Abstract: An arbiter is provided for resolving contention on synchronous packet switched busses, including busses composed of a plurality of pipelined segments, to ensure that all devices serviced by such a bus are given fair, bounded time access to the bus and to permit such devices to fill all available bus cycles with packets. Flow control for shared memory multiprocessors is readily implemented with this arbiter because the arbiter supports different types of arbitration requests and the prioritization of such arbitration requests by type.

Abstract: An improved memory model and implementation is disclosed. The memory model includes a Total Store Ordering (TSO) and Partial Store Ordering (PSO) memory model to provide a partial order for the memory operations which are issued by multiple processors. The TSO memory model includes a FIFO Store Buffer for Store, and Atomic Load-Store operations. The Load operations are not placed in the FIFO Store Buffer. The Load operation checks for a value stored in the same location in the FIFO Store Buffer; if no such value is found, then requested value is returned from memory. The PSO model also includes a Store Buffer for Store, and Atomic Load-Store operations. However, unlike the TSO model, the Store Buffer in the PSO model is not FIFO. The processors in the PSO model may issue the Store and Atomic Load-Store in a certain order; however, such operations may be executed by memory out of the order issued by the processors. The execution order is assured only by address matching and the STBAR operation.

Abstract: A high speed, synchronous, packet-switched inter-chip bus apparatus and method for transferring data between multiple system buses and a cache controller. In the preferred embodiment, the bus connects a cache controller client within the external cache of a processor to a plurality of bus watcher clients, each of which is coupled to a separate system bus. The bus allows the cache controller to provide independent processor-side access to the cache and allows the bus watchers to handle functions related to bus-snooping. An arbiter is employed to allow the bus to be multiplexed between the bus watchers and cache controller. Flow control mechanisms are also employed to ensure that queues receiving packets or arbitration requests over the bus never overflow. A default grantee mechanism is employed to minimize the arbitration latency due to a request for the bus when the bus is idle.