Abstract: Various methods and apparatuses are described in which an interconnect couples to a plurality of initiator network resources and a plurality of target network resources. The interconnect may include a first stage of circuitry, a second stage of circuitry, and an arbitration controller. The first stage of circuitry receives incoming transactions from the plurality of initiator network resources. The second stage of circuitry passes outgoing transactions to the plurality of target network resources connecting to the interconnect. The arbitration controller arbitrates transactions from the plurality of initiator network resources destined to one or more of the target network resources. The target network resources supply their availability to service a transaction to the arbitration controller.

Abstract: A method for communicating between nodes of a plurality of nodes is disclosed. Each node includes a plurality of processors and an interconnect chipset. The method issues a request for data from a processor in a first node and passes the request for data to other nodes through an expansion port (or scalability port). The method also starts an access of a memory in response to the request for data and snoops a processor cache of each processor in each node. The method accordingly identifies the location of the data in either the processor cache or memory in the node having the processor issuing the request or in a processor cache or memory of another node. A method for requesting data between two directly coupled nodes in a router system is also disclosed. A method for requesting data between three or more nodes in an interconnect system is also disclosed. A method for resolving crossing cases in an interconnect system is also disclosed.

Abstract: A method for communicating between nodes of a plurality of nodes is disclosed. Each node includes a plurality of processors and an interconnect chipset. The method issues a request for data from a processor in a first node and passes the request for data to other nodes through an expansion port (or scalability port). The method also starts an access of a memory in response to the request for data and snoops a processor cache of each processor in each node. The method accordingly identifies the location of the data in either the processor cache or memory in the node having the processor issuing the request or in a processor cache or memory of another node. A method for requesting data between two directly coupled nodes in a router system is also disclosed. A method for requesting data between three or more nodes in an interconnect system is also disclosed. A method for resolving crossing cases in an interconnect system is also disclosed.

Abstract: A method and apparatus for a round robin resource arbitration scheme is described. An apparatus to provide round robin token arbitration comprises at least two token arbiters, each token arbiter associated with a node to which at least two sub-trees are connected, each sub-tree comprising a token arbiter or a finite state machine requestor.

Abstract: Various methods and apparatuses are described in which an arbitration controller cooperates with arbitration logic. The arbitration controller has a plurality of inputs that receive one or more transactions from a plurality of blocks of functionality. The arbitration controller arbitrates requests for access to a shared resource amongst the plurality of blocks of functionality by implementing an arbitration policy. The arbitration policy groups the transactions from the plurality of blocks of functionality into global groups of transactions for servicing by that shared resource. All of the transactions in a first global group are serviced by that shared resource prior to servicing transactions in a next global group of transactions. The arbitration logic facilitates the arbitration policy. The arbitration logic includes cascaded arbitration units that hierarchically arbitrate for the shared resource.

Abstract: The present invention provides for the scheduling of requests to one resource from a plurality of initiator devices. In one embodiment, scheduling of requests within threads and scheduling of initiator device access is performed wherein requests are only reordered between threads.

Abstract: The present invention is directed to a method and apparatus for scheduling a resource to meet quality of service guarantees. In one embodiment of three levels of priority, if a channel of a first priority level is within its bandwidth allocation, then a request is issued from that channel. If there are no requests in channels at the first priority level that are within the allocation, requests from channels at the second priority level that are within their bandwidth allocation are chosen. If there are no requests of this type, requests from channels at the third priority level or requests from channels at the first and second levels that are outside of their bandwidth allocation are issued. The system may be implemented using rate-based scheduling.

Abstract: A method for communicating between nodes of a plurality of nodes is disclosed. Each node includes a plurality of processors and an interconnect chipset. The method issues a request for data from a processor in a first node and passes the request for data to other nodes through an expansion port (or scalability port). The method also starts an access of a memory in response to the request for data and snoops a processor cache of each processor in each node. The method accordingly identifies the location of the data in either the processor cache or memory in the node having the processor issuing the request or in a processor cache or memory of another node. A method for requesting data between two directly coupled nodes in a router system is also disclosed. A method for requesting data between three or more nodes in an interconnect system is also disclosed. A method for resolving crossing cases in an interconnect system is also disclosed.

Abstract: A pipelined network is disclosed which provides for at least one mode to control the state of a response flag and when the target device is unable to respond to an initiator device request.

Abstract: A method and apparatus for automatic configuration of multiple on-chip interconnects have been described. In one embodiment, the invention reduces the configuration time of several on-chip network features, and also ensures that these features are configured correctly to minimize errors in a design.

Abstract: A method and apparatus for error handling in networks have been described.

Abstract: A method and apparatus of a configurable address mapping and protection architecture and hardware for on-chip systems have been described.

Abstract: A method and apparatus for multicast handling in mixed core systems have been described.

Abstract: A communication system and method with configurable posting points have been described.

Abstract: A method and apparatus for composing on-chip interconnects with configurable interfaces have been described.

Abstract: A method and apparatus for on-chip inter-network performance optimization using configurable performance parameters have been described.

Abstract: The present invention provides for the scheduling of requests to one resource from a plurality of initiator devices. In one embodiment, scheduling of requests within threads and scheduling of initiator device access is performed wherein requests are only reordered between threads.

Abstract: The present invention consists of a cache coherence protocol within a cache coherence unit for use in a data processing system. The data processing system is comprised of multiple nodes, each node having a plurality of processors with associated caches, a memory, and input/output. The processors within the node are coupled to a memory bus operating according to a “snoopy” protocol. This invention includes a cache coherence protocol for a sparse directory in combination with the multiprocessor nodes. In addition, the invention has the following features: the current state and information from the incoming bus request are used to make an immediate decision on actions and next state; the decision mechanism for outgoing coherence is pipelined to follow the bus; and the incoming coherence pipeline acts independently of the outgoing coherence pipeline.

Abstract: An algorithm for selecting a directory entry in a multiprocessor-node system. In response to a memory request from a processor in a processor node, the algorithm finds an available entry to store information about the requested memory line. If at least one entry is available, then the algorithm uses one of the available entries. Otherwise, the algorithm searches for a “shared” entry. If at least one shared entry is available, then the algorithm uses one of the shared entries. Otherwise, the algorithm searches for a “dirty” entry. If at least one dirty entry is available, then the algorithm uses one of the dirty entries. In selecting a directory entry, the algorithm uses a “least-recently-used” (LRU) algorithm because an entry that was not recently used is more likely to be stale. Further, to improve system performance, the algorithm preferably uses a shared entry before using a dirty entry.

Abstract: The present invention provides for the scheduling of requests to one resource from a plurality of initiator devices. In one embodiment, scheduling of requests within threads and scheduling of initiator device access is performed wherein requests are only reordered between threads.