Abstract: A method and system for speculatively pre-fetching data from a memory. A memory controller on a data bus “snoops” data requests put on the data bus by a bus control logic. Based on information in the header of the data request, such as transaction type, tag, transaction size, etc., a speculative pre-fetch is made to read data from the memory associated with the memory controller. If the speculative fetch turns out to be correct, then the memory controller makes an assumption that the pre-fetch was too conservative (non-speculative), and a pre-fetch for a next data request is performed at an earlier more speculative time. If the speculative fetch turns out to be incorrect, then the memory controller makes an assumption that the pre-fetch was too speculative (made early), and a pre-fetch for a next data request is performed at a later less speculative time.

Abstract: The present invention provides an improved method, an system, and a set of computer implemented instructions for handling a cache containing multiple single-bit hard errors on multiple addresses within a data processing system. Such handles will prevent any down time by logging in the parts to be replaced by an operator when certain level of bit errors is reached. When a hard error exists on a cache address for the first time, serviceable first hard error, that cache line is deleted. Thus the damaged memory device is no longer used by the system. As a result, the system is running with “N?x” lines wherein “N” constitutes the total number of existing lines and “x” is less than “N”. An alternative method is to exchange the damaged memory device to a spare memory device. In order to provide such services, the system must first differentiate whether an error is a soft or hard error.

Abstract: A data processing system includes a requester having a request queue and a recipient. The requester, which buffers a request in an entry of the request queue, transmits the request to the recipient for servicing. According to the request-and-forget protocol, the requester removes the request from the entry of the request queue without receipt of any indication that the request has been serviced.

Abstract: A computer system includes a home node and one or more remote nodes coupled by a node interconnect. The home node includes a local interconnect, a node controller coupled between the local interconnect and the node interconnect, a home system memory, a memory directory including a plurality of entries, and a memory controller coupled to the local interconnect, the home system memory and the memory directory. The memory directory includes a plurality of entries that each provide an indication of whether or not an associated data granule in the home system memory has a corresponding cache line held in at least one remote node. The memory controller includes demand invalidation circuitry that, responsive to a data request for a requested data granule in the home system memory, reads an associated entry in the memory directory and issues an invalidating command to at least one remote node holding a cache line corresponding to the requested data granule.

Abstract: A non-uniform memory access (NUMA) computer system includes at least one remote node and a home node coupled by a node interconnect. The home node contains a home system memory and a memory controller. In response to receipt of a data request from a remote node, the memory controller determines whether to grant exclusive or non-exclusive ownership of requested data specified in the data request by reference to history information indicative of prior data accesses originating in the remote node. The memory controller then transmits the requested data and an indication of exclusive or non-exclusive ownership to the remote node.

Abstract: A data processing system includes a plurality of nodes, which each contain at least one agent and each have an associated node identifier, and memory distributed among the plurality of nodes. The data processing system further includes an interconnect containing a segmented data channel, where each node contains a segment of the segmented data channel and each segment is coupled to at least one other segment by destination logic. In response to snooping a write request of a master agent on the interconnect, a target agent that will service the write request places its node identifier in a snoop response. When the master agent receives the combined response, which contains the node identifier of the target agent, the master agent issues on the segmented data channel a write data transaction specifying the node identifier of the target agent as a destination identifier.

Abstract: A method, apparatus, and computer implemented instructions for processing an error in a multiprocessor data processing system. An error is detected within the data processing system. A chip, causing the error, is identified within a plurality of chips to form an identified chip. Data is collected from the identified chip and hardware associated with the identified chip.

Abstract: Upon snooping an operation in which an intervention is permitted or required, an intervening cache may elect to source only that portion of a requested cache line which is actually required, rather than the entire cache line. For example, if the intervening cache determines that the requesting cache would likely be required to invalidate the cache line soon after receipt, less than the full cache line may be sourced to the requesting cache. The requesting cache will not cache less than a full cache line, but may forward the received data to the processor supported by the requesting cache. Data bus bandwidth utilization may therefore be reduced. Additionally, the need to subsequently invalidate the cache line within the requesting cache is avoided, together with the possibility that the requesting cache will retry an operation requiring invalidation of the cache line.

Abstract: A computer system includes a processing unit, a system memory, and a memory controller coupled to the processing unit and the system memory. According to the present invention, the memory controller accesses the system memory to obtain prefetch data and transmits the prefetch data to the processing unit in a prefetch write operation specifying the processing unit in a destination field. In one embodiment, the memory controller transmits the prefetch write operation in response to receipt of a prefetch hint from the processing unit, which may accompany a read-type request by the processing unit. This prefetch methodology may advantageously be implemented imprecisely, with the memory controller responding to the prefetch hint only if a prefetch queue is available and ignoring the prefetch hint otherwise. The processing unit may similarly ignore the prefetch write operation if no snoop queue is available.

Abstract: A non-uniform memory access (NUMA) computer system and associated method of operation are disclosed. The NUMA computer system includes at least a remote node and a home node coupled to an interconnect. The remote node contains at least one processing unit coupled to a remote system memory, and the home node contains at least a home system memory. To reduce access latency for data from other nodes, a portion of the remote system memory is allocated as a remote memory cache containing data corresponding to data resident in the home system memory. In one embodiment, access bandwidth to the remote memory cache is increased by distributing the remote memory cache across multiple system memories in the remote node.

Abstract: A non-uniform memory access (NUMA) computer system includes a first node and a second node coupled by a node interconnect. The second node includes a local interconnect, a node controller coupled between the local interconnect and the node interconnect, and a controller coupled to the local interconnect. In response to snooping an operation from the first node issued on the local interconnect by the node controller, the controller signals acceptance of responsibility for coherency management activities related to the operation in the second node, performs coherency management activities in the second node required by the operation, and thereafter provides notification of performance of the coherency management activities.

Abstract: A cache controller for a processor in a remote node of a system bus in a multiway multiprocessor link sends out a cache deallocate address transaction (CDAT) for a given cache line when that cache line is flushed and information from memory in a home node is no longer deemed valid for that cache line of that remote node processor. A local snoop of that CDAT transaction is then performed as a background function by other processors in the same remote node. If the snoop results indicate that same information is valid in another cache, and that cache decides it better to keep it valid in that remote node, then the information remains there. If the snoop results indicate that the information is not valid among caches in that remote node, or will be flushed due to the CDAT, the system memory directory in the home node of the multiprocessor link is notified and changes state in response to this.

Abstract: In addition to an address tag, a coherency state and an LRU position, each cache directory entry includes historical processor access, snoop operation, and system controller hint information for the corresponding cache line. Each entry includes different subentries for different processors which have accessed the corresponding cache line, with subentries containing a processor access sequence segment, a snoop operation sequence segment, and a system controller hint history segment. In addition to an address tag, within each system controller bus transaction sequence log directory entry is contained one or more opcodes identifying bus operations addressing the corresponding cache line, a processor identifier associated with each opcode, and a timestamp associated with each opcode.

Abstract: A non-uniform memory access (NUMA) computer system includes a remote node coupled by a node interconnect to a home node including a home system memory. The remote node includes a plurality of snoopers coupled to a local interconnect. The plurality of snoopers includes a cache that caches a cache line corresponding to but modified with respect to data resident in the home system memory. The cache has a cache controller that issues a deallocate operation on the local interconnect in response to deallocating the modified cache line. The remote node further includes a node controller, coupled between the local interconnect and the node interconnect, that transmits the deallocate operation to the home node with an indication of whether or not a copy of the cache line remains in the remote node following the deallocation. In this manner, the local memory directory associated with the home system memory can be updated to precisely reflect which nodes hold a copy of the cache line.

Abstract: System bus snoopers within a multiprocessor system in which dynamic application sequence behavior information is maintained within cache directories append the dynamic application sequence behavior information for the target cache line to their snoop responses. The system controller, which may also maintain dynamic application sequence behavior information in a history directory, employs the available dynamic application sequence behavior information to append “hints” to the combined response, appends the concatenated dynamic application sequence behavior information to the combined response, or both. Either the hints or the dynamic application sequence behavior information may be employed by the bus master and other snoopers in cache management.

Abstract: An apparatus for allocating data usage in an embedded dynamic random access memory (DRAM) device is disclosed. The apparatus for allocating data usages within an embedded dynamic random access memory (DRAM) device comprises a control analysis circuit, a data/command flow circuit, and a partition management control. The control analysis circuit generates an allocation signal in response to processing performances of a processor. Coupled to an embedded DRAM device, the data/command flow circuit controls data flow from the processor to the embedded DRAM device. The partition management control, coupled to the control analysis circuit, partitions the embedded DRAM device into a first partition and a second partition. The data stored in the first partition are different from the data stored in the second partition according to their respective usage. The allocation percentages of the first and second partitions are dynamically allocated by the allocation signal from the control analysis circuit.

Abstract: A data processing system includes a plurality of nodes, which each contain at least one agent, and data storage accessible to agents within the nodes. The plurality of nodes are coupled by a non-hierarchical interconnect including multiple non-blocking uni-directional address channels and at least one uni-directional data channel. The agents, which are each coupled to and snoop transactions on all of the plurality of address channels, can only issue transactions on an associated address channel. The uni-directional channels employed by the present non-hierarchical interconnect architecture permit high frequency pumped operation not possible with conventional bi-directional shared system buses. In addition, access latencies to remote (cache or main) memory incurred following local cache misses are greatly reduced as compared with conventional hierarchical systems because of the absence of inter-level (e.g., bus acquisition) communication latency.

Abstract: A non-uniform memory access (NUMA) data processing system includes a plurality of nodes coupled to a node interconnect. The plurality of nodes contain a plurality of processing units and at least one system memory having a table (e.g., a page table) resident therein. The table includes at least one entry for translating a group of non-physical addresses to physical addresses that individually specifies control information pertaining to the group of non-physical addresses for each of the plurality of nodes. The control information may include one or more data storage control fields, which may include a plurality of write through indicators that are each associated with a respective one of the plurality of nodes. When a write through indicator is set, processing units in the associated node write modified data back to system memory in a home node rather than caching the data.

Abstract: A computer system includes a home node and one or more remote nodes coupled by a node interconnect. The home node includes a local interconnect, a node controller coupled between the local interconnect and the node interconnect, a home system memory, and a memory controller coupled to the local interconnect and the home system memory. In response to receipt of a data request from the remote node, the memory controller transmits requested data from the home system memory to the remote node and, in a separate transfer, conveys responsibility for global coherency management for the requested data from the home node to the remote node. By decoupling responsibility for global coherency management from delivery of the requested data in this manner, the memory controller queue allocated to the data request can be deallocated earlier, thus improving performance.

Abstract: A non-uniform memory access (NUMA) computer system includes a node interconnect to which a remote node and a home node are coupled. The home node contains a home system memory, and the remote node includes at least one processing unit and a cache. In response to the cache deallocating an unmodified cache line that corresponds to data resident in the home system memory, a cache controller of the cache issues a deallocate operation on a local interconnect of the remote node. In one embodiment, the deallocate operation is further transmitted to the home node via the node interconnect only in response to an indication, such as a combined response, that no other cache in the remote node caches the cache line. In response to receipt of the deallocate operation, a memory controller in the home node updates a local memory directory associated with the home system memory to indicate that the remote node does not hold a copy of the cache line.