Abstract: A circuit and corresponding method enable mining for digital currency in a blockchain network. The circuit comprises a controller and at least one partial hash engine that (i) implements a hash function, partially, to compute a partial hash digest of a final hash digest for a block header of a block candidate and (ii) generates a notification based on determining that the partial hash digest satisfies a criterion. The controller includes a complete hash engine that implements the hash function, completely. In response to the notification generated, the controller activates the complete hash engine to compute, in its entirety, the final hash digest for the block header, effectuating a decision for submission of the block candidate with the block header to the blockchain network for mining the digital currency. Power savings and reduction in area are achieved relative to multiple hash engines that compute the entire final hash digest.

Abstract: A first communication device generates and transmits a frame that is configured to cause one or more second communication devices in a wireless local area network (WLAN) to refrain from transmitting during a set of repeating time segments, and the frame is generated to include an indication of a time period of the time segments in the set of repeating time segments, the time period being less than a duration of a beacon interval of the WLAN such that multiple ones of the time segments occur within one beacon interval. Alternatively, the frame is configured to cause one or more second communication devices in the WLAN to refrain from transmitting during a time segment that begins in conjunction with an end of transmission of i) the frame or ii) a packet that includes the frame, and the frame is generated to include an indication of a time duration of the time segment.

Abstract: A method of converting a data stored in a memory from a first format to a second format is disclosed. The method includes extending a number of bits in the data stored in a double data rate (DDR) memory by one bit to form an extended data. The method further includes determining whether the data stored in the DDR is signed or unsigned data. Moreover, responsive to determining that the data is signed, a sign value is added to the most significant bit of the extended data and the data is copied to lower order bits of the extended data. Responsive to determining that the data is unsigned, the data is copied to lower order bits of the extended data and the most significant bit is set to an unsigned value, e.g., zero. The extended data is stored in an on-chip memory (OCM) of a processing tile of a machine learning computer array.

Abstract: A system to support validation and debugging of compiled low-level instructions for a machine learning (ML) network model on an ML-specific hardware. A compiler identifies well-defined boundaries in the ML network model based on primitives used to generate low-level instructions for the hardware. The ML network model is partitioned into units/layers/sub-graphs based on the plurality of well-defined boundaries. The compiler then generates an internal representation for each of the units wherein the internal representation is mapped to components in the hardware. Each of the units is compiled into a first set to be executed on the ML-specific hardware and a second set to be executed on a second computing device. The output results from executing the two sets of low-level instructions are compared to validate the first set of low-level instructions. If the outputs do not match fully, the first set of low-level instructions is debugged and recompiled.

Abstract: In a vehicular communication network, a communication device generates a physical layer (PHY) preamble of a PHY protocol data unit (PPDU) for transmission in the vehicular communication network. The communication device generates a plurality of PHY data segments of the PPDU, and one or more PHY midambles, each PHY midamble to be transmitted between a respective pair of adjacent PHY data segments, and each PHY midamble including one or more training signal fields. Generating the one or more PHY midambles includes, when the PPDU is to be transmitted according to an extended range (ER) mode, generating each training signal field to include i) a first portion based on a very high throughput long training field (VHT-LTF) defined by the IEEE 802.11ac Standard and ii) a second portion based on the VHT-LTF defined by the IEEE 802.11ac Standard; and transmitting, by the communication device, the PPDU in the vehicular communication network.

Abstract: A network switch includes a plurality of ports for communicating over a network. Processing circuitry processes inbound frames received from the network via the ports and sends outbound frames to the network. Remote management circuitry (RMU) is responsive to commands received from a host device external to the network switch. The RMU receives via one of the ports a remote access request frame from the host device, wherein at least part of the remote access request frame is encrypted, and decrypts the remote access request frame. In response to successful decryption of the part of the remote access request frame, the RMU accesses one or more configuration registers of the network switch in accordance with the remote access request frame, composes a remote access response frame, at least a portion of the remote access response frame being encrypted, and sends the remote access response frame to the host device.

Abstract: The present invention is directed to data communication systems and methods thereof. According to various embodiments, the present invention provides a communication with a reconfigurable forward-error-correction (FEC) module. The FEC module processes data received from two or more communication lanes, and depending on the mode of operation, the FEC module can combine data from the two or more communication lanes and perform error correction on the combined data, or the FEC module can processes data from the two communications lanes separately and perform error correction independently for the each of the data communication lanes. There are other embodiments as well.

Abstract: A first access point (AP), which is associated with one or more first client stations, generates an announcement frame that announces a coordinated multi-user (MU) transmission involving multiple APs including the first AP and one or more second APs. Each of the second APs is associated with a respective one or more second client stations. The announcement frame is generated to indicate one or more respective sets of communication parameters to be used by the one or more second APs for communicating with the respective one or more second client stations during the coordinated MU transmission. The first AP transmits the announcement frame to the one or more second APs to initiate the coordinated MU transmission, and participates in the coordinated MU transmission while the one or more second APs also participate in the coordinated MU transmission.

Abstract: An Ethernet communication device includes a data interface and circuitry. The data interface is configured for communicating with a neighbor device. The circuitry is configured to exchange Ethernet data frames with the neighbor device over the data interface, wherein successive data frames are separated in time by an Inter-Packet Gap (IPG) having at least a predefined minimal duration, and to further exchange with the neighbor device, over the data interface, during the IPG between Ethernet frames exchanged on the data interface, a wake-up/sleep command that instructs switching between an active mode and a sleep mode.

Abstract: A system and corresponding method employ an object-oriented memory device. The object-oriented memory device includes at least one physical memory and a hardware controller. The hardware controller is coupled intra the object-oriented memory device to the at least one physical memory. The hardware controller (i) decodes an object-oriented message received from a hardware client of the object-oriented memory device and (ii) performs an action for the hardware client based on the object-oriented message received and decoded. The object-oriented message is associated with an object instantiated or to-be-instantiated in the at least one physical memory. The action is associated with the object. The object-oriented memory device alleviates the hardware client(s) from having to manage structure of respective data stored in the at least one physical memory, obviating duplication of code among the hardware clients for managing same and efforts for design and verification thereof.

Abstract: A processing unit to support inference acceleration for machine learning (ML) comprises an inline post processing unit configured to accept and maintain one or more lookup tables for performing a tanh and/or sigmoid operation/function. The inline post processing unit is further configured to accept data from a set of registers configured to maintain output from a processing block instead of streaming the data from an on-chip memory (OCM), perform the tanh and/or sigmoid operation on each element of the data from the processing block on a per-element basis via the one or more lookup tables, and stream post processing result of the per-element tanh and/or sigmoid operation back to the OCM after the tanh and/or sigmoid operation is complete.

Abstract: Methods and apparatus for receiving a user message in a communication network are disclosed. In an exemplary embodiment, a method includes receiving data samples in an uplink transmission from user equipment, performing preamble detection on the data samples, generating a trigger signal that indicates when a preamble is detected, and decoding a user message in response to the trigger signal, wherein the user message follows the detected preamble.

Abstract: A method for cancelling, from servo signals read in a read channel while a write channel is active, interference caused by write signals in the write channel, includes generating a predicted channel response signal from the write signals in a data clock domain, resampling the generated predicted channel response signal using a clock in the data clock domain having a rate corresponding to a servo clock from a servo clock domain, transferring the resampled predicted channel response signal from the data clock domain to the servo clock domain and aligning phase of the transferred resampled predicted channel response signal with phase of the servo clock, determining a domain-boundary-crossing delay incurred in the transferring, based on the domain-boundary-crossing delay, synchronizing the phase-aligned transferred resampled predicted channel response signal with the servo signals, and subtracting the synchronized phase-aligned transferred resampled predicted channel response signal from the servo signals.

Abstract: A first communication device generates a PHY preamble of a PHY data unit to include a first orthogonal frequency division multiplexing (OFDM) symbol corresponding to a legacy signal field. The legacy signal field includes i) a length subfield, and ii) a rate subfield. The length subfield and the rate subfield indicate a duration of the PHY data unit, and the legacy signal field is formatted according to a legacy second communication protocol. The first communication device generates the PHY preamble of a PHY data unit to include a second OFDM symbol corresponding to a duplicate of the legacy signal field, and a plurality of additional OFDM symbols corresponding to a non-legacy signal field. The first communication device sets the length subfield of the legacy signal field to a length value such that a remainder value resulting from dividing the length value by three, indicates that the PHY data unit conforms to the first communication protocol.

Abstract: A system and corresponding method perform large memory transaction (LMT) stores. The system comprises a processor associated with a data-processing width and a processor accelerator. The processor accelerator performs a LMT store of a data set to a coprocessor in response to an instruction from the processor targeting the coprocessor. The data set corresponds to the instruction. The LMT store includes storing data from the data set, atomically, to the coprocessor based on a LMT line (LMTLINE). The LMTLINE is wider than the data-processing width. The processor accelerator sends, to the processor, a response to the instruction. The response is based on completion of the LMT store of the data set in its entirety. The processor accelerator enables the processor to perform useful work in parallel with the LMT store, thereby improving processing performance of the processor.

Abstract: A semiconductor device includes circuitry configured to derive a physical unclonable function. The circuitry includes a plurality of bitcells, each bitcell being readable as one of a ‘0’ value and a ‘1’ value, and sense amplifier circuitry configurable to read values from the plurality of bitcells. The sense amplifier circuitry includes margin circuitry configurable (i) to selectably bias reading of the plurality of bitcells toward one of ‘0’ values and ‘1’ values, (ii) to identify addresses of bitcells having a stable ‘1’ value when the margin circuitry is configured to bias reading of the plurality of bitcells toward ‘0’ values, and (iii) to identify addresses of bitcells having a stable ‘0’ value when the margin circuitry is configured to bias reading of the plurality of bitcells toward ‘1’ values. Each bitcell in the plurality of bitcells may include a differential transistor pair.

Abstract: A computer device in a network receives data units sent by a remote computer device over a network link in a sequenced order. A traffic monitor evaluates traffic on the network link, and selectively sends congestion notifications in response to determining that certain data units sent by the remote computer device have been received or will likely be received out of the sequenced order. The notifications cause the remote computer device to pause sending further data units. An ingress processor, separate from a central processing unit (CPU) of the computer device, detects a header segment of a received data unit and a corresponding payload segment of the received data unit. A storage controller stores the header segment of the received data unit in a first memory location, and stores the payload segment of the received data unit in a second memory location separate from the first memory location. The second memory location corresponds to a next location in a storage queue.

Abstract: A communication device generates a first packet and a second packet. The first packet includes a first physical layer (PHY) preamble having: a first legacy signal field (L-SIG) having first duration information that indicates a first duration of the first packet; and first non-legacy signal field information having first modulation information that indicates a first modulation used in the first packet. The second packet includes a second PHY preamble having: a second L-SIG having second duration information that indicates a second duration of the second packet, wherein the second duration is different than the first duration; and second non-legacy signal field information having second modulation information that indicates a second modulation used in the second packet, wherein the second modulation is different than the first modulation. The communication device simultaneously transmits the first packet in a first frequency segment and the second packet in a second frequency segment.

Abstract: The present disclosure describes apparatuses and methods for contextual search of a storage system. In some aspects, a metadata manager of a storage system receives a query to search the data stored on the storage media of the apparatus. The metadata manager identifies an entry in a relational database of the metadata manager that includes a label that is relevant to the query and determines, based on the entry in the relational database, a reference address of a target node in a navigational database of the metadata manager that corresponds to the label. As results for the query to search, the metadata manager returns an object of the target node at the reference address in the navigational database and corresponding objects of relative nodes connected to the target node via respective links. By so doing, the metadata database may enable contextual or implicit search of data in the storage system.

Abstract: Examples of a carry chain for performing an operation on operands each including elements of a selectable size is provided. Advantageously, the carry chain adapts to elements of different sizes. The carry chain determines a mask based on a selected size of an element. The carry chain selects, based on the mask, whether to carry a partial result of an operation performed on corresponding first portions of a first operand and a second operand into a next operation. The next operation is performed on corresponding second portions of the first operand and the second operand, and, based on the selection, the partial result of the operation. The carry chain stores, in a memory, a result formed from outputs of the operation and the next operation.