Abstract: In a microprogrammed data processing pipeline system comprising a plurality of stages, microinstructions for controlling the stages are stored as templates in an addressable template micromemory store and are provided automatically and sequentially to the stage of the pipeline system. Each template is associated with an individual set of data and includes microinstructions for each stage, whether real or virtual, through which the associated set of data passes. The template micromemory store is segmented into a plurality of individually addressable micromemory units with each unit therein storing microinstructions for an individually associated stage in the data processing pipeline system.

Abstract: A high speed parallel operation, multiplication circuit is provided having a multiplier multiplexor which may function in combination with a column compressor for providing a resultant product, wherein, preferably, the multiplier multiplexor has been implemented using a modified Booth's algorithm, and wherein the column compressor operates to process every column within the same propagation delay whereby every input may create an output in essentially the same propagation time, i.e., true parallel operation requiring preferably no more than an average column propagation delay time.

Abstract: Modulo M translation is performed on a large binary number of n bits by grouping the binary number in contiguous sets of approximately K bits each, storing the modulo M residues for each K bit set in an individually associated pre-stored ROM, reading the modulo M residues for a particular K bit segment of a binary number out of the ROMs, and performing modulo M addition on the read-out residues. Thus, modulo M translation of a positive number is accomplished in n/k modulo M additions and a table look-up, with the look-up table being stored in n/k ROMs. A subsequent modulo M subtraction is performed if the binary number is negative.