Xilinx, Inc.

Netlogic Microsystems, Inc

Warren E Cory

Atul V Ghia

Steven P Young

David P Schultz

Suresh M Menon

Hare Krishna Verma

Paul T Sasaki

Riyas Noorudeen Remla

Philip M Freidin

Chee Chong Chan

Donald C Stark

Dean Liu

Vasisht Mantra Vadi

Adebabay M Bekele

Philip D Costello

Jason R Bergendahl

Clemenz Lenard Portmann

Joseph Neil Kryzak

Santiago G Asuncion

Embodiments herein describe a solution for deterministic de-assertion of write and read resets of an asynchronous gearbox FIFO having unequal write and read data bit widths. Proposed approaches look for a stable region between read and write clock phases by sweeping one of the clock phases until the leading edges (phases) of both clocks are aligned then releasing the write and read resets deterministically based upon a change in cyclic behavior of detected logic levels of a reset beacon waveform.

Deterministic reset mechanism for asynchronous gearbox FIFOs for predictable latency

Electrical circuits and associated methods relate to performing a phase alignment by providing N copies of clock alignment circuits, enabling and selecting different clock alignment circuits to achieve an initial phase alignment. In an illustrative example, a phase alignment circuit may include a first clock alignment circuit configured to find a first phase alignment point and a second clock alignment circuit configured to find a second phase alignment point. A control circuit may be configured to select a primary clock alignment circuit from the first clock alignment circuit and the second clock alignment circuit and generate a digital command signal to control a phase interpolator. In various embodiments, by setting the control circuit, the same phase alignment circuit may be used to perform phase alignments between clock domains with different frequencies.

Low latency data transfer technique for mesochronous divided clocks

An integrated circuit comprises programmable resources; a plurality of hard blocks; and a programmable connector coupled to the programmable resources, the plurality of hard blocks; wherein the programmable connector is configurable to route signals between a first hard block and a second hard block in a first mode of operation and to route signals between the first hard and the programmable resources in a second mode of operation.

Circuit for and method of providing a programmable connector of an integrated circuit device

A circuit for transmitting data in an integrated circuit device is described. The circuit comprises a first data width conversion circuit configured to receive a first portion of transmit data to be transmitted in parallel; a first parallel-in, serial-out circuit configured to receive an output of the first data width conversion circuit; a first reset timer configured to provide a first reset signal to enable resetting the first data width conversion circuit; a second data width conversion circuit configured to receive a second portion of the transmit data; a second parallel-in, serial-out circuit configured to receive an output of the second data width conversion circuit; and a second reset timer configured to provide a second reset signal to enable resetting the second data width conversion circuit.

Circuits for and methods of transmitting data in an integrated circuit

A circuit for measuring latency in an integrated circuit device is described. The circuit comprises a transmitter circuit having signal generator configured to generate a test signal having a marker for determining a latency in a path associated with the integrated circuit device; and a latency calculation circuit coupled to the signal generator and having a latency adjustment circuit and a unit interval (UI) adjustment circuit; wherein the latency calculation circuit generates a latency value (LATENCY) based upon a latency count from the latency adjustment circuit and a UI adjustment from the UI adjustment circuit.

Circuit for and method of measuring latency in an integrated circuit

A method and apparatus for a multiple lane transmission system that provides a fixed, low-latency mode of operation with reduced lane-lane skew while process, voltage, and temperature (PVT) variation, as well as other sources of variation, occur over time. Multiplexing techniques are utilized within each transmission lane to allow programmably adaptive use of phase alignment circuitry for various modes of operation. As a result, power consumption and semiconductor die area are reduced because multiple copies of phase alignment circuitry within each transmission lane are not required. Also, injection of additional jitter on the serial outputs due to continuous operation of phase alignment circuitry is prevented. Rather, multiplexers within the phase alignment circuitry selectively adapt the timing architecture to that required by the selected mode of operation.

Method and apparatus for a reduced lane-lane skew, low-latency transmission system

An apparatus for error checking is described. The apparatus includes a matrix having a plurality of bit position columns and rows, where the bit position columns are equal in number to data bits of a word length, the word length for a word serial transmission of a data vector, where the bit position columns are one each for each data bit. The bit position rows are equal in number to syndrome bits, and the bit position rows are one each for each syndrome bit. A portion of the bit position columns are allocated to parity bits for a selected word of the data vector, where the portion of the bit position columns for the selected word are one each for each parity bit allocated to the selected word.

Error checking parity and syndrome of a block of data with relocated parity bits

Method and apparatus for error checking information is described. Configuration data includes data bits and parity bits. Notably, parity bits may be relocated for determining a syndrome value. Syndrome bits are determined by computing a partial syndrome value for each word serially transmitted of the configuration data, where the configuration data includes one or more data vectors. Location of each word of the configuration data is identified. It is determined whether a partial syndrome value is an initial partial syndrome value or other partial syndrome value responsive to word location. An initial partial syndrome value is stored, and subsequent partial syndrome values are cumulatively added for each word of a data vector to arrive at a syndrome value for the data vector.

Data and clock synchronization within a gigabit receiver is maintained throughout the data byte processing logic of the receiver by utilizing the same byte clock signal. The deserialization clock signal that is used to deserialize the received serial data stream is phase coherent with the distributed byte clock signal used within the physical coding sublayer (PCS), thus establishing reliable data transfer across the physical media attachment (PMA) and PCS layers of the gigabit receiver while maintaining a known, fixed latency. The phase relationship between a derived bit clock signal and the byte clock signal is shifted in a manner that achieves coarse data alignment within each data byte without affecting the latency. Conversely, the coarse data alignment is combined with a data alignment toggling procedure to reduce data alignment granularity with minimized latency changes.

Method and apparatus for receive channel data alignment with minimized latency variation

A method and system for pipelined decryption is disclosed. One embodiment includes a circuit having an iterative calculation section and a cipher text storage section in support of cipher block chaining (CBC) encryption mode. The iterative calculation section may be pipelined and configured to process multiple ciphertexts at once for increased throughput.

Method and system for pipelined decryption

A variable latency elastic buffer comprises a plurality of memory locations in which to hold data. A write and read pointer may point to respective write and read addresses of the plurality of locations in which to write and read data. A controller may hold or increment the address of the read pointer upon determining that the amount of data within the buffer differs from a nominal fill level. In a particular embodiment, initialization circuitry may be operable to initialize the read and write addresses of the respective pointers responsive to an initialization request. The read and write addresses may differ from one another by an offset value equal to a value programmed for the nominal value.

Variable latency buffer and method of operation

A design is used for coordinating channel bonding operations of a set of transceivers. The set include a master transceiver and a plurality of first level slave transceivers that perform channel bonding operations. Each first level transceiver is controlled by the master transceiver. The set also comprises a plurality of second level slave transceivers that perform channel bonding operations. Each second level transceiver is controlled by one of the plurality of first level transceivers. Any transceiver can be set as either a master, a first level slave or a second level slave. The design comprises a plurality of flip-flops and multiplexers, and is controlled by a MODE signal that determines the mode of operation of the design.

Channel bonding control logic architecture

A method and apparatus for improving tolerance of inter-channel skew in channel bonded communications links includes designating a master channel and one or more slave channels. Each slave channel develops its own model of skew relative to the master channel. When its skew model is validated, a slave channel can perform channel bonding on its own. The skew models are developed over time, and therefore improve tolerance of inter-channel skew over prior art channel bonding methods.

Growing community of inventors

Redwood City, CA, United States of America

Warren E Cory