Network interface management system for physical interface remote management

Abstract

The application relates to a network port management system for remote management of a physical interface. Instead of passively relying on the original support of the RMII interface by the baseboard management controller, the system actively configures the required RMII interface of the NCSI based on the signals, clock and rate conversion mechanism inside the baseboard management controller, flexibly sets the MAC address in the network interface controller through the NCSI protocol, and further establishes a communication connection based on the MAC address, so as to realize the remote management of the physical interface. The system enables the management controller to control the network interface in a more fine and flexible manner, better cope with different network interfaces and management requirements, and significantly reduce the complexity and cost of hardware design.

Description

Technical Field

[0001] This application relates to the field of network port management system technology, and in particular to a network port management system for remote management of physical interfaces. Background Technology

[0002] Developed by Intel, the NCSI protocol is a key communication protocol in the server management field, primarily used for data interaction between the baseboard management controller and the network interface controller. The NCSI protocol establishes a standardized control communication framework, supporting diverse transmission media and physical interface configurations. Its core function is to enable the management controller to remotely monitor, configure, and maintain devices through the network interface controller.

[0003] In practical applications, the deployment of the NCSI protocol faces significant challenges, primarily due to the strict dependence of existing implementations on the hardware interface of the baseboard management controller (BMC). Specifically, current technology requires the BMC to natively integrate an RMII interface to meet the physical layer transmission requirements of the NCSI protocol. When the hardware platform lacks a compatible RMII interface, system designers are forced to introduce additional signal conversion circuitry or develop complex software simulation layers. This not only significantly increases the complexity of hardware design and manufacturing costs but also leads to timing mismatches during protocol conversion. Furthermore, interface differences between devices from different manufacturers force each new platform to undergo customized development, extending product launch cycles and limiting the technology's universal applicability. These limitations severely hinder the efficient deployment of the NCSI protocol in diverse server environments. Summary of the Invention

[0004] The following is an overview of the subject matter described in detail herein. This overview is not intended to limit the scope of the claims.

[0005] The main objective of this disclosure is to propose a network interface management system for remote management of physical interfaces, which can control network interfaces in a more precise and flexible manner, better meet different network interface and management needs, and significantly reduce the complexity and cost of hardware design.

[0006] A first aspect of this application provides a network port management system for remote management of physical interfaces, the system comprising:

[0007] The network interface card (NIC) is equipped with a first link node controller, which is communicatively connected to at least one physical interface, the physical interface being used to connect to an external device.

[0008] A network interface controller, which is communicatively connected to the network card;

[0009] The second link node controller is located in the baseboard management controller, and the second link node controller is connected to the first link node controller via the NCSI protocol.

[0010] A management controller is disposed in the baseboard management controller, and the management controller is communicatively connected to the second link node controller; the management controller remotely manages the at least one physical interface through the network interface controller based on the NCSI protocol of the first link node controller and the second link node controller.

[0011] The process of configuring the second link node controller to communicate with the first link node controller via the NCSI protocol includes:

[0012] The management controller generates the RMII interface required by NCSI based on the signal, clock, and rate conversion mechanism in the baseboard management controller, and performs NCSI protocol stack source code porting and NCSI protocol configuration according to the RMII interface.

[0013] The management controller, based on the NCSI protocol of the first link node controller and the second link node controller, remotely manages the at least one physical interface through the network interface controller, including:

[0014] The management controller sets a MAC address in the network interface controller through the NCSI protocol of the first link node controller and the second link node controller, and establishes a communication connection with the network interface controller based on the MAC address, so as to remotely manage the at least one physical interface through the network interface controller.

[0015] The network port management system for remote management of physical interfaces provided in this embodiment has at least the following beneficial effects:

[0016] This system no longer passively relies on the baseboard management controller's native support for the RMII interface. Instead, it proactively and flexibly configures the RMII interface required by NCSI based on the signal, clock, and rate conversion mechanism within the baseboard management controller. It then sets the MAC address in the network interface controller via the NCSI protocol and establishes a communication connection based on the MAC address, thereby enabling remote management of the physical interface. This system allows the management controller to control the network interface in a more precise and flexible manner, better addressing different network interface and management needs, and significantly reducing the complexity and cost of hardware design.

[0017] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a schematic diagram of the structure of a network port management system for remote management of physical interfaces provided in an embodiment of this application;

[0020] Figure 2 This is a schematic diagram of the service flow of a network port management system provided in an embodiment of this application. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0022] In the description of this application, the use of terms such as "first," "second," etc., is for the purpose of distinguishing technical features only and should not be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated or the order of the technical features indicated.

[0023] In the description of this application, it should be understood that the orientation descriptions, such as up, down, etc., are based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed or function in a specific orientation, and therefore should not be construed as a limitation of this application.

[0024] In traditional NCSI protocol applications, communication between the baseboard management controller (BMC) and the network interface controller (NIC) is limited by the requirement that the BMC chip must support the RMII interface. When a non-RMII physical interface is used, incompatibility between the interface protocols necessitates complex software and hardware adaptations and protocol conversions, increasing system integration difficulty and negatively impacting the stability and real-time performance of the communication link. This, in turn, hinders the reliable execution of remote server management functions. For example, in data center server management scenarios, the BMC needs to access multiple physical network ports through the NIC to perform remote configuration and maintenance tasks. When the BMC chip does not integrate an RMII interface, signal conversion circuits and protocol adaptation layers are additionally deployed, introducing timing deviations and noise interference during signal transmission. This frequently triggers errors in management command execution or communication link interruptions, disrupting the continuity of system maintenance operations.

[0025] like Figure 1 As shown, a network port management system for remote management of physical interfaces is provided. The system includes:

[0026] The network interface card (NIC) is equipped with a first link node controller, which communicates with at least one physical interface, which is used to connect to external devices.

[0027] Network interface controller, which communicates with the network card;

[0028] The second link node controller is located in the substrate management controller, and the second link node controller is connected to the first link node controller via the NCSI protocol.

[0029] A management controller is disposed in the baseboard management controller and is communicatively connected to the second link node controller. The management controller is based on the NCSI protocol of the first link node controller and the second link node controller to remotely manage at least one physical interface through the network interface controller.

[0030] The process of configuring the second link node controller to communicate with the first link node controller via the NCSI protocol includes:

[0031] The management controller generates the RMII interface required by NCSI based on the signal, clock, and rate conversion mechanism in the baseboard management controller, and performs NCSI protocol stack source code porting and NCSI protocol configuration based on the RMII interface;

[0032] The management controller, based on the NCSI protocol of the first and second link node controllers, remotely manages at least one physical interface through the network interface controller, including:

[0033] The management controller sets the MAC address in the network interface controller through the NCSI protocol of the first link node controller and the second link node controller, and establishes a communication connection with the network interface controller based on the MAC address, so as to remotely manage at least one physical interface through the network interface controller.

[0034] This embodiment provides a network interface management system for remote management of physical interfaces. The system aims to utilize NCSI to enable the management controller to remotely manage the network interfaces of network cards. The system completes management tasks through the collaborative work of multiple functional modules.

[0035] In this embodiment, the network interface card (NIC) is a hardware device for connecting to an external network, and it internally includes a first link node controller. The first link node controller is responsible for handling the data flow between the NIC and the external physical interface. The first link node controller can be a dedicated hardware logic unit. The first link node controller establishes a communication connection with at least one physical interface, which is the channel for data exchange between the system and external devices. For example, it can be an RJ45 interface used to connect an Ethernet cable. The physical interface provides a standardized connection point, enabling the system to interconnect with various external network devices. The network interface controller communicates with the NIC. The network interface controller can also be a dedicated hardware logic unit, and communication between the network interface controller and the NIC can be implemented in various ways, such as through a PCIe bus or a dedicated internal bus.

[0036] The second link node controller is located within the baseboard management controller. As part of the baseboard management controller, its function is to act as an endpoint of the NCSI protocol and communicate with the first link node controller in the network interface card. The second link node controller and the first link node controller communicate via the NCSI protocol, which enables the baseboard management controller to manage network functions through the network interface controller.

[0037] The management controller is also located within the baseboard management controller and communicates with the second link node controller. The management controller is the core logic within the baseboard management controller, responsible for parsing and executing management commands. Its communication with the second link node controller can be via an internal bus, ensuring accurate transmission of management commands. Based on the NCSI protocol between the first and second link node controllers, the management controller remotely manages at least one physical interface through the network interface controller. This means the management controller can utilize the NCSI protocol mechanism to remotely configure, monitor, and diagnose the functionality of the network interface. For example, the management controller can send NCSI commands to query the status of the physical interface.

[0038] Specifically, the management controller, based on the signal, clock, and rate conversion mechanisms within the baseboard management controller, configures the RMII interface required by NCSI, and performs the porting of the NCSI protocol stack source code and NCSI protocol configuration. The baseboard management controller typically contains multiple interfaces and clock sources; through flexible configuration, the signal timing required by the RMII interface can be simulated or generated. Porting the NCSI protocol stack source code involves adapting general protocol code to the specific hardware and software environment of the baseboard management controller, while NCSI protocol configuration includes setting various protocol parameters, such as channel ID and timeout.

[0039] The management controller configures the MAC address in the network interface controller via the NCSI protocol between the first and second link node controllers. A MAC address is a unique identifier for a network device at the data link layer. The management controller can send specific commands via the NCSI protocol to write a preset MAC address into the register of the network interface controller, thereby achieving dynamic MAC address configuration.

[0040] Finally, the management controller establishes a communication connection with the network interface controller based on the MAC address, enabling remote management of at least one physical interface through the network interface controller. After the MAC address is set, the management controller can utilize protocols such as SMBUS for deeper communication with the network interface controller. The SMBUS protocol provides a low-speed serial communication method suitable for management operations such as configuration, status queries, and event notifications between the management controller and the network interface controller. For example, the management controller can use the SMBUS protocol to read temperature sensor data from the network interface controller or send commands to restart it.

[0041] In this embodiment, the management controller no longer passively relies on the baseboard management controller's native support for the RMII interface. Instead, it actively and flexibly configures the RMII interface required by NCSI based on the baseboard management controller's internal signal, clock, and rate conversion mechanisms. This embodiment achieves interface compatibility through software configuration and internal resource reuse, significantly reducing the complexity and cost of hardware design. Furthermore, this embodiment sets the MAC address in the network interface controller via the NCSI protocol and further establishes a communication connection based on the MAC address, thereby enabling remote management of the physical interface. This phased, multi-protocol collaborative management approach allows the management controller to control the network interface in a more granular and flexible manner, better addressing different network interface and management needs.

[0042] This embodiment provides a method for achieving NCSI remote management without complex hardware modifications, simplifying system design, reducing implementation costs, and improving system manageability and flexibility. This has significant practical implications for applications such as servers that require efficient remote management.

[0043] In some embodiments of this application, establishing a communication connection with the network interface controller includes:

[0044] The management controller polls the network interface controller by sending messages at preset time intervals, so that the network interface controller can complete the master-slave-master switching process of the SMBUS protocol.

[0045] The management controller and network interface controller send and respond to requests based on the SMBUS protocol's block write with pec, and establish communication connections.

[0046] The management controller polls the network interface controller by sending specific messages at preset time intervals. Its purpose is to proactively detect the status of the network interface controller and trigger or coordinate the master-slave-master switchover process in the SMBUS protocol. This polling mechanism ensures that the management controller can promptly detect changes in the communication status of the network interface controller. For example, it can be achieved by configuring a hardware timer to periodically generate interrupts, performing message sending and status checks in the interrupt service routine; or by using a software loop to check whether the preset time interval has been reached in each loop and send a polling message. This switchover can be implemented through specific command sequences or status register configurations defined by the SMBUS protocol, ensuring communication flexibility and efficiency.

[0047] The management controller and network interface controller establish a stable communication connection by exchanging requests and responses through the SMBUS protocol's block write with PEC mechanism. The management controller sends request messages, the network interface controller receives and processes the requests, and then sends response messages. This request-response mode, combined with PEC verification, ensures that both parties can reliably exchange configuration information or handshake signals when establishing a communication connection, laying the foundation for subsequent remote management operations.

[0048] This embodiment not only improves the efficiency of SMBUS communication connection establishment, but also greatly enhances the stability and reliability of the connection, providing a solid and error-free communication foundation for the management controller to remotely manage the physical interface through the network interface controller.

[0049] In some embodiments of this application, the NCSI protocol configuration includes at least:

[0050] NCSI protocol network card management interface configuration, NCSI protocol driver interface configuration, and NCSI protocol command configuration.

[0051] In this embodiment, the management controller can perform comprehensive and fine-grained configuration of the NCSI protocol, thereby ensuring the stability and reliability of remote management functions. This detailed configuration enables the management controller to more effectively control and monitor the network interface controller and its physical interfaces, significantly improving the efficiency and security of remote management.

[0052] In some embodiments of this application, the management controller is also used to send request instructions to the network interface controller, receive judgment signals fed back by the network interface controller, send IP address configuration instructions to the network interface controller to complete the IP communication connection when the judgment signal is a safe information, and send a warning signal to the host computer connected to the management controller when the judgment signal is an unsafe information.

[0053] The network interface controller is used to perform security checks on request commands and to send the check signals back to the management controller.

[0054] Request commands are instructions issued by the management controller to the network interface controller to request the execution of specific operations or the retrieval of specific information. In remote management scenarios, these commands may involve network configuration, status queries, or enabling / disabling functions. A request command can be a structured data packet containing fields such as command type, parameters, and target address, transmitted via a predefined communication protocol. Alternatively, a request command can be a specific register write operation, whereby the management controller triggers the corresponding operation or request information by writing to a specific register within the network interface controller.

[0055] The judgment signal is an indication signal sent by the network interface controller to the management controller after performing a security assessment on a received request command. It informs the management controller whether the command is safe and can be executed. The judgment signal can be a Boolean value (e.g., 0 for insecure, 1 for secure) or a short message containing a status code, sent back to the management controller via the communication link. The judgment signal can also be represented by specific hardware pin level changes, such as high level for secure and low level for insecure.

[0056] Security information is a specific state of a signal, indicating that the network interface controller has confirmed that the received request command is legitimate, authorized, and poses no security threat to the system. An IP address configuration command is a command sent by the management controller to the network interface controller after confirming the security of the request command, used to set or modify the IP address of the network interface. This is a crucial step in establishing an IP communication connection. An IP address configuration command can be a data structure containing network parameters such as IP address, subnet mask, and gateway, sent to the network interface controller through a specific protocol interface. Alternatively, the IP address configuration command can also be implemented by writing to specific configuration registers within the network interface controller, which store the IP configuration information of the network interface.

[0057] Insecure information is another specific state of the judgment signal, indicating that the network interface controller determines that the received request command has a security risk, such as incorrect command format, unauthorized, or containing malicious payload.

[0058] An early warning signal is a signal sent by the management controller to the connected host computer after receiving unsafe information. It notifies the host computer system that a potential security threat or anomaly has been detected. Early warning signals can be sent to the host computer via network protocols or triggered by specific hardware interfaces. They can also be pop-up messages or log entries pushed to the host computer's management interface. The host computer is a computer system or terminal device connected to the management controller for system monitoring, management, and human-computer interaction. The host computer can be a PC server or workstation running specific management software. Alternatively, it can be an embedded management terminal that communicates with the management controller via serial port, USB, or Ethernet.

[0059] Security assessment is the process by which a network interface controller evaluates received request commands to determine whether the command conforms to preset security policies and specifications. Security assessment may include authenticating the source of the command, verifying the legality of the command content, and validating command permissions. Alternatively, security assessment may involve comparing the command with a preset security policy database or using behavioral analysis to identify abnormal or malicious commands.

[0060] In this embodiment, the commands sent by the management controller undergo effective security checks at the network interface controller. This prevents unauthorized or malicious requests from directly performing critical configurations on the network interface, such as IP address settings, thereby significantly improving the security of the entire network interface management system. When an insecure command is detected, the system can promptly issue a warning to the host computer, enabling administrators to respond quickly and take measures to effectively prevent potential security threats and ensure the stability of network communication and the integrity of data. This mechanism provides a necessary security barrier for critical network configuration operations without compromising the convenience of remote management, enhancing the system's reliability and resistance to attacks.

[0061] In some embodiments of this application, the system further includes:

[0062] The logic communication controller communicates with the management controller. The logic communication controller encapsulates at least the TCP, NETBIOS, and ARP protocols. The logic communication controller is used to encapsulate data packets according to the encapsulated protocols and establish a connection with the network card through a handshake mechanism.

[0063] The logic communication controller is a module or unit specifically designed to handle network communication logic. Its main function is to provide support for higher-level network protocol stacks, facilitating packet encapsulation and decapsulation, and managing network connections. The communication connection between the logic communication controller and the management controller aims to enable the exchange of data and control information between them. The logic communication controller integrates at least the Transmission Control Protocol (TCP), Network Basic Input / Output System (NETBIOS), and Address Resolution Protocol (ARP) stack. The encapsulation of these protocols means that the logic communication controller can handle the message formats, state machines, and related logic of these protocols, thereby supporting higher-level network communication functions. Encapsulating data packets according to the encapsulated protocol stack means that after receiving data from the upper-layer application, the logic communication controller adds the corresponding protocol header and trailer information to the data according to the specifications of protocols such as TCP, NETBIOS, and ARP, forming data packets that conform to network transmission standards.

[0064] The handshake mechanism between the network interface card (NIC) and the logic communication controller refers to a series of predetermined information exchanges between the NIC and the logic communication controller before data transmission begins, in order to establish, synchronize, and confirm communication parameters and status. For example, a handshake based on hardware signal lines or a handshake based on software protocols can be used to ensure a stable and reliable data transmission channel between the two, avoiding data loss or conflicts.

[0065] This embodiment overcomes the limitations of relying solely on low-level protocols for basic communication, enabling the system to support higher-level, standardized network communication, such as reliable data stream transmission, network resource naming, and address resolution. This provides more robust data exchange capabilities for remote management functions. This not only improves the reliability and integrity of data transmission, preventing data loss or out-of-order delivery, but also enhances system interoperability, allowing seamless communication with external devices that conform to standard network protocols. Therefore, the management controller can perform complex remote management tasks more efficiently and flexibly, such as remote file transfer, software updates, or diagnostics, greatly expanding the application scope and functional depth of the remote management system.

[0066] In some embodiments of this application, the network interface card is further provided with a first network layer controller, which is communicatively connected to a first link node controller;

[0067] The system also includes: a second network layer controller, and a communication connection between the second network layer controller, the second link node controller, and the management controller;

[0068] The first network layer controller is used to receive data packets sent by the second network layer controller, parse MAC addresses, configure the network protocol stack if the MAC address matches successfully, and send data packets to the second network layer controller through the NCSI protocol between the first link node controller and the second link node controller if the MAC address does not match successfully.

[0069] The first network layer controller is a logical or hardware module that processes network layer data packets. Its main functions are receiving, parsing, filtering, and forwarding data packets. As one implementation, the first network layer controller can be integrated inside the network interface card (NIC) chip, existing as a dedicated hardware logic circuit. Alternatively, it can be a software module running in the NIC firmware, executed by the NIC's internal processor, working in conjunction with the hardware to implement data packet parsing and control functions.

[0070] This embodiment introduces a first network layer controller and a second network layer controller. The second network layer controller communicates with the management controller and can generate or forward data packets according to the instructions of the management controller. These data packets are then sent to the first network layer controller in the network interface card (NIC). After receiving the data packets from the second network layer controller, the first network layer controller performs MAC address resolution. When the destination MAC address of the data packet matches the address configured in the NIC, the first network layer controller delivers the data packet to the network protocol stack inside the NIC for normal processing, enabling the host system to receive and process data initiated by the management controller and destined for the host. However, when the destination MAC address of the data packet does not match, the first network layer controller does not deliver the data packet to the host. Instead, it redirects these mismatched data packets back to the second network layer controller through the NCSI protocol channel established between the first and second link node controllers. This mechanism allows the management controller to flexibly test the data processing path of the NIC, or to perform cyclic processing or diagnosis of specific data packets. It solves the limitations of traditional management systems in the routing and processing of data packets initiated by the management controller, and enhances the management controller's control and testing capabilities over the NIC data path.

[0071] In some embodiments of this application, the network card is further provided with a first buffer module, which is communicatively connected to the first network layer controller;

[0072] The system also includes a second buffer module, which is communicatively connected to the second network layer controller;

[0073] The first and second buffer modules are used for flow control between the first and second network layer controllers.

[0074] The first buffer module is a buffer for temporary data storage. Its concept is to provide an intermediate storage area to coordinate possible differences in data rates or processing capabilities between the data sender and receiver. This module can be, for example, a dedicated SRAM (Static Random Access Memory) or DRAM (Dynamic Random Access Memory) area inside the network card, or a first-in-first-out (FIFO) queue implemented in a field-programmable gate array (FPGA).

[0075] The second Buffer module is similar to the first Buffer module; it is also a buffer used for temporary data storage.

[0076] In this embodiment, a first buffer module is configured in the network interface card (NIC) and communicates with the first network layer controller, while a second buffer module is configured in the system and communicates with the second network layer controller. These two buffer modules are used for flow control between the first and second network layer controllers, effectively solving the data congestion and loss problems that may occur due to processing speed mismatch during data transmission between the first and second network layer controllers. This buffering and flow control mechanism ensures that data packets can be transmitted smoothly and reliably between the two network layer controllers. Even under conditions of large instantaneous data volumes or unbalanced processing loads, it can maintain stable system operation and efficient data exchange, thereby improving the reliability and performance of the entire network interface management system.

[0077] In some embodiments of this application, the first link node controller is used to redirect data packets and send the redirected data packets to the management controller via the NCSI protocol between the first link node controller and the second link node controller.

[0078] The NCSI protocol is used not only for the management controller to configure and query the network interface card (NIC), but also for transmitting specific data between the NIC and the baseboard management controller. When a data packet is marked by the first link node controller as needing to be redirected to the management controller, it is encapsulated or forwarded to the management controller via a specific channel of the NCSI protocol through the second link node controller. This can be achieved through data transmission commands defined in the NCSI protocol, or by utilizing the out-of-band data channel of the NCSI protocol.

[0079] This embodiment enables flexible routing and efficient processing of specific data packets. The packet redirection capability of the first link node controller allows the management controller to directly receive and process specific data packets that are unsuitable or should not be processed by the host system, such as out-of-band management commands, diagnostic information, or security event notifications. This avoids unnecessary forwarding and processing of data packets in the host system, significantly reducing the host CPU load and minimizing management operation latency. Simultaneously, utilizing the NCSI protocol as the transmission channel for redirected data packets ensures the independence and reliability of communication between the management controller and the network interface card (NIC). Even in the event of a host system failure or shutdown, the management controller can still effectively acquire and process critical information, thereby improving the robustness and management efficiency of the entire network interface management system.

[0080] In some embodiments of this application, the management controller, the first network layer controller, the second network layer controller, the first link node controller, and the second link node controller are all any one of hardware circuits, field-programmable gate logic, or software logic.

[0081] Specifically, hardware circuits refer to physical circuits built using electronic components, which perform logical operations and data processing through changes in current and voltage. Their implementation can include Application-Specific Integrated Circuits (ASICs), highly optimized for specific functions, providing high performance and low power consumption; or circuits built based on standard logic gates or discrete components, suitable for scenarios with low performance requirements or requiring high customization. Field-Programmable Gate Arrays (FPGAs) are logic circuits containing a large number of programmable logic units and programmable interconnect resources. Users can program and configure their internal logic to implement various digital circuit functions. Implementation can be done through hardware description languages ​​(such as VHDL and Verilog), mapping logic functions to FPGA resources such as lookup tables (LUTs) and flip-flops; or by using High-Level Synthesis (HLS) tools to convert C / C++ code into FPGA logic. Software logic refers to the execution of instructions on a general-purpose processor (such as a CPU or microcontroller) by writing computer program code to achieve specific functions. Implementation can be an application or driver running on an operating system (such as Linux or RTOS); or firmware running directly on bare metal.

[0082] This embodiment allows key controllers in the system to flexibly choose hardware circuits, field-programmable gate logic, or software logic for implementation based on their functional characteristics and performance requirements. This flexibility enables system designers to achieve the optimal balance between performance, cost, power consumption, and scalability according to actual needs. It effectively solves the technical problem of how to select a suitable implementation method in complex network port management systems to balance system flexibility, performance, development costs, and ease of maintenance and upgrades, thereby improving the adaptability and competitiveness of the entire network port management system.

[0083] In some embodiments of this application, the physical interface includes any one of circuit wiring, hub, twisted pair cable, and network port.

[0084] A physical interface is a port or medium that physically connects a system to an external network or device, responsible for the physical transmission of data, including the conversion and transmission of electrical, optical, or radio signals. Its implementation can include various standardized connectors, transmission media, or network devices. A circuit connection refers to the physical wire or circuit path used to transmit electrical signals; it can be a trace on a printed circuit board (PCB) or a bundle of wires used for short-distance connections within or between devices. A hub is a network device used to connect multiple network devices together to form a network segment sharing a medium. Hubs can be passive, providing only physical connection, or active, with signal amplification and shaping functions. Twisted-pair cable is a commonly used network transmission medium, consisting of two insulated wires twisted together at a certain density. This twisted structure effectively resists external electromagnetic interference and reduces signal attenuation and crosstalk. Twisted-pair cables can be divided into unshielded twisted-pair (UTP) and shielded twisted-pair (STP) based on their shielding, and are widely used in Ethernet connections. A network port, usually referring to an Ethernet port, is a standard interface used to connect Ethernet devices. It typically uses an RJ45 connector and supports data transmission via twisted-pair cable. A network port can be a physical interface integrated into a device, such as an Ethernet port on a computer motherboard, or a port on a standalone network interface card (NIC). For example, servers, routers, and switches are equipped with one or more network ports for accessing local area networks (LANs) or wide area networks (WANs).

[0085] This example's network interface management system significantly enhances its adaptability and flexibility across different network environments and device deployment scenarios. This diverse physical interface support allows the system to seamlessly integrate into various existing or new network infrastructures, whether through direct circuit connections, hub-extended connections, or standard Ethernet connections. Therefore, the system provides unified remote management capabilities for a wider range of external devices, reduces deployment complexity, and improves system versatility and scalability, effectively addressing potential compatibility issues when facing diverse physical connectivity requirements.

[0086] like Figure 1 and Figure 2 As shown in one embodiment of this application, a network port management system is provided, the system including the following:

[0087] This solution primarily addresses the issues of traditional network port management systems, such as over-reliance on a centralized processing platform, limited scalability of network port monitoring, and compatibility problems when new and old devices coexist. In a shared mode, it separates service traffic and manages the status and configuration of physical ports. NCSI technology is implemented through software configuration, increasing flexibility and reducing hardware design complexity.

[0088] The network port management system disclosed in this invention includes a network interface controller, a management controller, a logic communication controller, two network layer controllers, two link node controllers, and several physical interfaces. All controllers refer to hardware circuits, field-programmable gate circuits, or software logic capable of implementing this function; physical interfaces refer to circuit connections or connections such as hubs, twisted-pair cables, and network ports that enable data interaction.

[0089] The network interface management system is a complex system comprised of a management controller and a network interface controller working together via NCSI, encompassing functions at the application layer, transport layer, network layer, data link layer, and physical layer. Its architecture is as follows: Figure 1 As shown.

[0090] At the application layer, communication interfaces are provided through the network interface controller and management controller. These interfaces handle file transfer, domain name resolution, user authentication, and are the modules that ultimately enable remote management. The functional implementation steps are as follows:

[0091] In step S110, the management controller actively initiates a request, and the network interface controller distinguishes the request type, which can be mainly divided into two categories: configuration requests and management requests.

[0092] In step S120, the network interface controller determines the security information. If the security information is incorrect, it returns a request failure result to the management controller; if the security information is correct, it returns a request success result to the management controller.

[0093] Step S130: If the management controller fails to make a request, it sends an alarm message to the host computer (remote management computer) through the out-of-band management dedicated network port, notifying the administrator to log in to the web terminal of BMC through the network to troubleshoot.

[0094] In step S140, if the management controller's request is successful, it continues to send configuration commands to complete the IP address configuration and establish a communication channel.

[0095] At the transport layer, a dedicated logic communication control module encapsulates the TCP / NETBIOS / ARP protocols, providing reliable and efficient transmission services through the communication mechanism. Its functional steps are as follows:

[0096] Step S210: The logic communication controller encapsulates the data layer by layer according to the protocol stack.

[0097] Step S220: Establish a connection through a handshake mechanism;

[0098] At the network layer, communication between different networks is mainly achieved through the network layer controller. The functional implementation steps are as follows:

[0099] Step S310: The network layer controller receives the data packet and resolves the MAC address;

[0100] In step S320, if the MAC address of the data packet matches successfully, it is uploaded to the network protocol stack; otherwise, it is sent back to the management controller via the NCSI interface.

[0101] At the link layer, the link node controller primarily adapts to differences in physical layer hardware. The steps involved in implementing this functionality are as follows:

[0102] In step S410, the link node redirects the data packets;

[0103] Step S420: The redirected data packet is sent back to the management controller via the NCSI channel.

[0104] The management controller can query the status of each physical interface through NCSI:

[0105] Step S510: Configure the RMII interface required by NCSI for the BMC using the signal, clock, and rate conversion functions in the hardware. Port the NCSI protocol stack source code to adapt it to the underlying Hardware Abstraction Layer (HAL) code, including pin mapping, interrupt handling, cache refresh, and other interfaces;

[0106] Step S520 involves implementing the NIC management interface of the NCSI configuration method, which identifies, initializes, and configures the bus (such as SMBus and PCIe sideband channels) of BMC / NIC devices in the BMC platform. The NIC management interface supports multi-channel and multi-packet type (Control Packet, Pass-through Packet) processing; specifically, the NIC management interface is a means of identifying and processing network interface devices in the BMC kernel. In other words, it enables the BMC to identify, initialize, and configure the bus for the NIC's management interface.

[0107] Step S530 involves driver interface adaptation for the NCSI configuration method, establishing communication between the NCSI protocol stack and the Linux network subsystem (netdev) or character device interface ( / dev) to enable ioctl / sysfs control. Furthermore, user-mode state is exposed via netlink or the proc filesystem.

[0108] Step S540: NCS configuration method supports configuration commands, implementing NCSI standard commands such as Get Link Status, Set Link, EnableChannel, Select Package, Get Version ID, and Reset Channel. It also supports key control commands such as link query, channel selection, and power management.

[0109] Step S550: After completing the above steps, the management controller sets the MAC address in the network interface controller via NCSI.

[0110] In step S560, the management controller sends a message and polls at intervals of at least 20ms to ensure that the network interface controller completes the SMBUS master-slave-master switchover process and that the management controller receives a response.

[0111] In step S570, the management controller and the network interface controller respectively send requests and responses based on smbus block write with pec to establish a connection.

[0112] The beneficial effects of the network management system provided in this embodiment include:

[0113] (1) It reduces the difficulty of hardware design and improves the dynamic and flexible instruction allocation and management methods;

[0114] (2) The network port management system can establish a physical-level security mechanism between the network port and the BMC, with strong anti-tampering capabilities;

[0115] (3) A large amount of data can be polled between the network port controller and the management controller via NCSI, resulting in better performance, and the architecture is not limited by the use case.

[0116] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A network port management system for remote management of physical interfaces, characterized in that, The system includes: The network interface card (NIC) is equipped with a first link node controller, which is communicatively connected to at least one physical interface, the physical interface being used to connect to an external device. A network interface controller, which is communicatively connected to the network card; The second link node controller is located in the baseboard management controller, and the second link node controller is connected to the first link node controller via the NCSI protocol. A management controller is disposed in the baseboard management controller, and the management controller is communicatively connected to the second link node controller; the management controller remotely manages the at least one physical interface through the network interface controller based on the NCSI protocol of the first link node controller and the second link node controller. The process of configuring the second link node controller to communicate with the first link node controller via the NCSI protocol includes: The management controller generates the RMII interface required by NCSI based on the signal, clock, and rate conversion mechanism in the baseboard management controller, and performs NCSI protocol stack source code porting and NCSI protocol configuration according to the RMII interface. The management controller, based on the NCSI protocol of the first link node controller and the second link node controller, remotely manages the at least one physical interface through the network interface controller, including: The management controller sets a MAC address in the network interface controller through the NCSI protocol of the first link node controller and the second link node controller, and establishes a communication connection with the network interface controller based on the MAC address, so as to remotely manage the at least one physical interface through the network interface controller.

2. The network port management system for remote management of physical interfaces according to claim 1, characterized in that, The process of establishing a communication connection with the network interface controller includes: The management controller sends messages at preset time intervals to poll the network interface controller so that the SMBUS protocol master-slave-master switching process is completed. The management controller and the network interface controller respectively send and respond to requests based on the SMBUS protocol's block write with pec, and establish a communication connection.

3. The network port management system for remote management of physical interfaces according to claim 2, characterized in that, The NCSI protocol configuration includes at least: NCSI protocol network card management interface configuration, NCSI protocol driver interface configuration, and NCSI protocol command configuration.

4. The network port management system for remote management of physical interfaces according to claim 1, characterized in that: The management controller is also used to send request instructions to the network interface controller, receive judgment signals fed back by the network interface controller, send IP address configuration instructions to the network interface controller to complete the IP communication connection when the judgment signal is a safe information, and send a warning signal to the host computer connected to the management controller when the judgment signal is an unsafe information. The network interface controller is used to perform security checks on the request command and to send the check signal of the request back to the management controller.

5. A network port management system for remote management of physical interfaces according to claim 4, characterized in that, The system also includes: A logical communication controller is communicatively connected to the management controller. The logical communication controller encapsulates at least TCP, NETBIOS, and ARP protocols. The logical communication controller is used to encapsulate data packets according to the encapsulated protocols and establish a connection with the network card through a handshake mechanism.

6. A network port management system for remote management of physical interfaces according to claim 5, characterized in that, The network interface card is also equipped with a first network layer controller, which is communicatively connected to the first link node controller. The system further includes: a second network layer controller, which is communicatively connected to the second link node controller and the management controller; The first network layer controller is used to receive data packets sent by the second network layer controller, parse MAC addresses, configure the network protocol stack if the MAC address matches successfully, and send data packets to the second network layer controller through the NCSI protocol between the first link node controller and the second link node controller if the MAC address does not match successfully.

7. A network port management system for remote management of physical interfaces according to claim 6, characterized in that, The network card is also provided with a first buffer module, which is communicatively connected to the first network layer controller; The system also includes a second buffer module, which is communicatively connected to the second network layer controller. The first buffer module and the second buffer module are used for flow control between the first network layer controller and the second network layer controller.

8. A network port management system for remote management of physical interfaces according to claim 6, characterized in that, The first link node controller is used to redirect data packets and send the redirected data packets to the management controller via the NCSI protocol of the first link node controller and the second link node controller.

9. A network port management system for remote management of physical interfaces according to claim 6, characterized in that, The management controller, the first network layer controller, the second network layer controller, the first link node controller, and the second link node controller are all any one of hardware circuits, field-programmable gate logic, or software logic.

10. A network port management system for remote management of physical interfaces according to claim 1, characterized in that, The physical interface includes any one of the following: circuit wiring, hub, twisted pair cable, and network port.

Images