Abstract: The present invention provides a hybrid programmable logic device which includes a programmable field programmable gate array logic fabric and a many-core distributed processing subsystem. The device integrates both a fabric of programmable logic elements and processors in the same device, i.e., the same chip. The programmable logic elements may be sized and arranged such that place and route tools can address the processors and logic elements as a homogenous routing fabric. The programmable logic elements may provide hardware acceleration functions to the processors that can be defined after the device is fabricated. The device may include scheduling circuitry that can schedule the transmission of data on horizontal and vertical connectors in the logic fabric to transmit data between the programmable logic elements and processor in an asynchronous manner.

Abstract: In one aspect, a method includes receiving a differential strobe signal including first and second components; buffering, by a first buffer, both the first and second components; and buffering, by a second buffer, the first component. The method includes receiving, by a control logic block, the output of the second buffer. The method includes, after a period when the values of both the first and second components are at a first logic state, but before receiving a burst of clock edges in the differential strobe signal, detecting a transition in the first component from the first logic state to a second logic state, and in response to the detected transition, asserting an enable signal. The method further includes receiving, by a gating logic block, the enable signal and the output of the first buffer, and, when the enable signal is asserted, un-gating the output of the first buffer.

Abstract: An apparatus includes a transmitter adapted to transmit encoded information to a communication link. The transmitter includes a DC balance skew generator. The DC balance skew generator is adapted to skew a DC balance of the information before information is provided to the communication link.

Abstract: An apparatus includes a substrate that includes electronic circuitry. The apparatus further includes a first die that includes electronic circuitry, and at least one shielded interconnect. The shielded interconnect(s) couple(s) electronic circuitry in the substrate to electronic circuitry in the first die.

Abstract: A hybrid programmable logic is described that performs packet processing functions on received data packets using programmable logic elements, and processors interleaved with the programmable logic elements. The header data may be scheduled for distribution to processing threads associated with the processors by the programmable logic elements. The processors may perform packet processing functions on the header data using both the processing threads and hardware acceleration functions provided by the programmable logic elements.

Abstract: One embodiment relates a method for communicating data using an optical transport network. Multiple sub-rate client data signals are received from client sources. The sub-rate client data signals each have a data rate which is less than a data rate capacity of a lowest-order data unit. A predetermined number of tributary slots are provided in the lowest-order optical channel data unit, and each sub-rate client data signal are mapped to at least one of the tributary slots. Another embodiment relates to an optical data communication server that includes a sub-rate mapper for mapping multiple sub-rate client data streams to a predetermined number of tributary slots. Other embodiments and features are also disclosed.

Abstract: The present invention is an improvement in a molded semiconductor package and the method for its manufacture. The package comprises a substrate, a semiconductor die mounted on the substrate, a molding compound encircling the die on the substrate, a lid on the molding compound, and a heat pipe extending between the semiconductor die and the lid. Preferably, the heat pipe is formed so that it encircles the die. The package is assembled by mounting the die on the substrate, applying the molding compound to the substrate while a channel is formed in the molding compound adjacent the semiconductor die, inserting a heat pipe material in the channel, and mounting the lid on the molding compound and the heat pipe material.

Abstract: A molded semiconductor package comprises a substrate, a semiconductor die mounted on the substrate, a molding compound encircling the die on the substrate, and one or more heat conductors in the molding compound that are thermally coupled to the substrate. Advantageously, the heat conductors are mounted in the molding compound near one or more of the corners of the die. The package may also include a lid. The heat conductors produce a more uniform distribution of heat in the substrate. The package is assembled by mounting the die on the substrate, mounting the heat conductors on the substrate and applying the molding compound to the substrate, the die, and the heat conductors mounted on the substrate. For packages that use a lid, the lid is then secured to the package and coupled to the heat conductors.

Abstract: Systems, methods, apparatus, and techniques are presented for processing a codeword. A Reed-Solomon mother codeword n symbols in length and having k check symbols is received, and the n symbols of the received Reed-Solomon mother codeword are separated into v Reed-Solomon daughter codewords, where v is a decomposition factor associated with the Reed-Solomon mother codeword. The v Reed-Solomon daughter codewords are processed in a respective set of v parallel processes to output v decoded codewords.

Abstract: Receiver circuits in serial lanes each generate a synchronous clock signal that is aligned with a master clock signal to allow synchronous transfer of data onto the master clock domain without corruption. A serial-to-parallel converter circuit in each receiver circuit converts a serial data signal into parallel data signals in response to one of the synchronous clock signals. Phase detection circuitry generates an indication of a phase shift based on a phase offset between the synchronous and master clock signals. A clock signal generation circuit provides an adjustment to a phase of the synchronous clock signal based on the indication of the phase shift. The serial-to-parallel converter circuit adjusts positions of bits indicated by the parallel data signals based on the adjustment to the phase of the synchronous clock signal.

Abstract: An integrated circuit package may include a substrate and an interposer. The interposer is disposed over the substrate. The interposer may include embedded switching elements that may be used to receive different power supply signals. An integrated circuit with multiple logic blocks is disposed over the substrate. The switching elements embedded in the interposer may be used to select a power supply signal from the power supply signals and may be used to provide at least one circuit block in the integrated circuit with a selected power supply signal.

Abstract: Integrated circuits with sequential logic circuitry are provided. Sequential logic circuitry may include a chain of bypassable clocked storage elements coupled between a speed critical input terminal and a speed critical output terminal. Combinational logic circuits may be interposed between each adjacent pair of bypassable clocked storage elements in the chain. Dynamic voltage-frequency scaling (DVFS) control circuitry may provide an adjustable power supply voltage to the combinational logic circuits and may provide an adjustable clock signal to control the clocked storage elements. The DVFS control circuitry may be used to selectively enable at least some of the bypassable clocked storage elements while disabling other bypassable clocked storage elements so that the power supply voltage can be reduced while maintaining the same operating frequency. The power supply voltage and the frequency of the clock signal can be adjusted to provide the desired voltage-frequency tradeoff.

Abstract: A method of aligning clock signals in multiple transceiver channels on an integrated circuit may include adjusting a slave clock signal at a slave transceiver channel based on a master clock signal received from a master transceiver channel. A clock generation circuit and/or a delay circuit in the slave transceiver channel may be used to adjust the slave clock signal to produce an intermediate slave clock signal. The master clock signal may be adjusted based on the intermediate slave clock signal received at the master transceiver channel to obtain a total adjustment value. The phase of the intermediate slave clock signal may further be adjusted at the slave transceiver channel based on the total adjustment made at the master transceiver channel.

Abstract: A metal-programmable integrated circuit may include an array of metal-programmable cells. Each cell may include multi-gate transistor structures in which multiple surfaces of a gate structure serve to control current flow through at least one channel structure. The multi-gate transistor structures may form one or more fin-shaped field effect transistors. The gate structure may at least partially enclose multiple channel structures. Pairs of source-drain structures may be coupled to the channel structures. The transistor structures of each cell may be formed in a substrate covered with one or more metal interconnect layers. Paths formed in the metal interconnect layers may configure the cells to perform desired logic functions. The paths associated with a given cell may be selectively coupled to transistor structures of the cell to configure the cell for a desired logic function and/or for desired output drive strength.

Abstract: This disclosure relates generally to field-programmable gate arrays (FPGAs). Some implementations relate to methods and systems for partitioning FPGA circuit designs to facilitate certification. In one aspect, a method includes generating a hardware description language (HDL) implementation of a circuit design. The method additionally includes partitioning the design into a first portion and a second portion. In some implementations, the second portion corresponds to a safety-critical portion of the design while the first portion corresponds to a non-safety-critical portion. The method additionally includes generating first configuration settings for the first portion and generating second configuration settings for the second portion. The method additionally includes verifying, or providing to a third-party certification body for verification, the first configuration settings for the first portion and the second configuration settings for the second portion.

Abstract: Techniques and mechanisms are provided for an improved built in self-test (BIST) mechanism for 3D assembly defect detection. According to an embodiment of the present disclosure, the described mechanisms and techniques can function to detect defects in interconnects which vertically connect different layers of a 3D device, as well as to detect defects on a 2D layer of a 3D integrated circuit. Additionally, according to an embodiment of the present disclosure, techniques and mechanisms are provided for determining not only the presence of a defect in a given set of interfaces of an integrated circuit, but the particular interface at which a defect may exist.

Abstract: Systems, methods, and devices are provided to enable an integrated circuit device of relatively higher capacity. Such an integrated circuit device may include at least two component integrated circuits that communicate with one another. Specifically, the component integrated circuits may communicate through a “stitched silicon interposer” that is larger than a reticle limit of the lithography system used to manufacture the interposer. To achieve this larger size, the stitched silicon interposer may be composed of at least two component interposers, each sized within the reticle limit and each separated from one another by a die seal structure.

Abstract: One embodiment relates to an equalizer circuit for a data link. The equalizer circuit including a continuous-time linear equalizer, a first circuit loop, and a second circuit loop. The continuous-time linear equalizer receives a received signal and outputs an equalized signal. The first circuit loop determines a first average signal amplitude. The first average signal amplitude may be an average signal amplitude of the equalized signal. The second circuit loop a second average signal amplitude. The second average signal amplitude may be an average signal amplitude of a high-frequency portion of the equalized signal. Other embodiments and features are also disclosed.

Abstract: Integrated circuits with memory cells are provided. A memory cell may include first and second cross-coupled inverting circuits configured to store a single data bit. The first inverting circuit may have an output serving as a first data storage node for the memory cell, whereas the second inverting circuit may have an output serving as a second data storage node for the memory cell. Access transistors may be coupled between the first and second data storage nodes and corresponding data lines. Each of the first and second inverting circuit may have a pull-down transistor and at least two pull-up transistors stacked in series. The pull-down transistors may have body terminals that are reverse biased to help reduce leakage current through the first and second inverting circuits. The memory cell may be formed using a narrower two-gate configuration or a wider four-gate configuration.

Abstract: An integrated circuit with an antenna diode is described. The integrated circuit includes a substrate, a transistor, first and second diffusion regions, and a dummy gate. The transistor and the first and second diffusion regions may be formed within the substrate. The transistor has its gate structure disposed on the substrate. The dummy gate structure may be disposed on a region of the substrate such that it separates the first diffusion region from the second diffusion region. The dummy gate structure may also be coupled to the transistor gate structure.