Abstract: Embodiments may comprise N-path filter circuitry with tunable radio frequency selectivity and up to 80 decibels per decade roll-off. The N-path filter may comprise at least one input transistor, wherein the at least one input transistor comprises a channel and a gate. A first end of the channel is coupled with a receiver circuitry input, wherein a second end of the channel is coupled with a load. The gate of the at least one input transistor is coupled with a clock circuitry input. The load may comprise a fourth order, all-pole driving point impedance. The impedance may shunt the second end of the channel to a circuit ground or a low voltage circuit rail via the impedance. And the impedance may comprise a first active impedance circuit coupled in series with a second active impedance circuit.

Abstract: This disclosure describes systems, methods, and devices related to traffic indications for multi-link devices (MLDs). A device may generate a first traffic indication map (TIM) with a first bitmap including a first indication that traffic is to be sent by a first access point (AP) device of the MLD to a first non-AP device of a second MLD using a first communication link. The device may generate a second TIM with a second bitmap including a second indication that no traffic is to be sent by a second AP device of the MLD to a second non-AP device of the second MLD using a second communication link. The device may send, using the first communication link, the beacon, the beacon including the first TIM and the second TIM. The device may send, using the first communication link, a data frame to the first non-AP device of the second MLD.

Abstract: System, method, and apparatus embodiments for creating, using, and managing protected cryptography keys are described. In an embodiment, an apparatus includes a decoder, an execution unit, and a cache. The decoder is to decode a single instruction into a decoded single instruction, the single instruction having a first source operand to specify encrypted data and a second source operand to specify a handle including a first including ciphertext of an encryption key, an integrity tag, and additional authentication data.

Abstract: Methods, apparatuses, and computer readable media for communicating elements between multi-link devices are disclosed. Apparatuses of a multi-link device (MLD) are disclosed, where the apparatuses comprise processing circuitry configured to encode a management frame, the management frame comprising a link information field, the link information field indicating a first link of the MLD for which management information is applicable and configure a non-access point (AP) of the MLD or a station (STA) of the MUD to transmit the management frame on a second link of the MLD. The processing circuitry is further configured to decode a second management frame, the second management frame comprising a second link information field, the second link information field indicating a second link of the MLD for which second management information is applicable.

Abstract: An Automatic Gain Control (AGC) SERDES circuit may be used to provide improved gain control for SERDES operation. This AGC SERDES circuit uses an initial gain convergence to determine and store an initial gain level. Once the initial gain convergence is complete, the AGC SERDES circuit uses a signal peak tracking to reduce or prevent saturation events. By setting the gain target based on tracked changes in the equalizer coefficients, the AGC SERDES circuit adapts the gain target to reduce or prevent saturation events and provide the improved communication throughput. A SERDES receiver circuit also provides improved performance using an improved convergence flow within its subcomponent blocks. The improved convergence flow also provides the ability to track environmental changes, voltage changes, and changes to input parameters, and can be performed while data is running on the link to provide continuously improved communication channel performance.

Abstract: Technologies for dynamic accelerator selection include a compute sled. The compute sled includes a network interface controller to communicate with a remote accelerator of an accelerator sled over a network, where the network interface controller includes a local accelerator and a compute engine. The compute engine is to obtain network telemetry data indicative of a level of bandwidth saturation of the network. The compute engine is also to determine whether to accelerate a function managed by the compute sled. The compute engine is further to determine, in response to a determination to accelerate the function, whether to offload the function to the remote accelerator of the accelerator sled based on the telemetry data. Also the compute engine is to assign, in response a determination not to offload the function to the remote accelerator, the function to the local accelerator of the network interface controller.

Abstract: Technologies for offloading acceleration task scheduling operations to accelerator sleds include a compute device to receive a request from a compute sled to accelerate the execution of a job, which includes a set of tasks. The compute device is also to analyze the request to generate metadata indicative of the tasks within the job, a type of acceleration associated with each task, and a data dependency between the tasks.

Abstract: In an embodiment, a host system for selecting a power supply includes a processor, a bus interface to connect to a peripheral device, and a power controller. The power controller may be to: determine whether the processor has entered a reduced power mode; determine, via one or more bus messages, whether charging is to be performed for a battery of the peripheral device; and in response to a determination that the processor has entered the reduced power mode and that charging is not to be performed for the battery of the peripheral device, switch from a first power supply to a second power supply as an active power source of the host system. Other embodiments are described and claimed.

Abstract: A timer intellectual property (IP) block that automatically determines an interval on which a processor circuitry is to be woken up to service periodic events, when it is given details about the requirements for those events (e.g., approximately how often they must occur, if it's important that they not happen too frequently or too infrequently, if the total number of events over a long average is important, etc.). For each periodic event that firmware must handle, the IP provides an Application Programming Interface (API) to register details of that event. The firmware configures all the events that it requires during system configuration, although it is possible to add, remove or modify individual events at runtime. At runtime, the optimized timer IP will interrupt the processor circuitry whenever one or more events need to be handled, based on a batching algorithm.

Abstract: Embodiments include semiconductor packages and a method to form such semiconductor packages. A semiconductor package includes a plurality of dies on a substrate, and an encapsulation layer over the substrate. The encapsulation layer surrounds the dies. The semiconductor package also includes a plurality of dummy silicon regions on the substrate. The dummy silicon regions surround the dies and encapsulation layer. The plurality of dummy silicon regions are positioned on two or more edges of the substrate. The dummy silicon regions have a top surface substantially coplanar to a top surface of the dies. The dummy silicon regions include materials that include silicon, metals, or highly-thermal conductive materials. The materials have a thermal conductivity of approximately 120 W/mK or greater, or is equal to or greater than the thermal conductivity of silicon. An underfill layer surrounds the substrate and the dies, where the encapsulation layer surrounds portions of the underfill layer.

Abstract: A system can predict memory device failure through identification of correctable error patterns based on the memory architecture. The failure prediction can thus account for the circuit-level of the memory rather than the mere number or frequency of correctable errors. A failure prediction engine correlates hardware configuration of the memory device with correctable errors (CEs) detected in data of the memory device to predict an uncorrectable error (UE) based on the correlation.

Abstract: An apparatus may comprise multiplexing circuitry to select an ingress lane from among a plurality of ingress lanes to couple to an egress lane; and retiming circuitry to retime a signal received on the selected ingress lane and transmit the retimed signal on the egress lane.

Abstract: Embodiments detailed herein relate to matrix operations. In particular, support for matrix (tile) addition, subtraction, and multiplication is described. For example, circuitry to support instructions for element-by-element matrix (tile) addition, subtraction, and multiplication are detailed. In some embodiments, for matrix (tile) addition, decode circuitry is to decode an instruction having fields for an opcode, a first source matrix operand identifier, a second source matrix operand identifier, and a destination matrix operand identifier; and execution circuitry is to execute the decoded instruction to, for each data element position of the identified first source matrix operand: add a first data value at that data element position to a second data value at a corresponding data element position of the identified second source matrix operand, and store a result of the addition into a corresponding data element position of the identified destination matrix operand.

Abstract: Universal Serial Bus (USB) Power Delivery is augmented by allowing devices that attach to the USB to include and/or have access to an enhanced device policy manager (eDPM) so that device information such as status, state, or requirements, such as power requirements, may be at least be shared by the eDPMs between, for example, a host device on the bus, a secondary device providing power to devices on the bus, and a new device attaching to the bus. Sharing device information facilitates the host having contextual awareness for the attaching device and assists with determining whether the attaching device may be enumerated on the bus. If not, such as due to insufficient power available from the secondary device, the host and/or secondary device may seek to influence bus devices to change an operating mode to accommodate the attaching device.

Abstract: Techniques for real-time updating of encryption keys are disclosed. In the illustrative embodiment, an encrypted link is established between a local and remote processor over a point-to-point interconnect. The encrypted link is operated for some time until the encryption key should be updated. The local processor sends a key update message to the remote processor notifying the remote processor of the change. The remote processor prepares for the change and sends a key update confirmation message to the local processor. The local processor then sends a key switch message to the remote processor. The local processor pauses transmission of encrypted message while the remote processor completes use of the encrypted message. After a pause, the local processor continues sending encrypted messages with the updated encryption key.

Abstract: Contact over active gate (COAG) structures with etch stop layers, and methods of fabricating contact over active gate (COAG) structures using etch stop layers, are described. In an example, an integrated circuit structure includes a plurality of gate structures above substrate, each of the gate structures including a gate insulating layer thereon. A plurality of conductive trench contact structures is alternating with the plurality of gate structures, each of the conductive trench contact structures including a trench insulating layer thereon. A first dielectric etch stop layer is directly on and continuous over the trench insulating layers and the gate insulating layers. A second dielectric etch stop layer is directly on and continuous over the first dielectric etch stop layer, the second dielectric etch stop layer distinct from the first dielectric etch stop layer. An interlayer dielectric material is on the second dielectric etch stop layer.

Abstract: An apparatus to facilitate disaggregated computing for a distributed confidential computing environment is disclosed. The apparatus includes a graphics processing unit (GPU) to: provide a virtual GPU monitor (VGM) to interface over a network with a middleware layer of a client platform, the VGM to interface with the middleware layer using a message passing interface; configure and expose, by the VGM, virtual functions (VFs) of the GPU to the middleware layer of the client platform; intercept, by the VGM, request messages directed to the GPU from the middleware layer, the request messages corresponding to VFs of the GPU to be utilized by the client platform; and generate, by the VGM, a response to the request messages for the middleware client.

Abstract: Various examples provide a semiconductor package. The semiconductor package includes a substrate having first and second opposed substantially planar major surfaces extending in an x-y direction. The package further includes a bridge die having third and fourth opposed substantially planar major surfaces extending in the x-y direction. The third substantially planar major surface of the bridge die is in direct contact with the second substantially planar major surface of the substrate. The semiconductor package further includes a through silicon via extending in a z-direction through the first substantially planar major surface of the substrate and the fourth substantially planar major surface of the bridge die. The semiconductor package further includes a power source coupled to the through silicon via, a first electronic component electronically coupled to the bridge die, and a second electronic component electronically coupled to the bridge die.

Abstract: Systems and techniques for mobility-as-a-service for user experience are described herein. An orchestration log may be maintained that includes current orchestration data. An orchestration backup record may be generated that includes alternate MaaS nodes on the MaaS network. It may be determined that connectivity is lost to a first orchestration container hosted by a first MaaS node. An orchestration container is generated using the orchestration log to maintain orchestration functionality. An available second MaaS node is identified from the alternate MaaS nodes. The orchestration container may be transferred to the second MaaS node.

Abstract: Methods, apparatus, systems, and articles of manufacture are disclosed to implement parallel architectures for neural network classifiers. An example non-transitory computer readable medium comprises instructions that, when executed, cause a machine to at least: process a first stream using first neural network blocks, the first stream based on an input image; process a second stream using second neural network blocks, the second stream based on the input image; fuse a result of the first neural network blocks and the second neural network blocks; perform average pooling on the fused result; process a fully connected layer based on the result of the average pooling; and classify the image based on the output of the fully connected layer.