Abstract: A cache coherence manager, disposed in a multi-core microprocessor, includes a request unit, an intervention unit, a response unit and an interface unit. The request unit receives coherent requests and selectively issues speculative requests in response. The interface unit selectively forwards the speculative requests to a memory. The interface unit includes at least three tables. Each entry in the first table represents an index to the second table. Each entry in the second table represents an index to the third table. The entry in the first table is allocated when a response to an associated intervention message is stored in the first table but before the speculative request is received by the interface unit. The entry in the second table is allocated when the speculative request is stored in the interface unit. The entry in the third table is allocated when the speculative request is issued to the memory.

Abstract: A processor with coherency-leveraged support for low latency message signaled interrupt handling includes multiple execution cores and their associated cache memories. A first cache memory associated a first of the execution cores includes a plurality of cache lines. The first cache memory has a cache controller including hardware logic, microcode, or both to identify a first cache line as an interrupt reserved cache line and map the first cache line to a host physical memory address translated from a guest physical memory address in the address space of a virtual machine to which an I/O device has been assigned. The controller may set a coherency state of the first cache line to shared and, in response to detecting an I/O transaction including I/O data from the I/O device and containing a reference to the host physical memory address, emulate a first message signaled interrupt identifying the host physical memory address.

Abstract: A method for providing a transactional memory is described. A cache coherency protocol is enforced upon a cache memory including cache lines, wherein each line is in one of a modified state, an owned state, an exclusive state, a shared state, and an invalid state. Upon initiation of a transaction accessing at least one of the cache lines, each of the lines is ensured to be either shared or invalid. During the transaction, in response to an external request for any cache line in the modified, owned, or exclusive state, each line in the modified or owned state is invalidated without writing the line to a main memory. Also, each exclusive line is demoted to either the shared or invalid state, and the transaction is aborted.

Abstract: Technologies for dynamic home tile mapping are described. an address request can be received from a processing core, the processing core being associated with a home tile table, the home tile table including respective mappings of one or more directory addresses to one or more home tiles. A buffer can be scanned to identify a presence of the address within the buffer. Based on an identification of the presence of the address within the buffer, a home tile identifier corresponding to the address can be provided from the buffer.

Abstract: A coherent attached processor proxy (CAPP) includes transport logic having a first interface configured to support communication with a system fabric of a primary coherent system and a second interface configured to support communication with an attached processor (AP) that is external to the primary coherent system and that includes a cache memory that holds copies of memory blocks belonging to a coherent address space of the primary coherent system. The CAPP further includes one or more master machines that initiate memory access requests on the system fabric of the primary coherent system on behalf of the AP, one or more snoop machines that service requests snooped on the system fabric, and a CAPP directory having a precise directory having a plurality of entries each associated with a smaller data granule and a coarse directory having a plurality of entries each associated with a larger data granule.

Abstract: A coherent attached processor proxy (CAPP) includes transport logic having a first interface configured to support communication with a system fabric of a primary coherent system and a second interface configured to support communication with an attached processor (AP) that is external to the primary coherent system and that includes a cache memory that holds copies of memory blocks belonging to a coherent address space of the primary coherent system. The CAPP further includes one or more master machines that initiate memory access requests on the system fabric of the primary coherent system on behalf of the AP, one or more snoop machines that service requests snooped on the system fabric, and a CAPP directory having a precise directory having a plurality of entries each associated with a smaller data granule and a coarse directory having a plurality of entries each associated with a larger data granule.

Abstract: A data processing system includes a system interconnect, a processor coupled to the system interconnect, and a cache coherency manager (CCM) coupled to the system interconnect. The processor includes a cache. A method includes generating, by the CCM, one or more snoop requests to the cache of the processor; storing the one or more snoop requests to the cache of the processor into a snoop queue; setting a cache enable indicator to indicate that the cache of the processor is to be disabled; in response to setting the cache enable indicator to indicate that the cache of the processor is to be disabled, selectively invalidating the one or more snoop requests to the cache of the processor, wherein the selectively invalidating is performed based on an invalidate snoop queue indicator of the processor; and disabling the cache.

Abstract: In one embodiment, a method includes receiving a read request from a first caching agent, determining whether a directory entry associated with the memory location indicates that the information is not present in a remote caching agent, and if so, transmitting the information from the memory location to the first caching agent before snoop processing with respect to the read request is completed. Other embodiments are described and claimed.

Abstract: A computer system is provided including a central processing unit having an internal cache, a memory controller is coupled to the central processing unit, and a closely coupled peripheral is coupled to the central processing unit. A coherent interconnection may exist between the internal cache and both the memory controller and the closely coupled peripheral, wherein the coherent interconnection is a bus.

Abstract: A snoop-based kernel integrity monitoring apparatus and a method thereof are provided. More particularly, provided are a kernel integrity monitoring apparatus which is provided as a hardware device independent of a host system, and snoops traffic occurring in a system bus of the host system and by detecting a write attempt in a kernel immutable region, monitors integrity of the kernel, and a method thereof. According to the apparatus and method, by analyzing traffic of the system bus of the host system, a write attempt in the kernel immutable region is detected. Thus, a transient attack which is difficult for a snapshot method to detect can be detected.

Abstract: Embodiments of the invention describe a cache coherency protocol that eliminates the need for ordering between message classes and also eliminates home tracker preallocation. Embodiments of the invention describe a less complex conflict detection and resolution mechanism (at the home agent) without any performance degradation in form of bandwidth or latency compared to prior art solutions. Embodiments of the invention describe a home agent that may receive request messages, e.g., data ownership request messages and data request messages, which include issuance data indicating an order of the respective message issued. Said home agent may determine whether an early or late conflict exists based, at least in part, on a received conflict response message and the issuance data of a most recent completed transaction.

Abstract: A context-free (stateless) dataport may allow multiple processors to perform read and write operations on a shared memory. The operations may include, for example, structured data operations such as image and video operations. The dataport may perform addressing computations associated with block memory operations. Therefore, the dataport may be able, for example, to relieve the processors that it serves from this duty. The dataport may be accessed using a message interface that may be implemented in a standard and generalized manner and that may therefore be easily transportable between different types of processors.

Abstract: Methods and apparatus relating to ring protocols and techniques are described. In one embodiment, a first agent generates a request to write to a cache line of a cache over a first ring of a computing platform. A second agent that receives the write request forwards it to a third agent over the first ring of the computing platform. In turn, a third agent (e.g., a home agent) receives data corresponding to the write request over a second, different ring of the computing platform and writes the data to the cache. Other embodiments are also disclosed.

Abstract: A cache controller receives a reference request from a CPU executing a program in which information indicative of a reference request specifying in shared memory, an area not having an update request and information indicative of a snoop reference request are distinguished from one another. When the reference request specifying an area not having the update request is received, the cache controller acquires from the shared memory and without performing a snoop process, information stored in the specified area. The cache controller stores the information acquired from the shared memory to the cache memory of the CPU executing the program.

Abstract: Methods and apparatus are provided for reusing snoop responses and data phase results in a bus controller. A bus controller receives an incoming bus transaction BTR1 corresponding to an incoming cache transaction CTR1 for an entry in at least one cache; issues a snoop request with a cache line address of the incoming bus transaction BTR1 for the entry to a plurality of cache controllers; collects at least one snoop response from the plurality of cache controllers; broadcasts a combined snoop response to the plurality of cache controllers, wherein the combined snoop response is a combination of the snoop responses from the plurality of cache controllers; and broadcasts cache line data from a source cache for the entry during a data phase to the plurality of cache controllers, wherein a subsequent cache transaction CTR2 for the entry is processed based on the broadcast combined snoop response and the broadcast cache line data.

Abstract: A computer-implemented method for adaptively configuring a cache includes: implementing a cache adaptation agent in a system that has multiple applications, the system including a memory and a disk storage, wherein the system creates a cache in the memory for use by each of the applications; monitoring, by the cache adaptation agent, the cache in use by at least one of the applications, the monitoring covering at least a size of the cache used by the application, how many objects are in the application's cache, and sizes of the objects in the application's cache; and configuring, by the cache adaptation agent and based on the monitoring, at least one of the cache and system behavior regarding at least one of the applications.

Abstract: Systems, methods, and devices for reducing snoop traffic in a central processing unit are provided. In accordance with one embodiment, an electronic device includes a central processing unit having a plurality of cores. A cache memory management system may be associated with each core that includes a cache memory device configured to store a plurality of cache lines, a page status table configured to track pages of memory stored in the cache memory device and to indicate a status of each of the tracked pages of memory, and a cache controller configured to determine, upon a cache miss, whether to broadcast a snoop request based at least in part on the status of one of the tracked pages in the page status table.

Abstract: In response to snooping a read-type memory access request of a requestor on a system fabric of a data processing system, a memory channel interface forwards the request to a memory buffer and starts a timer. In response to the forwarded request, the memory buffer performs a lookup of a target address of the request in a memory controller cache. In response to the target address hitting in a coherence state permitting provision of early data, the memory buffer provides a response indicating early data and provides a copy of a target memory block of the request to the memory channel interface. The memory channel interface, responsive to receipt prior to expiration of the timer of the response indicating early data, transmits the copy of the target memory block to the requestor via the system fabric prior to receiving a combined response of the data processing system to the request.

Abstract: A data processing apparatus is provided in which a processor unit accesses data values stored in a memory and a cache stores local copies of a subset of the data values. The cache maintains a status value for each local copy stored in the cache. When the processor unit executes a load-exclusive operation, a first data value is loaded from a specified memory location and an exclusive use monitor begins monitoring the specified memory location for accesses. When the processor unit executes a store-exclusive operation, a second data value is stored to the specified memory location if the exclusive use monitor indicates that the first data value has not been modified since the load-exclusive operation was executed.

Abstract: In response to snooping a read-type memory access request of a requestor on a system fabric of a data processing system, a memory channel interface forwards the request to a memory buffer and starts a timer. In response to the forwarded request, the memory buffer performs a lookup of a target address of the request in a memory controller cache. In response to the target address hitting in a coherence state permitting provision of early data, the memory buffer provides a response indicating early data and provides a copy of a target memory block of the request to the memory channel interface. The memory channel interface, responsive to receipt prior to expiration of the timer of the response indicating early data, transmits the copy of the target memory block to the requestor via the system fabric prior to receiving a combined response of the data processing system to the request.

Abstract: An address is received. One or more neighbors associated with the received address is/are determined. One or more neighboring hot metrics is/are determined for the one or more neighbors associated with the received address. A hot metric for the received address is determined based at least in part on the neighboring hot metrics.

Abstract: In a system including a plurality of CPU units having a cache memory of different capacity each other and a system controller that connects to the plurality of CPUs and controls cache synchronization, the system controller includes a cache synchronization unit which monitors an address contention between a preceding request and a subsequent request and a setting unit which sets different monitoring range of the contention between the preceding request and the subsequent request for each capacity of the cache memory in each of the CPU units.

Abstract: Systems, methods, and devices for efficient cache coherence between memory-sharing devices are provided. In particular, snoop traffic may be suppressed based at least partly on a table of block tracking entries (BTEs). Each BTE may indicate whether groups of one or more cache lines of a block of memory could potentially be in use by another memory-sharing device. By way of example, a memory-sharing device may employ a table of BTEs that each has several cache status entries. When a cache status entry indicates that none of a group of one or more cache lines could possibly be in use by another memory-sharing device, a snoop request for any cache lines of that group may be suppressed without jeopardizing cache coherence.

Abstract: Efficient techniques are described for tracking a potential invalidation of a data cache entry in a data cache for which coherency is required. Coherency information is received that indicates a potential invalidation of a data cache entry. The coherency information in association with the data cache entry is retained to track the potential invalidation to the data cache entry. The retained coherency information is kept separate from state bits that are utilized in cache access operations. An invalidate bit, associated with a data cache entry, may be utilized to represents a potential invalidation of the data cache entry. The invalidate bit is set in response to the coherency information, to track the potential invalidation of the data cache entry. A valid bit associated with the data cache entry is set in response to the active invalidate bit and a memory synchronization command. The set invalidate bit is cleared after the valid bit has been cleared.

Abstract: Methods and apparatus relating to an inter-queue anti-starvation mechanism with dynamic deadlock avoidance in a retry based pipeline are described. In one embodiment, logic may arbitrate between two queues based on various rules. The queues may store data including local or remote requests, data responses, non-data responses, external interrupts, etc. Other embodiments are also disclosed.

Abstract: A data processing system 3 employing a coherent memory system comprises multiple main cache memories 8. An inclusive snoop directory memory 14 stores directory lines 22. Each directory line includes a directory tag and multiple snoop vectors. Each snoop vector relates to a span of memory addresses corresponding to the cache line size within the main cache memories 8.

Abstract: In one embodiment, a conflict detection logic is configured to receive a plurality of memory requests from an arbiter of a coherent fabric of a system on a chip (SoC). The conflict detection logic includes snoop filter logic to downgrade a first snooped memory request for a first address to an unsnooped memory request when an indicator associated with the first address indicates that the coherent fabric has control of the first address. Other embodiments are described and claimed.

Abstract: Various aspects provide for implementing a cache coherence protocol. A system comprises at least one processing component and a centralized controller. The at least one processing component comprises a cache controller. The cache controller is configured to manage a cache memory associated with a processor. The centralized controller is configured to communicate with the cache controller based on a power state of the processor.

Abstract: Apparatuses and methods for adaptive control of memory are disclosed. One example apparatus includes a processing unit configured to run an operating system, and a memory coupled to the processing unit. The memory configured to communicate with the processing unit via a memory bus. The example apparatus may further include an adaptive memory controller configured to receive monitored statistical data from the memory and from the processing unit. The adaptive memory controller is configured to manage the memory based on the monitored statistical data.

Abstract: A method, apparatus, computer program product, and computer readable medium to perform receipt of a snoop notification indicating a write to a memory address associated with a cache, determination that the snoop notification signifies receipt of a message based, at least in part, on the memory address, and performance of an operation based, at least in part, on the message is disclosed.

Abstract: A data structure includes a plurality of entries each corresponding to a different systemwide combined response of a data processing system. A particular entry includes identifiers of multiple possible actions that can be taken in response to a systemwide combined response. Master logic issues a memory access request on a system fabric of the data processing system. The master logic, responsive to receiving the systemwide combined response and a selection of one of the multiple possible actions from a source of the memory access request prior to receipt of the systemwide combined response, selects the particular entry based on the systemwide combined response and selects one of the multiple possible actions identified in the particular entry based on the received selection. The master logic services the memory access request in accordance with the systemwide combined response by performing the selected one of the multiple possible actions.

Abstract: A data structure includes a plurality of entries each corresponding to a different systemwide combined response of a data processing system. A particular entry includes identifiers of multiple possible actions that can be taken in response to a systemwide combined response. Master logic issues a memory access request on a system fabric of the data processing system. The master logic, responsive to receiving the systemwide combined response and a selection of one of the multiple possible actions from a source of the memory access request prior to receipt of the systemwide combined response, selects the particular entry based on the systemwide combined response and selects one of the multiple possible actions identified in the particular entry based on the received selection. The master logic services the memory access request in accordance with the systemwide combined response by performing the selected one of the multiple possible actions.

Abstract: Techniques for escalating a real time agent's request that has an address conflict with a best effort agent's request. A best effort request can be allocated in a memory controller cache but can progress slowly in the memory system due to its low priority. Therefore, when a real time request has an address conflict with an older best effort request, the best effort request can be escalated if it is still pending when the real time request is received at the memory controller cache. Escalating the best effort request can include setting the push attribute of the best effort request or sending another request with a push attribute to bypass or push the best effort request.

Abstract: Embodiments of the invention address deficiencies of the art in respect to cache coherency management and provide a novel and non-obvious method, system and apparatus for silent invalid state transition handling in an SMP environment. In one embodiment of the invention, a cache coherency method can be provided. The cache coherency method can include identifying an invalid state transition for a cache line in a local node, evicting a corresponding cache directory entry for the cache line, forwarding an invalid state transition notification to a node controller for a home node for the cache line in order for the home node to evict a corresponding cache directory entry for the cache line, and relinquishing ownership of the cache line to the home node.

Abstract: A coherent attached processor proxy (CAPP) of a primary coherent system receives a memory access request specifying a target address in the primary coherent system from an attached processor (AP) external to the primary coherent system. The CAPP includes a CAPP directory of contents of a cache memory in the AP that holds copies of memory blocks belonging to a coherent address space of the primary coherent system. In response to the memory access request, the CAPP performs a first determination of a coherence state for the target address and allocates a master machine to service the memory access request in accordance with the first determination. Thereafter, during allocation of the master machine, the CAPP updates the coherence state and performs a second determination of the coherence state. The master machine services the memory access request in accordance with the second determination.

Abstract: A coherent attached processor proxy (CAPP) of a primary coherent system receives a memory access request specifying a target address in the primary coherent system from an attached processor (AP) external to the primary coherent system. The CAPP includes a CAPP directory of contents of a cache memory in the AP that holds copies of memory blocks belonging to a coherent address space of the primary coherent system. In response to the memory access request, the CAPP performs a first determination of a coherence state for the target address and allocates a master machine to service the memory access request in accordance with the first determination. Thereafter, during allocation of the master machine, the CAPP updates the coherence state and performs a second determination of the coherence state. The master machine services the memory access request in accordance with the second determination.

Abstract: A multiprocessor data processing system includes a plurality of cache memories including a cache memory. The cache memory issues a read-type operation for a target cache line. While waiting for receipt of the target cache line, the cache memory monitors to detect a competing store-type operation for the target cache line. In response to receiving the target cache line, the cache memory installs the target cache line in the cache memory, and sets a coherency state of the target cache line installed in the cache memory based on whether the competing store-type operation is detected.

Abstract: A system comprises a first node operative to provide a source broadcast requesting data. The first node associates an F-state with a copy of the data in response to receiving the copy of the data from memory and receiving non-data responses from other nodes in the system. The non-data responses include an indication that at least a second node includes a shared copy of the data. The F-state enabling the first node to serve as an ordering point in the system capable of responding to requests from other nodes in the system with a shared copy of the data.

Abstract: In general, the present invention relates to data cache processing. Specifically, the present invention relates to a system that provides reconfigurable dynamic cache which varies the operation strategy of cache memory based on the demand from the applications originating from different external general processor cores, along with functions of a virtualized hybrid core system. The system includes receiving a data request, selecting an operational mode based on the data request and a predefined selection algorithm, and processing the data request based on the selected operational mode.

Abstract: A Region Coherence Array (RCA) having subregions and subregion prefetching for shared-memory multiprocessor systems having a single-level, or a multi-level interconnect hierarchy architecture.

Abstract: Methods and apparatus relating to memory mirroring and migration at a Home Agent (HA) are described. In one embodiment, a home agent may mirror its data at a slave agent. In some embodiments, a bit in a directory may indicate status of cache lines. Other embodiments are also disclosed.

Abstract: A multiprocessor data processing system includes a plurality of cache memories including a cache memory. In response to the cache memory detecting a storage-modifying operation specifying a same target address as that of a first read-type operation being processed by the cache memory, the cache memory provides a retry response to the storage-modifying operation. In response to completion of the read-type operation, the cache memory enters a referee mode. While in the referee mode, the cache memory temporarily dynamically increases priority of any storage-modifying operation targeting the target address in relation to any second read-type operation targeting the target address.

Abstract: A multiprocessor data processing system includes a plurality of cache memories including a cache memory. The cache memory issues a read-type operation for a target cache line. While waiting for receipt of the target cache line, the cache memory monitors to detect a competing store-type operation for the target cache line. In response to receiving the target cache line, the cache memory installs the target cache line in the cache memory, and sets a coherency state of the target cache line installed in the cache memory based on whether the competing store-type operation is detected.

Abstract: Color-based caching allows each cache line to be distinguished by a specific color, and enables the manipulation of cache behavior based upon the colors of the cache lines. When multiple threads are able to share a cache, effective cache management is critical to overall performance. Color-based caching provides an effective method to better utilize caches and avoid unnecessary cache thrashing and pollution. Hardware maintains color-based counters relative to the cache lines to monitor and obtain feedback on cache line events. These counters are utilized for cache coherence transactions in multiple processor systems.

Abstract: A system and method for supporting targeted stores in a shared-memory multiprocessor. A targeted store enables a first processor to push a cache line to be stored in a cache memory of a second processor. This eliminates the need for multiple cache-coherence operations to transfer the cache line from the first processor to the second processor. More specifically, the disclosed embodiments provide a system that notifies a waiting thread when a targeted store is directed to monitored memory locations. During operation, the system receives a targeted store which is directed to a specific cache in a shared-memory multiprocessor system. In response, the system examines a destination address for the targeted store to determine whether the targeted store is directed to a monitored memory location which is being monitored for a thread associated with the specific cache. If so, the system informs the thread about the targeted store.

Abstract: A method and apparatus for using a tree-structured cluster as a library for a computing grid. In one embodiment, a request for computation is received at a cache node of the cluster. The computation requires data from an other cache node of the cluster, and not present in the cache node receiving the request. The other cache nodes of the cluster are polled for the required data. An instance of the required data stored in the other cache node of the cluster is replicated to the cache node receiving the computation request.

Abstract: A multiprocessor data processing system includes a plurality of cache memories including a cache memory. In response to the cache memory detecting a storage-modifying operation specifying a same target address as that of a first read-type operation being processed by the cache memory, the cache memory provides a retry response to the storage-modifying operation. In response to completion of the read-type operation, the cache memory enters a referee mode. While in the referee mode, the cache memory temporarily dynamically increases priority of any storage-modifying operation targeting the target address in relation to any second read-type operation targeting the target address.

Abstract: A method of accelerating memory operations using virtualization information includes executing a hypervisor on hardware resources of a computing system. A plurality of domains are created under the control of the hypervisor. Each domain is allocated memory resources that include accessible memory space that is exclusively accessible by that domain. Each domain is allocated one or more processor resources. The hypervisor identifies domain layout information that includes a boundary of accessible memory space of each domain. The hypervisor provides the domain layout information to each processor resource. Each processor resource is configured to implement, on a per domain basis, a restricted coherency protocol based on the domain layout information. The restricted coherency protocol bypasses, relative to the domain, downstream caches when a cache line falls within the accessible memory space of that domain.

Abstract: A computer system includes but is not limited to a primary processing circuitry, a bus coupled to the primary processing circuitry, and memory circuitry coupled to the bus. The memory circuitry is physically separated from the primary processing circuitry. The memory circuitry includes at least one integrated memory circuit and computational circuitry. The at least one integrated memory circuit configured to store and retrieve data and to provide to the bus, during accessing intervals, requested data for the primary processing circuitry. The computational circuitry co-located with the at least one integrated memory circuit, the computational circuitry co-located with integrated memory circuit can be configured for performing supplemental functions at least partially during time periods that are not accessing intervals.

Abstract: A coherent attached processor proxy (CAPP) that participates in coherence communication in a primary coherent system on behalf of an external attached processor maintains, in each of a plurality of entries of a CAPP directory, information regarding a respective associated cache line of data from the primary coherent system cached by the attached processor. In response to initiation of recovery operations, the CAPP transmits, in a generally sequential order with respect to the CAPP directory, multiple memory access requests indicating an error for addresses indicated by the plurality of entries. In response to a snooped memory access request that targets a particular address hitting in the CAPP directory during the transmitting, the CAPP performs a coherence recovery operation for the particular address prior to a time indicated by the generally sequential order.