configuration by the user, reduces the possibility of configuration errors, and ensures that real-time tasks are responded to in a timely manner, thereby improving the stability and reliability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] In the attached picture: Figure 1 4 is a block diagram of a Linux real-time development platform according to an embodiment of the present application.

[0019] Figure 2 The present invention is a block diagram of a Linux system configuration option determination module in a Linux real-time development platform according to an embodiment of the present application.

[0020] Figure 3 The present invention is a data flow diagram of a Linux system configuration option determination module in a Linux real-time development platform according to an embodiment of the present application.

[0021] Figure 4 It is a block diagram of a unit for significantly enhancing the descriptiveness of hardware platform characteristics in a Linux real-time development platform according to an embodiment of the present application.

[0022] Figure 5 It is a block diagram of a Linux system installation module in a Linux real-time development platform according to an embodiment of the present application. DETAILED DESCRIPTION

[0023] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0024] With the rapid development of technologies such as the Internet of Things, industrial automation, and driverless cars, embedded systems are increasingly demanding real-time performance. The application scenarios of these technologies require that the system must respond to external events within a strict time frame to ensure the reliability and security of the system. The Linux operating system has been widely used and recognized in the field of embedded systems due to its open source characteristics, flexibility, and scalability.

[0025] However, as multi-core processors become the dominant force in the market, although they provide more powerful processing capabilities, they also bring new challenges to the real-time and deterministic nature of the system. Multi-core processors enable multiple tasks to be executed in parallel, thereby improving the overall processing efficiency of the system. However, this parallelism also increases the complexity of task scheduling and synchronization, which may make it impossible to effectively guarantee the response time of real-time tasks. Specifically, in a multi-core environment, task scheduling not only needs to consider the priority and required execution time of tasks, but also needs to solve the data dependency and resource sharing problems between tasks. In addition, frequent interrupt processing and context switching add additional burden to the system, making the execution and response time of real-time tasks more unpredictable, which may affect the reliability and stability of the system.

[0026] Based on this, this application proposes a Linux real-time development platform, which ensures the response speed of real-time tasks by optimizing scheduling strategies and system configurations, can effectively cope with the challenges of real-time and determinism in a multi-core processor environment, and provides strong support for the development of real-time systems. Figure 1 FIG. 1 is a block diagram of a Linux real-time development platform according to an embodiment of the present application. Figure 1 As shown, in the Linux real-time development platform 100, it includes: a hardware platform feature description data acquisition module 110, which is used to obtain the feature description of the hardware platform; a Linux system configuration option determination module 120, which is used to determine the configuration options of the Linux system adapted to the hardware platform based on the feature description of the hardware platform; a Linux system installation module 130, which is used to install the Linux system on the hardware platform based on the configuration options of the Linux system to obtain a hardware platform with the Linux system installed; a performance test result generation module 140, which is used to perform real-time performance testing on the hardware platform with the Linux system installed to obtain performance test results.

[0027] In an embodiment of the present application, the hardware platform feature description data acquisition module 110 is used to obtain a feature description of the hardware platform. It should be understood that the hardware platform referred to in the present application may be a quad-core A40I platform, a dual-core T113 platform, or a single-core MP1 platform. The feature description of the hardware platform here specifically refers to the characteristics of the hardware platform in terms of processor architecture, memory management, interrupt handling, etc. By obtaining the feature description of the hardware platform, it can help understand the performance limitations and capabilities of the hardware platform, and then select the appropriate system configuration options according to specific conditions.

[0028] In the embodiment of the present application, the Linux system configuration option determination module 120 is used to determine the configuration options of the Linux system adapted to the hardware platform based on the characteristic description of the hardware platform.

[0029] It should be understood that the powerful functions of the Linux system also mean that its configuration options are very rich. For non-professional users, choosing the configuration options of the appropriate Linux kernel version is a challenge. However, the traditional Linux kernel configuration method usually relies on the user's professional knowledge and experience. The user needs to manually select the appropriate kernel configuration according to the characteristics of the hardware platform. This method is not only time-consuming and labor-intensive, but also prone to improper configuration due to human errors, which in turn affects the real-time performance and stability of the system.

[0030] Accordingly, in the Linux system configuration option determination module 120, the technical concept of the present application is to perform word segmentation and word embedding coding on the feature description of the hardware platform by adopting the natural language processing model and technology of artificial intelligence, and then significantly enhance the description of the embedding coding features of each platform feature description word after embedding, so as to intelligently recommend the kernel version type of the Linux system according to the contextual semantic association of each enhanced expression platform feature description word embedding coding feature. In this way, the feature description of the hardware platform can be automatically analyzed, and the configuration options of the Linux kernel version that best suits the hardware platform can be intelligently recommended. This significantly reduces the time and expertise required for manual configuration by the user, reduces the possibility of configuration errors, ensures that real-time tasks are responded to in a timely manner, and thus improves the stability and reliability of the system.

[0031] Specifically, Figure 2 The present invention is a block diagram of a Linux system configuration option determination module in a Linux real-time development platform according to an embodiment of the present application. Figure 3 Schematic diagram of data flow of the Linux system configuration option determination module in the Linux real-time development platform according to the embodiment of the present application. Figure 2 and Figure 3As shown, the Linux system configuration option determination module 120 includes: a hardware platform feature description word embedding encoding unit 121, which is used to perform word embedding encoding on the feature description of the hardware platform after word segmentation processing to obtain a sequence of platform feature description word embedding encoding vectors; a hardware platform feature feature descriptiveness significant enhancement unit 122, which is used to perform platform feature feature descriptiveness significant enhancement on the sequence of platform feature description word embedding encoding vectors to obtain a sequence of enhanced expression platform feature description word embedding encoding vectors; a configuration option recommendation result generation unit 123, which is used to generate a configuration option recommendation result based on the enhanced expression hardware platform feature description word granularity context semantic association features obtained by context encoding the sequence of enhanced expression platform feature description word embedding encoding vectors, and the configuration option recommendation result is used to represent the kernel version recommendation type label of the Linux system.

[0032] In an embodiment of the present application, the hardware platform feature description word embedding coding unit 121 is used to perform word segmentation processing on the feature description of the hardware platform and then perform word embedding coding to obtain a sequence of platform feature description word embedding coding vectors. Specifically, in an embodiment of the present application, the hardware platform feature description word embedding coding unit is used to: perform word segmentation processing on the feature description of the hardware platform to obtain the word granularity distribution of the platform feature description; use the feature description word embedding matrix to perform word embedding coding on the word granularity distribution of the platform feature description to obtain a sequence of platform feature description word embedding coding vectors.

[0033] It should be understood that the feature description of the hardware platform contains key words about the hardware, such as processor model, memory size, storage capacity and other key hardware parameters. Therefore, in order to be able to analyze and process the structure and meaning of the text in each key word in a more fine-grained manner, in the technical solution of the present application, the feature description of the hardware platform is segmented to obtain the word granularity distribution of the platform feature description. Next, considering that each word or phrase in the word granularity of the platform feature description contains its own semantic information and meaning, based on this, in the technical solution of the present application, the word granularity distribution of the platform feature description is word-embedded and encoded using a feature description word embedding matrix to obtain a sequence of platform feature description word embedding encoding vectors. That is, the semantic relationship between words can be captured through word embedding, and the feature description word embedding matrix obtained through training will make similar words close to each other in the vector space, which is of great significance for understanding the words in the hardware description.

[0034] In the embodiment of the present application, the hardware platform characteristic feature descriptiveness significant enhancement unit 122 is used to significantly enhance the platform characteristic feature descriptiveness of the sequence of the platform characteristic descriptor embedding coding vectors to obtain a sequence of enhanced expression platform characteristic descriptor embedding coding vectors. Specifically, Figure 4 FIG. 1 is a block diagram of a unit for significantly enhancing the descriptiveness of hardware platform characteristics in a Linux real-time development platform according to an embodiment of the present application. Figure 4 As shown, the hardware platform characteristic feature descriptive significant enhancement unit 122 includes: a platform characteristic significant description factor calculation first-level sub-unit 1221, which is used to calculate the feature significant description factor of each platform characteristic descriptor word embedding coding vector in the sequence of the platform characteristic descriptor word embedding coding vector to obtain a sequence of platform characteristic significant description factors; a platform characteristic descriptor word embedding coding vector enhancement first-level sub-unit 1222, which is used to perform hierarchical modulation enhancement on the sequence of the platform characteristic descriptor word embedding coding vector based on the sequence of the platform characteristic significant description factors to obtain the sequence of the enhanced expression platform characteristic descriptor word embedding coding vector.

[0035] Accordingly, considering that different platform characteristic description word embedding coding features in the sequence of the platform characteristic description word embedding coding vectors have different feature significances, in order to further refine and enhance the useful information in the platform characteristic description word embedding coding vectors, make the vector representation of each word richer, and extract more discriminative and semantically rich feature representations, in the technical solution of the present application, the sequence of the platform characteristic description word embedding coding vectors is significantly enhanced in terms of platform characteristic feature descriptiveness to obtain a sequence of enhanced expression platform characteristic description word embedding coding vectors.

[0036] Specifically, first, the feature significance description factor of each platform feature description word embedding coding vector in the sequence of the platform feature description word embedding coding vector is calculated to obtain a sequence of platform feature significance description factors. Specifically, each platform feature description word embedding coding vector in the sequence of the platform feature description word embedding coding vector is mapped to a hyperbolic space to obtain a sequence of platform feature description word embedding coding vectors. It should be understood that the hardware feature description may contain multiple hierarchical relationships (such as different components on a motherboard), and the hyperbolic space can more effectively represent such hierarchical structures than the Euclidean space, because the hyperbolic space has an exponentially growing volume, which can better reflect the sparsity between distant nodes in the tree structure. Next, the feature significance description factor of each platform feature description word embedding coding vector is calculated to obtain a sequence of platform feature significance description factors. In other words, the significance description factor can be regarded as giving a weight to each hardware feature description vocabulary, indicating its importance in describing the hardware platform characteristics. Specifically, in this technical solution, the significance of the feature is modeled by calculating the squared Poincare norm of each mapping vector, thereby revealing its significance and more accurately evaluating the importance of each feature.

[0037] Then, based on the sequence of the platform characteristic saliency descriptor, the sequence of the platform characteristic descriptor embedding coding vector is hierarchically modulated and enhanced to obtain the sequence of the enhanced expression platform characteristic descriptor embedding coding vector. That is, by significantly enhancing the original features, those features that are more important in the classification task can be highlighted, thereby improving the accuracy of subsequent kernel version recommendation. Specifically, first, by setting different levels for the hierarchical modulation of each platform characteristic saliency descriptor, important features can be strengthened in a targeted manner and the influence of secondary features can be reduced, which helps to highlight the key feature information more clearly in the feature space, while ignoring those features that are noise or irrelevant. In particular, in the present technical solution, the hierarchical mask modulation strategy adopted divides the sequence of feature saliency descriptors into three different levels by setting two critical values. And different degrees of feature scaling are applied to the feature saliency descriptor in each level, so that different degrees of attention are given while retaining the feature diversity, thereby highlighting the key mode and reducing the interference of irrelevant features. Finally, the sequence of platform characteristic description word embedding coding vectors is weighted by each modulated weight to obtain a sequence of enhanced expression platform characteristic description word embedding coding vectors.

[0038] Specifically, in an embodiment of the present application, the first-level sub-unit for calculating the platform characteristic significance description factor includes: a second-level sub-unit for mapping the characteristic description word embedding coding vector to a hyperbolic space, which is used to map each platform characteristic description word embedding coding vector in the sequence of the platform characteristic description word embedding coding vector to a hyperbolic space to obtain a sequence of platform characteristic description word embedding coding mapping vectors; a second-level sub-unit for generating the platform characteristic significance description factor, which is used to calculate the characteristic significance description factor of each platform characteristic description word embedding coding mapping vector in the sequence of the platform characteristic description word embedding coding mapping vector to obtain a sequence of the platform characteristic significance description factors; wherein the second-level sub-unit for mapping the characteristic description word embedding coding vector to a hyperbolic space is used to: after matrix-vector multiplication of the first weight matrix and the platform characteristic description word embedding coding vector, vector-matrix multiplication of the obtained platform characteristic description word embedding coding weight vector and the second weight matrix to obtain the platform characteristic description word embedding coding mapping vector.

[0039] More specifically, in an embodiment of the present application, the platform characteristic significance description factor generates a secondary sub-unit, which is used to: calculate the square of the norm of the platform characteristic descriptor embedding coding mapping vector to obtain the platform characteristic descriptor embedding coding mapping norm value; add and subtract the constant one from the platform characteristic descriptor embedding coding mapping norm value to obtain a first platform characteristic descriptor embedding coding mapping norm modulation value and a second platform characteristic descriptor embedding coding mapping norm modulation value, and divide the first platform characteristic descriptor embedding coding mapping norm modulation value by the second platform characteristic descriptor embedding coding mapping norm modulation value to obtain the platform characteristic descriptor embedding coding mapping norm modulation value; calculate the logarithmic function value of the logarithm of the platform characteristic descriptor embedding coding mapping norm modulation value with base two to obtain the platform characteristic significance description factor.

[0040] Specifically, in an embodiment of the present application, the platform characteristic descriptor word embedding coding vector enhancement first-level sub-unit includes: a platform characteristic significant modulation weight calculation second-level sub-unit, which is used to perform hierarchical mask modulation on the sequence of the platform characteristic significant description factors to obtain a sequence of platform characteristic significant modulation weights; a platform characteristic descriptor word embedding coding vector weighted enhancement second-level sub-unit, which is used to use each platform characteristic significant modulation weight in the sequence of the platform characteristic significant modulation weights as a weight, and respectively perform weighted enhancement on the sequence of the platform characteristic descriptor word embedding coding vectors to obtain the sequence of the enhanced expression platform characteristic descriptor word embedding coding vectors.

[0041] More specifically, in the embodiment of the present application, the secondary subunit for calculating the significant modulation weight of the platform characteristic includes: a tertiary subunit for normalizing the significant description factor of the platform characteristic, which is used to normalize the sequence of the significant description factor of the platform characteristic to obtain a sequence of normalized significant description factors of the platform characteristic; a tertiary subunit for generating significant modulation weights of the platform characteristic, which is used to hierarchically modulate the sequence of the normalized significant description factor of the platform characteristic to obtain a sequence of significant modulation weights of the platform characteristic, wherein the tertiary subunit for generating significant modulation weights of the platform characteristic is used to: compare each normalized significant description factor of the platform characteristic in the sequence of the significant description factor of the normalized platform characteristic with a preset threshold value; A sequence of significant modulation weights of the platform characteristics is obtained by comparison; wherein, in response to the normalized platform characteristic significance description factor being greater than a second preset threshold, the product of the normalized platform characteristic significance description factor and the first feature scaling factor is used as the significant modulation weight of the platform characteristics; in response to the normalized platform characteristic significance description factor being greater than the first preset threshold and less than or equal to the second preset threshold, the normalized platform characteristic significance description factor is used as the significant modulation weight of the platform characteristics; in response to the normalized platform characteristic significance description factor being less than or equal to the first preset threshold, the product of the normalized platform characteristic significance description factor and the second feature scaling factor is used as the significant modulation weight of the platform characteristics.

[0042] In the embodiment of the present application, specifically, the hardware platform characteristic feature descriptive significant enhancement unit is used to: perform platform characteristic feature descriptive significant enhancement processing on the sequence of the platform characteristic descriptor embedding coding vectors according to the following formula, so as to obtain a sequence of enhanced expression platform characteristic descriptor embedding coding vectors; wherein the formula is: ;in, is a sequence of embedding encoding vectors for the platform feature description words, are the first, second, ..., and third words in the sequence of embedding encoding vectors of the platform feature description words. ,..., Platform feature description word embedding encoding vector, and are the first weight matrix and the second weight matrix respectively, The first in the sequence of embedding encoding mapping vectors for platform feature description words Platform feature description word embedding encoding mapping vector, is the square of the one-norm of the vector, represents the logarithmic function value with base 2, is the first in the sequence of factors describing the significance of platform characteristics. The significant descriptive factors of platform characteristics, yes function, is the first in the sequence of normalized platform characteristic significance descriptors Normalized platform characteristic significance descriptor factor, For masking, is the preset threshold, and is the characteristic scaling factor, and , is the first in the sequence of platform characteristics that significantly modulate the weights The platform characteristics significantly modulate the weights, is a sequence of embedding encoding vectors of the enhanced expression platform feature description words, is the first, second, ..., first in the sequence of embedding encoding vectors of the enhanced expression platform feature description words ,..., An enhanced expression platform features word embedding encoding vectors.

[0043] In an embodiment of the present application, the configuration option recommendation result generation unit 123 is used to generate a configuration option recommendation result based on the enhanced expression hardware platform feature descriptor granularity contextual semantic association features obtained by context encoding the sequence of the enhanced expression platform feature descriptor embedding coding vectors, and the configuration option recommendation result is used to represent the kernel version recommendation type label of the Linux system. Specifically, in an embodiment of the present application, the configuration option recommendation result generation unit is used to: pass the sequence of the enhanced expression platform feature descriptor embedding coding vectors through a converter-based context encoder to obtain an enhanced expression hardware platform feature descriptor granularity contextual semantic association feature vector as the enhanced expression hardware platform feature descriptor granularity contextual semantic association feature; pass the enhanced expression hardware platform feature descriptor granularity contextual semantic association feature vector through a classifier-based configuration option recommender to obtain the configuration option recommendation result, and the configuration option recommendation result is used to represent the kernel version recommendation type label of the Linux system.

[0044] It should be understood that in order to capture the relationship between all platform feature words in the sequence of the enhanced expression platform feature description word embedding coding vectors, so as to fully understand the semantic information between the contexts in the hardware feature description, because there may be cross-lexical dependencies between hardware features, in the technical solution of the present application, the sequence of the enhanced expression platform feature description word embedding coding vectors is passed through a converter-based context encoder to obtain an enhanced expression hardware platform feature description word granular context semantic association feature vector. Then, the enhanced expression hardware platform feature description word granular context semantic association feature vector is passed through a classifier-based configuration option recommender to obtain a configuration option recommendation result, and the configuration option recommendation result is used to represent the recommended type label of the kernel version of the Linux system. That is, the enhanced expression hardware platform feature description word granular context semantic association features obtained by context semantic association of the sequence of the enhanced expression platform feature description word embedding coding vectors are used for classification processing, so as to intelligently recommend the kernel version type of the Linux system. In addition, the configuration option recommendation result in the present application can also be the type of real-time patch, scheduling strategy or driver. By using training data with different labels to train the classifier-based configuration option recommender, different configuration option recommendation results can be obtained.

[0045] In particular, when the sequence of the platform feature description word embedding coding vectors represents the embedded coding features of the word granularity distribution of the platform feature description, when performing significant enhancement of the platform feature descriptiveness and context association encoding, the sparsity of the explicit modeling of the uneven word granularity distribution of the embedding coding will lead to sequence enhancement gaps and missing local context associations, resulting in simple repetition of semantic enhancement and association in the local semantic domain, affecting the semantic logical dependency of the sequence set and reducing the classification accuracy based on the semantic distribution of the sequence set.

[0046] Preferably, the enhanced expression hardware platform feature description word granularity context semantic association feature vector is passed through a classifier-based configuration option recommender to obtain a configuration option recommendation result, including: determining an enhanced expression hardware platform feature description word granularity context semantic association response matrix and an enhanced expression hardware platform feature description word granularity context semantic distance response matrix based on the square root of the association value of the eigenvalue of the enhanced expression hardware platform feature description word granularity context semantic association feature vector and the L2 distance value: wherein the process can be expressed as: ;in, A word-granular context semantic association feature vector for the enhanced expression hardware platform feature description, express The number of eigenvalues ​​in , The first character in the context semantic association feature vector of the enhanced expression hardware platform feature description word granularity The eigenvalues ​​at the positions, The first character in the context semantic association feature vector of the enhanced expression hardware platform feature description word granularity The eigenvalues ​​at the positions, The enhanced expression hardware platform feature description word granularity context semantic association response matrix is The eigenvalues ​​at the positions, The enhanced expression hardware platform feature description word granularity context semantic distance response matrix The feature value at each position.

[0047] The enhanced expression hardware platform feature description word granularity context semantic association feature vector is matrix-multiplied with the enhanced expression hardware platform feature description word granularity context semantic association response matrix to obtain the enhanced expression hardware platform feature description word granularity context semantic association response vector; the enhanced expression hardware platform feature description word granularity context semantic distance response matrix is ​​matrix-multiplied with the transposed vector of the enhanced expression hardware platform feature description word granularity context semantic association feature vector to obtain the enhanced expression hardware platform feature description word granularity context semantic distance response vector; the enhanced expression hardware platform feature description word granularity context semantic association response matrix is ​​dot-multiplied with the enhanced expression hardware platform feature description word granularity context semantic distance response matrix Afterwards, matrix multiplication is performed with the transposed vector of the enhanced expression hardware platform feature description word granularity context semantics association feature vector to obtain the enhanced expression hardware platform feature description word granularity context semantics bias vector; the enhanced expression hardware platform feature description word granularity context semantics association response vector, the enhanced expression hardware platform feature description word granularity context semantics distance response vector and the enhanced expression hardware platform feature description word granularity context semantics bias vector are added to obtain an optimized enhanced expression hardware platform feature description word granularity context semantics association feature vector; the optimized enhanced expression hardware platform feature description word granularity context semantics association feature vector is passed through the classifier-based configuration option recommender to obtain the configuration option recommendation result.

[0048] Here, the optimized enhanced expression hardware platform feature description word granularity context semantic association feature vector is expressed as: ;in, A word-granular context semantic association feature vector for the enhanced expression hardware platform feature description, express The number of eigenvalues ​​in , represents matrix multiplication, The enhanced expression hardware platform feature description word granularity context semantic association response matrix, Indicates adding by position point. The enhanced expression hardware platform feature description word granularity context semantic distance response matrix, for The transposed vector of The point-add vector between the enhanced expression hardware platform feature description word granularity context semantic association response vector and the enhanced expression hardware platform feature description word granularity context semantic distance response vector, It means point multiplication by position. A word-granular contextual semantic bias vector for the enhanced expression hardware platform feature description, A word-granular contextual semantic association feature vector is described for the optimized enhanced expression hardware platform characteristics.

[0049] That is, by taking the enhanced expression hardware platform feature description word granular context semantic association feature vector as its statistical reference-free distribution response framework based on the self-association matrix and the self-distance matrix, and constructing the retrieval-enhanced reverse response for the enhanced expression hardware platform feature description word granular context semantic association feature vector to avoid simple repetition of feature distribution, and by ensuring the intrinsic mapping logic of the enhanced expression hardware platform feature description word granular context semantic association feature vector based on the retrieval-response context to avoid its surface combinatorial mapping, so as to achieve the logical dependency mapping of the enhanced expression hardware platform feature description word granular context semantic association feature vector to the classification target domain while maintaining the intuitive response relevance, thereby improving the accuracy of the configuration option recommendation result obtained by the enhanced expression hardware platform feature description word granular context semantic association feature vector through the classifier-based configuration option recommender.

[0050] In summary, the Linux system configuration option determination module 120 is explained, which performs word segmentation and word embedding coding on the feature description of the hardware platform by adopting the natural language processing model and technology of artificial intelligence, and then describes the embedding coding features of each platform feature description word after embedding, and significantly enhances it, so as to intelligently recommend the kernel version type of the Linux system according to the contextual semantic association of each enhanced expression platform feature description word embedding coding feature. In this way, the feature description of the hardware platform can be automatically analyzed, and the configuration options of the Linux kernel version that best suits the hardware platform can be intelligently recommended. This significantly reduces the time and expertise required for manual configuration by the user, reduces the possibility of configuration errors, and ensures that real-time tasks are responded to in a timely manner, thereby improving the stability and reliability of the system.

[0051] In the embodiment of the present application, the Linux system installation module 130 is used to install the Linux system on the hardware platform based on the configuration options of the Linux system to obtain a hardware platform with the Linux system installed. Specifically, Figure 5 FIG. 1 is a block diagram of a Linux system installation module in a Linux real-time development platform according to an embodiment of the present application. Figure 5 As shown, the Linux system installation module 130 includes: a Linux system download patch transplantation unit 131, which is used to download the Linux system adapted to the hardware platform on the hardware platform, and transplant the real-time patch to the Linux system; a Linux system configuration unit 132, which is used to configure the Linux system after the real-time patch is transplanted based on the configuration options of the Linux system using the make menuconfig or make xconfig tool, and select the real-time scheduler and set the interrupt priority according to the hardware characteristics of the hardware platform; a Linux system compilation unit 133, which is used to compile the configured Linux system and generate an image file; an image file deployment unit 134, which is used to deploy the image file on the hardware platform; and a Linux system inspection unit 135, which is used to perform an inspection after starting the system to ensure that the Linux system can run normally and load the real-time patch.

[0052] In an embodiment of the present application, the Linux system download patch transplantation unit 131 is used to download a Linux system adapted to the hardware platform on the hardware platform, and transplant the real-time patch to the Linux system. It should be understood that downloading the Linux system source code compatible with the hardware platform from a reliable download source (such as the official website of the Linux distribution or a specific Linux kernel repository) can make full use of the performance of the hardware platform to meet the needs of subsequent development applications. Real-time patches usually include modifications to the kernel, aiming to improve the real-time performance of the system, such as reducing delays, improving scheduling accuracy, etc. By transplanting real-time patches, the Linux system can better handle real-time tasks, reduce delays in task scheduling and interrupt processing, and is suitable for applications that require high real-time performance.

[0053] In the embodiment of the present application, the Linux system configuration unit 132 is used to configure the Linux system after the real-time patch is transplanted using the make menuconfig or make xconfig tool based on the configuration options of the Linux system, and select the real-time scheduler and set the interrupt priority according to the hardware characteristics of the hardware platform. It should be understood that the two commands makemenuconfig and make xconfig provide a graphical configuration interface (menuconfig is a text-based interface, and xconfig is a GUI-based interface). Running the make menuconfig or makexconfig command in the Linux source code directory allows the user to configure the system configuration options recommended according to the characteristics of the hardware platform. Optimizing the configuration of the Linux system according to the characteristics of the hardware platform can ensure that the Linux system can run efficiently, make full use of hardware resources, and avoid unnecessary resource consumption. In the configuration interface, the user can also select a scheduler suitable for real-time applications and set the interrupt priority as needed.

[0054] In the embodiment of the present application, the Linux system compiling unit 133 is used to compile the configured Linux system and generate an image file. It should be understood that after the configuration is completed, the make command is used to compile the Linux system source code and generate an image file (such as zImage, uImage, etc.) that can be used for deployment. The generated image file contains all necessary kernel modules and configurations and can be directly deployed to the hardware platform for operation.

[0055] In the embodiment of the present application, the image file deployment unit 134 is used to deploy the image file on the hardware platform. It should be understood that deploying the compiled image file to the storage medium (such as eMMC) of the hardware platform can enable the hardware platform to load and run the configured Linux system.

[0056] In the embodiment of the present application, the Linux system checking unit 135 is used to check after starting the system to ensure that the Linux system can run normally and load the real-time patch. It should be understood that after starting the system, by checking the system log and the running status, it is ensured that the system runs stably and normally, and the real-time patch has been successfully loaded, so as to provide guarantee for subsequent performance testing and development work.

[0057] In an embodiment of the present application, the performance test result generation module 140 is used to perform real-time performance testing on the hardware platform with the Linux system installed to obtain performance test results. Specifically, in an embodiment of the present application, the performance test result generation module is used to: use latencytop or rt-tests to perform real-time performance testing on the hardware platform with the Linux system installed to obtain the performance test results, and the performance test results include system response time, jitter data and throughput. It should be noted that latencytop and rt-tests are both tools for real-time performance testing of Linux systems. It should be understood that by designing a reasonable real-time performance testing scheme and testing it on a hardware platform, and performing performance analysis based on the obtained performance test results including indicators such as system response time, jitter data and throughput, performance bottlenecks and optimization points can be found, and performance optimization can be performed. For example, if the maximum jitter of the real-time performance of the expected hardware platform is A40I platform within 30us, the maximum jitter of the real-time performance of the hardware platform is T113 platform within 40us, and the maximum jitter of the real-time performance of the hardware platform is MP1 platform within 80us, if the result of the real-time performance test of the corresponding hardware platform does not meet the expected standard, it can be optimized by modifying kernel parameters, optimizing algorithms, etc. In particular, in the embodiment of the present application, an implementation method of using rt-tests to perform real-time performance testing on the hardware platform installed with the Linux system can be: a. Install rt-tests in the Linux system; b. Configure test parameters to determine the task cycle, number of tasks, and priority to be tested; c. Run rt-tests in the terminal and output the execution status of the real-time task in the terminal, including the response time, jitter data, and throughput of each execution; d. Save the data output by rt-tests into a file, which contains specific indicator data such as response time, jitter data, and throughput, so as to facilitate subsequent optimization analysis.

[0058] In summary, the Linux real-time development platform 100 based on the embodiment of the present application is explained, which first obtains the characteristic description of the hardware platform, and determines the adapted Linux system configuration options based on the characteristic description of the hardware platform, then installs the adapted Linux system on the hardware platform, and finally performs a real-time performance test on the hardware platform with the adapted Linux system installed to obtain the performance test results. In this way, the Linux system can be configured and optimized according to the characteristics of the hardware platform, thereby improving the reliability and stability of the Linux system.

[0059] The basic principles of the present application are described above in conjunction with specific embodiments. However, it should be noted that the advantages, strengths, effects, etc. mentioned in the present application are only examples and not limitations, and it cannot be considered that these advantages, strengths, effects, etc. are required by each embodiment of the present application. In addition, the specific details of the above application are only for the purpose of illustration and ease of understanding, rather than limitation, and the above details do not limit the present application to the use of the above specific details to be implemented.

Claims

1. A Linux real-time development platform, characterized in that: include: The hardware platform feature description data acquisition module is used to obtain the feature description of the hardware platform; A Linux system configuration option determination module, used to determine configuration options of the Linux system adapted to the hardware platform based on the characteristic description of the hardware platform; A Linux system installation module, used for installing the Linux system on the hardware platform based on the configuration options of the Linux system to obtain a hardware platform with the Linux system installed; The performance test result generating module is used to perform real-time performance test on the hardware platform installed with the Linux system to obtain performance test results.

2. The Linux real-time development platform according to claim 1, characterized in that: The Linux system installation module includes: a Linux system download patch transplantation unit, which is used to download a Linux system adapted to the hardware platform on the hardware platform and transplant a real-time patch to the Linux system; a Linux system configuration unit, which is used to configure the Linux system after the real-time patch is transplanted using a make menuconfig or make xconfig tool based on the configuration options of the Linux system, and select a real-time scheduler and set an interrupt priority according to the hardware characteristics of the hardware platform; a Linux system compilation unit, which is used to compile the configured Linux system and generate an image file; an image file deployment unit, which is used to deploy the image file on the hardware platform; and a Linux system inspection unit, which is used to perform an inspection after starting the system to ensure that the Linux system can run normally and load the real-time patch.

3. The Linux real-time development platform according to claim 2, characterized in that: The Linux system configuration option determination module includes: a hardware platform feature description word embedding encoding unit, which is used to perform word segmentation processing on the feature description of the hardware platform and then perform word embedding encoding to obtain a sequence of platform feature description word embedding encoding vectors; a hardware platform feature feature descriptive significant enhancement unit, which is used to perform platform feature feature descriptive significant enhancement on the sequence of platform feature description word embedding encoding vectors to obtain a sequence of enhanced expression platform feature description word embedding encoding vectors, wherein the hardware platform feature feature descriptive significant enhancement unit includes: a platform feature significant description factor calculation first-level subunit, which is used to calculate each platform feature description word embedding encoding in the sequence of platform feature description word embedding encoding vectors. The feature significance description factor of the vector is used to obtain a sequence of platform characteristic significance description factors; the platform characteristic description word embedding coding vector enhancement first-level sub-unit is used to perform hierarchical modulation enhancement on the sequence of platform characteristic description word embedding coding vectors based on the sequence of platform characteristic significance description factors to obtain the sequence of enhanced expression platform characteristic description word embedding coding vectors; the configuration option recommendation result generation unit is used to generate a configuration option recommendation result based on the enhanced expression hardware platform characteristic description word granularity context semantic association features obtained by context encoding the sequence of enhanced expression platform characteristic description word embedding coding vectors, and the configuration option recommendation result is used to represent the kernel version recommendation type label of the Linux system.

4. The Linux real-time development platform according to claim 3, characterized in that: The hardware platform feature description word embedding encoding unit is used to: perform word segmentation processing on the feature description of the hardware platform to obtain the word granularity distribution of the platform feature description; use the feature description word embedding matrix to perform word embedding encoding on the word granularity distribution of the platform feature description to obtain a sequence of the platform feature description word embedding encoding vectors.

5. The Linux real-time development platform according to claim 4, characterized in that: The platform characteristic significance description factor calculation primary subunit includes: a characteristic description word embedding coding vector hyperbolic space mapping secondary subunit, which is used to map each platform characteristic description word embedding coding vector in the sequence of the platform characteristic description word embedding coding vector to a hyperbolic space to obtain a sequence of platform characteristic description word embedding coding mapping vectors; a platform characteristic significance description factor generation secondary subunit, which is used to calculate the feature significance description factor of each platform characteristic description word embedding coding mapping vector in the sequence of the platform characteristic description word embedding coding mapping vector to obtain the sequence of the platform characteristic significance description factors; wherein the characteristic description word embedding coding vector hyperbolic space mapping secondary subunit is used to: after matrix-vector multiplication of a first weight matrix and the platform characteristic description word embedding coding vector, vector-matrix multiplication of the obtained platform characteristic description word embedding coding weight vector and a second weight matrix to obtain the platform characteristic description word embedding coding mapping vector.

6. The Linux real-time development platform according to claim 5, characterized in that: The platform characteristic significance description factor generation secondary subunit is used to: calculate the square of the norm of the platform characteristic descriptor embedding coding mapping vector to obtain the platform characteristic descriptor embedding coding mapping norm value; add and subtract the constant one from the platform characteristic descriptor embedding coding mapping norm value to obtain a first platform characteristic descriptor embedding coding mapping norm modulation value and a second platform characteristic descriptor embedding coding mapping norm modulation value, and divide the first platform characteristic descriptor embedding coding mapping norm modulation value by the second platform characteristic descriptor embedding coding mapping norm modulation value to obtain the platform characteristic descriptor embedding coding mapping norm modulation value; calculate the logarithmic function value of the logarithm of the platform characteristic descriptor embedding coding mapping norm modulation value with base two to obtain the platform characteristic significance description factor.

7. The Linux real-time development platform according to claim 6, characterized in that: The platform characteristic description word embedding coding vector enhancement first-level sub-unit includes: a platform characteristic significant modulation weight calculation second-level sub-unit, which is used to perform hierarchical mask modulation on the sequence of the platform characteristic significant description factors to obtain a sequence of platform characteristic significant modulation weights; a platform characteristic description word embedding coding vector weight enhancement second-level sub-unit, which is used to use each platform characteristic significant modulation weight in the sequence of platform characteristic significant modulation weights as a weight, and respectively perform weighted enhancement on the sequence of the platform characteristic description word embedding coding vectors to obtain the sequence of the enhanced expression platform characteristic description word embedding coding vectors.

8. The Linux real-time development platform according to claim 7, characterized in that: The secondary subunit for calculating the significant modulation weight of the platform characteristic comprises: a tertiary subunit for normalizing the significant description factors of the platform characteristic, which is used to normalize the sequence of the significant description factors of the platform characteristic to obtain a sequence of normalized significant description factors of the platform characteristic; a tertiary subunit for generating significant modulation weights of the platform characteristic, which is used to hierarchically modulate the sequence of the normalized significant description factors of the platform characteristic to obtain a sequence of significant modulation weights of the platform characteristic, wherein the tertiary subunit for generating significant modulation weights of the platform characteristic is used to compare each normalized significant description factor of the platform characteristic in the sequence of the significant description factors of the platform characteristic with a preset threshold value to obtain the significant description factors of the platform characteristic; A sequence of platform characteristic significant modulation weights; wherein, in response to the normalized platform characteristic significant description factor being greater than a second preset threshold, the product of the normalized platform characteristic significant description factor and a first feature scaling factor is used as the platform characteristic significant modulation weight; in response to the normalized platform characteristic significant description factor being greater than the first preset threshold and less than or equal to the second preset threshold, the normalized platform characteristic significant description factor is used as the platform characteristic significant modulation weight; in response to the normalized platform characteristic significant description factor being less than or equal to the first preset threshold, the product of the normalized platform characteristic significant description factor and a second feature scaling factor is used as the platform characteristic significant modulation weight.

9. The Linux real-time development platform according to claim 8, characterized in that: The configuration option recommendation result generating unit is used to: pass the sequence of the enhanced expression platform feature description word embedding encoding vectors through a converter-based context encoder to obtain an enhanced expression hardware platform feature description word granularity context semantic association feature vector as the enhanced expression hardware platform feature description word granularity context semantic association feature; The enhanced expression hardware platform feature description word granularity context semantic association feature vector is passed through a classifier-based configuration option recommender to obtain the configuration option recommendation result, and the configuration option recommendation result is used to represent the kernel version recommendation type label of the Linux system.

10. The Linux real-time development platform according to claim 9, characterized in that: The performance test result generating module is used to: use latencytop or rt-tests to perform real-time performance testing on the hardware platform installed with the Linux system to obtain the performance test results, and the performance test results include system response time, jitter data and throughput.

Images