Abstract: A method of providing accelerated post-silicon testing for a silicon hardware includes computing a simulation cumulative record of state using a plurality of test instructions and a cycle breakpoint, performing a simulation of an instrumented logic design using the plurality of test instructions and the cycle breakpoint, manufacturing the silicon hardware using the instrumented logic design, computing a silicon cumulative record of state by executing the plurality of instructions using the silicon hardware; and comparing the simulation cumulative record of state to the silicon cumulative record of state.

Abstract: A method of providing ad hoc verification for a simulation includes generating a cumulative record of a state value for the simulation of a circuit design, comparing the cumulative record of the state value to a golden record of the state value to obtain a comparison result, and performing ad hoc verification of the circuit design using the comparison result.

Abstract: A method and apparatus for computation is provided. A main cluster crossbar is connected to a plurality of statically scheduled routing processors. A first sub-cluster crossbar is associated with a first one of the plurality of statically scheduled routing processors where the first sub-cluster crossbar is connected to a first plurality of execution processors. A second sub-cluster crossbar is associated with a second one of the plurality of statically scheduled routing processors where the second sub-cluster crossbar is connected to a second plurality of execution processors.

Abstract: A computer system for cycle-based computation includes a processor array, a translation component adapted to translate a cycle-based design, a host computer operatively connected to the processor array and to the translation component, a data connection component interconnecting a plurality of members of the processor array using static routing, a synchronization component enabling known timing relationships among the plurality of members of the processor array, a host service request component adapted to send a host service request from a member of the processor array to the host computer, and an access component adapted to access a portion of a state of the processor array and a portion of a state of the data connection.

Abstract: A method for performing debugging of an executable source program in a massively parallel processing environment involves associating a major cycle counter and a minor cycle counter with each of a plurality of execution processors in the massively parallel processing environment, obtaining a first stopping point value associated with the major cycle counter and a second stopping point value associated with the minor cycle counter, executing instructions of the executable source program on each of the plurality of execution processors, modifying the major cycle counter and the minor cycle counter, and halting each of the plurality of execution processors and returning control to the user if the major cycle counter reaches the first stopping point value and the minor cycle counter reaches the second stopping point value.

Abstract: A method for controlling a processor array by a host computer involves creating a graph of a plurality of nodes using a data connection component, configuring a broadcast tree from a spanning tree of the graph, propagating a first command from the host computer to a member of the processor array using the broadcast tree, configuring a reply tree from a spanning tree of the graph, transmitting a response from the member of the processor array to the host computer using the reply tree, and configuring the data connection component to send at least one message selected from the first command and the response on at least one run mode communication path.

Abstract: According to the invention, a first processor chip (10) comprising a processing core (12) and at least one bank of memory (14). The at least one bank of memory (14) preferably includes a mode control input (32) for controlling the mode of the at least one bank of memory (14) between physical memory and cache memory. In addition, the first processor chip (10) may further comprise an I/O link (26) configured to facilitate communication between the first processor chip (10) and other processor chips, and a communication and memory controller (20, 22) in electrical communication with the processing core (12), the at least one bank of memory (14), and the I/O link (26).

Abstract: A moving object identification method (10) for identifying and tracing an object (20) within a video image (14) such that the object (20) can act as a hot spot (30) as for an interactive computer/user interface (70). A plurality of tags (28) define the edges (26) of the object (20) and a plurality of hot spot borders (32) define the hot spot (30) such that the hot spot (30) and the object (20) generally coincide. A physical tag (28b) is optionally used to position the tags (28). Sensitivity to disappearance of the edges (26) is adjustable according to the relative size of a subtag (68) to the corresponding tag (28).

Abstract: Disclosed is an instruction-level method and system for prefetching data or instructions of variable size to specified cache sets. A prefetch instruction containing binary fields allows the compiler, loader or runtime software to control cache prefetching and reduce thrashing by providing the prefetch hardware with information as to the optimal cache set location and the optimal amount of data to be prefetched. Support of Harvard architectures with separate instruction and data caches is provided by separate software control of instruction and data caches. The cache set number is identified to indicate into which set the information is to be preloaded. The size field provides a variable prefetch size. An address field indicates the address where prefetching begins.

Abstract: Disclosed is a method and system for providing for the prefetching of data or instructions. A prefetch instruction which is in an instruction stream is processed by memory management unit (MMU) where prefetch cache control information is placed as part of the already existing prefetch instruction. Once processed by the MMU, the prefetch instruction thus contains binary fields allowing the operating system or runtime software to control cache prefetching by assigning values to the binary fields which provide the optimal cache set location and the optimal amount of data to be prefetched and thus reduces thrashing.

Abstract: Disclosed is a prefetch execution unit, prefetch instruction buffer and a prefetch victim buffer which operate to optimize prefetching by recording a particular cache miss' history. To record cache misses, victimized (overwritten) lines and/or a prefetch tag are stored in a prefetch victim buffer. When the processor experiences a cache miss, it accesses the prefetch victim buffer to retrieve information relating to the prefetch victim. The prefetch execution unit then modifies the values of the additional field or fields and then stores the modified augmented prefetch instruction in the prefetch instruction buffer. The next time a prefetch instruction for the victimized lines is executed by the processor, the new values of the modified augmented prefetch instruction will dictate where the prefetch information is stored or what size increment it has. By continuous modification of the augmented prefetch instructions, eventually thrashing may be eliminated.