摘要 |
A pipelined paralled vector processor decreases the time required to process the elements of a single vector stored in a vector register. Each vector register of a plurality of vector registers is subdivided into a plurality of smaller registers. A vector, stored in a vector register, includes N elements; however, each of the smaller registers store M elements of the vector, where M is less than N. A pipelined element processor is associated with each smaller register for processing the M elements of the vectors stored in the smaller register and storing a result of the processing in a result register. Each of the smaller registers of the vector registers, and its corresponding element processor, comprise a unit. A plurality of units are connected in a parallel configuration. The element processors, associated with each unit, have been loaded with the result, the result being stored in a result register. Each of the results are processed in parallel fashion, as a result of the parallel configuration of the plurality of units. Therefore, the time required to process the elements of a single vector, stored in a vector register, is decreased.
|