Information processing apparatus, cache apparatus, and data processing method

A data processing method executed in a data processing apparatus, comprising:

• controlling a first pipeline, which has a plurality of first nodes, to shift first data held in a first node of interest to a first node which is located in a first direction with respect to the first node of interest;
• controlling a second pipeline, which has a plurality of second nodes respectively corresponding to the first nodes of the first pipeline, to shift second data held in a second node of interest to a second node which is located in a second direction opposite to the first direction with respect to the second node of interest;
• executing data processing using the first data and the second data;
• and
• holding second data output from the second pipeline, and deciding priority order of the second data held in the holding step based on attribute information of the second data held in the holding step, the attribute information being used for determining whether the second data held in the holding step is to be discarded or to be returned to the second pipeline, wherein the second data held in the holding step and the second data held by the second pipeline are exchanged in accordance with the priority order.

A program stored in a non-transitory computer-readable recording medium, said program controlling, when said program is executed by a processor, a computer to execute:

• controlling a first pipeline, which has a plurality of first nodes, to shift first data held in a first node of interest to a first node which is located in a first direction with respect to the first node of interest;
• controlling a second pipeline, which has a plurality of second nodes respectively corresponding to the first nodes of the first pipeline, to shift second data held in a second node of interest to a second node which is located in a second direction opposite to the first direction with respect to the second node of interest;
• executing data processing using the first data and the second data;
• and
• holding second data output from the second pipeline, and deciding priority order of the second data held in the holding step based on attribute information of the second data held in the holding step, the attribute information being used for determining whether the second data held in the holding step is to be discarded or to be returned to the second pipeline, wherein the second data held in the holding step and the second data held by the second pipeline are exchanged in accordance with the priority order.