Abstract: Devices, systems, and methods for temporary link performance elevation are disclosed herein. In one embodiment, a link performance elevation (LPE) request is received. The LPE request is a request to temporarily boost performance of a network link between an endpoint and a service provider for a finite duration. Based on the LPE request, a temporary performance boost is activated for the network link at the start of the finite duration, and the temporary performance boost is deactivated for the network link at the end of the finite duration.

Abstract: Method, systems and apparatuses may include technology that identifies a second computing platform that includes a second hardware device that satisfies one or more conditions. The second hardware device is associated with a hardware abstraction layer on the second computing platform. The second computing platform is coupled to a first computing platform. The technology may further include generating a virtual hardware abstraction layer that is to represent the hardware abstraction layer on the second computing platform.

Abstract: Apparatus and method for processing virtual graphics processor telemetry data based on quanta.

Abstract: Embodiments of the present disclosure are based on using transistors with back-side contacts. Such transistors enable back-side power delivery to IC components (e.g., transistors, etc.) of an IC structure, which may be more advantageous than front-side power delivery in some implementations. Embodiments of the present disclosure are further based on recognition that using a glass support structure at the front side of an IC structure with back-side power delivery may advantageously reduce parasitic effects in the IC structure, e.g., compared to using a silicon-based support structure at the front.

Abstract: A port is to couple to another die over a die-to-die (D2D) link and includes physical layer (PHY) circuitry including a first number of sideband lanes to carry data for use in training and management of the D2D link, and a second number of mainband lanes to implement a main data path of the D2D link. The mainband lanes include a forwarded clock lane, a valid lane, and a plurality of data lanes. A logical PHY coordinates functions of the sideband lanes and the mainband lanes.

Abstract: An integrated circuit interconnect structure includes a first interconnect in a first metallization level and a first dielectric adjacent to at least a portion of the first interconnect, where the first dielectric having a first carbon content. The integrated circuit interconnect structure further includes a second interconnect in a second metallization level above the first metallization level. The second interconnect includes a lowermost surface in contact with at least a portion of an uppermost surface of the first interconnect. A second dielectric having a second carbon content is adjacent to at least a portion of the second interconnect and the first dielectric. The first carbon concentration increases with distance away from the lowermost surface of the second interconnect and the second carbon concentration increases with distance away from the uppermost surface of the first interconnect.

Abstract: Techniques for wireless audio device selection are disclosed. In the illustrative embodiment, a compute device uses visual training to recognize headphones worn by a user. When the user puts on the headphones, the compute device may recognize the headphones and automatically route audio output to the headphones. In some embodiments, when selecting from several possible wireless audio output devices, the compute device may determine a distance to each of the wireless audio output devices and select one of the wireless audio output devices based on the distance to the wireless audio output device.

Abstract: Techniques for package loading mechanisms are disclosed. In the illustrative embodiment, a base portion of a laptop includes a circuit board on which an integrated circuit component is mounted. A heat sink is mated with the integrated circuit component. A spring presses against part of the chassis of the laptop, pressing the integrated circuit component and the heat sink together, providing strong thermal coupling between them.

Abstract: An apparatus includes a memory to store subscription data for access to a network, the subscription data including at least terms and conditions information for the network; circuitry linked to the memory; and a connection component for execution on the circuitry to identify an access point connected to the network and automatically forward at least a portion of the subscription data to the identified access point in an association message. Other embodiments are disclosed and claimed.

Abstract: A circuit system includes a processing integrated circuit die comprising a first die-to-die interface circuit and a memory interface circuit. The circuit system also includes a second integrated circuit die comprising a second die-to-die interface circuit and a compute circuit that performs computations for the processing integrated circuit die. The first and the second die-to-die interface circuits are coupled together. The compute circuit is coupled to exchange information with the memory interface circuit through the first and the second die-to-die interface circuits.

Abstract: The present disclosure is directed to systems and methods for detecting side-channel exploit attacks such as Spectre and Meltdown. Performance monitoring circuitry includes first counter circuitry to monitor CPU cache misses and second counter circuitry to monitor DTLB load misses. Upon detecting an excessive number of cache misses and/or load misses, the performance monitoring circuitry transfers the first and second counter circuitry data to control circuitry. The control circuitry determines a CPU cache miss to DTLB load miss ratio for each of a plurality of temporal intervals. The control circuitry the identifies, determines, and/or detects a pattern or trend in the CPU cache miss to DTLB load miss ratio. Upon detecting a deviation from the identified CPU cache miss to DTLB load miss ratio pattern or trend indicative of a potential side-channel exploit attack, the control circuitry generates an output to alert a system user or system administrator.

Abstract: For example, an apparatus may include an interference detector configured to detect interference in a radar Received (Rx) signal. The interference detector may include an input to receive Transmit (Tx) parameter information of a radar Tx signal, wherein the radar Rx signal is based on the radar Tx signal; and a processor configured to determine interference detection information of an interfering signal based on the radar Rx signal and the Tx parameter information. The interference detection information may include one or more signal parameters corresponding to a shape of the interfering signal, and interference level information to indicate a level of noise caused by the interfering signal in the radar Rx signal.

Abstract: Methods and apparatus relating to a gold finger design for differential edge cards are described. In one embodiment, a signal finger comprises a first portion to communicate electrical signals between a signal pin of a card connector and a card; and a second portion to provide a mechanical wiping surface for the signal pin. Other embodiments are also claimed and disclosed.

Abstract: Particular embodiments described herein provide for an electronic device can include a support structure, a radiation source on the support structure, a radiation shield around the radiation source, and a hook and loop radiation shield securing mechanism to removably secure the radiation shield to the support structure, where the hook and loop radiation shield securing mechanism includes a hook portion with a plurality of hooks and a loop portion that includes a plurality of loops, where an angle of a retention hook for each of the plurality of hooks is less than about eighty degrees.

Abstract: A circuit system includes an accelerator circuit and a compute circuit. The accelerator circuit generates a request in response to receiving packets of data. The accelerator circuit generates an indication of a low power state based on a reduced number of the packets of data being received. The compute circuit performs a processing operation for the accelerator circuit using the packets of data in response to receiving the request. The compute circuit comprises a power management circuit that decreases a supply voltage in the compute circuit and decreases a frequency of a clock signal in the compute circuit in response to the indication of the low power state from the accelerator circuit.

Abstract: Methods and apparatus relating to a high density cable structure and wire termination are described. In one embodiment, a plug structure includes a paddle card to couple two wires to two gold fingers and a first add-on plug to couple a first wire to a first gold finger. The paddle card and the first add-on plug are to be stacked to form a single plug structure having a first row of gold fingers and a second row of gold fingers. Other embodiments are also claimed and disclosed.

Abstract: An integrated circuit includes logic circuits, first buffer circuits coupled to external ports of the integrated circuit, second buffer circuits that are each coupled to one of the logic circuits, and a crossbar circuit coupled to the first and the second buffer circuits. The crossbar circuit is configurable to provide data transfer between the logic circuits and the external ports of the integrated circuit through the first buffer circuits and the second buffer circuits.

Abstract: On-demand paging support for confidential computing is described. An example of an apparatus includes circuitry including one or more processors including a first processor, the first processor including a TEE and registers, wherein the one or more processors are to: receive a memory access request associated with a trust domain (TD), wherein one or more direct memory access payloads associated with the request being generated by a protocol engine (PE) of a peripheral device and written to a host interface (HIF), the HIF including an address translation engine (ATE); and, in response to a page fault being identified for a payload, divert the payload and forward a payload fault to one or more TD fault buffers in a set of registers, and resolve the page fault by an ATE driver and a virtual machine manager using the TEE.

Abstract: For example, an Access Point (AP) may be configured to determine a plurality of beamformed sectors for communication with a plurality of wireless communication stations (STAs), wherein a beamformed sector for communication with a STA is based on a beamforming training between the AP and the STA; to determine a grouping sector for communication between the AP and a group of STAs including two or more grouped STAs of the plurality of STAs, wherein the grouping sector covers two or more beamformed sectors corresponding to the two or more grouped STAs; to listen over the grouping sector for a transmission from a grouped STA of the group of STAs; and, based on the transmission from the grouped STA, to communicate a beamformed transmission with the grouped STA via a beamformed sector corresponding to the grouped STA.

Abstract: Data integrity logic is executable by a processor to generate a data integrity code using a hardware-based secret. A container manager, executable by the processor, creates a secured container including report generation logic that determines measurements of the secured container, generates a report according to a defined report format, and sends a quote request including the report. The defined report format includes a field to include the measurements and a field to include the data integrity code, and the report format is compatible for consumption by any one of a plurality of different quote creator types.

Abstract: Techniques for cryptographic computing isolation are described. A processor includes circuitry to be coupled to memory configured to store one or more instructions. The circuitry is to execute the one or more instructions to instantiate a first process based on an application. To instantiate the first process is to include creating a context table to be used by the first process, identifying a software component to be invoked during the first process, encrypting the software component using a first cryptographic key, and creating a first entry in the context table. The first entry is to include first context information identifying the encrypted software component and second context information representing the first cryptographic key. In more specific embodiments, third context information representing a first load address of the encrypted software component is stored in the first entry of the context table.

Abstract: Stacked finFET structures including a fin having at least a first layer of semiconductor material stacked over or under a second layer of semiconductor material. The first and second layers may include a Group IV semiconductor material layer and a Group III-V semiconductor material layer, for example. A stacked finFET may include an N-type finFET stacked over or under a P-type finFET, the two finFETs may have channel portions within the different semiconductor material layers. Channel portions of the first and second layers of semiconductor material may be coupled to separate gate electrodes that are vertically aligned. Channel portions of the first and second layers of semiconductor material may be vertically separated by subfin portions of the first and second layers. Different layers of dielectric material adjacent to the subfin portions may improve electrical isolation between the channel portions, for example as a source of fixed charge or impurity dopants.

Abstract: Embodiments provide mechanisms to facilitate compute operations for deep neural networks. One embodiment comprises a graphics processing unit comprising one or more multiprocessors, at least one of the one or more multiprocessors including a register file to store a plurality of different types of operands and a plurality of processing cores. The plurality of processing cores includes a first set of processing cores of a first type and a second set of processing cores of a second type. The first set of processing cores are associated with a first memory channel and the second set of processing cores are associated with a second memory channel.

Abstract: A mechanism is described for facilitating optimization of cache associated with graphics processors at computing devices. A method of embodiments, as described herein, includes introducing coloring bits to contents of a cache associated with a processor including a graphics processor, wherein the coloring bits to represent a signal identifying one or more caches available for use, while avoiding explicit invalidations and flushes.

Abstract: Methods and apparatus relating to phase heterogeneous interconnects for crosstalk reduction are described. In one embodiment, an interconnect includes a plurality of links. A first set of links from the plurality of links communicates signals and a second set of links from the plurality of links provides a return path. The interconnect also includes one or more links from the first set of links that include one or more structures with a larger diameter than a minimum diameter of the one or more links. The larger diameter modifies an inductance or capacitance of the one or more links to provide a heterogenous phase delay amongst the plurality of links. Other embodiments are also claimed and disclosed.

Abstract: An embodiment of an electronic apparatus may comprise one or more substrates and a controller coupled to the one or more substrates, the controller including circuitry to apply a reliability, availability, and serviceability (RAS) policy for access to a memory in accordance with a first RAS scheme, change the applied RAS policy in accordance with a second RAS scheme at runtime, where the second RAS scheme is different from the first RAS scheme, and access the memory in accordance with the applied RAS policy. Other embodiments are disclosed and claimed.

Abstract: In one example an apparatus comprises a computer readable memory, an XMSS verification manager logic to manage XMSS verification functions, a one-time signature and public key generator logic, a chain function logic to implement chain function algorithms, a low latency SHA3 hardware engine, and a register bank communicatively coupled to the XMSS verification manager logic. Other examples may be described.

Abstract: Methods and apparatus relating to a Universal Serial Bus (USB) cable paddle card design for wire termination are described. In one embodiment, a cable paddle card includes: a first plug connector pad coupled to a first transmission line; a second plug connector pad coupled to a second transmission line; a third plug connector pad coupled to a third transmission line; and a fourth plug connector pad coupled to a fourth transmission line. The first transmission line and the second transmission line are to transmit signals received from a cable and the third transmission line and the fourth transmission line are to receive signals to be transmitted to the cable. Other embodiments are also claimed and disclosed.

Abstract: A flattening portal bridge (FPB) is provided to support addressing according to a first addressing scheme and a second, alternative addressing scheme. The FPB comprises a primary side and a secondary side, the primary side connects to a first set of devices addressed according to a first addressing scheme, and the secondary side connects to a second set of devices addressed according to a second addressing scheme. The first addressing scheme uses a unique bus number within a Bus/Device/Function (BDF) address space for each device in the first set of devices, and the second bus addressing scheme uses a unique bus-device number for each device in the second set of devices.

Abstract: Particular embodiments described herein provide for a privacy cover that include a main body and a privacy cover orientation magnet. The privacy cover orientation magnet includes a plurality of positively biased areas and a plurality of negatively biased areas arranged in a specific orientation that allows the plurality of positively biased areas and the plurality of negatively biased areas in the privacy cover orientation magnet interact with a plurality of positively biased areas and a plurality of negatively biased areas in a display orientation magnet to self-orient the privacy cover over a sensor.

Abstract: For example, an apparatus may include a digitally-controlled frequency multiplier, which may be controllable according to a digital control input, to generate an output frequency signal having an output frequency, for example, by multiplying an input frequency of an input frequency signal. For example, the digitally-controlled frequency multiplier may include a phase generator configured to generate a plurality of phase-shifted signal groups corresponding to a respective plurality of first phase-shifts applied to the input frequency signal, a plurality of digital clock multipliers controllable according to the digital control input to generate a respective plurality of frequency-multiplied signals based on the plurality of phase-shifted signal groups, and a combiner to generate the output frequency signal based on a combination of the plurality of frequency-multiplied signals.

Abstract: Disclosed herein are tunneling field effect transistors (TFETs), and related methods and computing devices. In some embodiments, a TFET may include: a first source/drain material having a p-type conductivity; a second source/drain material having an n-type conductivity; a channel material at least partially between the first source/drain material and the second source/drain material, wherein the channel material has a first side face and a second side face opposite the first side face; and a gate above the channel material, on the first side face, and on the second side face.

Abstract: An apparatus to facilitate workload scheduling is disclosed. The apparatus includes one or more clients, one or more processing units to processes workloads received from the one or more clients, including hardware resources and scheduling logic to schedule direct access of the hardware resources to the one or more clients to process the workloads.

Abstract: An apparatus to verify firmware in a computing system, comprising a non-volatile memory, including firmware memory to store agent firmware associated with each of a plurality of interconnect protocol (IP) agents and version memory to store security version numbers (SVNs) included in the agent firmware, a security controller comprising verifier logic to verify an integrity of the version memory by applying a hash algorithm to contents of the version memory to generate a SVN hash, and a trusted platform module (TPM) to store the SVN hash.

Abstract: A circuit system includes a processing circuit, an accelerator circuit, and a buffer circuit that stores packets of data and that is coupled to the processing circuit and to the accelerator circuit. The buffer circuit functions as a crossbar circuit by allowing each of the accelerator circuit and the processing circuit to access at least one of the packets of data stored in the buffer circuit during access to another one of the packets of data stored in the buffer circuit.

Abstract: An apparatus comprising a memory to store an observed trajectory of a pedestrian, the observed trajectory comprising a plurality of observed locations of the pedestrian over a first plurality of timesteps; and a processor to generate a predicted trajectory of the pedestrian, the predicted trajectory comprising a plurality of predicted locations of the pedestrian over the first plurality of timesteps and over a second plurality of timesteps occurring after the first plurality of timesteps; determine a likelihood of the predicted trajectory based on a comparison of the plurality of predicted locations of the pedestrian over the first plurality of timesteps and the plurality of observed locations of the pedestrian over the first plurality of timesteps; and responsive to the determined likelihood of the predicted trajectory, provide information associated with the predicted trajectory to a vehicle to warn the vehicle of a potential collision with the pedestrian.

Abstract: Protocol layer logic in a protocol stack receives an indication that a particular mode is to be utilized on a die-to-die (D2D) link connecting a first device to a second device. The protocol layer logic generates data to be sent on the D2D link to adapt the particular data format to a flit format defined for use on the D2D link in the particular mode, the flit format comprises providing a set of reserved fields to be completed by an adapter block positioned between the protocol circuitry and a physical layer block. The data in the flit format is sent to the data to the adapter block to prepare the data for transmission over the D2D link.

Abstract: An apparatus and method for performing a transform on complex data.

Abstract: Described is a thin film transistor which comprises: a dielectric comprising a dielectric material; a first structure adjacent to the dielectric, the first structure comprising a first material; a second structure adjacent to the first structure, the second structure comprising a second material wherein the second material is doped; a second dielectric adjacent to the second structure; a gate comprising a metal adjacent to the second dielectric; a spacer partially adjacent to the gate and the second dielectric; and a contact adjacent to the spacer.

Abstract: One embodiment provides for a graphics processing unit comprising a processing cluster to perform coarse pixel shading and output shaded coarse pixels for processing by a pixel processing pipeline and a render cache to store coarse pixel data for input to or output from a pixel processing pipeline.

Abstract: One embodiment provides for a graphics processing unit to perform computations associated with a neural network, the graphics processing unit comprising a hardware processing unit having a dynamic precision fixed-point unit that is configurable to quantize elements of a floating-point tensor to convert the floating-point tensor into a dynamic fixed-point tensor.

Abstract: Embodiments described herein provide for an instruction and associated logic to enable a vector multiply add instructions with automatic zero skipping for sparse input. One embodiment provides for a general-purpose graphics processor comprising logic to perform operations comprising fetching a hardware macro instruction having a predicate mask, a repeat count, and a set of initial operands, where the initial operands include a destination operand and multiple source operands. The hardware macro instruction is configured to perform one or more multiply/add operations on input data associated with a set of matrices.

Abstract: For example, an apparatus may be configured to cause a wireless communication station (STA) to generate, process and/or communicate one or more frames and/or messages based on a Quality of Service (QoS) index value to indicate a predefined setting of a set of a plurality QoS parameters. In one example, a STA may be configured to transmit a frame including a QoS index value to another STA. In another example, a STA may be configured to process a frame including a QoS index value from another STA.

Abstract: An apparatus comprising a sorting unit to sort primitives of a graphics image, the primitives to be grouped, each group to form a first level node of a hierarchical acceleration structure; a parallel reconfigurable clustering array to construct the hierarchical acceleration structure, the parallel reconfigurable clustering array comprising a plurality of processing clusters, each cluster comprising: parallel efficiency analysis circuitry to evaluate different groupings of the first level nodes for a next level of the hierarchical acceleration structure to determine efficiency values for the different groupings; and node merge circuitry to merge the first level nodes based on the efficiency values to form second level nodes.

Abstract: Described herein are integrated circuit (IC) structures, devices, and methods associated with device layer interconnects. For example, an IC die may include a device layer including a transistor array along a semiconductor fin, and a device layer interconnect in the transistor array, wherein the device layer interconnect is in electrical contact with multiple different source/drain regions of the transistor array.

Abstract: An apparatus to facilitate supporting 8-bit floating point format operands in a computing architecture is disclosed. The apparatus includes a processor comprising: a decoder to decode an instruction fetched for execution into a decoded instruction, wherein the decoded instruction is a matrix instruction that operates on 8-bit floating point operands to cause the processor to perform a parallel dot product operation; a controller to schedule the decoded instruction and provide input data for the 8-bit floating point operands in accordance with an 8-bit floating data format indicated by the decoded instruction; and systolic dot product circuitry to execute the decoded instruction using systolic layers, each systolic layer comprises one or more sets of interconnected multipliers, shifters, and adder, each set of multipliers, shifters, and adders to generate a dot product of the 8-bit floating point operands.

Abstract: Embodiments are directed to homomorphic encryption for machine learning and neural networks using high-throughput Chinese remainder theorem (CRT) evaluation. An embodiment of an apparatus includes a hardware accelerator to receive a ciphertext generated by homomorphic encryption (HE) for evaluation, decompose coefficients of the ciphertext into a set of decomposed coefficients, multiply the decomposed coefficients using a set of smaller modulus determined based on a larger modulus, and convert results of the multiplying back to an original form corresponding to the larger modulus by performing a reverse Chinese remainder theorem (CRT) transform on the results of multiplying the decomposed coefficients.

Abstract: Techniques for memory tagging are disclosed. In the illustrative embodiment, 16 bits of a virtual memory address are used as memory tag bits. In a page table entry corresponding to the virtual memory address, page tag bits indicate which of the 16 bits of the virtual memory address are to be sent to the memory as memory tag bits when a memory operation is requested on the virtual memory address. The memory can then compare the memory tag bits sent with the physical memory address to memory tag bits stored on the memory that correspond to the physical memory address. If the memory tag bits match, then the operation is allowed to proceed.

Abstract: Technologies for controlling operation of a compute node coupled to a computer network via a computing device that includes communications for communicating with the computer network and persistent instructions such as firmware for providing control functions to the computing device, wherein the control functions being defined at least in part by protocol data. An update control module of the computing device may receive update data from a remote node in the computer network via the communications, wherein the update data comprising new protocol data for the persistent instructions. A protocol parser module may parse the update data and generate metadata relating to the update data. A classifier module may receive rules for the control functions, wherein the rules are based at least in part on the update data and metadata. A compiler may compile the parsed update data to the persistent instructions for providing new control functions to the computing device based at least in part on the received rules.

Abstract: A variable capacitance III-N device having multiple two-dimensional electron gas (2DEG) layers are described. In some embodiments, the device comprises a first source and a first drain; a first polarization layer adjacent to the first source and the first drain; a first channel layer coupled to the first source and the first drain and adjacent to the first polarization layer, the first channel layer comprising a first 2DEG region; a second source and a second drain; a second polarization layer adjacent to the second source and the second drain; and a second channel layer coupled to the second source and the second drain and adjacent to the second polarization layer, the second channel layer comprising a second 2DEG region, wherein the second channel layer is over the first polarization layer.