Abstract: A system and method of broadcasting data to multiple targets across a system bus, such as the peripheral component interconnect (PCI) bus, that does not normally support broadcast transfers, in which one target responds to the bus transaction and the remaining targets listen in on the bus transaction to receive data from the system bus. The responding target stalls the bus transaction when any of the listening targets communicate to the responding target that they are temporarily unable to accept the data on the bus.

Abstract: An alternation network for use with a content addressable memory for implementing a divide and conquer algorithm is described.

Abstract: A method of generating a stream of non-contiguous memory addresses representing contiguous points in logical space is described. The method comprises: generating initializing parameters describing the contiguous points in the logical space; configuring a memory address engine with the initializing parameters; performing an algorithm in the memory address engine according to the initialising parameters to produce a plurality of non-contiguous memory addresses; and collating the non-contiguous memory addresses into the stream of memory addresses for output to a data memory. The present invention has particular application to SIMD processing techniques where there are a plurality of memory address engines.

Abstract: A method of carrying out a data fetch operation for a data-parallel processor such as a SIMD processor is described. The operation is specifically involving the use of a plurality of non-sequential data addresses. The method comprises constructing a linear address vector from the non-sequential addresses, and using the address vector in a block fetch command to a data store.

Abstract: A compound associative memory for use with a data-parallel computer, and a method of storing/retrieving data in the compound associative memory is disclosed. The memory comprises a bit-parallel word-organized associative memory having an array of associative memory cells arranged to be capable of bit-parallel search and write operations. A bit-serial associative memory having an array of memory cells arranged to be capable of bit-serial search and write operations, but not word bit-parallel search and write operations, is also included. The bit-serial memory is operatively connected to the bit-parallel memory and arranged to operate as an extension of the same. The method comprises searching the bit-parallel word-organized associative memory and/or the bit-serial associative memory coupled to the bit-parallel memory for data matching search data, and marking the memory cells having stored data matching the search data.

Abstract: A method of generating a stream non-contiguous memory addresses representing contiguous points in logical space is described. The method comprises: generating initialising parameters describing the contiguous points in the logical space; configuring a memory address engine with the initialising parameters; performing an algorithm in the memory address engine according to the initialising parameters to produce a plurality of non-contiguous memory addresses; and collating the non-contiguous memory addresses into the stream of memory addresses for output to a data memory. The present invention has particular application to SIMD processing techniques where there are a plurality of memory address engines.

Abstract: A data processor controller comprising a first processor for generating data processor instructions at a first rate and an instruction multiplying circuit for receiving the data processor instructions at the first rate and being a arranged to multiply the instructions and forward the multiplied instructions to a data processor at a second rate substantially greater than the first rate is disclosed. The first processor outputs a stream of compounded data processor instructions and the multiplying circuit separates the compounded instructions into a single stream of individual instructions in a non-compounded format. Multiplication is effectively achieved by repeating both single and blocks of data processor instructions. The effective bandwidth between the first processor and the data processor is multiplied by the multiplying circuit which takes advantage of the different sizes of data pathways available between the first processor and the data processor.