A method, device, chip and storage medium for determining routing information

By performing router information determination and port arbitration in parallel in the on-chip network, the problem of large access delay of Master nodes to Slave nodes in the prior art is solved, and access efficiency is improved.

CN116248574BActive Publication Date: 2024-12-06SHANGHAI SUIYUAN TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202211573729.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-12-08
Publication Date
2024-12-06
Estimated Expiration
2042-12-08

AI Technical Summary

Technical Problem

In the prior art, routers in the on-chip network use serial methods to determine routing information and port arbitration, resulting in a large delay in access to the Slave node by the Master node and low efficiency.

Method used

The initial router receives the access request and initial routing information of the Master node, performs input port arbitration and determines the lower-level routing information of the lower-level router. If the lower-level router is the target router, the input port arbitration is performed through the lower-level router and the access request is sent to the Slave node. Implement parallel execution of the router information determination process.

Benefits of technology

It reduces the time-consuming access of Master nodes to Slave nodes, improves access efficiency, and reduces access latency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN116248574B_ABST
    Figure CN116248574B_ABST
Patent Text Reader

Abstract

The present invention discloses a method, device, chip and storage medium for determining routing information, including: receiving an access request and initial routing information triggered by a Master node through a target input port of an initial router; performing input port arbitration according to the initial routing information through the initial router, and simultaneously determining the lower routing information corresponding to the lower router; sending the lower routing information and the access request to the lower router through the initial router according to the input port arbitration result; if the lower router is determined to be the target router corresponding to the access request, performing input port arbitration according to the lower routing information through the lower router, and sending the access request to the target Slave node according to the input port arbitration result. The technical solution of the embodiment of the present invention can reduce the time consumed by the Master node to access the Slave node and improve the access efficiency.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a method, device, chip and storage medium for determining routing information. Background Art

[0002] The System On Chip (SOC) includes a Network On Chip (NOC). The NOC network can connect multiple Master nodes and Slave nodes, and realize information interaction between the Master nodes and the Slave nodes through multiple routers.

[0003] In the prior art, when a Master node accesses a Slave node, it needs to send an access request to the corresponding router, which calculates the routing information corresponding to the router based on the access request, then arbitrates the input port based on the routing information, and finally sends the access request to the subsequent router based on the arbitration result. The subsequent router determines the final router corresponding to the Slave node by repeatedly executing the above process, and sends the access request to the Slave node through the router.

[0004] However, since the router in the prior art uses a serial method to execute the routing information determination process and the port arbitration process, the access process of the Master node to the Slave node is delayed significantly. Summary of the invention

[0005] The present invention provides a method, device, chip and storage medium for determining routing information, which can reduce the time consumed by a Master node to access a Slave node and improve access efficiency.

[0006] According to one aspect of the present invention, a method for determining routing information is provided, which is applied to a chip, wherein routers in the chip are connected via an on-chip network, and the method comprises:

[0007] Receive the access request triggered by the Master node and the initial routing information corresponding to the initial router through the target input port of the initial router;

[0008] By means of the initial router, input port arbitration is performed according to the initial routing information, and the lower layer routing information corresponding to the lower layer router is determined at the same time;

[0009] Sending the lower-layer routing information and the access request to the lower-layer router through the initial router according to the input port arbitration result;

[0010] If it is determined that the lower layer router is the target router corresponding to the access request, the lower layer router performs input port arbitration according to the lower layer routing information, and sends the access request to the target Slave node according to the input port arbitration result.

[0011] Optionally, performing input port arbitration according to the initial routing information by the initial router includes:

[0012] Determine, by the initial router, a target output port according to the initial routing information, and obtain a plurality of input ports corresponding to the target output port;

[0013] The initial router arbitrates the multiple input ports to obtain execution orders corresponding to the input ports.

[0014] Optionally, sending the lower-layer routing information and the access request to the lower-layer router through the initial router according to the input port arbitration result includes:

[0015] The lower-layer routing information and the access request are sent to the lower-layer router through the target output port of the initial router according to the execution order corresponding to the target input port.

[0016] Optionally, determining the lower-layer routing information corresponding to the lower-layer router includes:

[0017] Determine, through the initial router, a lower layer router and a target router according to a target Slave node corresponding to the access request and a composition structure of the on-chip network;

[0018] The lower-layer routing information corresponding to the lower-layer router is determined through the initial router according to the connection modes corresponding to the lower-layer router and the target router.

[0019] Optionally, after the initial router sends the lower-layer routing information and the access request to the lower-layer router according to the input port arbitration result, the method further includes:

[0020] If it is determined that the lower layer router is not the target router corresponding to the access request, then performing input port arbitration according to the lower layer routing information through the lower layer router, and simultaneously determining subsequent routing information corresponding to the subsequent cascade router;

[0021] Sending the subsequent routing information and the access request to the subsequent cascade router through the lower router according to the input port arbitration result;

[0022] If it is determined that the subsequent cascade router is the target router corresponding to the access request, the subsequent cascade router performs input port arbitration according to the subsequent routing information, and sends the access request to the target Slave node according to the input port arbitration result.

[0023] Optionally, after sending the access request to the target slave node, the process further includes:

[0024] Determine the target routing information corresponding to the target router through the target Slave node, and send the target routing information and the response data corresponding to the access request to the target router;

[0025] Through the target router, perform input port arbitration according to the target routing information, and simultaneously determine the upper-layer routing information corresponding to the upper-layer router;

[0026] Sending the upper-layer routing information and response data to the upper-layer router through the target router according to the input port arbitration result;

[0027] If it is determined that the upper-layer router is the initial router, the upper-layer router performs input port arbitration according to the upper-layer routing information, and sends the response data to the Master node according to the input port arbitration result.

[0028] According to another aspect of the present invention, a routing information determination device is provided, which is applied to a chip, wherein routers in the chip are connected via an on-chip network, and the device comprises:

[0029] A request receiving module, used to receive the access request triggered by the Master node and the initial routing information corresponding to the initial router through the target input port of the initial router;

[0030] An information determination module, used to perform input port arbitration according to the initial routing information through the initial router, and simultaneously determine the lower-layer routing information corresponding to the lower-layer router;

[0031] An information sending module, used for sending the lower layer routing information and the access request to the lower layer router through the initial router according to the input port arbitration result;

[0032] The request sending module is used to perform input port arbitration according to the lower-layer routing information through the lower-layer router if it is determined that the lower-layer router is the target router corresponding to the access request, and send the access request to the target Slave node according to the input port arbitration result.

[0033] According to another aspect of the present invention, a chip is provided, the chip comprising:

[0034] at least one processor; and

[0035] a memory communicatively connected to the at least one processor; wherein,

[0036] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the method for determining routing information according to any embodiment of the present invention.

[0037] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the method for determining routing information according to any embodiment of the present invention when executed.

[0038] The technical solution provided by the embodiment of the present invention receives the access request and initial routing information triggered by the Master node through the target input port of the initial router, performs input port arbitration according to the initial routing information through the initial router, and simultaneously determines the lower routing information of the lower router, and sends the lower routing information and the access request to the lower router according to the input port arbitration result through the initial router. If the lower router is determined to be the target router corresponding to the access request, the lower router performs input port arbitration according to the lower routing information, and sends the access request to the target Slave node according to the input port arbitration result. This technical solution can reduce the time consumed by the Master node to access the Slave node and improve the access efficiency.

[0039] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0041] Figure 1a is a flow chart of a method for determining routing information provided according to an embodiment of the present invention;

[0042] Figure 1b is a schematic structural diagram of a router provided according to an embodiment of the present invention;

[0043] Figure 1c is a schematic diagram of the structure of a network on chip provided according to an embodiment of the present invention;

[0044] Figure 2 is a flow chart of another method for determining routing information provided according to an embodiment of the present invention;

[0045] Figure 3 is a flow chart of another method for determining routing information provided according to an embodiment of the present invention;

[0046] Figure 4 is a schematic diagram of the structure of a routing information determination device provided according to an embodiment of the present invention;

[0047] Figure 5 It is a schematic diagram of the chip structure for implementing the routing information determination method of the embodiment of the present invention. DETAILED DESCRIPTION

[0048] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0049] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0050] Figure 1a This is a flowchart of a method for determining routing information provided in the first embodiment of the present invention. This embodiment is applicable to the case where a router in a chip determines routing information. The method can be executed by a routing information determination device. The routing information determination device can be implemented in the form of hardware and / or software. The routing information determination device can be configured in a chip. The routers in the chip are connected via an on-chip network. Figure 1a As shown, the method includes:

[0051] Step 110: Receive the access request triggered by the Master node and the initial routing information corresponding to the initial router through the target input port of the initial router.

[0052] In this embodiment, the on-chip network can connect multiple Master nodes and Slave nodes. If a Master node triggers an access request to a target Slave node, the router corresponding to the Master node can be used as the initial router, and the routing information corresponding to the initial router (i.e., the initial routing information) can be determined by the Master node. The target input port can be the connection port between the Master node and the initial router.

[0053] In a specific embodiment, Figure 1b This is a schematic diagram of the structure of a router in this embodiment. Figure 1b As shown, each router may include five ports (port 0, port 1, port 2, port 3 and port 4). Port 0, port 1, port 2 and port 3 may be input or output ports of the router, and port 4 may be a local port corresponding to the router, and the local port may be connected to a master node or a slave node corresponding to the router. If the master node triggers an access request to a target slave node, the initial routing information and the access request may be sent to the initial router via port 4.

[0054] In this embodiment, the Master node may optionally determine the initial routing information of the initial router according to the path information between the initial router and the target Slave node. The initial routing information may include the output port information of the initial router to the target Slave node.

[0055] Step 120: Perform input port arbitration according to the initial routing information through the initial router, and simultaneously determine the lower-layer routing information corresponding to the lower-layer router.

[0056] In this embodiment, if the output port of the router receives requests from multiple input ports at the same time, it is necessary to arbitrate the multiple input ports so as to transmit the requests of each input port to the output port in sequence according to the arbitration result.

[0057] In this step, the initial router can arbitrate the matched multiple input ports according to the initial routing information, and simultaneously determine the lower-layer routing information corresponding to the lower-layer router. The lower-layer router can be the next router that the access request passes through in the on-chip network. The lower-layer routing information can include the output port information of the lower-layer router reaching the target slave node.

[0058] Step 130: Send the lower-layer routing information and the access request to the lower-layer router through the initial router according to the input port arbitration result.

[0059] Step 140: If it is determined that the lower-layer router is the target router corresponding to the access request, the lower-layer router performs input port arbitration according to the lower-layer routing information, and sends the access request to the target Slave node according to the input port arbitration result.

[0060] In this embodiment, the target router is the router corresponding to the target Slave node. If the lower router is the target router, the lower router can directly arbitrate the matched multiple input ports according to the lower routing information, and send the access request to the target Slave node according to the arbitration result.

[0061] In a specific embodiment, Figure 1c As an example, the network on chip in FIG. 1 includes the following routers: R0, R1, R2, R3, R4, R5, R6, R7, R8, R9, R10, R11, R12, R13, R14 and R15. Assuming that the Master node corresponding to R5 (i.e., M5) triggers an access request to S9 (i.e., the Slave node corresponding to R9), R5 can be used as the initial router, and M5 can determine that the initial routing information of R5 is "output port 2", and send the initial routing information to R5 through port 4 of R5. R5 can arbitrate multiple input ports corresponding to output port 2, and at the same time determine that the lower-layer routing information corresponding to the lower-layer router R9 is "output port 4". R5 can send the lower-layer routing information and the access request to R9 according to the arbitration result of the input port. Since R9 is the target router, multiple input ports corresponding to output port 4 can be arbitrated directly through R9, and the access request can be sent to S9 according to the arbitration result.

[0062] In one implementation of this embodiment, after the initial router sends the lower-layer routing information and the access request to the lower-layer router according to the input port arbitration result, it also includes: if it is determined that the lower-layer router is not the target router corresponding to the access request, the lower-layer router performs input port arbitration according to the lower-layer routing information, and simultaneously determines the subsequent routing information corresponding to the subsequent cascade router; the lower-layer router sends the subsequent routing information and the access request to the subsequent cascade router according to the input port arbitration result; if it is determined that the subsequent cascade router is the target router corresponding to the access request, the subsequent cascade router performs input port arbitration according to the subsequent routing information, and sends the access request to the target Slave node according to the input port arbitration result. The subsequent cascade router can be the next router in the network-on-chip that is located after the lower-layer router and is passed by the access request.

[0063] In a specific embodiment, Figure 1c Take the on-chip network in as an example, assuming that M5 triggers an access request to S13 (that is, the Slave node corresponding to R13), then R5 can be used as the initial router, and M5 can determine that the initial routing information of R5 is "output port 2", and send the initial routing information to R5 through port 4 of R5. R5 can arbitrate the multiple input ports corresponding to output port 2, and at the same time determine that the lower-level routing information corresponding to the lower-level router R9 is "output port 2". R5 can send the lower-level routing information and the access request to R9 based on the input port arbitration result. Since R9 is not the target router, R9 needs to arbitrate the multiple input ports corresponding to output port 2, and determine that the subsequent routing information corresponding to the subsequent cascade router R13 is "output port 4". Since R13 is the target router, the multiple input ports corresponding to output port 4 can be directly arbitrated through R13, and the access request can be sent to S13 based on the arbitration result.

[0064] In a specific implementation, if the subsequent cascade router is not the target router, the operations of steps 120-140 may be repeatedly performed by the subsequent cascade server until the final target router is determined, and the target router performs input port arbitration according to the corresponding routing information, and then sends the access request to the target Slave node according to the input port arbitration result.

[0065] For example, Figure 1cTake the on-chip network in as an example, assuming that M5 triggers an access request to S11 (that is, the Slave node corresponding to R11), the chip in the prior art needs to execute the following process "R5 determines local routing information - R5 performs input port arbitration - R6 determines local routing information - R6 performs input port arbitration - R7 determines local routing information - R7 performs input port arbitration - R11 determines local routing information - R11 performs input port arbitration". In contrast, in this embodiment, the previous router determines the routing information of the next router, so that the router can execute the routing information determination process and the port arbitration process in parallel, that is, through the technical solution of this embodiment, the chip only needs to execute the following process "R5 determines R6 routing information and performs input port arbitration - R6 determines R7 routing information and performs input port arbitration - R7 determines R11 routing information and performs input port arbitration - R11 performs input port arbitration", thereby reducing the time consumed by the Master node to access the Slave node and improving the access efficiency.

[0066] The technical solution provided by the embodiment of the present invention receives the access request and initial routing information triggered by the Master node through the target input port of the initial router, performs input port arbitration according to the initial routing information through the initial router, and simultaneously determines the lower routing information of the lower router, and sends the lower routing information and the access request to the lower router according to the input port arbitration result through the initial router. If the lower router is determined to be the target router corresponding to the access request, the lower router performs input port arbitration according to the lower routing information, and sends the access request to the target Slave node according to the input port arbitration result. This technical solution can reduce the time consumed by the Master node to access the Slave node and improve the access efficiency.

[0067] Figure 2 This is a flow chart of a method for determining routing information provided by Embodiment 2 of the present invention. This embodiment is a further refinement of the above embodiment. Figure 2 As shown, the method includes:

[0068] Step 210: Receive the access request triggered by the Master node and the initial routing information corresponding to the initial router through the target input port of the initial router.

[0069] Step 220: Determine a target output port according to the initial routing information through the initial router, and obtain multiple input ports corresponding to the target output port.

[0070] In this step, the initial router may determine the target output port according to the output port information specified in the initial routing information, and obtain multiple input ports that send requests to the target output port.

[0071] Step 230: arbitrate the multiple input ports through the initial router to obtain the execution order corresponding to each of the input ports, and simultaneously determine the lower-layer routing information corresponding to the lower-layer router.

[0072] In this embodiment, optionally, the initial router can arbitrate each input port according to the preset weight corresponding to each input port to obtain the execution order corresponding to each input port. The higher the weight corresponding to the input port, the higher the priority of the execution order. Specifically, the initial router can arbitrate each input port through a preset output port arbitration module.

[0073] Step 240: Send the lower-layer routing information and the access request to the lower-layer router through the target output port of the initial router according to the execution order corresponding to the target input port.

[0074] The advantage of such a setting is that congestion of the target output port of the initial router can be avoided, thereby improving the request transmission efficiency.

[0075] Step 250: If it is determined that the lower-layer router is the target router corresponding to the access request, the lower-layer router performs input port arbitration according to the lower-layer routing information, and sends the access request to the target Slave node according to the input port arbitration result.

[0076] The technical solution provided by the embodiment of the present invention receives the access request triggered by the Master node and the initial routing information through the target input port of the initial router, determines the target output port according to the initial routing information through the initial router, and obtains multiple input ports corresponding to the target output port, arbitrates the multiple input ports through the initial router to obtain the execution order corresponding to each input port, and simultaneously determines the lower-level routing information corresponding to the lower-level router, sends the lower-level routing information and the access request to the lower-level router through the target output port of the initial router according to the execution order corresponding to the target input port, and if the lower-level router is determined to be the target router corresponding to the access request, the input port arbitration is performed by the lower-level router according to the lower-level routing information, and the access request is sent to the target Slave node according to the input port arbitration result. The technical means can reduce the time consumed by the Master node to access the Slave node and improve the access efficiency.

[0077] Figure 3 This is a flowchart of a method for determining routing information provided by Embodiment 3 of the present invention. This embodiment is a further refinement of the above embodiment. Figure 3 As shown, the method includes:

[0078] Step 310: Receive the access request triggered by the Master node and the initial routing information corresponding to the initial router through the target input port of the initial router.

[0079] Step 320: perform input port arbitration according to the initial routing information through the initial router, and determine the lower-level router and the target router according to the target Slave node corresponding to the access request and the composition structure of the on-chip network, and determine the lower-level routing information corresponding to the lower-level router according to the corresponding connection mode of the lower-level router and the target router.

[0080] In a specific embodiment, Figure 1c Take the on-chip network in as an example, assuming that M5 triggers an access request to S6 (that is, the slave node corresponding to R6), then R5 can be used as the initial router, and M5 can determine that the initial routing information of R5 is "output port 1", and send the initial routing information to R5 through port 4 of R5. R5 can arbitrate multiple input ports corresponding to output port 1, and at the same time, according to S6 and the composition structure of the on-chip network, determine that the lower-layer router is R6, the target router is R6, and determine that the lower-layer routing information of the lower-layer router R6 is "output port 4".

[0081] Step 330: Send the lower-layer routing information and the access request to the lower-layer router through the initial router according to the input port arbitration result.

[0082] Step 340: If it is determined that the lower-layer router is the target router corresponding to the access request, the lower-layer router performs input port arbitration according to the lower-layer routing information, and sends the access request to the target Slave node according to the input port arbitration result.

[0083] Step 350: Determine the target routing information corresponding to the target router through the target Slave node, and send the target routing information and the response data corresponding to the access request to the target router.

[0084] In this embodiment, similar to the request transmission process, the target Slave node may determine the target routing information corresponding to the target router when feeding back the response data.

[0085] Step 360: Perform input port arbitration according to the target routing information through the target router, and simultaneously determine the upper-layer routing information corresponding to the upper-layer router.

[0086] Step 370: Send the upper-layer routing information and response data to the upper-layer router through the target router according to the input port arbitration result.

[0087] Step 380: If it is determined that the upper-layer router is the initial router, the upper-layer router performs input port arbitration according to the upper-layer routing information, and sends the response data to the Master node according to the input port arbitration result.

[0088] The advantage of this setting is that during the interaction between the Master node and the Slave node, the routing information of the next router is determined by the previous router, so that the router can execute the routing information determination process and the port arbitration process in parallel, thereby reducing the interaction time between the Master node and the Slave node and improving the interaction efficiency.

[0089] The technical solution provided by the embodiment of the present invention receives the access request and initial routing information triggered by the Master node through the initial router, performs input port arbitration and determines the lower-layer routing information at the same time through the initial router, sends the lower-layer routing information and the access request to the lower-layer router according to the input port arbitration result through the initial router, if the lower-layer router is determined to be the target router, performs input port arbitration through the lower-layer router, and sends the access request to the target Slave node according to the input port arbitration result, sends the target routing information and response data to the target router through the target Slave node, performs input port arbitration through the target router and determines the upper-layer routing information at the same time, sends the upper-layer routing information and response data to the upper-layer router according to the input port arbitration result through the target router, if the upper-layer router is determined to be the initial router, performs input port arbitration through the upper-layer router, and sends the response data to the Master node according to the input port arbitration result. The technical means can reduce the interaction time between the Master node and the Slave node and improve the interaction efficiency.

[0090] Figure 4 This is a schematic diagram of the structure of a routing information determination device provided by Embodiment 4 of the present invention, wherein the device is applied to a chip, and each router in the chip is connected via an on-chip network. Figure 4 As shown, the device includes: a request receiving module 410, an information determining module 420, an information sending module 430 and a request sending module 440.

[0091] The request receiving module 410 is used to receive the access request triggered by the Master node and the initial routing information corresponding to the initial router through the target input port of the initial router;

[0092] An information determination module 420, configured to perform input port arbitration according to the initial routing information through the initial router, and simultaneously determine the lower-layer routing information corresponding to the lower-layer router;

[0093] An information sending module 430, configured to send the lower layer routing information and the access request to the lower layer router through the initial router according to the input port arbitration result;

[0094] The request sending module 440 is used to perform input port arbitration according to the lower-layer routing information through the lower-layer router if it is determined that the lower-layer router is the target router corresponding to the access request, and send the access request to the target Slave node according to the input port arbitration result.

[0095] The technical solution provided by the embodiment of the present invention receives the access request and initial routing information triggered by the Master node through the target input port of the initial router, performs input port arbitration according to the initial routing information through the initial router, and simultaneously determines the lower routing information of the lower router, and sends the lower routing information and the access request to the lower router according to the input port arbitration result through the initial router. If the lower router is determined to be the target router corresponding to the access request, the lower router performs input port arbitration according to the lower routing information, and sends the access request to the target Slave node according to the input port arbitration result. This technical solution can reduce the time consumed by the Master node to access the Slave node and improve the access efficiency.

[0096] Based on the above embodiment, the information determination module 420 includes:

[0097] An input port acquisition unit, configured to determine a target output port according to the initial routing information through the initial router, and acquire a plurality of input ports corresponding to the target output port;

[0098] A sequence determination unit, configured to arbitrate the plurality of input ports through the initial router to obtain an execution sequence corresponding to each of the input ports;

[0099] A target router determination unit, configured to determine a lower router and a target router through the initial router according to a target Slave node corresponding to the access request and a composition structure of the on-chip network;

[0100] The routing information determining unit is used to determine the lower-layer routing information corresponding to the lower-layer router through the initial router according to the connection modes corresponding to the lower-layer router and the target router.

[0101] The information sending module 430 includes:

[0102] A request sending unit, configured to send the lower-layer routing information and the access request to the lower-layer router through the target output port of the initial router according to the execution order corresponding to the target input port;

[0103] A subsequent information determination unit, configured to perform input port arbitration according to the lower-layer routing information through the lower-layer router if it is determined that the lower-layer router is not the target router corresponding to the access request, and simultaneously determine the subsequent routing information corresponding to the subsequent cascade router;

[0104] A subsequent information sending unit, used to send the subsequent routing information and access request to the subsequent cascade router through the lower router according to the input port arbitration result;

[0105] The subsequent router execution unit is used to perform input port arbitration according to subsequent routing information through the subsequent cascade router if it is determined that the subsequent cascade router is the target router corresponding to the access request, and send the access request to the target Slave node according to the input port arbitration result.

[0106] The routing information determining device further comprises:

[0107] A response data sending module, used to determine the target routing information corresponding to the target router through the target Slave node, and send the target routing information and the response data corresponding to the access request to the target router;

[0108] An upper layer information determination module, used to perform input port arbitration according to the target routing information through the target router, and simultaneously determine the upper layer routing information corresponding to the upper layer router;

[0109] An upper layer information sending module, used for sending the upper layer routing information and response data to an upper layer router through the target router according to the input port arbitration result;

[0110] The initial router execution module is used to perform input port arbitration according to upper-layer routing information through the upper-layer router if it is determined that the upper-layer router is the initial router, and send the response data to the Master node according to the input port arbitration result.

[0111] The above device can execute the methods provided by all the above embodiments of the present invention, and has the corresponding functional modules and beneficial effects of executing the above methods. For technical details not fully described in the embodiments of the present invention, please refer to the methods provided by all the above embodiments of the present invention.

[0112] Figure 5 The schematic diagram of the structure of a chip 10 that can be used to implement an embodiment of the present invention is shown. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or claimed herein.

[0113] like Figure 5 As shown, the chip 10 includes a plurality of processor cores 11, a plurality of computing acceleration cores 12, and at least one memory connected to the processor core 11 and the computing acceleration core 12, such as a random access memory (RAM) 13, a read-only memory (ROM) 14, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor core 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 14 or the computer program loaded into the random access memory (RAM) 13. Various programs and data required for the operation of the chip 10 can also be stored in the RAM 13. The processor core 11, the computing acceleration core 12, the RAM 13 and the ROM 14 are connected to each other through a bus 15. The input / output (I / O) interface is also connected to the bus 15.

[0114] A number of components in the chip 10 are connected to the I / O interface, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; and a communication unit 18, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 18 allows the chip 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0115] The processor core 11 may be a variety of general and / or dedicated processing components with processing and computing capabilities. Some examples of the processor core 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor core 11 executes the various methods and processes described above, such as a method for determining routing information.

[0116] In some embodiments, the routing information determination method may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium. In some embodiments, part or all of the computer program may be loaded and / or installed on the chip 10 via the ROM 14 and / or the communication unit 18. When the computer program is loaded into the RAM 13 and executed by the processor core 11, one or more steps of the routing information determination method described above may be performed. Alternatively, in other embodiments, the processor core 11 may be configured to perform the routing information determination method in any other appropriate manner (e.g., by means of firmware).

[0117] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0118] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0119] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0120] To provide interaction with a user, the systems and techniques described herein may be implemented on a chip having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the chip. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0121] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0122] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0123] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0124] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for determining routing information, characterized in that: Applied in a chip, wherein routers in the chip are connected via an on-chip network, the method comprising: Receive the access request triggered by the Master node and the initial routing information corresponding to the initial router through the target input port of the initial router; By means of the initial router, input port arbitration is performed according to the initial routing information, and the lower layer routing information corresponding to the lower layer router is determined at the same time; Sending the lower-layer routing information and the access request to the lower-layer router through the initial router according to the input port arbitration result; If it is determined that the lower-layer router is the target router corresponding to the access request, the lower-layer router performs input port arbitration according to the lower-layer routing information, and sends the access request to the target Slave node according to the input port arbitration result; The step of performing input port arbitration according to the initial routing information through the initial router includes: Determine, by the initial router, a target output port according to the initial routing information, and obtain a plurality of input ports corresponding to the target output port; The initial router arbitrates the multiple input ports to obtain execution orders corresponding to the input ports.

2. The method according to claim 1, characterized in that The method includes sending the lower-layer routing information and the access request to the lower-layer router through the initial router according to the input port arbitration result, including: The lower-layer routing information and the access request are sent to the lower-layer router through the target output port of the initial router according to the execution order corresponding to the target input port.

3. The method according to claim 1, characterized in that Determine the lower-level routing information corresponding to the lower-level router, including: Determine, through the initial router, a lower layer router and a target router according to a target Slave node corresponding to the access request and a composition structure of the on-chip network; The lower-layer routing information corresponding to the lower-layer router is determined through the initial router according to the connection modes corresponding to the lower-layer router and the target router.

4. The method according to claim 1, characterized in that: After the initial router sends the lower layer routing information and the access request to the lower layer router according to the input port arbitration result, the method further includes: If it is determined that the lower layer router is not the target router corresponding to the access request, then performing input port arbitration according to the lower layer routing information through the lower layer router, and simultaneously determining subsequent routing information corresponding to the subsequent cascade router; Sending the subsequent routing information and the access request to the subsequent cascade router through the lower router according to the input port arbitration result; If it is determined that the subsequent cascade router is the target router corresponding to the access request, the subsequent cascade router performs input port arbitration according to the subsequent routing information, and sends the access request to the target Slave node according to the input port arbitration result.

5. The method according to claim 1, characterized in that After sending the access request to the target Slave node, the process further includes: Determine the target routing information corresponding to the target router through the target Slave node, and send the target routing information and the response data corresponding to the access request to the target router; Through the target router, perform input port arbitration according to the target routing information, and simultaneously determine the upper-layer routing information corresponding to the upper-layer router; Sending the upper-layer routing information and response data to the upper-layer router through the target router according to the input port arbitration result; If it is determined that the upper-layer router is the initial router, the upper-layer router performs input port arbitration according to the upper-layer routing information, and sends the response data to the Master node according to the input port arbitration result.

6. A routing information determination device, characterized in that: Applied in a chip, wherein routers in the chip are connected via an on-chip network, the device comprises: A request receiving module, used to receive the access request triggered by the Master node and the initial routing information corresponding to the initial router through the target input port of the initial router; An information determination module, used to perform input port arbitration according to the initial routing information through the initial router, and simultaneously determine the lower-layer routing information corresponding to the lower-layer router; An information sending module, used for sending the lower layer routing information and the access request to the lower layer router through the initial router according to the input port arbitration result; A request sending module, configured to perform input port arbitration according to the lower-layer routing information through the lower-layer router if it is determined that the lower-layer router is the target router corresponding to the access request, and send the access request to the target Slave node according to the input port arbitration result; Wherein, the information determination module includes: An input port acquisition unit, configured to determine a target output port according to the initial routing information through the initial router, and acquire a plurality of input ports corresponding to the target output port; The sequence determination unit is used to arbitrate the multiple input ports through the initial router to obtain the execution sequence corresponding to each of the input ports.

7. A chip, characterized in that: The chip comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the routing information determination method according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the method for determining routing information according to any one of claims 1 to 5 when executed.