Instruction format with designations for operand lengths of byte, half word, word, or double word encoded in address bits

摘要

A set of processors and instruction set for a pipelined multiprocessing architecture with separate code and data streams is disclosed. The overall architecture considered interlocks instruction execution with the destination transfer of results and includes automated input and output of array data. The data is dispensed from a central memory that is associated with the control unit and all results and inputs are automatically returned to this central memory over a second bus. Once the processors receive their data they operate independently and several instructions may be in the process of being executed at the same time with destination validity checking being used to coordinate the activity. The central memory may contain partitions, called circular partitions, or first-in/first-out buffers which make it possible for it to automatically prefetch or output array data. The instructions include overlapping address and operand length bits.