Abstract: A method is implemented by a network element to provide scalable hierarchical traffic management (HTM) over a plurality of network layers for a network and eliminate priority leaking caused by quick loopback batch scheduling that analyzes a subset of network layers to shorten processing time and resource requirements when the scalable HTM selects data packets to be forwarded. The method and system function as a low pass filter over the selected data packets to prevent low priority data packets being forwarded where higher priority data packets are available to be forwarded.

Abstract: Embodiments of the invention include a method for avoiding memory bandwidth utilization during packet processing. The packet processing core receives a plurality of packets. The packet processing core identifies the packet's quality of service (QoS) descriptor. The packet processing core determines that at least one packet should be moved to an off-chip packet stored prior to the packet being transmitted to the egress port. The packet processing core bases that determination, at least in part, on the packet's QoS descriptor. The packet processing core moves the determined packets to the off-chip packet store. The packet processing core determines that at least one packet should not be moved to the off-chip packet store prior to the packet being transmitted to the egress port. This determination is also made, at least in part, based on the packet's QoS descriptor.

Abstract: In one aspect, the present invention reduces the amount of low-latency memory needed for rules-based packet classification by representing a packet classification rules database in compressed form. A packet processing rules database, e.g., an ACL database comprising multiple ACEs, is preprocessed to obtain corresponding rule fingerprints. These rule fingerprints are much smaller than the rules and are easily accommodated in on-chip or other low-latency memory that is generally available to the classification engine in limited amounts. The rules database in turn can be stored in off-chip or other higher-latency memory, as initial matching operations involve only the packet key of the subject packet and the fingerprint database. The rules database is accessed for full packet classification only if a tentative match is found between the packet key and an entry in the fingerprint database. Thus, the present invention also advantageously minimizes accesses to the rules database.

Abstract: Embodiments of the invention a method for policing a packet at line rate. A hierarchical policer receives a policer request comprising packet characteristics and identifying request configuration information. The hierarchical policer retrieves meter states specified by the request configuration information. The hierarchical policer processes packet characteristics through meters to generate a meter result. The hierarchical policer generates a hierarchical policer table lookup address using a plurality of meter types, a plurality of input color controls, one or more of the packet characteristics, the meter results, and a plurality of coupling algorithm identifiers. The hierarchical policer reads a hierarchical meter result from a hierarchical policer result table, containing at least a final output packet attribute that classifies the packet. The hierarchical policer updates one or more of the meter states based on the plurality of meter state results.

Abstract: In one aspect, the present invention reduces the amount of low-latency memory needed for rules-based packet classification by representing a packet classification rules database in compressed form. A packet processing rules database, e.g., an ACL database comprising multiple ACEs, is preprocessed to obtain corresponding rule fingerprints. These rule fingerprints are much smaller than the rules and are easily accommodated in on-chip or other low-latency memory that is generally available to the classification engine in limited amounts. The rules database in turn can be stored in off-chip or other higher-latency memory, as initial matching operations involve only the packet key of the subject packet and the fingerprint database. The rules database is accessed for full packet classification only if a tentative match is found between the packet key and an entry in the fingerprint database. Thus, the present invention also advantageously minimizes accesses to the rules database.

Abstract: Embodiments of the invention a method for policing a packet at line rate. A hierarchical policer receives a policer request comprising packet characteristics and identifying request configuration information. The hierarchical policer retrieves meter states specified by the request configuration information. The hierarchical policer processes packet characteristics through meters to generate a meter result. The hierarchical policer generates a hierarchical policer table lookup address using a plurality of meter types, a plurality of input color controls, one or more of the packet characteristics, the meter results, and a plurality of coupling algorithm identifiers. The hierarchical policer reads a hierarchical meter result from a hierarchical policer result table, containing at least a final output packet attribute that classifies the packet. The hierarchical policer updates one or more of the meter states based on the plurality of meter state results.

Abstract: Embodiments of the invention include a method for avoiding memory bandwidth utilization during packet processing. The packet processing core receives a plurality of packets. The packet processing core identifies the packet's quality of service (QoS) descriptor. The packet processing core determines that at least one packet should be moved to an off-chip packet stored prior to the packet being transmitted to the egress port. The packet processing core bases that determination, at least in part, on the packet's QoS descriptor. The packet processing core moves the determined packets to the off-chip packet store. The packet processing core determines that at least one packet should not be moved to the off-chip packet store prior to the packet being transmitted to the egress port. This determination is also made, at least in part, based on the packet's QoS descriptor.

Abstract: A hierarchical pipelined distributed scheduling traffic manager includes multiple hierarchical levels to perform hierarchical winner selection and propagation in a pipeline including selecting and propagating winner queues of a lower level to subsequent levels to determine one final winning queue. The winner selection and propagation is performed in parallel between the levels to reduce the time required in selecting the final winning queue. In some embodiments, the hierarchical traffic manager is separated into multiple separate sliced hierarchical traffic managers to distributively process the traffic.

Abstract: A hierarchical pipelined distributed scheduling traffic manager includes multiple hierarchical levels to perform hierarchical winner selection and propagation in a pipeline including selecting and propagating winner queues of a lower level to subsequent levels to determine one final winning queue. The winner selection and propagation is performed in parallel between the levels to reduce the time required in selecting the final winning queue. In some embodiments, the hierarchical traffic manager is separated into multiple separate sliced hierarchical traffic managers to distributively process the traffic.

Abstract: A SGMII that operates to transfer data between MAC and PHY chips at 2500/1000/100/10 Mbps utilizes a unique frame extending technique in one embodiment where frames having multiples of 2 and 3 data bytes are utilized to change the data transfer rate by multiples of 2.5. In another embodiment different clock signals are utilized.

Abstract: An apparatus and method for performing logical operations on information in the communications protocol stack, such as the transport layer (L4) port numbers, characterizing a received packet or frame of data in a data communications device such as a router or switch. The results of the logical operations, along with other packet/frame-identifying data, are used to generate a more efficient lookup key. A content addressable memory (CAM) lookup is used to determine the action indicated by the rules defined by a rule-based routing or switching scheme, such as an access control list (ACL). The results of these logical operations extend the key space and thus provide a finer-grained match between the original, unextended input key and a rule action, thereby pointing to a rule action precisely tailored to packet processing.

Abstract: An apparatus and method for handling interrupt requests from any of a plurality of interrupters to any of a plurality of interrupt handlers. Each interrupt handler includes an interrupt input request queue for holding a plurality of incoming interrupt requests. A system controller is connected to the interrupters and the interrupt handlers, and includes an input queue coupled to each interrupter for receiving a plurality of interrupt requests. The system controller includes a processor and a memory storing instructions for controlling its operation. The system controller also includes an output queue coupled to each interrupt handler (which in many cases will also be interrupters), and a counter for monitoring the current number, at any given time, of interrupt requests pending in each of the interrupt input queues in the interrupt handlers.

Abstract: A virtual I/O processor (VIOP) is implemented using a programmed I/O (PIO) unit. The PIO unit is complemented by a VIOP interrupt, a VIOP interrupt handler, and a number of VIOP data structures. Preferably, the PIO unit is further complemented with a set of dedicated I/O global registers, a number of VIOP library read/write routines for various I/O device types, and non-blocking read and write operations. During execution, these elements cooperate with each other to perform multiple sequences of programmed I/Os as if they were being performed by a dedicated I/O processor.

Abstract: An apparatus and method for handling interrupt requests from any of a plurality of interrupters to any of a plurality of interrupt handlers. Each interrupt handler includes an interrupt input request queue for holding a plurality of incoming interrupt requests. A system controller is connected to the interrupters and the interrupt handlers, and includes an input queue coupled to each interrupter for receiving a plurality of interrupt requests. The system controller includes a processor and a memory storing instructions for controlling its operation. The system controller also includes an output queue coupled to each interrupt handler (which in many cases will also be interrupters), and a counter for monitoring the current number, at any given time, of interrupt requests pending in each of the interrupt input queues in the interrupt handlers.

Abstract: Deadlock detection and masking systems are incorporated into a bus coupler intercoupling at least two buses, wherein at least one master is coupled to each bus and at least one slave is coupled to at least one of the buses. The bus coupler also includes an arbiter coupled to the buses to determine which master may control each bus. The deadlock detection system detects a potential arbitration deadlock condition between two master devices seeking control of a bus and access to a slave. Once a potential arbitration deadlock is detected, the masking system is activated to prohibit the second master from gaining control of the second bus for a random period of time. The random time delay acts as a mask to provide the first master device an opportunity to reaccess the slave device and avoid the deadlock situation. By providing a random masking period complementary, synchronized arbitration deadlocks are avoided.

Abstract: A bus-to-bus interface preserves data coherence between masters and slaves operating within a multiple processor computer system. Two buses are connected via the interface. The first bus connects a number of self-identifying masters. The second bus connects a number of master devices and a number of slave devices. The second bus has no mechanism with which devices connected to the second bus may identify themselves. The interface contains a pair of registers for each slave device connected through the second bus. One register stores a busy bit if the corresponding slave is engaged on behalf of a master. The second register stores an identifying code for the master delegating a task to the corresponding slave. When a slave has accepted a task on behalf of a master and commanded the master to relinquish the bus, the busy register will be set and the master identification register will store the identifying code for the delegating master.