Data driven information processor having pipeline processing units connected in series including processing portions connected in parallel
AssigneeSharp Kabushiki Kaisha
InventorManabu Yumoto
Life of this patent
Estimating the term…
Abstract
A data driven information processor includes an input control processing portion having a plurality of outputs which receives an input data packet from an immediately preceding pipeline processing unit and outputs the input data packet to one output according to a prescribed allocation method, a plurality of processing portions connected in parallel to the plurality of outputs, and an output control portions which receives a data packet output from the plurality of processing portions and outputting the packet to an immediately succeeding pipeline processing unit. One of the plurality of processing units is selected by a counter or based on the content of the input data packet.
Claims21
A data driven information processor including a plurality of pipeline processing units connected in series, wherein
- at least one of said plurality of pipeline processing units comprises:
- allocation means receiving an input data packet from another pipeline processing unit for allocating the input data packet to one of a plurality of outputs thereof according to a prescribed allocation method;
- a plurality of processing means connected in parallel to said respective plurality of outputs of said allocation means for conducting prescribed processing to the data packet, and
- output means receiving the data packet output from said plurality of processing means for outputting the data packet to an immediately following pipeline processing unit.
A data driven information processor including a plurality of pipeline processing units connected in series, wherein
- at least one of said plurality of pipeline processing units comprises:
- an allocation unit receiving an input data packet from another pipeline processing unit that allocates the input data packet to one of a plurality of outputs thereof according to a prescribed allocation method;
- a plurality of processing portions connected in parallel to said respective plurality of outputs of said allocation unit that conducts prescribed processing to the data packet, and
- an output unit that receives the data packet output from said plurality of processing portions and outputs the data packet to an immediately following pipeline processing unit.
A method for processing information in a data driven information processor including a plurality of pipeline processing units connected in series, comprising the steps of:
- receiving an input data packet at one of the plurality of pipeline processing units from another pipeline processing unit;
- allocating the input data packet to one of a plurality of outputs thereof according to a prescribed allocation method;
- conducting prescribed processing to the data packet using a plurality of processing portions connected in parallel to said respective plurality of outputs, and
- outputting the data packet from said plurality of processing portions to an immediately following pipeline processing unit.
Dependencies are inferred from the claim text.