Implementations of the present disclosure are directed to systems and methods for reducing design complexity and critical path timing challenges of credit return logic. A wide bus supports simultaneous transmission of multiple flits, one per lane of the wide bus. A source device transmitting flits on a wide bus selects from among multiple credit return options to ensure that only one of the multiple flits being simultaneously transmitted includes a credit return value. In some example embodiments, the receiving device checks only the flit of one lane of the wide bus (e.g., lane 0) for credit return data. In other example embodiments, the receiving device uses a bitwise-OR to combine the credit return data of all received flits in a single cycle.

The embodiments herein describe a 3D SmartNIC that spatially distributes compute, storage, or network functions in three dimensions using a plurality of layers. That is, unlike current SmartNIC that can perform acceleration functions in a 2D, a 3D Smart can distribute these functions across multiple stacked layers, where each layer can communicate directly or indirectly with the other layers.

A building management system includes a communications bus, and devices coupled to the communications bus. The devices are coupled to the communications bus and configured to communicate on the communications bus using a master-slave token passing protocol. A first one of the devices has an active node table stored therein. The active node table includes multiple nodes. Each node represents one of the devices participating in a token passing ring used to exchange information among the devices via the communications bus using the master-slave token passing protocol. The first device is configured to monitor the active node table for new nodes and to identify a new device communicating on the communications bus in response to a determination that the active node table includes a new node.

A method for authenticating a peripheral device, that includes detecting, by a baseboard management controller (BMC), a presence of the peripheral device, receiving authentication credentials from the peripheral device, making a determination, based on the authentication credentials, that the peripheral device is authentic, and sending, in response to the determination, a command to open a peripheral communication channel with the peripheral device.

An electronic device is provided. The electronic device includes a first power management integrated circuit (PMIC) with a first fault controller connected to a first node and a first interface circuit connected to a second node; a second PMIC with a second fault controller connected to the first node and a second interface circuit connected to the second node; and a third PMIC with a third fault controller connected to the first node and a third interface circuit connected to the second node. The first fault controller is configured to, during a power on sequence or a power off sequence, detect a change in a voltage level of the first node. The first interface circuit is configured to communicate with any one or any combination of the second interface circuit and communication and the third interface circuit based on the change in the voltage level of the first node.

A DNN hardware accelerator and an operation method of the DNN hardware accelerator are provided. The DNN hardware accelerator includes: a network distributor for receiving an input data and distributing respective bandwidth of a plurality of data types of a target data amount based on a plurality of bandwidth ratios of the target data amount; and a processing element array coupled to the network distributor, for communicating data of the data types of the target data amount between the network distributor based on the distributed bandwidth of the data types.

Packet routing within a multi-chip processing system is shown. A first chip has a first interconnect bus, and a first microprocessor coupled to the first interconnect bus. The first interconnect bus has a first routing register. When the first microprocessor operates the first chip as a source node to output a packet to be transferred to a destination node, routing information indicating a routing path from the source node to the destination node is written into the first routing register and then loaded from the first routing register to a header of the packet. While being transferred within the multi-chip processing system from the source node to the destination node, the packet is guided along the routing path indicated in the routing information carried in the header of the packet.

A computing device includes a host device and a storage device. The host device is configured to receive instruction information via a code bus based on a code address comprised in a code address map of particular address maps and receive data via a system bus that is separate from the code bus based on a data address included in a data address map. The storage device is configured to store target instruction information via the system bus and provide the target instruction information to the host device via the code bus in response to a request from the host device for an object code address included in the code address map and corresponding to the target instruction information.

A PCIe-based data transmission method and an apparatus are provided. The method includes the following: A first node obtains and sends a first TLP; and a switch receives the first TLP, and sends a second TLP to a second node. The first TLP includes a first TLP header and a first data payload, the second TLP includes a second TLP header and the first data payload, the first TLP header and the second TLP header are used to indicate a data type of the first data payload, and the data type of the first data payload includes at least one of audio, an image, control, data stream write, or security. The second node obtains the first data payload based on the data type of the first data payload. This helps reduce communication complexity in a high-speed communication scenario.

An operation method of a sleep mode of an electronic device includes the following steps. A first sub-module of a first module sends a sleep command to a second sub-module of the first module and a third sub-module and a fourth sub-module of a second module, wherein the first sub-module includes first and second modes, the second sub-module includes third and fourth nodes, the third sub-module includes fifth and sixth nodes, and the fourth sub-module includes seventh and eighth nodes. The second sub-module, the third sub-module and fourth sub-module execute a sleep sequence in sequence to enter a sleep mode according to the sleep command. The first node sends the sleep command to the second node to execute the sleep sequence to enter the sleep mode. The first node sends the sleep command to the first node to execute the sleep sequence to enter the sleep mode.