摘要:
A multiprocessor data processing system includes at least first and second coherency domains, where the first coherency domain includes a system memory and a cache memory. According to a method of data processing, a cache line is buffered in a data array of the cache memory and a state field in a cache directory of the cache memory is set to a coherency state to indicate that the cache line is valid in the data array, that the cache line is held in the cache memory non-exclusively, and that another cache in said second coherency domain may hold a copy of the cache line.
摘要:
A data processing system includes at least a first through third processing nodes coupled by an interconnect fabric. The first processing node includes a master, a plurality of snoopers capable of participating in interconnect operations, and a node interface that receives a request of the master and transmits the request of the master to the second processing unit with a nodal scope of transmission limited to the second processing node. The second processing node includes a node interface having a directory. The node interface of the second processing node permits the request to proceed with the nodal scope of transmission if the directory does not indicate that a target memory block of the request is cached other than in the second processing node and prevents the request from succeeding if the directory indicates that the target memory block of the request is cached other than in the second processing node.
摘要:
A data processing system includes an interconnect fabric, a system memory coupled to the interconnect fabric and including a virtual barrier synchronization region allocated to storage of virtual barrier synchronization registers (VBSRs), and a plurality of processing units coupled to the interconnect fabric and operable to access the virtual barrier synchronization region of the system memory. Each of the plurality of processing units includes a processor core and a cache memory including a cache array that caches VBSR lines from the virtual barrier synchronization region of the system memory and a cache controller. The cache controller, responsive to a store request from the processor core to update a particular VBSR line, performs a non-blocking update of the cache array in each other of the plurality of processing units contemporaneously holding a copy of the particular VBSR line by transmitting a VBSR update command on the interconnect fabric.
摘要:
A data processing system includes a plurality of processing units coupled by an interconnect fabric. In response to a data request, a victim cache line is selected for castout from a first lower level cache of a first processing unit, and a target lower level cache of one of the plurality of processing units is selected based upon architectural proximity of the target lower level cache to a home system memory to which the address of the victim cache line is assigned. The first processing unit issues on the interconnect fabric a lateral castout (LCO) command that identifies the victim cache line to be castout from the first lower level cache and indicates that the target lower level cache is an intended destination. In response to a coherence response indicating success of the LCO command, the victim cache line is removed from the first lower level cache and held in the second lower level cache.
摘要:
A cache coherent data processing system includes at least first and second coherency domains. In a first cache memory within the first coherency domain of the data processing system, a memory block is held in a storage location associated with an address tag and a coherency state field. A determination is made if a home system memory assigned an address associated with the memory block is within the first coherency domain. If not, the coherency state field is set to a coherency state that indicates that the address tag is valid, that the storage location does not contain valid data, the first coherency domain does not contain the home system memory, and that, following formation of the coherency state, the memory block is cached outside of the first coherency domain.
摘要:
A cache coherent data processing system includes at least first and second coherency domains. In a first cache memory within the first coherency domain of the data processing system, a coherency state field associated with a storage location and an address tag is set to a first data-invalid coherency state that indicates that the address tag is valid and that the storage location does not contain valid data. In response to snooping an exclusive access operation, the exclusive access request specifying a target address matching the address tag and indicating a relative domain location of a requestor that initiated the exclusive access operation, the first cache memory updates the coherency state field from the first data-invalid coherency state to a second data-invalid coherency state that indicates that the address tag is valid, that the storage location does not contain valid data, and whether a target memory block associated with the address tag is cached within the first coherency domain upon successful completion of the exclusive access operation based upon the relative location of the requestor.
摘要:
A cache, system and method for reducing the number of rejected snoop requests. An incoming snoop request is entered in the first available latch in a pipeline of latches in a stall/reorder unit if the stall/reorder unit is not full. The entered snoop request is dispatched to a selector upon entering a bottom latch in the pipeline. The stall/reorder unit is not informed as to whether the dispatched snoop request is accepted by an arbitration mechanism for several clock cycles after the dispatch occurred. A copy of the dispatched snoop request is stored in a top latch in an overrun pipeline of latches in the first unit upon dispatching the snoop request. By maintaining information about the snoop request, the snoop request may be dispatched again to the selector in case the dispatched snoop request was rejected thereby increasing the chance that the snoop request will ultimately be accepted.
摘要:
A cache, system and method for reducing the number of rejected snoop requests. An incoming snoop request is entered in the first available latch in a pipeline of latches in a stall/reorder unit if the stall/reorder unit is not full. The entered snoop request is dispatched to a selector upon entering a bottom latch in the pipeline. The stall/reorder unit is not informed as to whether the dispatched snoop request is accepted by an arbitration mechanism for several clock cycles after the dispatch occurred. A copy of the dispatched snoop request is stored in a top latch in an overrun pipeline of latches in the first unit upon dispatching the snoop request. By maintaining information about the snoop request, the snoop request may be dispatched again to the selector in case the dispatched snoop request was rejected thereby increasing the chance that the snoop request will ultimately be accepted.
摘要:
In a data processing system, a plurality of agents communicate operations therebetween. Each operation includes a request and a combined response representing a system-wide response to the request. Latencies of requests and combined responses between the plurality of agents are observed. Each of the plurality of agents is configured with a respective duration of a protection window extension by reference to the observed latencies. Each protection window extension is a period following receipt of a combined response during winch an associated one of the plurality of agents protects transfer of coherency ownership of a data granule between agents. The plurality of agents employing protection window extensions in accordance with the configuration, and at least two of the agents have protection window extensions of differing durations.
摘要:
A cache, system and method for reducing the number of rejected snoop requests. A “stall/reorder unit” in a cache receives a snoop request from an interconnect. Information, such as the address, of the snoop request is stored in a queue of the stall/reorder unit. The stall/reorder unit forwards the snoop request to a selector which also receives a request from a processor. An arbitration mechanism selects either the snoop request or the request from the processor. If the snoop request is denied by the arbitration mechanism, information, e.g., address, about the snoop request may be maintained in the stall/reorder unit. The request may be later resent to the selector. This process may be repeated up to “n” clock cycles. By providing the snoop request additional opportunities (n clock cycles) to be accepted by the arbitration mechanism, fewer snoop requests may ultimately be denied.