CN114490194A

CN114490194A - Power failure processing method, functional node, processing system, device and storage medium

Info

Publication number: CN114490194A
Application number: CN202210407135.4A
Authority: CN
Inventors: 王凯
Original assignee: Haiguang Information Technology Co Ltd
Current assignee: Haiguang Information Technology Co Ltd
Priority date: 2022-04-19
Filing date: 2022-04-19
Publication date: 2022-05-13
Anticipated expiration: 2042-04-19
Also published as: CN114490194B

Abstract

A power failure processing method, a functional node, a processing system, a device and a storage medium are provided. The power failure processing method is used for a first node in a processing system and comprises the following steps: responding to the functional component of the first node in an idle state, and initiating a storage configuration negotiation request to perform storage configuration negotiation with at least one object node in N second nodes in the processing system; in response to the first node determining that the configuration storage negotiation is completed, providing the data to be backed up of the first node to at least one target node of the N second nodes for backup storage; after the data to be backed up of the first node is sent to at least one target node, executing power failure negotiation operation; and responding to the first node to determine that the power down negotiation is completed, and performing power down processing on the first node. The method can store the data to be backed up of the first node into a target node different from the first node before power failure negotiation, and avoids the loss of the data to be backed up of the first node caused by power failure.

Description

Power failure processing method, functional node, processing system, device and storage medium

Technical Field

Embodiments of the present disclosure relate to the field of integrated circuits, and in particular, to a power failure processing method, a functional node, a processing system, an electronic device, and a computer-readable storage medium.

Background

In an integrated circuit, a low power mode may be implemented by powering down high power consuming parts when the circuit is idle. For example, for a multi-node computing system, multiple nodes may be synchronized into a power down mode while the computing system is in an idle state to save system power consumption.

Disclosure of Invention

At least one embodiment of the present disclosure provides a power down processing method for a first node in a processing system, including: responding to the functional component of the first node in an idle state, and initiating a storage configuration negotiation request to perform storage configuration negotiation with at least one object node in N second nodes in the processing system; in response to the first node determining that the configuration storage negotiation is completed, providing the data to be backed up of the first node to at least one target node of the N second nodes for backup storage; after the data to be backed up of the first node is sent to the at least one target node, executing power failure negotiation operation; and responding to the first node to determine that the power down negotiation is completed, and performing power down processing on the first node, wherein N is a positive integer.

For example, in the power failure processing method provided in an embodiment of the present disclosure, in response to the first node determining that the negotiation for saving the configuration is completed, the method for providing the data to be backed up of the first node to at least one target node of the N second nodes for backup saving includes: sending a save configuration request to the at least one target node in response to the first node determining that a save configuration negotiation is complete; after receiving a response of the at least one target node to the storage configuration request, sending the data to be backed up of the first node to the at least one target node, so that the at least one target node backs up and stores the data to be backed up of the first node.

For example, in a power failure handling method provided in an embodiment of the present disclosure, the at least one target node includes a plurality of target nodes; in response to the first node determining that the negotiation for saving the configuration is completed, providing the data to be backed up of the first node to at least one target node of the N second nodes for backup saving, including: in response to the first node determining that the configuration storage negotiation is completed, dividing the data to be backed up of the first node into a plurality of data sets; and respectively providing the plurality of data sets to the plurality of target nodes for backup and storage.

For example, in a power failure processing method provided in an embodiment of the present disclosure, the first node does not include a storage device for performing backup storage; the at least one target node comprises a storage device for backup storage.

For example, in a power failure processing method provided in an embodiment of the present disclosure, the first node is a bus node, and the functional units of the bus node include a bus unit; the target node is a core node, and the functional components of the core node comprise data processing components.

For example, in the power down processing method provided in an embodiment of the present disclosure, the at least one object node and the first node are located on the same chip; determining that the save configuration negotiation is complete, comprising: determining that the at least one object node has responded to the save configuration negotiation request, wherein each of the at least one object node responds to the save configuration negotiation request with its respective feature in an idle state.

For example, in a power down processing method provided by an embodiment of the present disclosure, the processing system includes a plurality of chips, each of the chips includes a plurality of nodes; determining that the power down negotiation is complete comprises: determining that all nodes in a processing system where the first node is located respond to the power-down negotiation request, wherein each node in the processing system responds to the power-down negotiation request under the condition that respective functional units are in an idle state.

For example, the power down processing method provided in an embodiment of the present disclosure further includes: and responding to the first node to determine that the power-down negotiation is completed, and performing power-down processing on all nodes in the processing system.

For example, the power down processing method provided in an embodiment of the present disclosure further includes: in response to receiving a power-on request, obtaining the data to be backed up of the first node which is backed up and stored from the at least one target node; and configuring the first node based on the acquired data to be backed up of the first node so as to restore the first node to a state before power failure processing.

For example, in a power down processing method provided in an embodiment of the present disclosure, the power up request is generated in at least one of the following cases: in the process of the configuration storage negotiation, at least one node in a chip where the first node is located is changed from an idle state to a non-idle state; in the power failure negotiation process, at least one node in a processing system where the first node is located is changed from an idle state to a non-idle state; or when all the nodes in the processing system where the first node is located are in a power-down state, the processing system where the first node is located is abnormal.

At least one embodiment of the present disclosure provides another power down processing method for processing a target node in a system, including: receiving at least part of data in the data to be backed up sent by a first node in the processing system, and saving the at least part of data, wherein the at least part of data is sent by the first node under the condition of determining that the negotiation for saving the configuration is completed; after saving the at least part of the data, performing a power down negotiation operation; and performing power-down processing on the target node in response to the fact that the power-down negotiation is determined to be completed.

For example, the power down processing method provided by an embodiment of the present disclosure further includes: receiving a save configuration request sent by the first node, wherein the save configuration request is initiated by the first node when the save configuration negotiation is determined to be completed; and responding to the save configuration request to enable the first node to send the at least part of data.

For example, in a power failure processing method provided in an embodiment of the present disclosure, data to be backed up of the first node includes a plurality of data sets; receiving at least part of data in the data to be backed up of the first node, and storing the at least part of data, including: receiving one of the plurality of data sets and saving the received data set.

For example, the power down processing method provided by an embodiment of the present disclosure further includes: receiving a save configuration negotiation request issued by the first node prior to the target node receiving and saving the at least part of the data, wherein the save configuration negotiation request is initiated by the first node in response to a functional component of the first node being in an idle state; and responding to the storage configuration negotiation request.

For example, the power down processing method provided by an embodiment of the present disclosure further includes: and after responding to the storage configuration negotiation request, storing the data to be backed up of the target node into a storage device of the target node.

For example, in the power down processing method provided in an embodiment of the present disclosure, the target node and the first node are located on the same chip; responding the save configuration negotiation request, including: responding to the save configuration negotiation request with a feature of the target node in an idle state.

For example, in a power down processing method provided by an embodiment of the present disclosure, the processing system includes a plurality of chips, each of the chips includes a plurality of nodes; determining that the power down negotiation is complete, including: and determining that all nodes in the processing system where the target node is located respond to the power-down negotiation request, wherein each node in the processing system responds to the power-down negotiation request under the condition that respective functional components are in an idle state.

For example, the power down processing method provided by an embodiment of the present disclosure further includes: in response to receiving a power-up request, sending the at least part of the data to the first node to cause the first node to configure the first node based on the at least part of the data.

At least one embodiment of the present disclosure provides another power down processing method, including: initiating, by a first node in a processing system, a save configuration negotiation request in response to a functional component of the first node being in an idle state; receiving a configuration storage negotiation request through at least one object node in N second nodes in the processing system, and responding to the configuration storage negotiation request; in response to the first node determining that the configuration storage negotiation is completed, providing the data to be backed up of the first node to at least one target node of the N second nodes through the first node, and performing backup storage on the data to be backed up through the at least one target node; after the data to be backed up of the first node is sent to the at least one target node, performing power down negotiation at least through the first node and the N second nodes; and responding to the completion of power down negotiation, and performing power down processing on the first node and the N second nodes, wherein N is a positive integer.

At least one embodiment of the present disclosure provides a function node, which includes a storage negotiation module, a backup data module, a first power down negotiation module, and a first power down processing module. The saving negotiation module is configured to respond to the functional component of the functional node being in an idle state, initiate a saving configuration negotiation request, and perform saving configuration negotiation with at least one object node in N second nodes in the same processing system; the backup data module is configured to respond to the functional node determining that the storage configuration negotiation is completed, and provide the data to be backed up of the functional node to at least one target node of the N second nodes for backup storage; the first power down negotiation module is configured to execute power down negotiation operation after the data to be backed up of the function node is sent to the at least one target node; and the first power-down processing module is configured to respond to the functional node to determine that power-down negotiation is completed, and perform power-down processing, wherein N is a positive integer.

At least one embodiment of the present disclosure provides another function node, including a data receiving module, a second power down negotiation module, and a second power down processing module, where the data receiving module is configured to receive at least part of data to be backed up of a first node located in the same processing system as the function node, and store the at least part of data to be backed up, where the at least part of data to be backed up is sent by the first node when it is determined that the storage configuration negotiation is completed; the second power down negotiation module is configured to execute power down negotiation operation after the at least part of the data to be backed up is stored; the second power down processing module is configured to perform power down processing in response to determining that the power down negotiation is complete.

At least one embodiment of the present disclosure provides a processing system comprising a first node and N second nodes, the first node comprising a feature configured to initiate a save configuration negotiation request in response to the feature being in an idle state; the N second nodes are configured to at least one object node to receive the saved configuration negotiation request and respond to the saved configuration negotiation request; wherein the first node is further configured to: in response to determining that the configuration negotiation is finished, providing the data to be backed up of the first node to at least one target node in the N second nodes for backup and storage; the first node and the N second nodes are further configured to: after the data to be backed up of the first node is sent to the at least one target node, performing power failure negotiation operation; and responding to the completion of power failure negotiation, and performing power failure processing, wherein N is a positive integer.

At least one embodiment of the present disclosure provides an electronic device comprising a processor; a memory including one or more computer program modules; wherein the one or more computer program modules are stored in the memory and configured to be executed by the processor, the one or more computer program modules comprising instructions for implementing the power down handling method provided by any embodiment of the present disclosure.

At least one embodiment of the present disclosure provides a computer-readable storage medium storing non-transitory computer-readable instructions, which, when executed by a computer, implement a power-down processing method provided in any embodiment of the present disclosure.

Drawings

To more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings of the embodiments will be briefly introduced below, and it is apparent that the drawings in the following description relate only to some embodiments of the present disclosure and are not limiting to the present disclosure.

FIG. 1A shows a schematic diagram of a processing system;

FIG. 1B shows a schematic diagram of a node;

FIG. 2 shows a schematic diagram of a node distribution;

fig. 3 illustrates a flowchart of a power down processing method according to at least one embodiment of the present disclosure;

fig. 4 illustrates a schematic diagram of a processing system provided by at least one embodiment of the present disclosure;

fig. 5A illustrates a schematic diagram of a core node provided by at least one embodiment of the present disclosure;

fig. 5B illustrates a schematic diagram of a bus node provided by at least one embodiment of the present disclosure;

fig. 6 illustrates a schematic flow chart of saving configuration and powering down according to at least one embodiment of the present disclosure;

FIG. 7 illustrates a schematic diagram of another processing system provided by at least one embodiment of the present disclosure;

fig. 8 illustrates a schematic flow chart of power-up provided by at least one embodiment of the present disclosure;

fig. 9 illustrates a flow chart of another power down processing method provided by at least one embodiment of the present disclosure;

fig. 10 is a flowchart illustrating still another power loss processing method according to at least one embodiment of the disclosure;

fig. 11 illustrates a block diagram of a functional node provided by at least one embodiment of the present disclosure;

fig. 12 illustrates a block diagram of another functional node provided by at least one embodiment of the present disclosure;

fig. 13 illustrates a schematic block diagram of an electronic device provided by at least one embodiment of the present disclosure;

fig. 14 illustrates a schematic block diagram of another electronic device provided by at least one embodiment of the present disclosure; and

fig. 15 illustrates a schematic diagram of a computer-readable storage medium provided by at least one embodiment of the present disclosure.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present disclosure more apparent, the technical solutions of the embodiments of the present disclosure will be described clearly and completely with reference to the drawings of the embodiments of the present disclosure. It is to be understood that the described embodiments are only a few embodiments of the present disclosure, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the described embodiments of the disclosure without any inventive step, are within the scope of protection of the disclosure.

Unless otherwise defined, technical or scientific terms used herein shall have the ordinary meaning as understood by one of ordinary skill in the art to which this disclosure belongs. The use of "first," "second," and similar terms in this disclosure is not intended to indicate any order, quantity, or importance, but rather is used to distinguish one element from another. Also, the use of the terms "a," "an," or "the" and similar referents do not denote a limitation of quantity, but rather denote the presence of at least one. The word "comprising" or "comprises", and the like, means that the element or item listed before the word covers the element or item listed after the word and its equivalents, but does not exclude other elements or items. The terms "connected" or "coupled" and the like are not restricted to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "upper", "lower", "left", "right", and the like are used merely to indicate relative positional relationships, and when the absolute position of the object being described is changed, the relative positional relationships may also be changed accordingly.

FIG. 1A shows a schematic diagram of a processing system. As shown in fig. 1A, the processing system may include a plurality of chips, including, for example, a chip 10 and a chip 11. Each chip may include a plurality of nodes, with chip 10 including, for example, node A0 and node A1, and chip 11 including, for example, node A2 and node A3. For example, the processing system may be applied to Die-to-Die (D2D) for Die-to-Die connection mode, in which a chip is divided into a plurality of smaller Die (dies) packaged in a multi-chip module in which a plurality of dies are interconnected by Die-to-Die connections, and each node in the disclosed embodiment may be, for example, a Die.

In order to ensure the performance of the server, the power-down strategy can be that the nodes of the whole system synchronously enter a power-down mode. In the multi-node processing system, after any node is idle, a power failure negotiation request can be provided for an adjacent node or a superior node, if the adjacent node or the superior node is also in an idle state, the power failure negotiation request can be continuously transmitted to the other adjacent node or the superior node, and the nodes of the whole system can start power failure until the nodes of the whole system complete the negotiation so as to enter a low power consumption mode.

Fig. 1B shows a schematic diagram of a node. As shown in fig. 1A and 1B, each node may include a set of clients (clients) including at least one client (e.g., client a00 and client a01 in fig. 1A), for example, a kernel cluster 101 (kernel cluster including, for example, a processor core such as a CPU core or a GPU core), a bus interface (Hub) 102, a Peripheral on chip (Peripheral) 103, and the like. In addition, the nodes may further include a data network (df) (data fabric) 104, a power consumption control unit 105, a general Memory 108 (general Memory such as dynamic Random Access Memory (dram), a universal Memory control interface (umc) (unified Memory controller))106, and a chip external interconnect interface 107 (chip external interconnect interface such as gmi (global Memory interconnect), xgmi (external gmi), gop (gmi Over) pcie), where the client set accesses the general Memory 108 through the data network 104 and the chip external interconnect interface 107 (hereinafter also referred to as "interconnect interface" or "interface") and accesses other nodes through the chip external interconnect interface 107, and the power consumption control unit 105 is configured to complete a power down process.

For example, each node may include a plurality of configuration registers for storing the state and information of the hardware circuitry of the node, such as switch states, flag bits, etc. The power failure of the node can cause the data loss of the configuration register, in order to simplify and accelerate the process (namely, the power-on process) of the node exiting the low power consumption mode, and avoid the complex configuration which needs to be carried out when the node exits the low power consumption each time, the node can automatically back up the configuration register before the power failure, and thus, the data of the configuration register can be automatically recovered after the node is powered on again each time. For example, after a client set of a node enters an idle state, a power down negotiation request (i.e., a request to disconnect) may be sent to a neighboring node through the interconnect interface 107. After the synchronization of the power-down negotiation request of the whole system is completed through a certain arbitration strategy, the interconnection interfaces of all the nodes are disconnected, the data network of each node can write the data of the configuration register needing power down in the node into the general memory, all parts of the node start to power down, and the data network is used as a channel and needs to be powered down finally.

For example, as shown in fig. 1A, nodes located in the same chip are connected by using a first-level chip external interface, and nodes located in different chips are connected by using a second-level chip external interface. For example, when node A0 is ready to enter a low power state, a power down negotiation request may first be sent to neighboring node A1 via a primary chip external interface. After all nodes (for example, node a0 and node a 1) inside the chip 10 complete power down negotiation through the primary chip external interface, the nodes inside the chip 10 may send power down negotiation to the neighboring nodes of the neighboring chips through the secondary chip external interface. After all secondary external chip interfaces complete power down negotiation, each node automatically initiates a data network power down flow, stores the data of the configuration register and powers down.

For example, the power down negotiation request initiated by the chip external interconnection interface may be implemented by changing the connection state of the interconnection interface. The state of the interconnect interface can be divided into: the power consumption control unit is responsible for switching the states. The two connected nodes have the same interface state in the two directions back and forth, namely, the interconnection interface enters the corresponding working state. When the node works normally, the interconnection interface is in a working state. When each client in the node is idle and the interconnection interface does not have the data packet being transmitted, the interconnection interface of the node can be changed into a logic disconnection state, and a logic disconnection request is sent to the adjacent node by changing the state of the interconnection interface, namely, power failure negotiation is initiated to the adjacent node. When all the interconnection interfaces in the node complete power failure negotiation, namely all the interconnection interfaces are in a logic disconnection state, the power consumption control module initiates power failure operation of the interconnection interfaces and the data network according to an arbitration strategy, namely the interconnection interfaces are changed into a physical disconnection state after power failure, the data network backs up the data of the configuration register to the general memory, and then the data network is powered down.

For example, when the connection modes adopted by the nodes in the system are different, the power down negotiation modes are also different. The nodes in each chip shown in fig. 1A are all interconnected, a distributed power down negotiation may be adopted, and any node may initiate a power down negotiation request and transmit the power down negotiation request to the other nodes through the neighboring nodes. When the nodes cannot be connected completely, centralized power down negotiation can be adopted, the system can comprise a root node (also called a central node), all the other nodes are directly or indirectly connected with the root node, and a power down negotiation request can be initiated by the root node and transmitted to all the other nodes.

Fig. 2 shows a schematic diagram of a node distribution. As shown in fig. 2, node B0 is the root node, node B0 includes a low-level interface, end nodes (node B3 and B6) include a high-level interface, and the remaining nodes (node B1, B2, B4, and B5) each include a low-level interface and a high-level interface. The power down negotiation request may be transmitted first over a lower level interface, e.g., when node B0 is in an idle state, the power down negotiation request may be initiated over the lower level interface to neighboring node B1 and node B4 (the lower level interface of node B0 becomes logically disconnected). If node B1 is also in the idle state, node B1 continues to broadcast the power down negotiation request to node B2 via the low-level interface in response to the power down negotiation request sent by node B0 (the low-level interface of node B1 becomes a logical off state), and if node B2 is in the idle state, node B2 continues to broadcast the power down negotiation request to node B3 via the low-level interface (the low-level interface of node B2 becomes a logical off state), and similarly, in a case where both node B4 and node B5 are in the idle state, the power down negotiation request may be broadcast to node B6 via the low-level interfaces of node B4 and node B5 (the low-level interfaces of node B4 and node B5 become a logical off state).

And after the lower-level interfaces of the nodes complete the power down negotiation, the power down negotiation can be sent through the higher-level interfaces. For example, if node B3 is in an idle state, it may send back a power down negotiation request via the higher-level interface (the higher-level interface of node B3 becomes logically disconnected), and send back to node B2 as feedback information, and then feed back to node B0 via the state change of the higher-level interface of node B2 (becomes logically disconnected) and the state change of the higher-level interface of node B1 (becomes logically disconnected). If a node (e.g., node B1) adjacent to the root node (e.g., node B0) feeds back a power down negotiation request to node B0 via the higher-level interface (i.e., the higher-level interface becomes logically disconnected), it indicates that the series of nodes to which the adjacent node is connected are all in an idle state. Therefore, if the high-level interfaces of both node B1 and node B4 become logically disconnected, node B0 may determine that all nodes have entered the idle state, and the power down negotiation is completed, and the power down procedure may be started. The interconnection interface of the root node B0 becomes physically disconnected upon power down, which may serve as a power down request to notify neighboring nodes of power down, while the root node B0 stores configuration data and powers down the data network. After receiving the power-down request, the other nodes start to execute the power-down flow (interface power-down, configuration data storage, data network power-down and the like) of the own node, and continue to send the power-down request to the adjacent nodes. Finally, the power down request is broadcast to all nodes, and the nodes of the whole system are powered down.

Because each node initiates a power-down negotiation request by disconnecting the logic of the external interconnection interface of the chip, and the data network power-down flow is performed only after the logic of the interconnection interface is disconnected, the configuration register value required to be stored in the power-down operation of the data network can only be stored in the general memory of the node. With the development of small chip (chipset) technology, various types of nodes are often packaged in one chip. According to the power-down scheme, the nodes without the general memory cannot back up the data of the configuration register, the data of the configuration register is lost after the power-down, and complex configuration is required after the power-down is carried out again.

At least one embodiment of the present disclosure provides a power down processing method, a function node, a processing system, an electronic device, and a computer-readable storage medium. The power failure processing method is used for processing a first node in a system and comprises the following steps: responding to the functional component of the first node in an idle state, and initiating a storage configuration negotiation request to perform storage configuration negotiation with at least one object node in N second nodes in the processing system; in response to the first node determining that the configuration storage negotiation is completed, providing the data to be backed up of the first node to at least one target node of the N second nodes for backup storage; after the data to be backed up of the first node is sent to at least one target node, executing power failure negotiation operation; and responding to the first node to determine that the power down negotiation is completed, and performing power down processing on the first node, wherein N is a positive integer. The power failure processing method can store the data to be backed up of the first node into a target node different from the first node before power failure negotiation, so that the data to be backed up of the first node is prevented from being lost due to power failure, and the robustness of the system is improved.

Fig. 3 shows a flowchart of a power down processing method according to at least one embodiment of the present disclosure.

As shown in FIG. 3, the power-down processing method is used for processing a first node in a system, and may include steps S210-S240.

Step S210: and responding to the functional part of the first node in an idle state, and initiating a storage configuration negotiation request to perform storage configuration negotiation with at least one object node in N second nodes in the processing system, wherein N is a positive integer.

Step S220: and in response to the first node determining that the configuration storage negotiation is completed, providing the data to be backed up of the first node to at least one target node of the N second nodes for backup storage.

Step S230: and after the data to be backed up of the first node is sent to at least one target node, performing power down negotiation operation.

Step S240: and responding to the first node to determine that the power down negotiation is completed, and performing power down processing on the first node.

Fig. 4 illustrates a schematic diagram of a processing system provided by at least one embodiment of the present disclosure. As shown in fig. 4, the processing system includes a chip 20, and the chip 20 includes two types of nodes: a bus node and a core node. Chip 20 includes, for example, core node C0, core node C1, core node C2, core node C3, and bus node D0. The bus node D0 is connected to at least one core node through an interconnect interface, and a plurality of core nodes are interconnected. The processing system may also include other chips than chip 20, including, for example, core nodes and bus nodes, where bus nodes of different chips may be connected by an interconnect interface.

Fig. 5A illustrates a schematic diagram of a core node according to at least one embodiment of the present disclosure. As shown in fig. 5A, the kernel node includes a client set 301, and the client set 301 includes a kernel cluster, for example, a kernel cluster C00 and a kernel cluster C01. In addition, the core node further includes a data network 302, a power consumption control unit 303, a general memory control interface 304, a general memory 305, and an external chip interconnect interface 306, and the set of clients 301 access the general memory 305 through the data network 302 and through the general memory control interface 304, and access other nodes through the external chip interconnect interface 306. The power consumption control unit 303 is configured to complete a power down procedure.

Fig. 5B illustrates a schematic diagram of a bus node provided by at least one embodiment of the present disclosure. As shown in fig. 5B, the bus node includes a client set 401, and the client set 401 includes a bus interface and an on-chip peripheral (chip internal peripheral). Further, the bus node comprises a data network 402, a power consumption control unit 403 and a chip external interconnect interface 404. The client set 401 accesses other nodes through the data network 402 and the chip external interconnection interface 404, and the power consumption control unit 403 is configured to complete a power-down process.

For example, the first node may be a bus node, the functional unit of which comprises a bus unit, for example comprising a bus interface. In at least one embodiment, the first node does not include a storage device for performing backup storage, and the storage devices of other nodes are used for performing backup.

For example, in some examples, the N second nodes may be the remaining nodes in the processing system other than the first node, e.g., the N second nodes include all of the core nodes included in the processing system. The object node may be, for example, a node located on the same chip as the first node (for example, a core node), that is, the at least one object node and the first node are located on the same chip, in this example, the configuration storage negotiation may be performed by taking each chip as a unit, and after the negotiation of each node in the chip is completed, the subsequent operation of storing the data to be backed up may be performed (step S220), which does not need to perform the configuration storage negotiation within the range of the entire processing system, and may improve the efficiency of the configuration storage negotiation.

For example, in other examples, the object node may include a node in a chip where the first node is located and nodes in other chips except the chip where the first node is located, and in this example, the first node performs the save configuration negotiation with not only a node in the same chip but also nodes in other chips.

For example, determining that the save configuration negotiation is complete may include: determining that the at least one object node each responded to a save configuration negotiation request, wherein each node of the at least one object node responds to a save configuration negotiation request with the respective feature in an idle state. Based on the mode, the data to be backed up of the first node can be stored in the target node under the condition that each object node is determined to be in the idle state, so that frequent backup of the data to be backed up of the first node can be avoided, the operation is simplified, and the power consumption is saved.

For example, in some examples, each target node includes storage for backup storage. The target node is, for example, a core node, and the functional components of the core node include data processing components (e.g., a core cluster). The target node may be a core node located on the same chip as the first node, that is, the data to be backed up of the first node may be stored in the core node on the same chip. In other examples, the target node may also be a node located in a chip other than the chip where the first node is located, that is, the first node may store the data to be backed up in a core node of another chip.

For example, the object node and the target node may be the same or different, and may be located on the same chip or on different chips. For example, in some examples, the object node is a core node located on the same chip as the first node, and the target node is a core node located on a different chip from the first node, that is, after the first node completes the save configuration negotiation with the core node on the same chip, the first node may store the data to be backed up in the core node on another chip. In other examples, the object node and the target node may both be core nodes located on the same chip as the first node, that is, the first node may store the data to be backed up in the core node of the same chip after completing the saving configuration negotiation with the core node of the same chip.

Fig. 6 illustrates a flowchart of saving configuration and powering down according to at least one embodiment of the present disclosure. As shown in FIG. 6, for example, taking the first node as the bus node D0 and the at least one object node as the core nodes C0-C3 as examples, when the client set of the bus node D0 is in a power-on state (corresponding to the "power-on state" in FIG. 6), it is determined whether the client set is idle (corresponding to the "client set idle after the" power-on state "in FIG. 6)

"), after determining that the client set enters the idle state, a save configuration negotiation may be initiated to an adjacent core node in the same chip (corresponding to the" save configuration negotiation "in fig. 6), for example, a save configuration negotiation request may be initiated by changing the state of the interconnect interface, but the disclosure is not limited thereto, and in some embodiments, a save configuration negotiation request may also be initiated by sending a dedicated data packet, a dedicated connection, software configuration, and the like.

For example, when all the core nodes C0 to C3 are connected to the bus node D0, and when the client set of the bus node D0 is in an idle state, the bus node D0 may simultaneously initiate a save configuration negotiation request to the core nodes C0 to C3, and when the client set of the core nodes C0 to C3 is also in an idle state, the save configuration negotiation request may be responded to, for example, by changing the state of the interconnect interface or sending a dedicated packet. For example, in other examples, bus node D0 in the idle state may first initiate a save configuration negotiation request to some core nodes, and then broadcast the save configuration negotiation request to other core nodes by the some core nodes until all core nodes in the chip receive and respond to the save configuration negotiation request.

For example, after initiating the save configuration negotiation, it is determined whether the save configuration negotiation is completed (corresponding to "save configuration negotiation completed" in fig. 6)

"), the bus node D0 may determine that the saved configuration negotiation is complete after receiving the response information of the core nodes C0-C3 to the saved configuration negotiation request. After determining that the save configuration negotiation is complete, it may be determined whether the client set of bus node D0 is still idle (corresponding to "save configuration negotiation complete in fig. 6)

"later" client set idles

"), if the client set of the bus node D0 is still in an idle state, the data to be backed up (e.g., the data of the configuration register) of the bus node D0 may be sent to the general-purpose memory of any one or more core nodes for storage (corresponding to" save configuration register to general-purpose memory "in fig. 6). And after the storage configuration negotiation is completed, each core node may store its own data to be backed up (e.g., data of a configuration register) in its own storage device (e.g., a general-purpose memory).

For example, it is determined whether the save configuration is completed (corresponding to "save configuration completed" in fig. 6)

"), after the storage configuration is completed, the interconnect interface does not have a data packet to be transmitted, and power down negotiation (corresponding to" logic disconnection (power down negotiation) "in fig. 6) may be started. For example, the power-down negotiation may be a negotiation performed within the range of all nodes of the processing system, and the power-down operation is performed when all nodes of the processing system are in an idle state.

For example, the processing system includes a plurality of chips, each chip including a plurality of nodes, and determining that the power down negotiation is complete may include: and determining that all nodes in the processing system where the first node is located respond to the power-down negotiation request, wherein each node in the processing system responds to the power-down negotiation request under the condition that respective functional units are in an idle state.

For example, in the case that the processing system includes a plurality of chips, the power down negotiation process may first perform power down negotiation with each chip as a unit, and then perform power down negotiation between the chips, so that the efficiency of the power down negotiation may be improved. Fig. 7 illustrates a schematic diagram of another processing system provided by at least one embodiment of the present disclosure. As shown in fig. 7, the power down negotiation is described by taking as an example that the processing system includes a chip 20 and another chip 21 having the same composition as the chip 20 and connected to the chip 20, for example, the other chip 21 also includes a bus node F0 and a plurality of core nodes (e.g., core nodes E0-E3), and the bus node F0 in the other chip 21 is connected to the bus node D0 in the chip 20. In the chip 20, after the operation of saving the data to be backed up is completed, power down negotiation may be performed within the range of the chip 20, for example, a power down negotiation request may be initiated by any node (for example, the bus node D0), the power down negotiation request may be responded to when the remaining nodes are still in an idle state, and the power down negotiation inside the chip may be considered to be completed when it is confirmed that all the nodes in the chip respond to the power down negotiation request (that is, all the nodes are in an idle state). After completing the power down negotiation inside chip 20, chip 20 may perform the power down negotiation with chip 21 through the interconnect interface of bus node D0. Similarly, in another chip 21, after the operation of saving the data to be backed up is completed, the power down negotiation is performed within the range of the chip 21, and after the chip 21 completes the power down negotiation inside the chip, the chip 21 may respond to the inter-chip power down negotiation initiated by the chip 20 through the interconnection interface of its bus node F0, so as to complete the power down negotiation within the range of the processing system. For example, the power down negotiation may be performed by changing the interconnection interface into a logic disconnection mode, which may specifically refer to the above related contents, and will not be described herein again.

For example, it is determined whether the power down negotiation passes (corresponding to "power down negotiation pass" in fig. 6)

") and after the power down negotiation passes, determine whether the first node is still in an idle state (corresponding to" power down negotiation passes in FIG. 6)

"later" client set idles

") may begin a power down process if a first node (e.g., a bus node) is still in an idle state, including powering down the interconnect interface to physically disconnect other nodes and powering down components contained in nodes such as the data network (corresponding to" physical disconnect in fig. 6: internet interface power down, data network power down ").

For example, in response to the first node determining that the power down negotiation is complete, power down processing may be performed on all nodes in the processing system. For example, for the chip 20, when the interconnection interface of the bus node (i.e., the first node) changes to the physical disconnection state, the interconnection interface may serve as a power-down request initiated to the adjacent kernel node, and after receiving the power-down request, the adjacent node may also perform power-down processing, and broadcast the power-down request to another node, and so on until all the nodes complete the power-down processing. Similarly, each node inside the chip 21 also completes the power down processing successively, thereby completing the power down processing of all nodes in the processing system.

According to the power failure processing method of the embodiment of the disclosure, the data to be backed up of the first node can be stored in the target node with the storage device before power failure negotiation (namely before interconnection interface logic is disconnected), so that the data of the first node can be backed up, the data to be backed up of the first node is prevented from being lost, and further, the circuit configuration of the first node can be recovered after power on. Moreover, the storage configuration negotiation is carried out before the data to be backed up is stored, and the power failure negotiation process is carried out after the data to be backed up is stored, so that the behavior of storing the data to be backed up can be accurately and efficiently executed, and the system can be ensured to be in a real idle state in each power failure operation.

For example, step S220 may further include: sending a save configuration request to at least one target node in response to the first node determining that the save configuration negotiation is complete; after receiving a response of the at least one target node to the configuration storage request, sending the data to be backed up of the first node to the at least one target node, so that the at least one target node performs backup storage on the data to be backed up of the first node. For example, after the bus node D0 determines that the negotiation for saving the configuration is completed, it may send a request for saving the configuration to one or more core nodes, and after obtaining a response from the one or more core nodes to the request for saving the configuration, send the data to be backed up to the one or more core nodes for saving.

For example, the at least one target node in step S220 includes a plurality of target nodes, i.e., the number of target nodes may be plural. Step S220 may further include: in response to the first node determining that the configuration storage negotiation is completed, dividing the data to be backed up of the first node into a plurality of data sets; and respectively providing the plurality of data sets to a plurality of target nodes for backup and storage. For example, each of the core nodes C0, C1, and C2 may be a target node, the data to be backed up of the bus node D0 may be divided into three parts, three data sets are formed, and the three data sets are sent to the core nodes C0, C1, and C2 respectively for saving.

For example, in other examples, the data to be backed up may also be repeatedly stored in a plurality of target nodes, that is, the plurality of target nodes each store complete data to be backed up.

Based on the above manner of storing the data to be backed up to a plurality of target nodes, the problem that the data to be backed up is completely lost due to the failure of the target nodes in the scheme of storing the data to be backed up only in one target node can be avoided, and the safety of the data to be backed up is ensured. In addition, the data to be backed up is divided into a plurality of copies and stored in the plurality of target nodes respectively, so that the storage pressure of the plurality of target nodes can be balanced, partial data can be transmitted to the plurality of target nodes simultaneously during backup, and the backup efficiency is improved.

For example, the power-down processing method according to the embodiment of the present disclosure may further include: in response to receiving a power-on request, acquiring data to be backed up of a first node which is backed up and stored from at least one target node; and configuring the first node based on the acquired data to be backed up of the first node so as to restore the first node to the state before power failure processing.

Fig. 8 shows a schematic flowchart of power-up provided by at least one embodiment of the present disclosure. As shown in fig. 8, when a node is in a power-down state (corresponding to the "power-down state" in fig. 8), it is determined whether a power-down exit condition is triggered (corresponding to the "power-down exit condition" in fig. 8)

") when a power down exit condition occurs, such as system interrupt, exception, etc., the processing system needs to exit the power down state. For the first node, the power consumption control unit may preferentially power up the interconnect interface and the data network (corresponding to "power up the interconnect interface and power up the data network" in fig. 8), then read the values of the configuration registers from the general memory of one or more target nodes (corresponding to "read the values of the configuration registers from the general memory" in fig. 8) to restore the configuration of the data network, and finally control the set of clients to complete a power-up procedure (corresponding to "power up the client" in fig. 8) and restore the power-up status (corresponding to "power up status" in fig. 8). Having storage for kernel nodes or the likeThe power-on process of the node (b) is similar to that of the first node, except that the node with the storage device can read the data of the configuration register from the general memory of the node (b).

For example, the power-up request is generated in at least one of: in the process of storing configuration negotiation, at least one node in a chip where a first node is located is changed from an idle state to a non-idle state; in the power failure negotiation process, at least one node in a processing system where a first node is located is changed from an idle state to a non-idle state; or the processing system where the first node is located is abnormal under the condition that all nodes in the processing system where the first node is located are in the power-down state.

For example, in the process of executing the save configuration negotiation or the save configuration request, if a client set of a certain node (bus node or kernel node) jumps out of an idle state, the node jumps out of the process of executing the save configuration negotiation or the save configuration request, and meanwhile, initiates a power-on reconnection request to an adjacent node. The node receiving the power-on reconnection request is powered on again and broadcasts the power-on reconnection request to other nodes, and finally the whole processing system suspends the storage configuration process and switches back to the working state.

For example, in the power-down negotiation process, if a client set of a node jumps out of an idle state, the node jumps out of the power-down negotiation process, and meanwhile, a power-up reconnection request is initiated to an adjacent node. The node receiving the power-on reconnection request is powered on again and broadcasts the power-on reconnection request to other nodes, and finally the whole processing system suspends the storage configuration process and switches back to the working state.

For example, in the process of executing the power-down request of the interconnect interface or the power-down request of the data network, if the client set of a certain node goes out of the idle state, the power-down process of the interconnect interface or the power-down process of the data network may be completed first, and then the power-up process of the interconnect interface or the data network may be completed by waking up.

For example, after all nodes in the system are powered off, when the system is interrupted, abnormal, and the like, the system needs to be restored to a working state to process the fault. In this case, one or more nodes may initiate a power-on reconnection request, and the node receiving the power-on reconnection request is powered on again and broadcasts a power-on reconnection request to the remaining nodes, and finally the entire processing system may be restored to the operating state.

Another embodiment of the present disclosure provides another power down processing method executed by a target node, and fig. 9 shows a flowchart of another power down processing method provided in at least one embodiment of the present disclosure, and as shown in fig. 9, the another power down processing method includes steps S510 to S530.

Step S510: receiving at least part of data in the data to be backed up sent by a first node in the processing system, and saving the at least part of data, wherein the at least part of data is sent by the first node under the condition that the negotiation of saving configuration of the first node is determined to be completed.

Step S520: after saving the at least a portion of the data, a power down negotiation operation is performed.

Step S530: and responding to the determination that the power down negotiation is completed, and performing power down processing on the target node.

For example, before the target node receives and saves the at least part of the data, it may further perform: receiving a save configuration negotiation request issued by the first node, the save configuration negotiation request being initiated by the first node in response to a functional component of the first node being in an idle state; in response to the save configuration negotiation request.

For example, the first node may initiate a storage configuration negotiation in an idle state, and send at least part of the data to be backed up to the target node after determining that the storage configuration negotiation is completed. And after receiving the at least part of data to be backed up, the target node saves the at least part of data to be backed up. And after the target node responds to the save configuration negotiation request, the target node can save the data to be backed up of the target node to a storage device of the target node. Then, the target node may cooperate with other nodes to complete power down negotiation, and after the power down negotiation is completed, the target node may start power down processing, for example, an interconnection interface and a data network may be powered down.

For example, the target node may also perform: receiving a configuration saving request sent by the first node, wherein the configuration saving request is initiated by the first node under the condition that the configuration saving negotiation is determined to be completed; the at least a portion of the data is transmitted by the first node in response to the save configuration request.

For example, the data to be backed up of the first node may include a plurality of data sets, for example, the data to be backed up is divided into a plurality of data sets. Receiving at least part of data in the data to be backed up of the first node and saving the at least part of data may include: one of the plurality of data sets is received and the received data set is saved.

For example, the first node is a bus node, the functional components of which include bus components; the target node is a core node, and the functional components of the core node include data processing components.

For example, the target node is located on the same chip as the first node. Responding to the save configuration negotiation request may include: the saved configuration negotiation request is responded to with the feature of the target node in an idle state.

For example, the processing system includes a plurality of chips, each of which includes a plurality of nodes. Determining that the power down negotiation is complete may include: and determining that all nodes in the processing system where the target node is located respond to the power-down negotiation request, and responding to the power-down negotiation request by each node in the processing system under the condition that respective functional units are in an idle state.

For example, the target node may also perform: in response to receiving a power-up request, the at least a portion of the data is sent to the first node to cause the first node to configure the first node based on the at least a portion of the data.

For example, the power down processing method executed by the target node may refer to fig. 3 to fig. 8 and the above description about corresponding contents, which are not described herein again.

Another embodiment of the present disclosure provides another power down processing method, which is executed by a first node and a second node in common. Fig. 10 is a flowchart illustrating another power down processing method according to at least one embodiment of the present disclosure, where as shown in fig. 10, the another power down processing method includes steps S610 to S650.

Step S610: a save configuration negotiation request is initiated by a first node in a processing system in response to a feature of the first node being in an idle state.

Step S620: and receiving a storage configuration negotiation request through at least one object node in the N second nodes in the processing system, and responding to the storage configuration negotiation request (N is a positive integer).

Step S630: and in response to the first node determining that the negotiation for saving the configuration is completed, providing the data to be backed up of the first node to at least one target node of the N second nodes through the first node, and performing backup saving on the data to be backed up through the at least one target node.

Step S640: and after the data to be backed up of the first node is sent to the at least one target node, performing power down negotiation at least through the first node and the N second nodes.

Step S650: and responding to the completion of the power down negotiation, and performing power down processing on the first node and the N second nodes.

For example, the power down processing method executed by the first node and the second node together may refer to fig. 3 to fig. 8 and the description about the corresponding content, which are not described herein again.

Fig. 11 is a block diagram illustrating a functional node according to at least one embodiment of the present disclosure, and as shown in fig. 11, the functional node 700 includes a save negotiation module 710, a backup data module 720, a first power down negotiation module 730, and a first power down processing module 740.

The save negotiation module 710 is configured to initiate a save configuration negotiation request in response to a functional component of the functional node being in an idle state, so as to perform a save configuration negotiation with at least one object node of N second nodes located in the same processing system, where N is a positive integer. The save negotiation module 710 may, for example, perform step S210 described in fig. 3.

The backup data module 720 is configured to provide the data to be backed up of the functional node to at least one target node of the N second nodes for backup storage in response to the functional node determining that the storage configuration negotiation is completed. The backup data module 720 may perform step S220 described in fig. 3, for example.

The first power down negotiation module 730 is configured to perform a power down negotiation operation after the data to be backed up of the functional node is sent to the at least one target node. The first power down negotiation module 730 may, for example, perform step S230 described in fig. 3.

The first power down processing module 740 is configured to perform power down processing in response to the functional node determining that power down negotiation is complete. The first power down processing module 740 may perform, for example, step S240 described in fig. 3.

For example, save negotiation module 710, backup data module 720, first power down negotiation module 730, and first power down processing module 740 may be hardware, software, firmware, or any feasible combination thereof. For example, the save negotiation module 710, the backup data module 720, the first power down negotiation module 730, and the first power down processing module 740 may be dedicated or general circuits, chips, or devices, and may also be a combination of a processor and a memory. The embodiments of the present disclosure are not limited in this regard to the specific implementation forms of the above units.

It should be noted that, in the embodiment of the present disclosure, each module of the function node 700 corresponds to each step of the foregoing power failure processing method corresponding to fig. 3, and for a specific function of the function node 700, reference may be made to related description about the power failure processing method, which is not described herein again. The components and structure of the functional node 700 shown in fig. 11 are exemplary only, and not limiting, and the functional node 700 may include other components and structures as desired.

Another embodiment of the present disclosure provides another functional node, and fig. 12 shows a block diagram of another functional node provided in at least one embodiment of the present disclosure, and as shown in fig. 12, the functional node 800 includes a data receiving module 810, a second power down negotiation module 820, and a second power down processing module 830.

The data receiving module 810 is configured to receive at least part of data to be backed up of a first node located in the same processing system as the functional node, and save the at least part of data to be backed up, where the at least part of data to be backed up is sent by the first node when it is determined that the negotiation for saving the configuration is completed. The data receiving module 810 may perform, for example, step S510 described in fig. 9.

Second power down negotiation module 820 is configured to perform a power down negotiation operation after saving the at least part of the data to be backed up. The second power down negotiation module 820 may perform step S520 described in fig. 9, for example.

The second power down processing module 830 is configured to perform power down processing in response to determining that the power down negotiation is complete. The second power down processing module 830 may perform step S530 described in fig. 9, for example.

For example, data receiving module 810, second power down negotiation module 820, and second power down processing module 830 may be hardware, software, firmware, or any feasible combination thereof. For example, the data receiving module 810, the second power down negotiation module 820 and the second power down processing module 830 may be dedicated or general circuits, chips or devices, and may also be a combination of a processor and a memory. The embodiments of the present disclosure are not limited in this regard to the specific implementation forms of the above units.

It should be noted that, in the embodiment of the present disclosure, each module of the function node 800 corresponds to each step of the power down processing method corresponding to fig. 9, and for a specific function of the function node 800, reference may be made to related description about the power down processing method, which is not described herein again. The components and structure of functional node 800 shown in fig. 12 are exemplary only, and not limiting, and functional node 800 may include other components and structures as desired.

Another embodiment of the present disclosure also provides a processing system comprising a first node and N second nodes, the first node comprising a feature configured to initiate a save configuration negotiation request in response to the feature being in an idle state. The N second nodes are configured to receive the saved configuration negotiation request by at least one object node and respond to the saved configuration negotiation request. The first node is further configured to: and in response to determining that the negotiation for saving the configuration is completed, providing the data to be backed up of the first node to at least one target node in the N second nodes for backup saving. The first node and the N second nodes are further configured to: after the data to be backed up of the first node is sent to the at least one target node, performing power failure negotiation operation; and responding to the completion of power failure negotiation, and performing power failure processing, wherein N is a positive integer. For the processing system, reference may be made to the above-mentioned related description, which is not repeated herein.

At least one embodiment of the present disclosure also provides an electronic device comprising a processor and a memory, the memory including one or more computer program modules. One or more computer program modules are stored in the memory and configured to be executed by the processor, the one or more computer program modules including instructions for implementing the power down handling method described above. The electronic equipment can store the data to be backed up of the first node into a target node different from the first node before power failure negotiation, and the data to be backed up of the first node is prevented from being lost due to power failure.

Fig. 13 is a schematic block diagram of an electronic device provided in some embodiments of the present disclosure. As shown in fig. 13, the electronic device 900 includes a processor 910 and a memory 920. Memory 920 stores non-transitory computer-readable instructions (e.g., one or more computer program modules). The processor 910 is configured to execute non-transitory computer readable instructions, which when executed by the processor 910 may perform one or more of the steps of the power down handling method described above. The memory 920 and the processor 910 may be interconnected by a bus system and/or other form of connection mechanism (not shown).

For example, the processor 910 may be a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), or other form of processing unit having data processing capabilities and/or program execution capabilities. For example, the Central Processing Unit (CPU) may be an X86 or ARM architecture or the like. The processor 910 may be a general-purpose processor or a special-purpose processor that may control other components in the electronic device 900 to perform desired functions.

For example, memory 920 may include any combination of one or more computer program products that may include various forms of computer-readable storage media, such as volatile memory and/or non-volatile memory. Volatile memory can include, for example, Random Access Memory (RAM), cache memory (or the like). The non-volatile memory may include, for example, Read Only Memory (ROM), a hard disk, an Erasable Programmable Read Only Memory (EPROM), a portable compact disc read only memory (CD-ROM), USB memory, flash memory, and the like. One or more computer program modules may be stored on the computer-readable storage medium and executed by processor 910 to implement various functions of electronic device 900. Various applications and various data, as well as various data used and/or generated by the applications, and the like, may also be stored in the computer-readable storage medium.

It should be noted that, in the embodiment of the present disclosure, reference may be made to the above description on the power down processing method for specific functions and technical effects of the electronic device 900, and details are not described here again.

Fig. 14 is a schematic block diagram of another electronic device provided by some embodiments of the present disclosure. The electronic device 1000 is, for example, suitable for implementing the power down processing method provided by the embodiment of the present disclosure. The electronic device 1000 may be a terminal device or the like. It should be noted that the electronic device 1000 shown in fig. 14 is only one example, and does not bring any limitation to the functions and the scope of use of the embodiments of the present disclosure.

As shown in fig. 14, electronic device 1000 may include a processing means (e.g., central processing unit, graphics processor, etc.) 1010 that may perform various appropriate actions and processes in accordance with a program stored in a Read Only Memory (ROM) 1020 or a program loaded from storage device 1080 into a Random Access Memory (RAM) 1030. In the RAM1030, various programs and data necessary for the operation of the electronic apparatus 1000 are also stored. The processing device 1010, the ROM 1020, and the RAM1030 are connected to each other by a bus 1040. An input/output (I/O) interface 1050 is also connected to bus 1040.

Generally, the following devices may be connected to the I/O interface 1050: input devices 1060 including, for example, a touch screen, touch pad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, or the like; an output device 1070 including, for example, a Liquid Crystal Display (LCD), a speaker, a vibrator, or the like; storage 1080 including, for example, tape, hard disk, etc.; and a communication device 1090. The communication means 1090 may allow the electronic device 1000 to communicate wirelessly or by wire with other electronic devices to exchange data. While fig. 14 illustrates the electronic device 1000 with various means, it is to be understood that not all illustrated means are required to be implemented or provided, and that the electronic device 1000 may alternatively be implemented or provided with more or less means.

For example, according to an embodiment of the present disclosure, the above power down processing method may be implemented as a computer software program. For example, embodiments of the present disclosure include a computer program product comprising a computer program carried on a non-transitory computer readable medium, the computer program comprising program code for performing the power down handling method described above. In such embodiments, the computer program may be downloaded and installed from a network through communication device 1090, or from storage device 1080, or from ROM 1020. When executed by the processing device 1010, the computer program may implement the functions defined in the power failure processing method provided by the embodiment of the present disclosure.

At least one embodiment of the present disclosure also provides a computer-readable storage medium storing non-transitory computer-readable instructions that, when executed by a computer, implement the power-down processing method described above. By using the computer readable storage medium, the data to be backed up of the first node can be stored in a target node different from the first node before power failure negotiation, so that the data to be backed up of the first node is prevented from being lost due to power failure.

Fig. 15 is a schematic diagram of a computer-readable storage medium according to some embodiments of the present disclosure. As shown in fig. 15, computer-readable storage medium 1100 stores non-transitory computer-readable instructions 1110. For example, the non-transitory computer readable instructions 1110, when executed by a computer, perform one or more steps according to the power down handling methods described above.

The storage medium 1100 may be applied to the electronic apparatus 900 described above, for example. The storage medium 1100 may be, for example, the memory 920 in the electronic device 900 shown in fig. 13. For example, the related description about the storage medium 1100 may refer to the corresponding description of the memory 920 in the electronic device 900 shown in fig. 13, and is not repeated here.

The following points need to be explained:

(1) the drawings of the embodiments of the disclosure only relate to the structures related to the embodiments of the disclosure, and other structures can refer to common designs.

(2) Without conflict, embodiments of the present disclosure and features of the embodiments may be combined with each other to arrive at new embodiments.

The above description is only a specific embodiment of the present disclosure, but the scope of the present disclosure is not limited thereto, and the scope of the present disclosure should be subject to the scope of the claims.

Claims

1. A power failure processing method is used for processing a first node in a system, and is characterized by comprising the following steps:

responding to the functional component of the first node in an idle state, and initiating a storage configuration negotiation request to perform storage configuration negotiation with at least one object node in N second nodes in the processing system;

in response to the first node determining that the configuration storage negotiation is completed, providing the data to be backed up of the first node to at least one target node of the N second nodes for backup storage;

after the data to be backed up of the first node is sent to the at least one target node, executing power failure negotiation operation;

performing power down processing on the first node in response to the first node determining that power down negotiation is complete,

wherein N is a positive integer.

2. The power-down processing method according to claim 1, wherein in response to the first node determining that the negotiation for saving the configuration is completed, providing data to be backed up of the first node to at least one target node of the N second nodes for backup saving includes:

sending a save configuration request to the at least one target node in response to the first node determining that a save configuration negotiation is complete;

after receiving a response of the at least one target node to the storage configuration request, sending the data to be backed up of the first node to the at least one target node, so that the at least one target node backs up and stores the data to be backed up of the first node.

3. The power-down processing method according to claim 1, wherein the at least one target node includes a plurality of target nodes;

in response to the first node determining that the negotiation for saving the configuration is completed, providing the data to be backed up of the first node to at least one target node of the N second nodes for backup saving, including:

in response to the first node determining that the configuration storage negotiation is completed, dividing the data to be backed up of the first node into a plurality of data sets;

and respectively providing the plurality of data sets to the plurality of target nodes for backup and storage.

4. The power-down processing method according to claim 1,

the first node does not comprise a storage device for backup storage;

the at least one target node comprises a storage device for backup storage.

5. The power-down processing method according to claim 4,

the first node is a bus node, and the functional components of the bus node comprise bus components;

the target node is a core node, and the functional components of the core node comprise data processing components.

6. The power-down processing method according to claim 1,

the at least one object node and the first node are located on the same chip;

determining that the save configuration negotiation is complete, comprising:

determining that the at least one object node has responded to the save configuration negotiation request, wherein each of the at least one object node responds to the save configuration negotiation request with its respective feature in an idle state.

7. Power loss handling method according to any one of claims 1 to 5,

the processing system comprises a plurality of chips, each of the chips comprising a plurality of nodes;

determining that the power down negotiation is complete comprises:

determining that all nodes in a processing system where the first node is located respond to the power-down negotiation request, wherein each node in the processing system responds to the power-down negotiation request under the condition that respective functional units are in an idle state.

8. The power-down processing method according to claim 7, further comprising:

and responding to the first node to determine that the power-down negotiation is completed, and performing power-down processing on all nodes in the processing system.

9. The power failure handling method according to any one of claims 1 to 6, further comprising:

in response to receiving a power-on request, obtaining the data to be backed up of the first node which is backed up and stored from the at least one target node;

and configuring the first node based on the acquired data to be backed up of the first node so as to restore the first node to a state before power failure processing.

10. The power down processing method of claim 9, wherein the power up request is generated in at least one of:

in the process of the configuration storage negotiation, at least one node in a chip where the first node is located is changed from an idle state to a non-idle state;

in the power failure negotiation process, at least one node in a processing system where the first node is located is changed from an idle state to a non-idle state; or

And under the condition that all nodes in the processing system where the first node is located are in a power-down state, the processing system where the first node is located is abnormal.

11. A power failure processing method is used for processing a target node in a system and is characterized by comprising the following steps:

receiving at least part of data in the data to be backed up sent by a first node in the processing system, and saving the at least part of data, wherein the at least part of data is sent by the first node under the condition of determining that the negotiation for saving the configuration is completed;

after saving the at least part of the data, performing a power down negotiation operation;

and responding to the determination that the power down negotiation is completed, and performing power down processing on the target node.

12. The power-down processing method according to claim 11, further comprising:

receiving a save configuration request sent by the first node, wherein the save configuration request is initiated by the first node when the save configuration negotiation is determined to be completed;

and responding to the save configuration request to enable the first node to send the at least part of data.

13. The power-down processing method according to claim 11, wherein the data to be backed up of the first node includes a plurality of data sets;

receiving at least part of data in the data to be backed up sent by a first node in the processing system, and storing the at least part of data, including:

receiving one of the plurality of data sets and saving the received data set.

14. The power-down processing method according to claim 11, further comprising:

receiving a save configuration negotiation request issued by the first node prior to the target node receiving and saving the at least part of the data, wherein the save configuration negotiation request is initiated by the first node in response to a functional component of the first node being in an idle state;

and responding to the storage configuration negotiation request.

15. The power-down processing method according to claim 14, further comprising:

and after responding to the storage configuration negotiation request, storing the data to be backed up of the target node into a storage device of the target node.

16. The power-down processing method according to claim 14,

17. The power-down processing method according to claim 14, wherein the target node is located on the same chip as the first node;

responding the save configuration negotiation request, including:

responding to the save configuration negotiation request with a feature of the target node in an idle state.

18. The power loss processing method according to any one of claims 11 to 16,

determining that the power down negotiation is complete, including:

and determining that all nodes in the processing system where the target node is located respond to the power-down negotiation request, wherein each node in the processing system responds to the power-down negotiation request under the condition that respective functional components are in an idle state.

19. The power-down processing method according to any one of claims 11 to 16, further comprising:

in response to receiving a power-up request, sending the at least part of the data to the first node to cause the first node to configure the first node based on the at least part of the data.

20. A power failure processing method is characterized by comprising the following steps:

initiating, by a first node in a processing system, a save configuration negotiation request in response to a functional component of the first node being in an idle state;

receiving a configuration storage negotiation request through at least one object node in N second nodes in the processing system, and responding to the configuration storage negotiation request;

in response to the first node determining that the configuration storage negotiation is completed, providing the data to be backed up of the first node to at least one target node of the N second nodes through the first node, and performing backup storage on the data to be backed up through the at least one target node;

after the data to be backed up of the first node is sent to the at least one target node, performing power down negotiation at least through the first node and the N second nodes;

performing power down processing on the first node and the N second nodes in response to completion of the power down negotiation,

wherein N is a positive integer.

21. A functional node, comprising:

the saving negotiation module is configured to respond to that the functional component of the functional node is in an idle state, initiate a saving configuration negotiation request, and perform saving configuration negotiation with at least one object node in N second nodes in the same processing system;

the backup data module is configured to respond to the functional node determining that the storage configuration negotiation is completed, and provide the data to be backed up of the functional node to at least one target node of the N second nodes for backup storage;

the first power down negotiation module is configured to execute power down negotiation operation after the data to be backed up of the function node is sent to the at least one target node;

a first power down processing module configured to perform power down processing in response to the functional node determining that power down negotiation is complete,

wherein N is a positive integer.

22. A functional node, comprising:

a data receiving module configured to receive at least part of data to be backed up of a first node located in the same processing system as the functional node, and store the at least part of data to be backed up, wherein the at least part of data to be backed up is sent by the first node when the negotiation for storing configuration is determined to be completed;

the second power failure negotiation module is configured to execute power failure negotiation operation after the at least part of the data to be backed up is stored;

and the second power down processing module is configured to perform power down processing in response to the determination that the power down negotiation is completed.

23. A processing system, comprising:

a first node comprising a functional component configured to initiate a save configuration negotiation request in response to the functional component being in an idle state;

n second nodes, configured as at least one object node to receive the negotiation request of saving configuration and respond to the negotiation request of saving configuration;

wherein the first node is further configured to: in response to determining that the configuration negotiation is finished, providing the data to be backed up of the first node to at least one target node in the N second nodes for backup and storage;

the first node and the N second nodes are further configured to: after the data to be backed up of the first node is sent to the at least one target node, performing power failure negotiation operation; and performs power down processing in response to completion of the power down negotiation,

wherein N is a positive integer.

24. An electronic device, comprising:

a processor;

a memory including one or more computer program modules;

wherein the one or more computer program modules are stored in the memory and configured to be executed by the processor, the one or more computer program modules comprising instructions for implementing the power down handling method of any of claims 1-20.

25. A computer-readable storage medium storing non-transitory computer-readable instructions which, when executed by a computer, implement the power-down processing method of any one of claims 1-20.