Request transmission processing method and device based on feature recognition, chip and equipment
By monitoring the IP core status and bus status, calculating transmission characteristics in real time, and performing dynamic arbitration, the inflexibility of existing bus arbitration schemes is solved, thereby improving the performance and resource utilization efficiency of SoC systems.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- FUZHOU ROCKCHIP SEMICON
- Filing Date
- 2025-12-22
- Publication Date
- 2026-04-17
AI Technical Summary
In existing technologies, bus arbitration schemes lack the ability to dynamically perceive and intelligently respond to the behavioral characteristics of IP verification, resulting in resource waste and a decline in key performance indicators. They also fail to achieve on-demand scheduling, affecting the overall performance of complex SoC systems.
By monitoring the internal IP status and bus status, IP transmission characteristics are calculated in real time, classified according to user-configured characteristic tables, dynamically adjusted bus arbitration priority, and urgent demand characteristics are identified and prioritized for arbitration.
It enables precise arbitration and scheduling based on actual load conditions, solving the inflexibility problem of transmission based solely on fixed priorities or bandwidth, and improving system performance and resource utilization efficiency.
Smart Images
Figure CN121880237A_ABST
Abstract
Description
Technical Field
[0001] This disclosure belongs to the field of integrated circuit design technology, and relates to bus arbitration methods, particularly to a request transmission processing method and apparatus, chip and device based on feature recognition. Background Technology
[0002] With the rapid development of semiconductor technology, modern system-on-a-chip (SoC) integrates a large number of functional modules, namely IP cores, such as processors, graphics engines, video codecs, and neural network accelerators, onto a single silicon chip. These IP cores form a complex on-chip network through the Advanced Scalable Interface (AXI) bus, enabling massive data exchange. In this architecture, the bus arbitrator, acting as the scheduling center, directly determines the efficiency, latency, and overall system throughput of data transmission. Bus arbitration is a scheduling strategy that resolves conflicts between multiple master devices accessing shared resources (the bus). Based on a set of preset rules, it adjudicates multiple concurrent transmission requests, selecting one request to grant ownership of the bus, thus ensuring the orderliness and efficiency of data flow. Summary of the Invention
[0003] This disclosure provides a request transmission processing method, apparatus, chip, and device based on feature recognition, for improving data transmission efficiency at the sending end.
[0004] In a first aspect, this disclosure provides a request transmission processing method based on feature recognition. The method includes: acquiring the processing status of a target transmission request from a requesting end on a bus and a bus control signal; performing feature calculations on the processing status and the bus control signal using a configurable weight table to generate transmission feature points for the target transmission request; comparing the transmission feature points using the configurable feature table to identify specific features with urgent needs in the target transmission request and generating a priority selection signal; and arbitrating multiple transmission requests on the bus according to the priority selection signal, so that the target transmission request receives priority arbitration.
[0005] In one implementation of the first aspect, the processing state of the target sending request includes the request-side cache usage state and the request computation state.
[0006] In one implementation of the first aspect, using a configurable weight table to perform feature calculations on the processing state and the bus control signal to generate transmission feature points for the target transmission request includes: obtaining an input feature vector based on the request end cache usage state, the request calculation state, and the bus control signal; obtaining a weight vector based on the configurable weight table; and performing feature calculations on the input feature vector and the weight vector to obtain the transmission feature points.
[0007] In one implementation of the first aspect, obtaining the transmission feature points includes: performing feature detection and classification on the input feature vector and the weight vector using a multiplication-accumulation array to obtain the transmission feature points.
[0008] In one implementation of the first aspect, the configurable weight table and the configurable feature table are received, the configurable weight table containing constant vectors having specific transmission characteristics and the configurable feature table containing constant vectors for calculating the transmission characteristics.
[0009] In one implementation of the first aspect, comparing the transmission feature points using a configurable feature table to identify specific features with urgent needs in the target transmission request and generating a priority selection signal includes: comparing and classifying the transmission feature points using reference feature points in the configurable feature table to obtain specific features with urgent needs in the target transmission request; and generating the priority selection signal for the target transmission request based on the specific features.
[0010] In one implementation of the first aspect, arbitrating multiple transmission requests on the bus according to the priority selection signal includes: processing the priority of the target transmission request using a first multiplexer to obtain a new priority for the target transmission request according to the priority selection signal; comparing the new priority of the target transmission request with the priority of parallel transmission requests on the bus using a priority comparator to obtain a priority comparison result; and arbitrating the target transmission request and the parallel transmission requests using a second multiplexer based on the priority comparison result to obtain a priority arbitration result for the target transmission request.
[0011] In one implementation of the first aspect, processing the priority of the target transmission request using a first multiplexer to obtain a new priority for the target transmission request according to the priority selection signal includes: obtaining a basic priority of the target transmission request; obtaining a specific priority of the target transmission request according to the priority selection signal; and processing the basic priority and the specific priority using the first multiplexer to obtain a new priority for the target transmission request.
[0012] In one implementation of the first aspect, the transmission features include a send request buffer, an incomplete request count, and a send request delay.
[0013] Secondly, this disclosure provides a request transmission processing apparatus based on feature recognition. The apparatus includes: a request status acquisition module configured to acquire the processing status of a target transmission request from a requesting end on a bus and a bus control signal; a feature calculation module configured to perform feature calculations on the processing status and the bus control signal using a configurable weight table to generate transmission feature points for the target transmission request; a feature point comparison module configured to compare the transmission feature points using a configurable feature table to identify specific features with urgent needs in the target transmission request and generate a priority selection signal; and an arbitration module configured to arbitrate multiple transmission requests on the bus according to the priority selection signal, so that the target transmission request receives priority arbitration.
[0014] Thirdly, this disclosure provides a chip. The chip includes: a feature detection module configured to acquire the processing status of a target transmission request from a requesting end on a bus and a bus control signal; to perform feature calculations on the processing status and the bus control signal using a configurable weight table to generate transmission feature points of the target transmission request; and to compare the transmission feature points using a configurable feature table to identify specific features in the target transmission request that have urgent needs and generate a priority selection signal; and an arbitrator configured to receive the target transmission request and the parallel transmission request, and to arbitrate the target transmission request and the parallel transmission request on the bus according to the priority selection signal, so that the target transmission request receives priority arbitration.
[0015] In one implementation of the third aspect, the arbitrator includes: a first multiplexer configured to receive a priority selection signal and process the priority of the target transmission request according to the priority selection signal to obtain a new priority for the target transmission request; a priority comparator configured to compare the new priority of the target transmission request with the priority of the parallel transmission requests on the bus to obtain a priority comparison result; and a second multiplexer configured to arbitrate the target transmission request and the parallel transmission requests based on the priority comparison result to obtain a priority arbitration result for the target transmission request.
[0016] Fourthly, this disclosure provides an electronic device. The electronic device includes: a memory configured to store an executable program; and a processor configured to execute the program to implement the feature-based request transmission processing method described above.
[0017] According to embodiments of this disclosure, the priority of requests sent on the bus is arbitrated and adjusted based on the performance requirements of the requesting IP core and the bus status. By real-time monitoring of the IP core's cache status, computing status, and expansion interface bus control signals, IP transmission characteristics are calculated in real time. Specific transmission characteristics are categorized using a user-configured feature table. The current business intent (such as data starvation, high-throughput transmission, etc.) is intelligently extracted from dynamically changing transmission behavior, and the classification results are output to the bus for arbitration adjustment. This solution enables more precise arbitration and scheduling based on the actual load of the IP in real time, solving the problem of inflexible or even over-controlled bus arbitration or scheduling based solely on fixed priorities or bandwidth. It truly ensures on-demand scheduling, balancing and improving the overall system performance.
[0018] According to the embodiments of this disclosure, a transmission feature point is obtained by using a configurable weight table and the request end cache usage status, request calculation status and extended interface bus control signal of the target sending request. The transmission feature point is matched with the reference feature point of the configurable feature table. If the match is successful, it means that a waveform with specific features of urgent need has been detected, and priority adjustment is triggered.
[0019] According to embodiments of this disclosure, a priority selection signal is input to an arbitrator on the bus. A first multiplexer in the arbitrator generates a specific priority for the target transmission request based on the priority selection signal. Simultaneously, the first multiplexer fuses the base priority and the specific priority of the target transmission request to obtain a new priority for the target transmission request. A priority comparator then compares the new priority of the target transmission request with the priorities of other parallel transmission requests. If the new priority is higher than the priorities of other transmission requests, a second multiplexer performs priority arbitration on the target transmission request. Attached Figure Description
[0020] Figure 1 The diagram shown is an application scenario illustration of Embodiment 1 of this disclosure.
[0021] Figure 2 and Figure 3 The diagram shows a process of request transmission processing based on feature recognition as described in an embodiment of this disclosure.
[0022] Figure 4 The diagram shows a waveform curve of the feature indicator point as described in the embodiments of this disclosure.
[0023] Figure 5 The diagram shows a flowchart of the priority arbitration process as described in an embodiment of this disclosure.
[0024] Figure 6 The diagram shown is a structural schematic of the feature recognition-based request transmission processing apparatus according to an embodiment of this disclosure.
[0025] Figure 7 The diagram shown is a structural schematic of the electronic device described in an embodiment of this disclosure. Detailed Implementation
[0026] The following specific examples illustrate the implementation of this disclosure. Those skilled in the art can easily understand other advantages and effects of this disclosure from the content disclosed in this specification. This disclosure can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this disclosure. It should be noted that, unless otherwise specified, the following embodiments and features in the embodiments can be combined with each other.
[0027] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of this disclosure. Therefore, the drawings only show the components related to this disclosure and are not drawn according to the number, shape and size of the components in actual implementation. In actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.
[0028] Traditional arbitration schemes mostly employ static, fixed strategies, such as polling or strict priority arbitration based on the IP core's preset priority, or scheduling according to a pre-allocated fixed bandwidth percentage. This mechanical arbitration mechanism cannot perceive the actual transmission needs of IP cores at different times, and its drawbacks are significant: on the one hand, it may cause an IP core with high priority but no urgent data transmission needs to continuously occupy bus bandwidth, thus blocking other IP cores with actual transmission tasks and wasting bus resources; on the other hand, when an IP core handling an urgent task (such as a buffer about to overflow) urgently needs the bus, it may not receive a timely bus response because its inherent priority is lower than another higher-priority IP core performing non-critical data transmission, leading to a decline in key performance indicators or even system malfunctions. This "one-size-fits-all" arbitration mode lacks dynamic perception and intelligent response capabilities regarding the behavioral characteristics of IP cores during arbitration, and has become a bottleneck restricting the overall performance improvement of complex SoC systems.
[0029] To address at least the aforementioned issues, embodiments of this disclosure provide a request transmission processing method, apparatus, chip, and device based on feature recognition. According to the technical solution of this disclosure, the internal state of the IP and the bus state are monitored, and IP transmission characteristics are calculated in real time. Specific transmission characteristics are classified using a user-configured feature table, and the classification results are output to the bus module for arbitration. This solution enables more precise arbitration and scheduling based on the actual load of the IP in real time, thus solving the problem of inflexible or even over-controlled bus arbitration or scheduling based solely on fixed priorities or bandwidth. It truly ensures on-demand scheduling, balancing and improving the overall system performance.
[0030] The technical solutions in the embodiments of this disclosure will now be described in detail with reference to the accompanying drawings.
[0031] Figure 1 This diagram illustrates an application scenario according to an embodiment of the present disclosure. Figure 1 As shown, the IP bus arbitration device 100 includes an IP core 101, a bus module 102, a transmission feature detection module 103, and a matching and classification module 104. Specifically, the internal state of the IP core 101 and the bus state of the bus module 102 are monitored in real time. The internal IP state and the bus state are input to the transmission feature detection module 103 for feature calculation to obtain the transmission features of the sending request. The transmission features are input to the matching and classification module 104 for classification to obtain specific features with urgent needs. The specific features are fed back to the bus module for priority arbitration and scheduling. The IP bus arbitration device 100 can be applied to complex application scenarios with stringent requirements for multi-task concurrent processing and real-time response, such as electronic whiteboards, network video recorders, and smart cockpits.
[0032] Figure 2 and Figure 3 This diagram illustrates a request transmission process based on feature recognition, as shown in one embodiment of this disclosure. Figure 2 and Figure 3 As shown, the request transmission processing method based on feature recognition includes the following steps S11 to S14.
[0033] Step S11: Obtain the processing status of the target sending request and the bus control signal of the requesting end on the bus.
[0034] In some embodiments, the processing status of the target sending request includes the requesting end cache usage status and the request computation status. Furthermore, the requesting end is, for example, an IP core.
[0035] In some embodiments, the request-side cache usage state is either a preparation state or a consumption state of the data required for IP request calculation. The request-side cache usage state is the IP cache state; the IP core has an internal cache area used to store data to be processed or data waiting to be sent. When the IP core sends a request to the bus, it retrieves data from memory based on the characteristics of the request. When the data in the IP cache area is exhausted, the request cannot be effectively calculated.
[0036] When the transmission efficiency of the extended interface slows down, the bus service capacity is insufficient, or arbitration is inappropriate, leading to increased request latency or the data in the IP cache being nearly exhausted, the IP request cannot be effectively calculated, thus affecting the calculation status of the IP request. The calculation status of the IP request is related to factors such as the IP frame rate and its operating speed.
[0037] In some embodiments, the bus is an AXI (Advanced eXtensible Interface) bus, and the bus control signals are the relevant signals in the standard signals of the AXI bus used to indicate control information.
[0038] In some embodiments, the calculation state positioning of different sending requests is different. The calculation state of video decoding is the decoding frame rate, and the calculation state of video output display is whether the video display frame rate meets the real-time requirements. When bus data is not returned in time, and the decoding frame rate or video real-time performance does not meet the requirements, the sending request buffer will gradually become empty, the number of incomplete sending requests will stabilize at the maximum value, or the sending request delay will gradually increase.
[0039] Because different multimedia IPs (e.g., video decoders continuously read compressed video data streams from DDR, decode them, and output them; display controllers continuously read decoded image frames from DDR and send them to the screen for display) have different data demand patterns, the data saturation state of the IP send request buffer varies. By monitoring the data saturation state of the send request buffer in real time, the data state of the send request buffer is used as the processing state of the target send request, and the arbitration priority of the bus is dynamically adjusted by processing this state.
[0040] Step S12: Use a configurable weight table to perform feature calculations on the processing state and the bus control signal to generate transmission feature points for the target transmission request.
[0041] In some embodiments, the configurable weight table contains a constant vector with specific transmission characteristics.
[0042] In one embodiment of this disclosure, using a configurable weight table to perform feature calculations on the processing state and the bus control signal to generate transmission feature points for the target transmission request may include: obtaining an input feature vector based on the request end cache usage state, the request calculation state, and the bus control signal; obtaining a weight vector based on the configurable weight table; and performing feature calculations on the input feature vector and the weight vector to obtain the transmission feature points.
[0043] In some embodiments, the request-side cache usage status, request computation status, and bus control signals are different types of signals. The request-side cache usage status is the cache utilization rate, which can be quantified as a numerical value. The request computation status can be encoded according to the load type. The bus control signals are the bus request types, such as the extension interface bus control signal for a write operation. These three different signals are processed to obtain an input feature vector containing the request status information.
[0044] In some embodiments, the configurable weight table includes multiple sets of pre-computed weight vectors, each weight vector defining the importance of each element in the input feature vector, and the weight vectors are loaded according to the features of the currently sent request.
[0045] In some embodiments, obtaining the transmission feature points includes: using a MAC (Multiply-Accumulate) array to perform feature detection and classification on the input feature vector and the weight vector to obtain the transmission feature points.
[0046] Step S13: Use a configurable feature table to compare the transmission feature points to identify specific features with urgent needs in the target transmission request and generate a priority selection signal.
[0047] In one embodiment of this disclosure, comparing the transmission feature points using a configurable feature table to identify specific features with urgent needs in the target transmission request and generating a priority selection signal may include: comparing and classifying the transmission feature points using reference feature points in the configurable feature table to obtain specific features with urgent needs in the target transmission request; and generating the priority selection signal for the target transmission request based on the specific features.
[0048] In some embodiments, the transmission feature points include a send request buffer, an outstanding number of incomplete requests, and a send request latency. The configurable feature table contains a constant vector for calculating the transmission features, and the constant vector is... Figure 4 The dataset that matches the waveform curves in the dataset.
[0049] Figure 4The waveform curve represents the trend of buffer usage, computation status, and bus control signals changing over time during the request sending process. For example... Figure 4 As shown, "Outstanding" represents the total number of requests sent by an IP core to the bus but for which no completion signal has been received; that is, the number of requests that have been sent but not processed. "Latency" represents the time taken from when an IP core sends a request to when that request is processed, used to measure the system's response speed. For example, if the buffer utilization of a request rises from 20% to 80% over a period of time, this rise is represented by a waveform. When the waveform shows a rapid decline, it indicates that the bus is in a "data starvation" mode; when the waveform shows a sustained high load, it indicates that the bus is in a "big data transfer" mode.
[0050] Furthermore, each reference feature point corresponds to a known request mode, such as an emergency mode where data is about to run out and a normal mode where data status is normal. The transmission feature point of the target sending request is compared with the reference feature points in the configurable feature table to identify which reference feature point the current transmission feature point is most similar to, thereby extracting specific features to determine whether there is an urgent need.
[0051] In some embodiments, a priority selection signal is generated by direct mapping based on specific characteristics and sent to the bus. The bus arbitrator receives the priority selection signal and sends a request to the target corresponding to the priority selection signal to assign a specific, competitive priority, i.e., a specific priority.
[0052] Step S14: Arbitrate multiple transmission requests on the bus according to the priority selection signal, so that the target transmission request is given priority in arbitration.
[0053] Figure 5 This is a schematic diagram illustrating the priority arbitration process in one embodiment of this disclosure. Figure 5 As shown, arbitrating multiple transmission requests on the bus according to the priority selection signal includes the following steps S41 to S43.
[0054] Step S41: Based on the priority selection signal, the priority of the target transmission request is processed using the first multiplexer to obtain a new priority for the target transmission request.
[0055] In some embodiments, processing the priority of the target transmission request using a first multiplexer according to the priority selection signal to obtain a new priority for the target transmission request includes the following steps S51 to S53.
[0056] Step S51: Obtain the basic priority of the target sending request. Generally, the normal priority of the sending request is conveyed through the QoS (Quality of Service) of the AXI bus.
[0057] Step S52: Obtain the specific priority of the target sending request based on the priority selection signal. Upon detecting a specific characteristic of the target sending request, output the priority selection signal to assign a higher specific priority.
[0058] Step S53: The first multiplexer is used to process the basic priority and the specific priority to obtain a new priority for the target sending request.
[0059] In some embodiments, the base priority of the target sending request is a static priority pre-set based on the functional importance of the IP core. A priority selection signal is generated according to the aforementioned process to obtain a specific priority for the target sending request. This specific priority is dynamically generated in real-time based on the sending request buffer level, reflecting the current urgency of the target sending request. A first multiplexer is used to fuse the base priority and the specific priority to generate a higher priority value. This ensures critical functional services are maintained while flexibly responding to sudden high-urgency demands, achieving a combination of static strategy and dynamic adjustment.
[0060] Step S42: Use a priority comparator to compare the new priority of the target transmission request with the priority of the parallel transmission request on the bus to obtain the priority comparison result.
[0061] Step S43: Based on the priority comparison result, the second multiplexer is used to arbitrate the target transmission request and the parallel transmission request to obtain the priority arbitration result of the target transmission request.
[0062] In some embodiments, after a priority selection signal is generated for a target transmission request based on the specific characteristics of the urgent need, the bus enters a priority arbitration phase. First, a first multiplexer obtains the new priority of the target transmission request based on the priority selection signal. This new priority is then input to a priority comparator, which comprehensively compares the priorities of all parallel transmission requests to determine the highest priority. The priority comparison result is then input to a second multiplexer, which compares all parallel transmission requests. If the new priority of the target transmission request is higher than the priorities of other parallel transmission requests, arbitration is performed on the target transmission request to obtain a priority arbitration result.
[0063] The following section will provide a detailed description of the feature-based request transmission processing method provided in this application through a specific example. It should be noted that the content of this example is only used to explain and illustrate the feature-based request transmission processing method provided in this application, and is not intended to limit the scope of protection of this application in any way. In specific applications, corresponding steps can be added or deleted based on this example according to actual needs.
[0064] In some embodiments, the request transmission processing method based on feature recognition includes the following steps S100 to S107.
[0065] Step S100: Based on the preset detection features, design a configurable feature table and weight table offline, and load and configure them during the chip operation phase.
[0066] Step S101 involves real-time monitoring and acquisition of the processing status of the request sent by the requesting end on the bus and the bus control signals. The input status includes the requesting end's buffer usage status and the request calculation status.
[0067] Step S102: The processing status of the target sending request and the bus control signal are calculated using a configurable weight table, and the transmission feature points characterizing the target sending request behavior are obtained by MAC array quantization.
[0068] Step S103: Compare the real-time obtained transmission feature points with the reference feature points in the configurable feature table to match specific features with urgent needs in the target transmission request, and generate a priority selection signal.
[0069] Step S104: Obtain the specific priority of the target sending request based on the priority selection signal.
[0070] Step S105: Use the first multiplexer to fuse the base priority and specific priority of the target transmission request to obtain a new priority for the target transmission request.
[0071] Step S106: Use a priority comparator to compare the new priority of the target transmission request with the priorities of multiple transmission requests on the bus to obtain the priority comparison result.
[0072] Step S107: When the new priority of the target transmission request is higher than the priority of other parallel transmission requests, the target transmission request is preferentially arbitrated using the second multiplexer.
[0073] It should be noted that the labels S100 to S107 are only used to indicate different steps, and not to restrict the execution order of these steps.
[0074] The scope of protection of the request transmission processing method based on feature recognition described in this disclosure is not limited to the execution order of the steps listed in this embodiment. Any solution implemented by adding, subtracting, or replacing steps in the prior art based on the principles of this disclosure is included within the scope of protection of this disclosure.
[0075] This disclosure also provides a feature-based request transmission processing apparatus, which can implement the feature-based request transmission processing method described in this disclosure. However, the apparatus for implementing the feature-based request transmission processing method described in this disclosure includes, but is not limited to, the structure of the feature-based request transmission processing apparatus listed in this embodiment. Any structural modifications and substitutions of the prior art made in accordance with the principles of this disclosure are included within the protection scope of this disclosure.
[0076] Figure 6 The diagram shown is a structural schematic of a feature-based request transmission processing apparatus according to an embodiment of this disclosure. Figure 6 As shown, the request transmission processing device 1 based on feature recognition includes: a request status acquisition module 11, a feature calculation module 12, a feature point comparison module 13, and a priority arbitration module 14.
[0077] The request status acquisition module 11 is configured to acquire the processing status of the target sending request and the bus control signals of the requesting end on the bus.
[0078] The feature calculation module 12 is configured to perform feature calculations on the processing state and the bus control signal using a configurable weight table to generate transmission feature points for the target transmission request.
[0079] The feature point comparison module 13 is configured to perform feature point comparison on the transmission feature points using a configurable feature table to identify specific features with urgent needs in the target transmission request and generate a priority selection signal.
[0080] The priority arbitration module 14 is configured to arbitrate multiple transmission requests on the bus according to the priority selection signal, so that the target transmission request is given priority arbitration.
[0081] It should be noted that, Figure 6 The modules in the feature-based request transmission processing device 1 shown are related to... Figure 2 The steps in the feature-based request transmission processing method correspond one-to-one, and will not be elaborated here.
[0082] In another aspect, this disclosure also provides a chip. The chip includes a feature detection module and an arbitrator.
[0083] The feature detection module is configured to acquire the processing status of the target transmission request and the bus control signal from the requesting end on the bus, and to perform feature calculations on the processing status and the bus control signal using a configurable weight table to generate transmission feature points for the target transmission request. Furthermore, the feature detection module is configured to compare the transmission feature points using a configurable feature table to identify specific features with urgent needs in the target transmission request and generate a priority selection signal.
[0084] The arbitrator is configured to receive the target transmission request and the parallel transmission request, and to arbitrate the target transmission request and the parallel transmission request on the bus according to the priority selection signal, such that the target transmission request is given priority in arbitration.
[0085] In one embodiment of this disclosure, the arbitrator includes a first multiplexer, a priority comparator, and a second multiplexer.
[0086] The first multiplexer is configured to receive a priority selection signal and process the priority of the target transmission request on the bus according to the priority selection signal to obtain a new priority for the target transmission request.
[0087] The priority comparator is configured to compare the new priority of the target transmission request with the priority of the parallel transmission requests on the bus to obtain a priority comparison result.
[0088] The second multiplexer is configured to arbitrate the target transmission request and the parallel transmission request based on the priority comparison result to obtain the priority arbitration result of the target transmission request.
[0089] In the several embodiments provided in this disclosure, it should be understood that the disclosed systems, apparatuses, or methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules / units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or units may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection of apparatuses or modules or units may be electrical, mechanical, or other forms.
[0090] The modules / units described as separate components may or may not be physically separate. The components shown as modules / units may or may not be physical modules; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules / units can be selected to achieve the objectives of the embodiments of this disclosure, depending on actual needs. For example, the functional modules / units in the various embodiments of this disclosure may be integrated into one processing module, or each module / unit may exist physically separately, or two or more modules / units may be integrated into one module / unit.
[0091] 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 disclosure.
[0092] This disclosure also provides a computer-readable storage medium storing a computer program thereon. When executed by a processor, the computer program implements the feature-based request transmission processing method provided in this disclosure. Those skilled in the art will understand that all or part of the steps in the methods of the above embodiments can be implemented by a program instructing a processor. The program can be stored in a computer-readable storage medium, which is a non-transitory medium, such as random access memory, read-only memory, flash memory, hard disk, solid-state drive, magnetic tape, floppy disk, optical disk, and any combination thereof. The storage medium can be any available medium accessible to a computer or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., digital video disc (DVD)), or a semiconductor medium (e.g., solid-state disk (SSD)).
[0093] This disclosure also provides an electronic device. Figure 7 The diagram shown is a structural schematic of an electronic device 200 according to an embodiment of this disclosure. Figure 7As shown, in this embodiment, the electronic device 200 includes a memory 201 and a processor 202.
[0094] The memory 201 is used to store computer programs. In some possible implementations, the memory 201 may include various media capable of storing program code, such as ROM, RAM, magnetic disk, USB flash drive, memory card, or optical disk.
[0095] In embodiments of this disclosure, memory 201 may include computer system readable media in the form of volatile memory, such as RAM and / or cache memory. Electronic device 200 may further include other removable / non-removable, volatile / non-volatile computer system storage media. Memory 201 may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of the embodiments of this disclosure.
[0096] The processor 202 is connected to the memory 201 and is used to execute the computer program stored in the memory 201 so that the electronic device 200 performs a request transmission processing method based on feature recognition.
[0097] In some embodiments, processor 202 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc. In other embodiments, processor 202 may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.
[0098] In some implementations, the electronic device 200 provided in this disclosure embodiment may further include a display 203. The display 203 is communicatively connected to the memory 201 and the processor 202, and is used to display a graphical user interface (GUI) related to the feature recognition-based request transmission processing method.
[0099] In this embodiment, the display 203 may include a display screen (display panel). In some implementations, the display panel may be configured using a liquid crystal display (LCD), an organic light-emitting diode (OLED), or similar methods. Furthermore, the display 203 may also be a touch panel (touchscreen, touch screen), which may include a display screen and a touch-sensitive surface. When the touch-sensitive surface detects a touch operation on or near it, it transmits the information to the processor 202 to determine the type of touch event. Subsequently, the processor 202 provides corresponding visual output on the display device based on the type of touch event.
[0100] The descriptions of the processes or structures corresponding to the above figures each have their own emphasis. For parts of a process or structure that are not described in detail, please refer to the relevant descriptions of other processes or structures.
[0101] The above embodiments are merely illustrative of the principles and effects of this disclosure and are not intended to limit this disclosure. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of this disclosure. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in this disclosure should still be covered by the claims of this disclosure.
Claims
1. A request transmission processing method based on feature recognition, characterized in that, include: Obtain the processing status of the target request sent by the requesting end on the bus and the bus control signals; The processing state and the bus control signal are used to perform feature calculations using a configurable weight table to generate transmission feature points for the target transmission request; The transmission feature points are compared using a configurable feature table to identify specific features with urgent needs in the target transmission request and generate a priority selection signal. as well as Arbitrate multiple transmission requests on the bus according to the priority selection signal, so that the target transmission request is given priority in arbitration.
2. The method according to claim 1, characterized in that, The processing status of the target sending request includes the request-side cache usage status and the request calculation status.
3. The method according to claim 2, characterized in that, The process of calculating the characteristics of the processing state and the bus control signals using a configurable weight table to generate the transmission feature points of the target transmission request includes: The input feature vector is obtained based on the request-end cache usage status, the request calculation status, and the bus control signal. Obtain the weight vector according to the configurable weight table; and Feature calculations are performed on the input feature vector and the weight vector to obtain the transmission feature points.
4. The method according to claim 3, characterized in that, Obtaining the transmission feature points includes: The input feature vector and the weight vector are used to perform feature detection and classification to obtain the transmission feature points.
5. The method according to claim 1, characterized in that, Also includes: The system receives the configurable weight table and the configurable feature table, wherein the configurable weight table contains constant vectors with specific transmission characteristics, and the configurable feature table contains constant vectors used to calculate the transmission characteristics.
6. The method according to claim 1, characterized in that, The process of comparing transmission feature points using a configurable feature table to identify specific features with urgent needs in the target transmission request and generating a priority selection signal includes: The transmission feature points are compared and classified using reference feature points in the configurable feature table to obtain specific features with urgent needs in the target transmission request; and The priority selection signal for sending the target request is generated based on the specific characteristics.
7. The method according to claim 1, characterized in that, Arbitration of multiple transmit requests on the bus based on the priority selection signal includes: Based on the priority selection signal, the priority of the target transmission request is processed using a first multiplexer to obtain a new priority for the target transmission request; The new priority of the target transmission request is compared with the priority of the parallel transmission requests on the bus using a priority comparator to obtain the priority comparison result; and Based on the priority comparison result, the second multiplexer is used to arbitrate the target transmission request and the parallel transmission request to obtain the priority arbitration result of the target transmission request.
8. The method according to claim 7, characterized in that, The process of processing the priority of the target transmission request using a first multiplexer to obtain a new priority for the target transmission request, based on the priority selection signal, includes: Obtain the base priority of the target sending request; The specific priority of the target sending request is obtained based on the priority selection signal; and The first multiplexer is used to process the base priority and the specific priority to obtain a new priority for sending the target request.
9. The method according to claim 1, characterized in that, The transmission characteristics include the send request buffer, the number of incomplete requests, and the send request delay.
10. A request transmission processing apparatus based on feature recognition, characterized in that, include: The request status acquisition module is configured to acquire the processing status of the target request sent by the requesting end on the bus and the bus control signals; The feature calculation module is configured to perform feature calculations on the processing state and the bus control signal using a configurable weight table to generate transmission feature points for the target transmission request. The feature point comparison module is configured to compare the transmission feature points using a configurable feature table to identify specific features with urgent needs in the target transmission request and generate a priority selection signal. as well as An arbitration module is configured to arbitrate multiple transmission requests on the bus according to the priority selection signal, such that the target transmission request is given priority in arbitration.
11. A chip, characterized in that, include: The feature detection module is configured to acquire the processing status of the target sending request and the bus control signal of the requesting end on the bus, perform feature calculation on the processing status and the bus control signal using a configurable weight table to generate transmission feature points of the target sending request, and compare the transmission feature points using a configurable feature table to identify specific features with urgent needs in the target sending request and generate a priority selection signal. as well as An arbitrator is configured to receive the target transmission request and the parallel transmission request, and to arbitrate the target transmission request and the parallel transmission request on the bus according to the priority selection signal, such that the target transmission request is given priority in arbitration.
12. The chip according to claim 11, characterized in that, The arbitrator includes: A first multiplexer is configured to receive a priority selection signal and process the priority of the target transmission request according to the priority selection signal to obtain a new priority for the target transmission request. A priority comparator is configured to compare the new priority of the target transmission request with the priority of the parallel transmission requests on the bus to obtain a priority comparison result; and The second multiplexer is configured to arbitrate the target transmission request and the parallel transmission request based on the priority comparison result, so as to obtain the priority arbitration result of the target transmission request.
13. An electronic device, characterized in that, include: The memory is configured to store executable programs; as well as A processor is configured to execute the program to implement the feature-based request transmission processing method according to any one of claims 1 to 9.