Patent application number | Description | Published |
20080209177 | Mechanism in a Multi-Threaded Microprocessor to Maintain Best Case Demand Instruction Redispatch - A method and system for maintaining a best-case demand redispatch of an instruction to allow for maximizing the time a rejected thread may execute in lookahead execution mode, while maintaining the smallest L1 cache miss penalty supported by the memory subsystem. In response to a demand miss, a load/store unit sends a fetch request to the next level cache. The cache line of the demand miss is examined to identify the critical sector. Once the critical sector is identified, a best-case data return time is determined based on the fastest time the next level cache is able to return the critical sector of the cache line. The load/store unit then sends a speculative warning to the dispatch unit to coincide with the best-case data return, wherein the speculative warning prepares the dispatch unit to resend the instruction for execution as soon as data is available to the processor core. | 08-28-2008 |
20100180081 | Adaptive Data Prefetch System and Method - A data processing system includes a processor, a unit that includes a multi-level cache, a prefetch system and a memory. The data processing system can operate in a first mode and a second mode. The prefetch system can change behavior in response to a desired power consumption policy set by an external agent or automatically via hardware based on on-chip power/performance thresholds. | 07-15-2010 |
20100262808 | MANAGING INSTRUCTIONS FOR MORE EFFICIENT LOAD/STORE UNIT USAGE - The illustrative embodiments described herein provide a computer-implemented method, apparatus, and a system for managing instructions. A load/store unit receives a first instruction at a port. The load/store unit rejects the first instruction in response to determining that the first instruction has a first reject condition. Then, the instruction sequencing unit activates a first bit in response to the load/store unit rejection the first instruction. The instruction sequencing unit blocks the first instruction from reissue while the first bit is activated. The processor unit determines a class of rejection of the first instruction. The instruction sequencing unit starts a timer. The length of the timer is based on the class of rejection of the first instruction. The instruction sequencing unit resets the first bit in response to the timer expiring. The instruction sequencing unit allows the first instruction to become eligible for reissue in response to resetting the first bit. | 10-14-2010 |
20150324204 | PARALLEL SLICE PROCESSOR WITH DYNAMIC INSTRUCTION STREAM MAPPING - A processor core having multiple parallel instruction execution slices and coupled to multiple dispatch queues by a dispatch routing network provides flexible and efficient use of internal resources. The dispatch routing network is controlled to dynamically vary the relationship between the slices and instruction streams according to execution requirements for the instruction streams and the availability of resources in the instruction execution slices. The instruction execution slices may be dynamically reconfigured as between single-instruction-multiple-data (SIMD) instruction execution and ordinary instruction execution on a per-instruction basis, permitting the mixture of those instruction types. Instructions having an operand width greater than the width of a single instruction execution slice may be processed by multiple instruction execution slices configured to act in concert for the particular instructions. When an instruction execution slice is busy processing a current instruction for one of the streams, another slice can be selected to proceed with execution. | 11-12-2015 |
20150324205 | PROCESSING OF MULTIPLE INSTRUCTION STREAMS IN A PARALLEL SLICE PROCESSOR - Techniques for managing instruction execution for multiple instruction streams using a processor core having multiple parallel instruction execution slices provide flexibility in execution of program instructions by a processor core. An event is detected indicating that either resource requirement or resource availability will not be met by the execution slice currently executing the instruction stream. In response to detecting the event, dispatch of at least a portion of the subsequent instruction is made to another instruction execution slice. The event may be a compiler-inserted directive, may be an event detected by logic in the processor core, or may be determined by a thread sequencer. The instruction execution slices may be dynamically reconfigured as between single-instruction-multiple-data (SIMD) instruction execution, ordinary instruction execution, wide instruction execution. When an instruction execution slice is busy processing a current instruction for one of the streams, another slice can be selected to proceed with execution. | 11-12-2015 |
20150324206 | PARALLEL SLICE PROCESSOR WITH DYNAMIC INSTRUCTION STREAM MAPPING - A method of operation of a processor core having multiple parallel instruction execution slices and coupled to multiple dispatch queues coupled by a dispatch routing network provides flexible and efficient use of internal resources. The dispatch routing network is controlled to dynamically vary the relationship between the slices and instruction streams according to execution requirements for the instruction streams and the availability of resources in the instruction execution slices. The instruction execution slices may be dynamically reconfigured as between single-instruction-multiple-data (SIMD) instruction execution and ordinary instruction execution on a per-instruction basis. Instructions having an operand width greater than the width of a single instruction execution slice may be processed by multiple instruction execution slices configured to act in concert for the particular instructions. When an instruction execution slice is busy processing a current instruction for one of the streams, another slice can be selected to proceed with execution. | 11-12-2015 |
20150324207 | PROCESSING OF MULTIPLE INSTRUCTION STREAMS IN A PARALLEL SLICE PROCESSOR - A method of managing instruction execution for multiple instruction streams using a processor core having multiple parallel instruction execution slices provides instruction processing flexibility. An event is detected indicating that either resource requirement or resource availability for a subsequent instruction of an instruction stream will not be met by the instruction execution slice currently executing the instruction stream. In response to detecting the event, dispatch of at least a portion of the subsequent instruction is made to another instruction execution slice. The event may be a compiler-inserted directive, may be an event detected by logic in the processor core, or may be determined by a thread sequencer. The execution slices may be dynamically reconfigured as between single-instruction-multiple-data (SIMD) instruction execution, ordinary instruction execution, wide instruction execution. When an execution slice is busy processing a current instruction for one of the streams, another slice can be selected to proceed with execution. | 11-12-2015 |
Patent application number | Description | Published |
20080267281 | METHOD, DEVICE AND NETWORK ELEMENT FOR DECODING AN INFORMATION WORD FROM A CODED WORD - It is disclosed a method for decoding an information word from a set of coded words. The method comprises the steps of receiving a coded word, of selecting a coded word having the minimum distance from the received coded word from a pre-configured sub-set of the set of the coded words, wherein the sub-set is configured to at least two coded words having each other a distance higher than the minimum distance between the coded words of the set, and of decoding the information word from the selected coded word. | 10-30-2008 |
20100138711 | Equipment protection method and apparatus - Equipment protection of a switch matrix (SM) in a network node, which contains a number of matrix modules (M | 06-03-2010 |
20110268133 | TRANSMISSION OF PARALLEL DATA FLOWS ON A PARALLEL BUS - There is described a method for transmitting N parallel data flows on a parallel bus. The method comprises, at a first communication device: generating a further parallel data flow comprising alignment words periodically distributed with a period; at each period, rotating the N of parallel data flows and the further parallel data flow thus generating N+1 rotated parallel data flows, each comprising part of the alignment words periodically distributed with a frame period; transmitting the N+1 rotated parallel data flows on respective physical connections of the parallel bus. The method further comprises, at a second communication device: aligning the N+1 rotated parallel data flows by using the alignment words, thus compensating skew and obtaining N+1 aligned parallel data flows; and at each period, de-rotating the N+1 aligned parallel data flows, thus generating N de-rotated parallel data flows corresponding to the N parallel data flows. | 11-03-2011 |
20130028265 | UPDATE OF A CUMULATIVE RESIDENCE TIME OF A PACKET IN A PACKET-SWITCHED COMMUNICATION NETWORK - It is disclosed a method for updating a cumulative residence time of a synchronization packet received at a node of a packet-switched communication network. The cumulative residence time is equal to a cumulative sum of residence times of the packet at nodes interposed between a further node which has generated the packet and the node. The node comprises an ingress circuit and an egress circuit. The method comprises: receiving the packet at the egress circuit from the ingress circuit; at a timestamp generator of the egress circuit, generating a timestamp; at the egress circuit, calculating a virtual timestamp based on the timestamp and on an estimated variable delay that will be undergone by the packet due to buffering in a buffer located downstream the timestamp generator; and, at the egress circuit, using the virtual timestamp for updating the cumulative residence time, before transmitting the packet to a still further node. | 01-31-2013 |
20130182716 | DEVICE AND METHOD FOR SWITCHING DATA TRAFFIC IN A DIGITAL TRANSMISSION NETWORK - A network element for a digital transmission network is proposed. The network element contains two switching matrices for switching data cells, as well as ingress ports that receive TDM traffic flow and packet traffic flow and segment the traffic flows into cells. A control system for controlling the configuration of the ingress ports and the switching matrices controls the ingress ports, in case of no failure of the switching matrices, to forward the TDM traffic flows to both switching matrices and to split the packet traffic flow over the two switching matrices. | 07-18-2013 |