摘要:
Output logic generates read requests using a programmable schedule that controls read bandwidth for multiple data streams and stores the read requests in a queuing device. The output logic also dequeues the read requests based on a similar programmable schedule, forwards the read requests to the memory, and reads data units from the memory based on the read requests.
摘要:
A data read/write system receives a key associated with a data read request. The data read/write system hashes the key to obtain a first hash value and hashes the key to obtain a second hash value, where the second hash value is different than the first hash value. The data read/write system obtains a pointer from a pointer array using the first and second hash values, and uses one or more bits of the pointer and the first hash value to retrieve data from a data look-up array.
摘要:
A system manages a buffer having a group of entries. The system receives information relating to a read request for a memory. The system determines whether an entry in the buffer contains valid information. If the entry is determined to contain valid information, the system transmits the information in the entry in an error message. The system may then store the received information in the entry. In another implementation, the system stores data in one of the entries of the buffer, removes an address corresponding to the one entry from an address list, and starts a timer associated with the one entry. The system also determines whether the timer has exceeded a predetermined value, transferring the data from the one entry when the timer has exceeded the predetermined value, and adds the address back to the address list.
摘要:
Methods and systems consistent with the present invention provide a programmable table which allows software to define a plurality of branching functions, each of which maps a vector of condition codes to a branch offset. This technique allows for a flexible multi-way branching functionality, using a conditional branch outcome table that can be specified by a programmer. Any instruction can specify the evaluation of arbitrary conditional expressions to compute the values for the condition codes, and can choose a particular branching function. When the processor executes the instruction, the processor's arithmetic/logical functional units evaluate the conditional expressions and then the processor performs the branch operation, according to the specified branching function.
摘要:
Methods and systems consistent with the present invention provide a programmable table which allows software to define a plurality of branching functions, each of which maps a vector of condition codes to a branch offset. This technique allows for a flexible multi-way branching functionality, using a conditional branch outcome table that can be specified by a programmer. Any instruction can specify the evaluation of arbitrary conditional expressions to compute the values for the condition codes, and can choose a particular branching function. When the processor executes the instruction, the processor's arithmetic/logical functional units evaluate the conditional expressions and then the processor performs the branch operation, according to the specified branching function.
摘要:
A standalone router is integrated into a multi-chassis router. Integrating the standalone router into a multi-chassis router requires replacing switch cards in the standalone router with multi-chassis switch cards. The multi-chassis switch cards forward packets to a central switch card chassis for routing within the multi-chassis router. By incrementally replacing standalone switch cards with multi-chassis switch cards in the standalone router, packet forwarding performance is maintained during the integration.
摘要:
A system having multiple arbiters is constructed to reduce the chances of arbiters synchronizing with one another. Each arbiter includes a random process that introduces randomness into an arbitration scheme performed by the arbiter. Because of the randomness, the arbiters will not tend to synchronize with one another, even when receiving an identical stream of input values.
摘要:
A network device includes systems and methods for processes streams of data. The network device stores data and addresses corresponding to the streams in a memory. The addresses store pointers to the data. Output logic within the network device determines whether an address is required to be fetched. When no address is required to be fetched, then data is read from the memory. When an address is required to be fetched, the address is fetched from the memory and data is read from the memory using the fetched address. To facilitate this, notifications may be stored corresponding to the streams and notification pointers may be used to identify ones of the notifications to be processed. A prefetch pointer may also be used to identify a notification with one or more associated addresses to be prefetched.
摘要:
Data units received by a network device may be classified into traffic flow classes in which the determined traffic flow class for a data unit may be dynamically refined as the data unit is processed by the network device. A dispatch component of the network device may receive data units associated with traffic flow classes. Parallel processing engines of the network device may receive the data units from the dispatch component and may generate, for a least one of the data units, a plurality of dynamically refined indications of the traffic flow class to which the data unit belongs. Additionally, an ordering component of the network device may include a plurality of re-order queues, where the at least one data unit successively progresses through at least two of the re-order queues in an order defined by the plurality of dynamically refined indications of the traffic flow class.
摘要:
Data units received by a network device may be classified into traffic flow classes in which the determined traffic flow class for a data unit may be dynamically refined as the data unit is processed by the network device. A dispatch component of the network device may receive data units associated with traffic flow classes. Parallel processing engines of the network device may receive the data units from the dispatch component and may generate, for a least one of the data units, a plurality of dynamically refined indications of the traffic flow class to which the data unit belongs. Additionally, an ordering component of the network device may include a plurality of re-order queues, where the at least one data unit successively progresses through at least two of the re-order queues in an order defined by the plurality of dynamically refined indications of the traffic flow class.