Abstract: Systems and methods are provided to improve flexibility of optical signal transmission between integrated circuit devices, and more specifically data utilization circuits. More specifically, the integrated circuit devices may include a data utilization circuit communicatively coupled to a field programmable optical array (FPOA). In some embodiments, the FPOA may convert an electrical signal received from the data utilization to an optical signal, route the optical signal to an optical channel, and multiplex the optical signal with other optical signals routed to the optical channel. Additionally or alternatively, the FPOA may de-multiplex a multiplexed optical signal based on wavelength, route an optical signal included in the multiplexed optical signal to an electrical channel, convert the optical signal into an electrical signal, and output the electrical signal to the data utilization circuit via an electrical channel.

Abstract: A clocking arrangement for transceivers in an integrated circuit device includes a plurality of fractionally adjustable phase-locked loops. Each respective one of the fractionally adjustable phase-locked loops generates a respective transmit frequency for a respective one of the transceivers. There is a respective clock-data recovery module in a receive portion of each respective one of the transceivers, and each respective clock-data recovery module includes a respective fractionally adjustable frequency-lock loop. There is a reference clock input providing a reference clock for a plurality of the fractionally adjustable phase-locked loops and the fractionally adjustable frequency-lock loops. The reference clock input can be a sole reference clock input providing a reference clock for all of the adjustable phase-locked frequency-lock loops.

Abstract: Circuitry for receiving a high-speed serial data signal (e.g., having a bit rate in the range of about 10 Gbps and higher) includes a two-stage, continuous-time, linear equalizer having only two serially connected stages. Phase detector circuitry may be provided for receiving the serial output of the equalizer and for converting successive pairs of bits in that output to successive parallel-form bit pairs. Further demultiplexing circuitry may be provided to demultiplex successive groups of the parallel-form bit pairs to final groups of parallel bits, which can be quite large in terms of number of bits (e.g., 64 parallel bits). Another aspect of the invention relates to multiplexer circuitry for efficiently going in the opposite direction from such relatively large groups of parallel data bits to a high-speed serial data output signal.

Abstract: One embodiment relates to a method of automated adaptation of a transmitter equalizer. A multi-dimensional search space of tap settings for the transmitter equalizer is divided into multiple single-dimensional search spaces, each single-dimensional search space being associated with a single tap of the transmitter equalizer. The multiple single-dimensional search spaces are searched in series, and a tap for a single-dimensional search space is set before searching a next single-dimensional search space. Another embodiment relates to a transceiver with adaptation circuitry configured to perform this method. Other embodiments, aspects, and features are also disclosed.

Abstract: One embodiment relates to a method of automated adaptation of a transmitter equalizer. A multi-dimensional search space of tap settings for the transmitter equalizer is divided into multiple single-dimensional search spaces, each single-dimensional search space being associated with a single tap of the transmitter equalizer. The multiple single-dimensional search spaces are searched in series, and a tap for a single-dimensional search space is set before searching a next single-dimensional search space. Another embodiment relates to a transceiver with adaptation circuitry configured to implement the above-described method. Other embodiments, aspects, and features are also disclosed.

Abstract: Circuitry for receiving a high-speed serial data signal (e.g., having a bit rate in the range of about 10 Gbps and higher) includes a two-stage, continuous-time, linear equalizer having only two serially connected stages. Phase detector circuitry may be provided for receiving the serial output of the equalizer and for converting successive pairs of bits in that output to successive parallel-form bit pairs. Further demultiplexing circuitry may be provided to demultiplex successive groups of the parallel-form bit pairs to final groups of parallel bits, which can be quite large in terms of number of bits (e.g., 64 parallel bits). Another aspect of the invention relates to multiplexer circuitry for efficiently going in the opposite direction from such relatively large groups of parallel data bits to a high-speed serial data output signal.

Abstract: One embodiment relates to a method of automated adaptation of a transmitter equalizer. A multi-dimensional search space of tap settings for the transmitter equalizer is divided into multiple single-dimensional search spaces, each single-dimensional search space being associated with a single tap of the transmitter equalizer. The multiple single-dimensional search spaces are searched in series, and a tap for a single-dimensional search space is set before searching a next single-dimensional search space. Another embodiment relates to a transceiver with adaptation circuitry configured to implement the above-described method. Other embodiments, aspects, and features are also disclosed.

Abstract: Circuitry for receiving a high-speed serial data signal (e.g., having a bit rate in the range of about 10 Gpbs and higher) includes a two-stage, continuous-time, linear equalizer having only two serially connected stages. Phase detector circuitry may be provided for receiving the serial output of the equalizer and for converting successive pairs of bits in that output to successive parallel-form bit pairs. Further demultiplexing circuitry may be provided to demultiplex successive groups of the parallel-form bit pairs to final groups of parallel bits, which can be quite large in terms of number of bits (e.g., 64 parallel bits). Another aspect of the invention relates to multiplexer circuitry for efficiently going in the opposite direction from such relatively large groups of parallel data bits to a high-speed serial data output signal.

Abstract: One embodiment relates to an integrated circuit including multiple PMA modules, a plurality of multiple-purpose PLLs, multiple reference clock signal inputs, and a programmable clock network. Another embodiment relates to a fracture-able PLL circuit. The fracture-able PLL circuit includes a first phase-locked loop circuit generating a first frequency output, a second phase-locked loop circuit; arranged to generate a second frequency output, and a plurality of shared output resources. Reconfigurable circuitry is arranged so that either of the first and second frequency outputs is receivable by each of the plurality of shared output resources. Another embodiment relates to an integrated circuit including a first strip of PLL circuits on a first side of the integrated circuit, and a second strip of PLL circuits on a second side of the integrated circuit which is opposite from the first side. Other embodiments and features are also disclosed.

Abstract: Systems and methods are provided to improve flexibility of optical signal transmission between integrated circuit devices, and more specifically data utilization circuits. More specifically, the integrated circuit devices may include a data utilization circuit communicatively coupled to a field programmable optical array (FPOA). In some embodiments, the FPOA may convert an electrical signal received from the data utilization to an optical signal, route the optical signal to an optical channel, and multiplex the optical signal with other optical signals routed to the optical channel. Additionally or alternatively, the FPOA may de-multiplex a multiplexed optical signal based on wavelength, route an optical signal included in the multiplexed optical signal to an electrical channel, convert the optical signal into an electrical signal, and output the electrical signal to the data utilization circuit via an electrical channel.

Abstract: An integrated circuit package includes an interposer with an embedded clock network formed by multiple clock trees. A die with first and second clock circuits is disposed over the interposer. At least one of the first and second clock trees is a resonant clock tree and both the first and second clock circuits may provide clock signals at different frequencies. The first clock circuit may provide clock signals at one frequency to a clock tree in the embedded clock network while the second clock circuit may provide clock signals at another frequency to another clock tree in the embedded clock tree network.

Abstract: A clocking arrangement for transceivers in an integrated circuit device includes a plurality of fractionally adjustable phase-locked loops. Each respective one of the fractionally adjustable phase-locked loops generates a respective transmit frequency for a respective one of the transceivers. There is a respective clock-data recovery module in a receive portion of each respective one of the transceivers, and each respective clock-data recovery module includes a respective fractionally adjustable frequency-lock loop. There is a reference clock input providing a reference clock for a plurality of the fractionally adjustable phase-locked loops and the fractionally adjustable frequency-lock loops. The reference clock input can be a sole reference clock input providing a reference clock for all of the adjustable phase-locked frequency-lock loops.

Abstract: One embodiment relates to a receiver circuit for a data link. The receiver circuit includes at least a first signal path, a second signal path, and a path selector circuit. The first signal path includes first equalization circuitry, and the second signal path includes second equalization circuitry. The path selector circuit is configured to select one signal path of the first and second signal paths. Other embodiments and features are also disclosed.

Abstract: An analog test network includes a conductor. The conductor is coupled to provide a first analog signal from a circuit under test to an analog-to-digital converter circuit. The analog-to-digital converter circuit is operable to generate a first digital signal based on the first analog signal. A control circuit is operable to generate a second digital signal based on the first digital signal. A digital-to-analog converter circuit is operable to generate a second analog signal based on the second digital signal. The conductor is coupled to provide the second analog signal from the digital-to-analog converter circuit to the circuit under test.

Abstract: Integrated circuits may include memory interface circuitry operable to communicate with system memory. The memory interface circuitry may receive data (DQ) and data strobe (DQS) signals from system memory during read operations. The memory interface circuitry may include startup calibration circuitry and runtime calibration circuitry. The startup calibration circuitry may be used upon device startup to perform a one-time data de-skew and DQ/DQS centering. The runtime calibration circuitry may include at least two data sampling circuits, a first of which is used in active mode to latch incoming data and a second of which is used in redundant mode to obtain data eye boundary information on a continuous basis. The received DQS signal may be adjusted based on the obtained eye boundary information so that DQS properly positioned within the data eye periodically or on an as-needed basis.

Abstract: One embodiment relates to an integrated circuit including multiple PMA modules, a plurality of multiple-purpose PLLs, multiple reference clock signal inputs, and a programmable clock network. Each PMA module includes multiple CDR circuits, receives multiple serial data signals, and outputs data from those signals in parallel form. The programmable clock network allows the reference clock signals to be selectively shared by the PMA modules and the multiple-purpose PLLs. Another embodiment relates to a method of providing clock signals for multiple purposes in an integrated circuit. Clock signals are generated by a plurality of multiple-purpose PLLs and are selectively distributed to PMA modules arranged at a side of the integrated circuit and to logic circuitry arranged in a core section of the integrated circuit. The clock signals are used by circuitry in the PMA modules for supporting a plurality of data communications channels. Other embodiments and features are also disclosed.

Abstract: A metal-oxide-metal (MOM) capacitor structure is disclosed. The MOM capacitor includes a plurality of layers, each layer having a plurality of electrodes. The plurality of electrodes, separated by oxide layers, forms a first plate and a second plate of the MOM capacitor. The plurality of electrodes on each of the layers is coupled to a plurality of electrodes on an adjacent layer through a plurality of vias. A shield layer is coupled to each of the electrodes that forms the second plate of the MOM capacitor on each of the plurality of layers.

Abstract: An IC that includes a jitter generator, where the jitter generator is integral with the IC and generates non-intrinsic jitter, is provided. In one implementation, the non-intrinsic jitter is used to measure a characteristic of the IC. In one implementation, the non-intrinsic jitter is used to test jitter tolerance of the IC. In yet another implementation, the non-intrinsic jitter is used to test another IC coupled to the IC that includes the jitter generator.

Abstract: Integrated circuits with high-speed communications capabilities are provided. Such types of integrated circuits may include clock data recovery (CDR) circuitry. The CDR circuitry may receive incoming data and may generate a recovered clock that is phase-aligned to the incoming data. The CDR circuitry may also include data latching circuitry for separately latching even and odd data bits in alternating clock cycles. During a first mode, a first portion of the data latching circuitry may be used to latch even data bits while a second portion of the data latching circuitry may be used to latch odd data bits. During a second mode, the second portion of the data latching circuitry may be used to latch the even data bits while the first portion of the data latching circuitry may be used to latch the odd data bits. The mode that yields the better link performance may be selected.

Abstract: A method of equalizing an input data signal using a multiple-stage continuous-time linear equalization (CTLE) circuit. A zero-forcing least-mean-square (ZF LMS) procedure is applied to adapt the settings of the CTLE stages. The amplitude settings and the frequency boost settings of the CTLE stages are adapted within the ZF LMS procedure. In an exemplary implementation, an error screening threshold may be applied to an error signal within the ZF LMS procedure to generate a reduced error signal such that weight updates do not occur if the error signal is below the error screening threshold. In addition, if an accumulated sign error signal within the ZF LMS procedure reaches a predetermined maximum indicative of a high loss channel, then a setting for a variable gain amplifier may be increased, and an amplitude setting for the CTLE circuit may be decreased. Other embodiments, aspects and features are also disclosed.