CN106330776A - Message processing method and device - Google Patents

Abstract

The invention provides a message processing method and device. Wherein, the method includes: creating a fast channel buffer between the user state application program and the network device driver; judging whether the message to be transmitted belongs to a preset message type; In this case, the to-be-transmitted message is transmitted between the user mode application program and the network device driver through the fast channel buffer. The invention solves the problem that the message cannot be processed rapidly through the socket interface and the network protocol stack, and further achieves the effect of rapidly processing the message.

Description

Message processing method and device

technical field

The present invention relates to the communication field, in particular to a method and device for processing a message.

Background technique

As a free and open-source operating system, Linux is popular among communication equipment manufacturers for its powerful network processing capability and stability. Now most network devices are developed based on the Linux operating system.

The Transmission Control Protocol (TCP)/Internet Protocol (IP) system is the current Internet communication protocol standard, which is fully realized in the Linux network kernel. The system is divided into four layers: application layer, transport layer, network layer, network medium layer. The Linux network kernel realizes the network interconnection function with reference to the network protocol system, including the following parts:

1. Socket interface, the top layer of the network kernel supports the interface of application program development, and the socket interface is a series of standard functions. The socket supports many different types of protocol families: UNIX domain protocol family, TCP/IP protocol family, Internet packet exchange (IPX) protocol family, among which the TCP/IP protocol family is the most widely used;

2. Transport layer and network layer, the transport layer includes standard TCP and user datagram (UDP) protocol modules, and the network layer includes standard IP protocol modules;

3. Data link layer. For networks that require logical links, the data link layer provides independent logical link protocol modules, such as Point-to-Point Protocol (PPP), Serial Line Internet Protocol (SLIP), and so on. For Ethernet, this layer is relatively simple, and the main Ethernet protocol implementation is integrated into the underlying network card driver;

4. Network device drivers, due to differences in physical characteristics, different network devices use different device drivers.

The inventors of the present invention found in the research process that the user-mode application program needs to open the network device and bind the socket to read the data message. Data packets are driven by network devices to the kernel network protocol stack, and then to user mode applications. The processing process is complicated. At the same time, the kernel mode to user mode space needs to go through a memory copy, which is not suitable for occasions with high performance requirements.

Aiming at the problem that the related technology cannot quickly process the message through the socket interface and the network protocol stack, no effective solution has been proposed yet.

Contents of the invention

The invention provides a message processing method and device to at least solve the problem in the related art that the message cannot be processed quickly through the socket interface and the network protocol stack.

According to one aspect of the present invention, a message processing method is provided, including: creating a fast channel buffer between the user mode application program and the network device driver; judging whether the message to be transmitted belongs to the preset message type; When the message to be transmitted belongs to the preset message type, the message to be transmitted is transmitted between the user state application program and the network device driver through the fast channel buffer.

Preferably, creating the fast channel buffer between the user state application and the network device driver includes: creating and registering a fast channel character node; through the fast channel character node, the user state application creates a shared queue through a system call, The fast channel buffer is obtained; the user mode application program sets the preset message type through the fast channel character node.

Preferably, when it is determined that the message to be transmitted belongs to the preset message type, the message to be transmitted is transmitted between the user mode application program and the network device driver through the fast channel buffer Before the text, the method also includes: judging whether the fast channel buffer is full; wherein, when it is judged that the fast channel buffer is not full, through the fast channel buffer, the user state application The message to be transmitted is transmitted with the network device driver.

Preferably, when it is determined that the fast channel buffer is full, the method further includes: discarding the message to be transmitted.

Preferably, when it is determined that the message to be transmitted does not belong to the preset message type, the method further includes: transmitting the message to be transmitted between the user state application program and the network device driver through the network protocol stack transmit message.

Preferably, transmitting the message to be transmitted between the user mode application program and the network device driver through the network protocol stack includes: storing the message to be transmitted in the network protocol stack; The network protocol stack reads the message to be transmitted.

According to another aspect of the present invention, a message processing device is also provided, including: a creation module, used to create a fast channel buffer between the user state application program and the network device driver; a first judging module, used to judge Whether the message to be transmitted belongs to the preset message type; the first transmission module is configured to pass through the fast channel buffer and send the message to the user through the fast channel buffer when it is determined that the message to be transmitted belongs to the preset message type. The message to be transmitted is transmitted between the state application program and the network device driver.

Preferably, the device further includes: a second judging module, configured to judge whether the fast-track buffer is full; wherein, when it is judged that the fast-track buffer is not full, the fast-track buffer area, and transmit the message to be transmitted between the user mode application program and the network device driver.

Preferably, the device further includes: a discarding module, configured to discard the message to be transmitted when it is judged that the fast channel buffer is full.

Preferably, the device further includes: a second transmission module, configured to communicate between the user mode application program and the network device through the network protocol stack when it is determined that the message to be transmitted does not belong to the preset message type The to-be-transmitted message is transmitted between drivers.

Through the present invention, a fast channel buffer is created between the user mode application program and the network device driver; it is judged whether the message to be transmitted belongs to the preset message type; when it is judged that the message to be transmitted belongs to the preset message type , through the fast channel buffer, the method of transmitting the message to be transmitted between the user state application program and the network device driver solves the problem that the message cannot be processed quickly through the socket interface and the network protocol stack, and thus achieves fast The effect of packet processing.

Description of drawings

The accompanying drawings described here are used to provide a further understanding of the present invention and constitute a part of the application. The schematic embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations to the present invention. In the attached picture:

Fig. 1 is a flowchart of a message processing method according to an embodiment of the present invention;

Fig. 2 is a structural block diagram of a message processing device according to an embodiment of the present invention;

Fig. 3 is a preferred structural block diagram 1 of a message processing device according to an embodiment of the present invention;

FIG. 4 is a second preferred structural block diagram of a message processing device according to an embodiment of the present invention;

FIG. 5 is a third preferred structural block diagram of a message processing device according to an embodiment of the present invention;

Fig. 6 is a structural block diagram of a fast message processing device according to a preferred embodiment of the present invention;

Fig. 7 is a flow chart of fast channel initialization according to a preferred embodiment of the present invention;

Fig. 8 is a block diagram of the data interaction structure between the kernel and the user mode shared queue according to a preferred embodiment of the present invention;

Fig. 9 is a flowchart of a fast message processing method according to a preferred embodiment of the present invention.

detailed description

Hereinafter, the present invention will be described in detail with reference to the drawings and examples. It should be noted that, in the case of no conflict, the embodiments in the present application and the features in the embodiments can be combined with each other.

It should be noted that the terms "first" and "second" in the description and claims of the present invention and the above drawings are used to distinguish similar objects, but not necessarily used to describe a specific sequence or sequence.

A method for processing a message is provided in this embodiment. FIG. 1 is a flowchart of a method for processing a message according to an embodiment of the present invention. As shown in FIG. 1 , the process includes the following steps:

Step S102, creating a fast channel buffer between the user mode application program and the network device driver;

Step S104, judging whether the message to be transmitted belongs to a preset message type;

Step S106, when it is determined that the message to be transmitted belongs to a preset message type, transmit the message to be transmitted between the user-mode application program and the network device driver through the fast channel buffer.

Through the above steps, according to the message type of the message to be transmitted, for the message to be transmitted belonging to the preset message type, the message is transmitted through the created fast channel buffer, so that the message to be transmitted belonging to the preset message type The message does not need to go through the complicated processing of the network protocol stack and the socket interface, and the rapid transmission of the preset type of message can be realized according to the need. It can be seen that by adopting the above steps, the problem that the packet cannot be processed quickly through the socket interface and the network protocol stack is solved, and the effect of fast packet processing is achieved.

Preferably, the preset message type may be a fragmented and reassembled message.

Preferably, in step S102, by creating and registering the fast channel character node; through the fast channel character node, the user mode application creates a shared queue through a system call to obtain the fast channel buffer; through the fast channel character node, the user mode application The program sets the preset message type, and creates a fast channel buffer between the user mode application program and the network device driver.

Preferably, the fast channel buffer may be a ring buffer, which can repeatedly use a fixed-size memory space, and information coordination and synchronization are required for writing and reading.

Preferably, the network device driver can connect the receiving function of the data message to the callback function of fast channel message processing, and judge the preset message type through the callback function; the user state application program opens the fast channel device node and creates a shared queue. The user-mode application program sets the type of fast packets to be processed, obtains the physical address of the queue through system calls, and maps it to the user-mode application program in the user-mode address space.

Preferably, when it is judged that the message to be transmitted belongs to the preset message type, before step S104, it can also be judged whether the fast channel buffer is full; wherein, when it is judged that the fast channel buffer is not full , transmit the message to be transmitted between the user mode application program and the network device driver through the fast channel buffer.

Preferably, when it is determined that the fast channel buffer is full, the message to be transmitted may be discarded.

Preferably, when it is determined that the fast channel buffer is full, the message to be transmitted may also be sent to the network protocol stack, and the network protocol stack processes the message.

Preferably, when it is determined that the message to be transmitted does not belong to the preset message type, the message to be transmitted can be transmitted between the user mode application program and the network device driver through the network protocol stack, for example, through the network protocol stack In the case of transmitting packets to be transmitted between the user-mode application and the network device driver: the pending transmission messages sent by the network device driver to the user-mode application can be stored in the network protocol stack, and the user-mode application can pass through the network device The socket interface bound to the driver reads the message to be transmitted from the network protocol stack; correspondingly, the message to be transmitted sent by the user mode application to the network device driver is also transmitted through the socket interface and the network protocol stack.

Through the description of the above embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be implemented by means of software plus a necessary general-purpose hardware platform, and of course also by hardware, but in many cases the former is better implementation. Based on such an understanding, the essence of the technical solution of the present invention or the part that contributes to the prior art can be embodied in the form of software products, and the computer software products are stored in a storage medium (such as ROM/RAM, disk, CD) contains several instructions to enable a terminal device (which may be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in various embodiments of the present invention.

This embodiment also provides a packet processing device, which is used to implement the above embodiments and preferred implementation manners, and what has been described will not be repeated here. As used below, the term "module" may be a combination of software and/or hardware that realizes a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementations in hardware, or a combination of software and hardware are also possible and contemplated.

Fig. 2 is a structural block diagram of a message processing device according to an embodiment of the present invention. As shown in Fig. 2, the device includes; a creation module 22, a first judgment module 24, and a first transmission module 26, wherein the creation module 22 uses Create a fast channel buffer between the user state application program and the network device driver; the first judgment module 24 is coupled to the creation module 22, and is used to judge whether the message to be transmitted belongs to the preset message type; the first transmission module 26 , coupled to the first judging module 24, configured to transmit the message to be transmitted between the user mode application program and the network device driver through the fast channel buffer when it is judged that the message to be transmitted belongs to a preset message type .

Preferably, the above-mentioned creating module 22 is used to create and register a fast channel character node; through the fast channel character node, the user mode application creates a shared queue through a system call to obtain a fast channel buffer; through the fast channel character node, the user mode application Set the default packet type.

FIG. 3 is a preferred structural block diagram 1 of a message processing device according to an embodiment of the present invention. As shown in FIG. time, used to determine whether the fast channel buffer is full; wherein, when it is judged that the fast channel buffer is not full, transmit the message to be transmitted between the user state application program and the network device driver through the fast channel buffer .

FIG. 4 is a second preferred structural block diagram of a message processing device according to an embodiment of the present invention. As shown in FIG. 4 , the device further includes: a discarding module 42 coupled to the second judging module 32 for discarding messages to be transmitted.

FIG. 5 is a third preferred structural block diagram of a message processing device according to an embodiment of the present invention. As shown in FIG. Transmit messages to be transmitted between the user mode application program and the network device driver.

Preferably, the second transmission module 52 can be used to store the message to be transmitted in the network protocol stack; read the message to be transmitted from the network protocol stack through the socket interface.

It should be noted that each of the above-mentioned modules can be implemented by software or hardware. For the latter, it can be implemented in the following manner, but not limited to this: the above-mentioned modules are all located in the same processor; or, the above-mentioned modules are respectively located in multiple in the processor.

Embodiments of the present invention also provide software, which is used to implement the technical solutions described in the above embodiments and preferred implementation manners.

The embodiment of the invention also provides a storage medium. In this embodiment, the above-mentioned storage medium may be configured to store program codes for performing the following steps:

S20, creating a fast channel buffer between the user mode application program and the network device driver;

S40, judging whether the message to be transmitted belongs to the preset message type;

S60. When it is determined that the message to be transmitted belongs to the preset message type, transmit the message to be transmitted between the application program in the user state and the network device driver through the fast channel buffer.

Optionally, the storage medium is also configured to store program codes for performing the following steps:

S22, creating and registering a fast channel character node;

S24, through the fast channel character node, the user mode application creates a shared queue through a system call to obtain a fast channel buffer;

S26, through the character node of the fast channel, the user mode application program sets a preset message type.

Optionally, the storage medium is also configured to store program codes for performing the following steps:

S62. Determine whether the fast channel buffer is full.

Optionally, the storage medium is also configured to store program codes for performing the following steps:

S622. Discard the message to be transmitted when it is determined that the fast channel buffer is not full.

Optionally, the storage medium is also configured to store program codes for performing the following steps:

S50. When it is determined that the message to be transmitted does not belong to the preset message type, transmit the message to be transmitted between the user state application program and the network device driver through the network protocol stack.

Optionally, the storage medium is also configured to store program codes for performing the following steps:

S52, storing the message to be transmitted in the network protocol stack;

S54. Read the message to be transmitted from the network protocol stack through the socket interface.

Optionally, in this embodiment, the above-mentioned storage medium may include but not limited to: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk Various media that can store program codes such as discs or optical discs.

In order to make the description of the embodiments of the present invention clearer, the following describes and illustrates in conjunction with preferred embodiments.

The preferred embodiment of the present invention provides a fast message processing method based on the Linux operating system to solve the problem that the Linux kernel protocol stack cannot efficiently process messages with high network card performance requirements, and at the same time give full play to the functions of the Linux protocol stack.

A kind of fast message processing scheme based on Linux operating system that preferred embodiment of the present invention provides, comprises following module:

The network device driver module is used for detecting network devices, receiving and sending data packets.

Linux network protocol stack, a standard module provided by Linux, used for network packet processing.

The fast channel module is used for creating a shared queue (equivalent to the above-mentioned fast channel buffer), registering for a fast channel, and receiving and processing messages.

The user mode receiving module is used to open the shared queue, map the shared queue to the user mode space, and receive and process the message.

A kind of fast message processing method based on Linux that the preferred embodiment of the present invention provides, comprises the following steps:

Step 1, the network device driver registers the fast message processing module during detection;

Step 2, the network device is driven, and the data message receiving function is connected to the callback function of fast channel message processing;

Step 3, the user mode application, open the fast channel device node, create a shared queue, set the fast message type to be processed (equivalent to the above preset message type), obtain the physical address of the queue through the system call, and map it to the user mode address space;

Step 4: After initialization according to the above steps, the data message received by the network device driver module can be processed normally, and the subsequent processing method is determined according to the set message type;

Step 5, for the fast message, store it directly in the shared queue of the fast channel module, and the application program obtains the data message by mapping the shared memory; for the unmatched message, all are sent to the Linux network protocol stack, and the application program The message needs to be read through the socket interface.

The preferred embodiments of the present invention will be described below in conjunction with the accompanying drawings.

In order to solve the problem of low performance of the Linux network protocol stack for processing high-speed messages, the preferred embodiment of the present invention provides a method for quickly processing messages, and at the same time, the Linux network protocol stack can be used for ordinary messages.

Fig. 6 is the structural block diagram of the fast message processing device according to the preferred embodiment of the present invention, as shown in Fig. 6, this device comprises: the fast path module created in Linux kernel state, is coupled between the application program and the network device driver module , used for high-speed message processing, the fast message of the application program accesses the message of the fast channel module by mapping the shared memory; the socket interface is coupled to the application program and the Linux network protocol stack, and is used for the ordinary message to pass through The Linux network protocol stack is accessed through the network socket; the Linux network protocol stack is coupled between the socket interface and the network device driver module, and is used for storing and processing ordinary messages.

In a preferred embodiment of the present invention, the fast message that needs to be processed can be a fragment reassembly message, and the common message can be other messages except the fragment reassembly message, such as Address Resolution Protocol (ARP) request message, Internet Protocol version 4 (IPv4) packets.

In a preferred embodiment of the present invention, the fast channel module is used to provide an interface for creating a shared queue, obtain a shared memory address, receive and process messages, and manage the shared queue.

Fig. 7 is a flow chart of fast channel initialization according to a preferred embodiment of the present invention. As shown in Fig. 7, the process includes the following steps:

Step S702, network device driver detection, driver loading;

Step S704, creating a fast channel character device, registering a device node, providing a system call interface, including creating a shared queue, obtaining a mapping relationship, and setting a fast message type;

Step S706, the message receiving and processing function driven by the network device is connected to the receiving function of the fast channel message;

Step S708, the Linux kernel is started, and the fast channel module is loaded;

In step S710, the user program creates a fast channel buffer through a system call, and sets a message type of the fast channel (equivalent to the aforementioned preset message type).

Preferably, the above-mentioned shared queue is a first-in-first-out circular queue, and Fig. 8 is a block diagram of the data interaction structure of the kernel and the user-mode shared queue according to a preferred embodiment of the present invention. As shown in Fig. 8, the circular queue does not need dynamic memory Freeing and allocating, the fixed-size memory space can be used repeatedly; the fast channel module inserts the message into the queue as the writing module; the user program pops the message out of the queue as the reading module; the writing module and the reading module need to coordinate information and synchronization; for multi-thread and multi-process writing or reading modules, critical section processing is required between writing modules and between reading modules.

In the actual queue insertion and pop-up operations, they are continuously interleaved. When inserting (push) operations, the queue head (head) will increase; while popping (pop) operations, the queue tail (tail) will increase. When the push speed is fast, it will catch up with the tail. At this time, it means that the queue is full and no more push operations can be performed. It is necessary to wait for the pop operation to free up space in the queue. When the pop operation is fast and the tail catches up with the head, it means that the queue is empty and the pop operation can no longer be performed, and it is necessary to wait for the push data to come in.

Fig. 9 is a flowchart of a fast message processing method according to a preferred embodiment of the present invention. As shown in Fig. 9, the process includes the following steps:

Step S902, the network card device driver obtains the data message from the hardware device;

Step S904, analyze the layer 2 header of the data message, if it matches the keyword of the fast channel (that is, judge whether it belongs to the preset message type), then execute step S906, enter the fast channel processing flow, otherwise, go to step S914 , the message enters the Linux network protocol stack for processing;

Step S906, judging the read/write pointer of the management node of the queue. If the ring queues are connected head to tail and the queue is full, then step S908 is executed, otherwise step S910 is executed;

Step S908, process the message that the queue is full, and directly discard the message;

Step S910, storing the message in the ring queue;

Step S912, adjusting the write pointer of the ring queue, and executing the next free buffer;

Step S914, processing ordinary packets, there is no requirement for high performance, and the data packets are sent to the Linux network protocol stack for processing.

In summary, through the above-mentioned embodiments and preferred embodiments of the present invention, a new message processing mode has been added, and the fast channel buffer can be used to selectively bypass the Linux network protocol stack, simplifying the process of receiving messages , reduce the loss of central processing unit (Central Processing Unit, referred to as CPU) resources, greatly improve the receiving performance, can meet the performance requirements of 10 Gigabit Ethernet, and at the same time take into account the powerful functions of the network protocol stack, so that ordinary messages can Processed directly through the Linux network protocol stack.

Obviously, those skilled in the art should understand that each module or each step of the above-mentioned present invention can be realized by a general-purpose computing device, and they can be concentrated on a single computing device, or distributed in a network formed by multiple computing devices Alternatively, they may be implemented in program code executable by a computing device so that they may be stored in a storage device to be executed by a computing device, and in some cases, in an order different from that shown here The steps shown or described are carried out, or they are separately fabricated into individual integrated circuit modules, or multiple modules or steps among them are fabricated into a single integrated circuit module for implementation. As such, the present invention is not limited to any specific combination of hardware and software.

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and changes. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims (10)

1. a message processing method, it is characterised in that including:

Express passway relief area is created between User space application program and network device driver；

Judge whether message to be transmitted belongs to default type of message；

Determining in the case of described message to be transmitted belongs to described default type of message, by described express passway Relief area, transmits described message to be transmitted between User space application program and network device driver.

Method the most according to claim 1, it is characterised in that between User space application program and network device driver Create described express passway relief area to include:

Create and register express passway character nodes；

By described express passway character nodes, User space application program calls establishment shared queue by system, To described express passway relief area；

By described express passway character nodes, User space application program arranges described default type of message.

Method the most according to claim 1, it is characterised in that belong to described default determining described message to be transmitted In the case of type of message, by described express passway relief area, drive with the network equipment at User space application program Before transmitting described message to be transmitted between Dong, described method also includes:

Judge that described express passway relief area is the fullest；

Wherein, determine described express passway relief area less than in the case of, by described express passway relief area, Described message to be transmitted is transmitted between User space application program and network device driver.

Method the most according to claim 3, it is characterised in that determining the feelings that described express passway relief area is the fullest Under condition, described method also includes:

Abandon described message to be transmitted.

Method the most according to any one of claim 1 to 4, it is characterised in that determining described message to be transmitted In the case of being not belonging to described default type of message, described method also includes:

Between User space application program and network device driver, described message to be transmitted is transmitted by network protocol stack.

Method the most according to claim 5, it is characterised in that by described network protocol stack at User space application program And transmit described message to be transmitted between network device driver to include:

By described packet storage to be transmitted to described network protocol stack；

Described message to be transmitted is read from described network protocol stack by sockets interface.

7. a message process device, it is characterised in that including:

Creation module, for creating express passway relief area between User space application program and network device driver；

First judge module, is used for judging whether message to be transmitted belongs to default type of message；

First transport module, for determining in the case of described message to be transmitted belongs to described default type of message, By described express passway relief area, transmit described to be transmitted between User space application program and network device driver Message.

Device the most according to claim 7, it is characterised in that described device also includes:

Second judge module, is used for judging that described express passway relief area is the fullest；

Wherein, determine described express passway relief area less than in the case of, by described express passway relief area, Described message to be transmitted is transmitted between User space application program and network device driver.

Device the most according to claim 8, it is characterised in that described device also includes:

Discard module, for determine described express passway relief area full in the case of, abandon described to be transmitted Message.

10. according to the device according to any one of claim 7 to 9, it is characterised in that described device also includes:

Second transport module, for determining described message to be transmitted and be not belonging to the situation of described default type of message Under, between User space application program and network device driver, transmit described message to be transmitted by network protocol stack.