Data routing management method and system with multiple AXI main equipment core particle controllers

By setting up a dynamic mapping table inside the core controller and expanding the ID field, the problem of ID information loss during cross-layer access of the AXI bus was solved, and accurate remote device access was achieved.

CN121509327APending Publication Date: 2026-02-10BEIJING XINLI TECH INNOVATION CENT CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202511562020.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-10-29
Publication Date
2026-02-10

AI Technical Summary

Technical Problem

In existing technologies, the ID field of the AXI bus has insufficient bits when crossing multiple layers of chip controllers, resulting in the loss of ID information and making it impossible to accurately access remote devices.

Method used

A dynamic mapping table is set up inside the core controller to map the master device number and ID number to a new ID number. Cross-layer access is achieved by extending the ID field. The dynamic mapping table is used to record and look up the master device number and ID number.

Benefits of technology

It enables correct access to remote devices across multiple core controllers, solves the problem of insufficient ID field digits, and ensures the accuracy of data transmission.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN121509327A_ABST
    Figure CN121509327A_ABST
Patent Text Reader

Abstract

The invention provides a data routing management method and system with a plurality of AXI main equipment core particle controllers, which are used for realizing correct access of the core particle controllers which cross multiple layers and accord with the national standard GBT 46280 to far-end equipment, and the method comprises the following steps: step S1, numbering all main equipment on a current AXI bus in sequence, and configuring numbers to the core particle controllers; s2, a dynamic mapping table is arranged in the core particle controller, an ID field has 8 bits, the number of the ID fields is 256, and each item corresponds to an independent ID number; and S3, setting all entries to be in an invalid state through a starting system. By utilizing the method and the device, the problem that the ID field bits are not enough after a plurality of main devices are arranged on the AXI bus and cross the core particle controller is effectively solved, and core particle data of the plurality of AXI main devices are correctly transmitted to the remote equipment across the multi-layer core particle controller.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention belongs to the field of semiconductor manufacturing technology, and in particular relates to a data routing management method and system for having multiple AXI master chip controllers. Background Technology

[0002] The commonly used algorithm in the current technology is the dynamic mapping algorithm of the AXI bus NOC, but this algorithm cannot be applied to the chip interconnect ecosystem.

[0003] The AXI bus uses an ID field to bind data to addresses. If there are multiple master devices on the AXI bus, the chip developer assigns a sequence number to each master device and concatenates the sequence numbers corresponding to multiple requests issued by that master device into an ID, which the receiver (slave device) will see. If the receiver is a chip controller, this ID will be packaged into the data packet.

[0004] The ID field has 8 bits. Within the same AXI bus, the master device number and the device number can be simply concatenated, usually not exceeding 8 bits. However, when crossing controllers, 8 bits are insufficient. Summary of the Invention

[0005] The purpose of this invention is to solve the above-mentioned problems in the prior art, so as to enable accurate and orderly access to remote devices even across multiple core controllers.

[0006] This invention provides a data routing management method for multiple AXI master device core controllers, used to enable correct access to remote devices across multiple core controllers conforming to the national standard GBT 46280. The method includes: Step S1, sequentially numbering all master devices on the current AXI bus and configuring the numbers to the core controllers; Step S2, setting up a dynamic mapping table inside the core controller, with an 8-bit ID field and a total of 256 entries, each entry corresponding to a unique ID number; Step S3, setting all entries to an invalid state by starting the system.

[0007] Furthermore, preferably, the data routing management method of the present invention further includes: step S4, after the core controller receives a read / write request from a master device, it selects an invalid entry, records the master device number and corresponding ID number of the request, overwrites the ID number of the current read / write with the sequence number of the entry, and sends it to the peer of the core link.

[0008] Furthermore, preferably, in the data routing management method of the present invention, after the peer has completed reading and writing, it returns rd data or wrsp response and brings back the ID number.

[0009] Furthermore, preferably, in the data routing management method of the present invention, when the chip controller receives the returned rd data or wrsp response, it finds the master device number and ID number of this read / write operation by looking up the dynamic mapping table.

[0010] Furthermore, preferably, in the data routing management method of the present invention, this ID number is overwritten with the corresponding field in the rd data or wrsp, and the data is sent to the corresponding master device through the AXI bus.

[0011] Furthermore, preferably, the present invention also provides a data routing management system having multiple AXI master device core controllers.

[0012] The data routing management method described in this invention enables correct access to remote devices across multiple layers of core controllers conforming to the national standard GBT 46280, effectively solving the problem of insufficient ID field bits after crossing core controllers when there are multiple master devices on the AXI bus. Attached Figure Description

[0013] Figure 1 This diagram illustrates the problem of ID information loss when data flows across multiple chip controllers in data transmission based on the national standard GBT 46280 chip protocol in the prior art.

[0014] Figure 2 This is a schematic diagram illustrating a scenario in a practical application where the master device accesses a remote device across multiple layers of chip controllers.

[0015] Figure 3 This is a schematic diagram illustrating a data routing management method with multiple AXI master chip controllers according to a preferred embodiment of the present invention. Detailed Implementation

[0016] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Other embodiments or modifications obtained by those skilled in the art based on the embodiments of this application without creative effort are all within the scope of protection of this application.

[0017] The data routing management model addresses the following problems: The national standard GB / T 46280 chip protocol only allocates 8 bits to the ID field. If the data stream flows across multiple chip controllers, ID information will be lost. Figure 1 As shown. Figure 1This diagram illustrates the problem of ID information loss when data flows across multiple chip controllers in data transmission based on the national standard GBT46280 chip protocol in the prior art.

[0018] In practical applications, the master device may access remote devices by bypassing several layers of chip controllers, such as... Figure 2 As shown, vendor 5's PCIe may go through vendor 4's compute chip, then through vendor 3's cache chip, and finally access vendor 1's HBM memory chip. Figure 2 This is a schematic diagram illustrating a scenario in a practical application where the master device accesses a remote device across multiple layers of chip controllers.

[0019] The details are as follows: The data in the chip protocol packet of the national standard GBT 46280 is an AXI packet, which has 5 packet types: rd, wd, and wrsp do not have address information (this is different from the design of UCIe, where every packet in UCIe carries address information).

[0020] The AXI bus uses an ID field to bind data to addresses. If there are multiple master devices on the AXI bus, the chip developer assigns a sequence number to each master device and concatenates the sequence numbers corresponding to multiple requests issued by that master device into an ID, which the receiver (slave device) will see. If the receiver is a chip controller, this ID will be packaged into the data packet.

[0021] The ID field has 8 bits. Within the same AXI bus, the master device number and the device number can be simply concatenated, usually not exceeding 8 bits. However, when crossing controllers, 8 bits are insufficient.

[0022] The proposed solution is to establish a dynamic mapping table within the chip controller, mapping the master device number and ID number to a new ID number, thus overwriting the ID field. For example... Figure 3 As shown. Figure 3 This is a schematic diagram illustrating a data routing management method with multiple AXI master devices according to a preferred embodiment of the present invention.

[0023] According to a preferred embodiment of the present invention, the method includes at least: (1) The id field in the core protocol of GBT 46280 is extended to dynamically map the static global ID to the local ID inside the current bus.

[0024] (2) The core controller is designed with a dynamic mapping table and a lookup table module.

[0025] (3) When the data transmission is complete, the completion flag is determined by the last field, and the mapping table entry is released.

[0026] The specific operating procedure is as follows: Assign numbers to all master devices on the current AXI bus and configure them to the chip controller.

[0027] A dynamic mapping table is set up inside the core controller. Since the ID field has 8 bits, there are a total of 2^8==256 entries, each corresponding to a unique ID number.

[0028] When the system starts up, all entries are in an invalid state.

[0029] After receiving a read / write request from the master device, the core controller randomly selects one from the invalid entries, records the master device number and ID number of the current request, overwrites the ID number of the current read / write request with the entry's sequence number, and sends it to the peer of the core link.

[0030] After the peer completes the read / write operation, it returns rd data or a wrsp response, bringing back the ID.

[0031] When the chip controller receives the returned rd data or wrsp response, it finds the master device number and ID number of this read / write operation by looking up the dynamic mapping table.

[0032] Overwrite the corresponding field in the rd data or wrsp with this ID number, and send the data to the corresponding master device via the AXI bus.

[0033] Furthermore, the present invention is not limited to the above-described embodiments, and may also include other embodiments as follows.

[0034] According to a preferred embodiment of the present invention, a data routing management method with multiple AXI master device kernel controllers can be used to achieve correct access to remote devices across multiple kernel controllers. The method includes: step S1, sequentially numbering all master devices on the current AXI bus and configuring the numbers to the kernel controllers; step S2, setting up a dynamic mapping table inside the kernel controller, with an 8-bit ID field and a total of 256 entries, each entry corresponding to an independent ID number; step S3, setting all entries to an invalid state by starting the system.

[0035] Furthermore, according to a preferred embodiment of the present invention, the data routing management method further includes: step S4, after the core controller receives a read / write request from a master device, it selects an invalid entry, records the master device number and the corresponding ID number of the request, overwrites the ID number of the current read / write with the sequence number of the entry, and sends it to the peer of the core link.

[0036] Furthermore, according to a preferred embodiment of the present invention, after the peer has completed the read / write operation, it returns rd data or a wrsp response, bringing back the ID number.

[0037] Furthermore, according to a preferred embodiment of the present invention, when the chip controller receives the returned rd data or wrsp response, it finds the master device number and ID number of this read / write operation by looking up the dynamic mapping table.

[0038] Furthermore, according to a preferred embodiment of the present invention, this ID number is overwritten with the corresponding field in the rd data or wrsp, and the data is sent to the corresponding master device via the AXI bus.

[0039] Furthermore, according to a preferred embodiment of the present invention, a data routing management system with multiple AXI master device kernel controllers is also provided, which utilizes the data routing management method described in the present invention to achieve correct access to remote devices across multiple kernel controllers.

[0040] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0041] The various embodiments in this specification are described in a related manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0042] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

[0043] The above embodiments of the present invention are merely examples for clearly illustrating the present invention and are not intended to limit the implementation of the present invention. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively describe all possible implementations here. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the claims of the present invention.

Claims

1. A data routing management method for multiple AXI master device chip controllers, used to enable correct access to remote devices across multiple layers of chip controllers conforming to the national standard GBT 46280, characterized in that, include: Step S1: Number all master devices on the current AXI bus sequentially and configure the numbers to the chip controller; Step S2: Set up a dynamic mapping table inside the core controller. The ID field has 8 digits and a total of 256 entries. Each entry corresponds to a unique ID number. Step S3: Start the system to invalidate all entries.

2. The data routing management method according to claim 1, characterized in that, Also includes: Step S4: After the core controller receives a read / write request from a master device, it selects an invalid entry, records the master device number and corresponding ID number of the request, overwrites the ID number of the current read / write with the sequence number of the entry, and sends it to the peer of the core link.

3. The data routing management method according to claim 2, characterized in that, After the peer completes the read / write operation, it returns rd data or a wrsp response, along with the ID number.

4. The data routing management method according to claim 3, characterized in that, When the chip controller receives the returned rd data or wrsp response, it finds the master device number and ID number of this read / write operation by looking up the dynamic mapping table.

5. The data routing management method according to claim 4, characterized in that, Overwrite the corresponding field in the rd data or wrsp with this ID number, and send the data to the corresponding master device via the AXI bus.

6. A core data routing management system with multiple AXI master core controllers, which utilizes the data routing management method according to any one of claims 1 to 5 to achieve correct access to remote devices across multiple core controllers.