초록 ▼

A processor cell that may be integrated with a multiplicity of dynamically reconfigurable 16-bit slices to enable and disable arbitrary collections of processing elements under software control according to the data being operating upon is provided. The structure allows a collection of word sizes to

A processor cell that may be integrated with a multiplicity of dynamically reconfigurable 16-bit slices to enable and disable arbitrary collections of processing elements under software control according to the data being operating upon is provided. The structure allows a collection of word sizes to be defined and then for certain processing elements to be enabled or disabled according to the data that they are operating upon. A slave mechanism is described wherein for words comprised of a multiplicity of slices the most significant slice is in control and the other slices are slaved or forced to go along with the operation of the most significant slice. This slaving is obtained automatically without the necessity to explicitly coordinate the operation of the multiplicity of slices cooperated together to form a word. The cell includes a Find and Lose operation wherein a scaler control means selects one from a multiplicity of processors satisfying some condition. The apparatus finds the first cell satisfying some condition and operates upon that processor cell when the operation upon that processor is completed, the apparatus looses that processor and then goes on to the next processor satisfying the condition and so on.

대표청구항 ▼

In a cellular array employing a plurality of processor cells arranged in an array of rows, each of said plurality of processor cells being capable of communicating with neighboring cells horizontally located in left and right directions, wherein the improvement comprises: said cellular array having

In a cellular array employing a plurality of processor cells arranged in an array of rows, each of said plurality of processor cells being capable of communicating with neighboring cells horizontally located in left and right directions, wherein the improvement comprises: said cellular array having control means for operating said processor cells of said array to process data in variable global words of selected longer bit lengths, and each processor cell processes data in local words of a fixed shorter bit length which is a fraction of said longer bit lengths, wherein said cellular array control means includes mask means for partitioning the global words into a number of local word slices to be processed by a corresponding number of processor cells and nesting signal means for providing nesting signals for selectively enabling the corresponding number of processor cells to operate together to respond to the respective local word slices of the global words; and each of said plurality of processor cells having: a multiport RAM having a plurality of input terminals for receiving data and first and second output terminals; an arithmetic logic unit (ALU) having first and second inputs and at least one outpout, said first and second inputs being coupled to said first and second output terminals of said multiport RAM; path logic means coupled to said at least one output of said arithmetic logic unit, and to corresponding path logic means in neighboring cells horizontally located in left and right directions for processing local word slices; and enable logic means responsive to said nesting signals for providing left and right enable signals for selectively enabling said neighboring cells horizontally located in left and right directions within said array to enable each of said neighboring processor cells to selectively respond to a predetermined nested processing operation together with it and thereafter transfer to other processor cells in said array, each said enable logic means being coupled to said path logic means and being responsive to the enable signals from neighboring processor cells to enable said path logic means to process the local word slices corresponding to each said processor cell.