Coherent Logix, Inc.

Rockwell International Corporation

Conexant Systems, Inc.

Mindspeed Technologies, Inc.

Hyperx Logic, Inc.

Keith M Bindloss

Carl S Dobbs

Michael R Trocino

Kenneth R Faulkner

Michael B Doerr

John Mark Beardslee

Michael B Solka

David A Gibson

Sumeer Arya

Kenneth E Garey

George A Watson

Faisal A Syed

Alex E Icaza

Frederick A Rush

Kevin A Shelby

John Earle

Zahir Raza

Moataz A Mohamed

Evgeny Mezhibovsky

Ricke Waylan Clark

Duncan M Fisher

Dale E Folwell

Donald D Harenberg

Rajiv Gupta

Wade Guthrie

Ching Long Su

Marty T Budrovic

Lawrence F Blank

Embodiments of a multi-processor array are disclosed that may include a plurality of processors, local memories, configurable communication elements, and direct memory access (DMA) engines, and a DMA controller. Each processor may be coupled to one of the local memories, and the plurality of processors, local memories, and configurable communication elements may be coupled together in an interspersed arrangement. The DMA controller may be configured to control the operation of the plurality of DMA engines.

Processing system with interspersed processors DMA-FIFO

Various embodiments are disclosed of a multiprocessor system with processing elements optimized for high performance and low power dissipation and an associated method of programming the processing elements. Each processing element may comprise a fetch unit and a plurality of address generator units and a plurality of pipelined datapaths. The fetch unit may be configured to receive a multi-part instruction, wherein the multi-part instruction includes a plurality of fields. First and second address generator units may generate, based on different fields of the multi-part instruction, addresses from which to retrieve first and second data for use by an execution unit for the multi-part instruction or a subsequent multi-part instruction. The execution units may perform operations using a single pipeline or multiple pipelines based on third and fourth fields of the multi-part instruction.

Memory-network processor with programmable optimizations

A multi-processor system with processing elements, interspersed memory, and primary and secondary interconnection networks optimized for high performance and low power dissipation is disclosed. In the secondary network multiple message routing nodes are arranged in an interspersed fashion with multiple processors. A given message routing node may receive messages from other message nodes, and relay the received messages to destination message routing nodes using relative offsets included in the messages. The relative offset may specify a number of message nodes from the message node that originated a message to a destination message node.

Memory network processor

Various embodiments are described of a system for improved processor instructions for a software-configurable processing element. In particular, various embodiments are described which accelerate functions useful for FEC encoding and decoding. In particular, the processing element may be configured to implement one or more instances of the relevant functions in response to receiving one of the processor instructions. The processing element may later be reconfigured to implement a different function in response to receiving a different one of the processor instructions. Each of the disclosed processor instructions may be implemented repeatedly by the processing element to repeatedly perform one or more instances of the relevant functions with a throughput approaching one or more solutions per clock cycle.

Processor instructions to accelerate FEC encoding and decoding

Various embodiments are disclosed of a multiprocessor system with processing elements optimized for high performance and low power dissipation and an associated method of programming the processing elements. Each processing element may comprise a fetch unit and a plurality of address generator units and a plurality of pipelined datapaths. The fetch unit may be configured to receive a multi-part instruction, wherein the multi-part instruction includes a plurality of fields. A first address generator unit may be configured to perform an arithmetic operation dependent upon a first field of the plurality of fields. A second address generator unit may be configured to generate at least one address of a plurality of addresses, wherein each address is dependent upon a respective field of the plurality of fields. A parallel assembly language may be used to control the plurality of address generator units and the plurality of pipelined datapaths.

An alternate multi-thread pipeline structure and method are provided. A deep pipeline is provided in which two threads of two separate pipeline stages are alternatively presented to the various logic and latch circuits for execution. The execution and latching of the threads alternates from one thread to the other within a single clock cycle. Thus, each thread is executed once per clock cycle and two threads are executed in a single clock cycle.

Method and system for execution and latching of data in alternate threads

Embodiments of systems, methods, and computer program products may facilitate translation of machine code associated with a first processor for execution on a second processor. Machine code associated with a first processor may be translated into a translated program that includes one or more translation instructions for execution on the second processor. The one or more translation instructions are used exclusively to translate machine code that is associated with a processor other than the second processor. The translated program may be stored in a storage medium where it may be executed using the second processor. Each translation instruction that involves access of the storage medium may be dispatched to one or more translation load-store units that are dedicated exclusively to processing the translation instructions. By translating machine code associated with a first processor into one or more translation instructions that are compatible with a second processor architecture, the second processor may execute both the translated program and programs compiled for execution on the second processor at the same time and would not have to run in separate modes (e.g., a translation mode and a non-translation mode).

Methods, systems, and computer program products for translating machine code associated with a first processor for execution on a second processor

The present invention minimizes power consumption and processing time in a very long instruction word digital signal processor by identifying certain blocks of instructions and placing them in a small, fast buffer for subsequent retrieval and execution. A decoder unit decodes a prefetch instruction flag bit that indicates when instructions are to be prefetched and placed into the buffer. The decoder unit signals a control unit, which sends the instruction code from a memory unit to the buffer and maintains an address mapping table and a program counter. The control unit also sets a select input on a multiplexer to indicate that the multiplexer is to output the prefetch instructions it receives from the buffer. The multiplexer outputs the prefetch instructions to an instruction register that sends the prefetch instructions to appropriate functional units for execution.

System for efficient operation of a very long instruction word digital signal processor

Growing community of inventors

Irvine, CA, United States of America

Keith M Bindloss