Abstract: A system side interface of a PHY chip used in conjunction with a 100 GBASE backplane, sends and receives data using an NRZ signal format, but at a data rate of between about 26.5 Gbps/per lane to 27.2 Gbps/per lane, which is consistent with the PAM 4 signaling protocol. Thus, chip-to-chip communications between a PHY chip and a switch or controller chip can use an “overclocked” NRZ signaling format, reducing the amount of logic needed, which in turn can reduce signal latency, and reduce the chip area and power consumption required to implement the logic.

Abstract: The present disclosure is directed to systems, apparatuses, and methods for performing continuous or periodic link training. Existing link training protocols generally perform link training only once during startup or initialization of a link and, as a result, are limited in their applications. After link training is performed and Open Systems Interconnect (OSI) data link layer and other high-layer data is transmitted across the link, no further link training is performed using these existing link training protocols. However, parameters of the link may change over time after link training is performed, such as temperature of the link and voltage levels of signals transmitted over the link by the transmitter of the transmitter-receiver pair.

Abstract: The present disclosure describes a serializer and a deserializer. The serializer can receive a sequence of information in a parallel format and control information over a serial interface from a host device. The serializer converts the sequence of information in the parallel format to provide the sequence of information in a serial format to the deserializer which converts the sequence of information in to serial format to the sequence of information in the parallel format. The serializer passes through the control information to provide the control information to the deserializer which is similarly passed through by the deserializer. The control information can include one or more control packets and/or one or more link pulses to train one or more other serializers and/or one or more other deserializers communicating with each other over a communication channel.

Abstract: A system side interface of a PHY chip used in conjunction with a 100 GBASE backplane, sends and receives data using an NRZ signal format, but at a data rate of between about 26.5 Gbps/per lane to 27.2 Gbps/per lane, which is consistent with the PAM 4 signaling protocol. Thus, chip-to-chip communications between a PHY chip and a switch or controller chip can use an “overclocked” NRZ signaling format, reducing the amount of logic needed, which in turn can reduce signal latency, and reduce the chip area and power consumption required to implement the logic.

Abstract: A system side interface of a PHY chip used in conjunction with a 100GBASE backplane, sends and receives data using an NRZ signal format, but at a data rate of between about 26.5 Gbps/per lane to 27.2 Gbps/per lane, which is consistent with the PAM 4 signaling protocol. Thus, chip-to-chip communications between a PHY chip and a switch or controller chip can use an “overclocked” NRZ signaling format, reducing the amount of logic needed, which in turn can reduce signal latency, and reduce the chip area and power consumption required to implement the logic.

Abstract: A system side interface of a PHY chip used in conjunction with a 100GBASE backplane, sends and receives data using an NRZ signal format, but at a data rate of between about 26.5 Gbps/per lane to 27.2 Gbps/per lane, which is consistent with the PAM 4 signaling protocol. Thus, chip-to-chip communications between a PHY chip and a switch or controller chip can use an “overclocked” NRZ signaling format, reducing the amount of logic needed, which in turn can reduce signal latency, and reduce the chip area and power consumption required to implement the logic.

Abstract: A system side interface of a PHY chip used in conjunction with a 100 GBASE backplane, sends and receives data using an NRZ signal format, but at a data rate of between about 26.5 Gbps/per lane to 27.2 Gbps/per lane, which is consistent with the PAM 4 signaling protocol. Thus, chip-to-chip communications between a PHY chip and a switch or controller chip can use an “overclocked” NRZ signaling format, reducing the amount of logic needed, which in turn can reduce signal latency, and reduce the chip area and power consumption required to implement the logic.

Abstract: A system and method for increasing input/output speeds in a network switch. A physical layer device is provided that includes a physical coding sublayer that insert data flow identifiers to data flows that are provided to a gearbox. In one embodiment, the gearbox is a 5 to 2 gearbox that can transport various combinations of 10G/40G data flows over a narrower interface to a second physical layer device having an inverse gearbox.

Abstract: Input/output of network switches and the like are improved by a system including a gearbox, an inverse-gearbox, and a Gigabit Ethernet link coupling them. The gearbox and inverse-gearbox interconnect data streams received through wider lower rate Gigabit Ethernet interfaces through narrower faster rate interfaces. The gearbox is configured to bit-multiplex physical-layer data streams received through input interfaces to generate bit-multiplexed data streams. The inverse-gearbox is configured to demultiplex the multiplexed data streams and to output the recovered data streams through output interfaces. One of the output interfaces is selected for each recovered data stream according to a respective embedded physical-layer data stream identifier.

Abstract: A system side interface of a PHY chip used in conjunction with a 100GBASE backplane, sends and receives data using an NRZ signal format, but at a data rate of between about 26.5 Gbps/per lane to 27.2 Gbps/per lane, which is consistent with the PAM 4 signaling protocol. Thus, chip-to-chip communications between a PHY chip and a switch or controller chip can use an “overclocked” NRZ signaling format, reducing the amount of logic needed, which in turn can reduce signal latency, and reduce the chip area and power consumption required to implement the logic.

Abstract: A system and method for increasing input/output speeds in a network switch. A physical layer device is provided that includes a physical coding sublayer that insert data flow identifiers to data flows that are provided to a gearbox. In one embodiment, the gearbox is a 5 to 2 gearbox that can transport various combinations of 10G/40G data flows over a narrower interface to a second physical layer device having an inverse gearbox.

Abstract: A system and method for increasing input/output speeds in a network switch. A physical layer device is provided that includes a physical coding sublayer that insert data flow identifiers to data flows that are provided to a gearbox. In one embodiment, the gearbox is a 5 to 2 gearbox that can transport various combinations of 10 G/40 G data flows over a narrower interface to a second physical layer device having an inverse gearbox.

Abstract: A system and method for using energy efficiency network refresh signals for exchanging link partner and device information. Energy savings can be realized through a usage of a energy saving state such as a low power idle (LPI) mode. In one embodiment, refresh signals used during the LPI mode can be used to encode information or state within the refresh signals. In general, such encoded information enables link partners to exchange information that would otherwise need to wait until the link partners have transitioned from an energy saving state back to the active state. In various examples, the messaging during the energy saving state can be used to facilitate synchronization during the energy saving state, transitions from the energy saving state, etc.

Abstract: Input/output of network switches and the like are improved by a system including a gearbox, an inverse-gearbox, and a Gigabit Ethernet link coupling them. The gearbox and inverse-gearbox interconnect data streams received through wider lower rate Gigabit Ethernet interfaces through narrower faster rate interfaces. The gearbox is configured to bit-multiplex physical-layer data streams received through input interfaces to generate bit-multiplexed data streams. The inverse-gearbox is configured to demultiplex the multiplexed data streams and to output the recovered data streams through output interfaces. One of the output interfaces is selected for each recovered data stream according to a respective embedded physical-layer data stream identifier.

Abstract: A system and method for increasing input/output speeds in a network switch. A physical layer device is provided that includes a physical coding sublayer that insert data flow identifiers to data flows that are provided to a gearbox. In one embodiment, the gearbox is a 5 to 2 gearbox that can transport various combinations of 10G/40G data flows over a narrower interface to a second physical layer device having an inverse gearbox.

Abstract: Aspects of a method and system for energy efficient networking over a serial communication channel are provided. In this regard, one or more circuits in an Ethernet PHY that communicates over one or more serial communication channels may transmit and/or receive physical layer signals to maintain and/or refresh synchronization and/or training parameters while operating in an energy saving mode. The Ethernet PHY may transition out of the energy saving mode upon transmitting and/or receiving a wake sequence via the serial communication channel(s), where the wake sequence comprises one or more deterministic forward error correction (FEC) block in instances that FEC is utilized for communications via the serial communication channel(s). The one or more circuits in the Ethernet PHY may be operable to perform forward error correction (FEC) functions and one or more of the FEC functions may be disabled while remaining ones of the FEC functions are enabled.

Abstract: According to an example embodiment, an apparatus may include logic. The apparatus may be configured to: determine, based on an error location polynomial, an error location syndrome corresponding to an actual location of a burst error in a data block; select a burst error pattern that is less than or equal to M bits, and having no more than Y consecutive zeros within the burst error, where M is greater than the order of the error location polynomial; determine an error pattern syndrome based on the selected burst error pattern and the error location polynomial; and determine an actual location of the burst error in the data block based on the error location syndrome and the error pattern syndrome.

Abstract: Aspects of a method and system for energy efficient networking over a serial communication channel are provided. In this regard, one or more circuits in an Ethernet PHY that communicates over one or more serial communication channels may transmit and/or receive physical layer signals to maintain and/or refresh synchronization and/or training parameters while operating in an energy saving mode. The Ethernet PHY may transition out of the energy saving mode upon transmitting and/or receiving a wake sequence via the serial communication channel(s), where the wake sequence comprises one or more deterministic forward error correction (FEC) block in instances that FEC is utilized for communications via the serial communication channel(s). The one or more circuits in the Ethernet PHY may be operable to perform forward error correction (FEC) functions and one or more of the FEC functions may be disabled while remaining ones of the FEC functions are enabled.

Abstract: According to an example embodiment, a method may include determining an actual location (N) of a burst error in a data block; selecting a burst error pattern that is a correctable error based on adjusting an error pattern syndrome by an adjustment amount (S); and determining a correction vector based on the burst error pattern; shifting the correction vector by an offset amount based on (N) and (S); and correcting the burst error in the data block based on the shifted correction vector.

Abstract: According to an example embodiment, an apparatus may include logic. The apparatus may be configured to: determine, based on an error location polynomial, an error location syndrome corresponding to an actual location of a burst error in a data block; select a burst error pattern that is less than or equal to M bits, and having no more than Y consecutive zeros within the burst error, where M is greater than the order of the error location polynomial; determine an error pattern syndrome based on the selected burst error pattern and the error location polynomial; and determine an actual location of the burst error in the data block based on the error location syndrome and the error pattern syndrome.