摘要:
A parallel data link includes a redundant line. A bank of switches permits any arbitrary line of the link to be enabled or disabled for carrying functional data, each line being dynamically calibrated in turn by disabling the line and allowing other lines to carry the functional data. The switches are located downstream of alignment mechanisms so that data input to the switches is compensated for data skew. Preferably, receiver synchronization circuitry in each line operates in a respective independent clock domain, while the switches and calibration mechanism operate in a common clock domain. Preferably, the receiver synchronization circuits provide an adjustable delay corresponding to a variable number of clock cycles to align the outputs of the receiver synchronization circuits with respect to one another, which can accommodate high data skew.
摘要:
A parallel data link includes a redundant line. A bank of switches permits any arbitrary line of the link to be enabled or disabled for carrying functional data, each line being dynamically calibrated in turn by disabling the line and allowing other lines to carry the functional data. The switches are located downstream of alignment mechanisms so that data input to the switches is compensated for data skew. Preferably, receiver synchronization circuitry in each line operates in a respective independent clock domain, while the switches and calibration mechanism operate in a common clock domain. Preferably, the receiver synchronization circuits provide an adjustable delay corresponding to a variable number of clock cycles to align the outputs of the receiver synchronization circuits with respect to one another, which can accommodate high data skew.
摘要:
A communications link of multiple parallel communications lines includes at least one redundant line. In a first aspect, the lines are periodically recalibrated one at a time while the others carry functional data. If a fault is detected, the faulty line is disabled and the remaining previously calibrated lines transmit functional data. In a second aspect, impending line malfunction is detected from anomalies during calibration. In a third aspect, line malfunction is detected from receiver circuit output by determining a logical lane upon which each detected error occurs, and by mapping the logical lane to a physical line currently carrying the logical lane data.
摘要:
A memory buffer, memory system and method for power-on initialization and test for a cascade interconnect memory system. The memory buffer includes a bus interface to links in a high-speed channel for communicating with a memory controller via a direct connection or via a cascade interconnection through an other memory buffer. The interface is operable in a SBC mode and a high-speed mode. The memory buffer also includes a field service interface (FSI) slave for receiving FSI signals from a FSI master. In addition, the memory buffer includes logic for executing a power-on and initialization training sequence initiated by the memory controller.
摘要:
An interface alignment pattern for de-skewing data bits received on an elastic interface is disclosed. The interface alignment pattern is “busy” in that it has a high number of logic state transitions. The busy interface alignment pattern can be used for scrambling and unscrambling operational data. The interface alignment pattern has a unique timing sequence for determining the location of a data bit's first data beat.
摘要:
A communications link of multiple parallel communications lines includes at least one redundant line. In a first aspect, the lines are periodically recalibrated one at a time while the others carry functional data. If a fault is detected, the faulty line is disabled and the remaining previously calibrated lines transmit functional data. In a second aspect, impending line malfunction is detected from anomalies during calibration. In a third aspect, line malfunction is detected from receiver circuit output by determining a logical lane upon which each detected error occurs, and by mapping the logical lane to a physical line currently carrying the logical lane data.
摘要:
A parallel data link includes a redundant line. The redundant line permits one line to be calibrated while the others carry functional data, a switching mechanism enabling each line to be selected in turn for calibration. Control information for controlling the link, which is preferably for coordinating calibration activity, is communicated on the line selected for calibration. Preferably, the link is bi-directional, having a separate redundant line in each direction, enabling a bi-directional handshaking protocol to be used for communicating control information. Preferably, the lines selected for calibration are time-multiplexed to carry calibration patterns and control information at different time intervals.
摘要:
An interface alignment pattern for de-skewing data bits received on an elastic interface is disclosed. The interface alignment pattern is “busy” in that it has a high number of logic state transitions. The busy interface alignment pattern can be used for scrambling and unscrambling operational data. The interface alignment pattern has a unique timing sequence for determining the location of a data bit's first data beat.
摘要:
A memory buffer, memory system and method for power-on initialization and test for a cascade interconnect memory system. The memory buffer includes a bus interface to links in a high-speed channel for communicating with a memory controller via a direct connection or via a cascade interconnection through an other memory buffer. The interface is operable in a SBC mode and a high-speed mode. The memory buffer also includes a field service interface (FSI) slave for receiving FSI signals from a FSI master. In addition, the memory buffer includes logic for executing a power-on and initialization training sequence initiated by the memory controller.
摘要:
An interface alignment pattern for de-skewing data bits received on an elastic interface is disclosed. The interface alignment pattern is “busy” in that it has a high number of logic state transitions. The busy interface alignment pattern can be used for scrambling and unscrambling operational data. The interface alignment pattern has a unique timing sequence for determining the location of a data bit's first data beat.