Patent application number | Description | Published |
20090006693 | Apparatus and Method for Fairness Arbitration for a Shared Pipeline in a Large SMP Computer System - A modification of rank priority arbitration for access to computer system resources through a shared pipeline that provides more equitable arbitration by allowing a higher ranked request access to the shared resource ahead of a lower ranked requester only one time. If multiple requests are active at the same time, the rank priority will first select the highest priority active request and grant it access to the resource. It will also set a ‘blocking latch’ to prevent that higher priority request from re-gaining access to the resource until the rest of the outstanding lower priority active requesters have had a chance to access the resource. | 01-01-2009 |
20090210629 | METHOD, SYSTEM AND COMPUTER PROGRAM PRODUCT FOR SELECTIVELY PURGING CACHE ENTRIES - A method, system and computer program product for selectively purging entries in a cache of a computer system. The method includes determining a starting storage address and a length of the storage address range to be purged, determining preset values for a congruence class and a compartment of a cache directory, accessing the cache directory based on the preset value of the congruence class, and selecting an entry in the cache directory based on the preset value of the compartment, determining validity of the entry accessed by examining an ownership tag of the entry, comparing a line address of the entry with the starting storage address and a sum of the starting storage address and the length of the storage address range, and selectively purging the entry based on the comparison result. | 08-20-2009 |
20090216928 | SYSTEM, METHOD AND COMPUTER PROGRAM PRODUCT FOR PROVIDING A NEW QUIESCE STATE - A system, method and computer program product for providing a new quiesce state. The method includes receiving a quiesce request at a system controller from an initiating processor. The quiesce request is sent to a plurality of processors. Notification is received at the system controller that the processors have finished purging their translation look aside buffers (TLBs). A fast quiesce reset command is received at the system controller from the initiating processor once updates to the system resources are complete. It is indicated to the processors that the block translation restriction can be dropped in response to receiving the fast quiesce reset command, thereby allowing the processors to continue processing without block translation restrictions. | 08-27-2009 |
20110314211 | RECOVER STORE DATA MERGING - Various embodiments of the present invention merge data in a cache memory. In one embodiment a set of store data is received from a processing core. A store merge command and a merge mask from are also received from the processing core. A portion of the store data to perform a merging operation thereon is identified based on the store merge command. A sub-portion of the portion of the store data to be merged with a corresponding set of data from a cache memory is identified based on the merge mask. The sub-portion is merged with the corresponding set of data from the cache memory. | 12-22-2011 |
20110314212 | MANAGING IN-LINE STORE THROUGHPUT REDUCTION - Various embodiments of the present invention manage a hierarchical store-through memory cache structure. A store request queue is associated with a processing core in multiple processing cores. At least one blocking condition is determined to have occurred at the store request queue. Multiple non-store requests and a set of store requests associated with a remaining set of processing cores in the multiple processing cores are dynamically blocked from accessing a memory cache in response to the blocking condition having occurred. | 12-22-2011 |
20110320665 | Managing Concurrent Serialized Interrupt Broadcast Commands In A Multi-Node, Symmetric Multiprocessing Computer - Managing concurrent serialized interrupt broadcast commands in a multi-node, symmetric multiprocessing computer including receiving, by a communications adapter in a compute node, a plurality of serialized interrupt broadcast commands; receiving, by the communications adapter, a plurality of interrupt tags for the plurality of serialized interrupt broadcast commands, each interrupt tag including an identification of an interrupt service order for a serialized interrupt broadcast command; assigning, by the communications adapter, to each serialized interrupt broadcast command its interrupt tag; and if an interrupt tag assigned to a serialized interrupt broadcast command has an interrupt service order that matches a value of a current operation tag that identifies the next serialized interrupt broadcast command to be exposed to the one or more processors, exposing, by the communications adapter, the serialized interrupt broadcast command to the one or more processors on the compute node to be serviced. | 12-29-2011 |
20110320695 | MITIGATING BUSY TIME IN A HIGH PERFORMANCE CACHE - Various embodiments of the present invention mitigate busy time in a hierarchical store-through memory cache structure. In one embodiment, a cache directory associated with a memory cache is divided into a plurality of portions each associated with a portion memory cache. Simultaneous cache lookup operations and cache write operations between the plurality of portions of the cache directory are supported. Two or more store commands are simultaneously processed in a shared cache pipeline communicatively coupled to the plurality of portions of the cache directory. | 12-29-2011 |
20110320731 | ON DEMAND ALLOCATION OF CACHE BUFFER SLOTS - Dynamic allocation of cache buffer slots includes receiving a request to perform an operation that requires a storage buffer slot, the storage buffer slot residing in a level of storage. The dynamic allocation of cache buffer slots also includes determining availability of the storage buffer slot for the cache index as specified by the request. Upon determining the storage buffer slot is not available, the dynamic allocation of cache buffer slots includes evicting data stored in the storage buffer slot, and reserving the storage buffer slot for data associated with the request. | 12-29-2011 |
20110320744 | DIAGNOSTIC DATA COLLECTION AND STORAGE PUT-AWAY STATION IN A MULTIPROCESSOR SYSTEM - A computer-implemented method for collecting diagnostic data within a multiprocessor system that includes capturing diagnostic data via a plurality of collection points disposed at a source location within the multiprocessor system, routing the captured diagnostic data to a data collection station at the source location, providing a plurality of buffers within the data collection station, and temporarily storing the captured diagnostic data on at least one of the plurality of buffers, and transferring the captured diagnostic data to a target storage location on a same chip as the source location or another storage location on a same node. | 12-29-2011 |
20110320779 | PERFORMANCE MONITORING IN A SHARED PIPELINE - A pipelined processing device includes: a device controller configured to receive a request to perform an operation; a plurality of subcontrollers configured to receive at least one instruction associated with the operation, each of the plurality of subcontrollers including a counter configured to generate an active time value indicating at least a portion of a time taken to process the at least one instruction; a pipeline processor configured to receive and process the at least one instruction, the pipeline processor configured to receive the active time value; and a shared pipeline storage area configured to store the active time value for each of the plurality of subcontrollers. | 12-29-2011 |
20130060997 | MITIGATING BUSY TIME IN A HIGH PERFORMANCE CACHE - Various embodiments of the present invention mitigate busy time in a hierarchical store-through memory cache structure. In one embodiment, a cache directory associated with a memory cache is divided into a plurality of portions each associated with a portion memory cache. Simultaneous cache lookup operations and cache write operations between the plurality of portions of the cache directory are supported. Two or more store commands are simultaneously processed in a shared cache pipeline communicatively coupled to the plurality of portions of the cache directory. | 03-07-2013 |
20130080705 | MANAGING IN-LINE STORE THROUGHPUT REDUCTION - Various embodiments of the present invention manage a hierarchical store-through memory cache structure. A store request queue is associated with a processing core in multiple processing cores. At least one blocking condition is determined to have occurred at the store request queue. Multiple non-store requests and a set of store requests associated with a remaining set of processing cores in the multiple processing cores are dynamically blocked from accessing a memory cache in response to the blocking condition having occurred. | 03-28-2013 |
20130339593 | REDUCING PENALTIES FOR CACHE ACCESSING OPERATIONS - A computer program product for reducing penalties for cache accessing operations is provided. The computer program product includes a tangible storage medium readable by a processing circuit and storing instructions for execution by the processing circuit for performing a method. The method includes respectively associating platform registers with cache arrays, loading control information and data of a store operation to be executed with respect to one or more of the cache arrays into the one or more of the platform registers respectively associated with the one or more of the cache arrays, and, based on the one or more of the cache arrays becoming available, committing the data from the one or more of the platform registers using the control information from the same platform registers to the one or more of the cache arrays. | 12-19-2013 |
20130339606 | REDUCING STORE OPERATION BUSY TIMES - A computer product for reducing store operation busy times is provided and relates to associating first and second platform registers with a cache array, determining that first and second store operations target a same wordline of the cache array, loading control information and data of the store operations into the platform registers and delaying a commit of the first store operation until the loading of the second platform register is complete. The method further includes committing the data from the platform registers using the control information from the platform registers to the wordline of the cache array at a same time to thereby reduce a busy time of the wordline of the cache array. | 12-19-2013 |
20130339607 | REDUCING STORE OPERATION BUSY TIMES - A computer product for reducing store operation busy times is provided. The computer product includes a tangible storage medium readable by a processing circuit and storing instructions for execution by the processing circuit for performing a method. The method includes associating first and second platform registers with a cache array, determining that first and second store operations target a same wordline of the cache array, loading control information and data of the first and second store operation into the first and second platform registers and delaying a commit of the first store operation until the loading of the second platform register is complete. The method further includes committing the data from the first and second platform registers using the control information from the first and second platform registers to the wordline of the cache array at a same time to thereby reduce a busy time of the wordline of the cache array. | 12-19-2013 |
20140095839 | MONITORING PROCESSING TIME IN A SHARED PIPELINE - A pipelined processing device includes: a pipeline controller configured to receive at least one instruction associated with an operation from each of a plurality of subcontrollers, and input the at least one instruction into a pipeline; and a pipeline counter configured to receive an active time value from each of the plurality of subcontrollers, the active time value indicating at least a portion of a time taken to process the at least one instruction, the pipeline controller configured to route the active time value to a shared pipeline storage for performance analysis. | 04-03-2014 |