Method and device for processing network packets

Abstract

The invention is applicable to the technical field of network equipment, and relates to a network message processing method and equipment. A network message processing method, comprising a network message receiving step and a network message sending step, the network message receiving step comprising: the network card driver module receives the network message from the DMA; when the message type of the network message has a registration process, Parsing and removing the vendor identification in the network message; sending the analysis result of the vendor identification of the network message and the network message from which the vendor identification has been removed to the user layer. The invention creates a channel independent of the network protocol stack of the operating system, which will not have any impact on the software development of the third layer or does not pay attention to the bottom layer information; it is especially suitable for the network message with the identification of the chip manufacturer.

Description

technical field [0001] The invention is applicable to the technical field of network equipment, and in particular relates to a network message processing method and equipment. Background technique [0002] At this stage, similar to the development process of Layer 2 or Layer 3 managed switches on the Linux platform, we often encounter a problem: for some Layer 2 network protocols, such as RSTP (rapid spanning Tree Protocol, Rapid Spanning Tree Protocol), or GVRP (Generic Attribute Registration Protocol, Generic Attribute Registration Protocol) and other developments, how the upper-layer software knows the underlying information such as the receiving port on the hardware of the protocol message; some chip manufacturers give a solution: Let the switch chip modify and send it to the CPU (Central Processing Unit, central processing unit) message structure, insert several bytes of manufacturer identification in front of the message type field to contain the underlying information...

AI Technical Summary

Problems solved by technology

However, this method also has certain problems: because the information after the removal of the chip manufacturer's logo is stored in the SKB structure, upper-layer applications can only obtain this part of information through some specific methods or custom function interfaces. These function interfaces require developers to define or modify the relevant network protocol stack processing parts in the kernel to be feasible, which greatly increases the difficulty of protocol porting in the future.

[0006] As mentioned above, the existing two-layer protocol development has the following main problems in the processing method of the message with the manufacturer's logo: 1) The attached manufacturer's logo changes the basic structure of the network message, causing the system network protocol stack to be unable to identify normally and cannot be used. The standard message processing method is used for sending and receiving processing, which increases the development difficulty and workload of the upper-layer application; 2) The way for the upper-layer application to obtain the underlying information provided by the manufacturer's logo is complicated, and the addition and modification of some kernel function interfaces require relatively large High technical level and experience are not conducive to the transplantation of applications and protocols between projects, resulting in a decrease in code versatility

Images