摘要:
An arithmetic processing device includes a first storage for storing processing contents in a state where the processing contents are associated with addresses, a second storage for storing each of the addresses of the processing contents stored in the first storage, a holding portion, a reading portion-for successively reading the addresses stored in the second storage and outputting the read addresses to the holding portion, and an execution portion for reading the processing content corresponding to the address output from the holding portion from the first storage and executing the read processing content. When the holding portion holds no address, the holding portion temporarily holds the address read by the reading portion and outputs the held address, whereas when the holding portion holds the address, the holding portion waits for completion of the execution of the processing content by the execution portion and outputs the held address after the completion of the execution.
摘要:
To provide a DMA transfer apparatus and a DMA transfer method capable of reducing traffic on a bus between an external shared memory and DMA controller with less additional hardware to effectively use a memory.A pattern generation section 11 is provided in a DMA controller 17 and generates data of a predetermined pattern, such as a zero matrix or unit matrix, in the DMA controller when data is transferred from an external shared memory 14 to an internal memory 15. Further, transfer data read out from the external shared memory is temporarily held in a queuing section 13 for queuing. At this time, switching between the transfer data from the queuing section and predetermined pattern data is made based on the number of the transfer data.
摘要:
To provide a DMA transfer apparatus and a DMA transfer method capable of reducing traffic on a bus between an external shared memory and DMA controller with less additional hardware to effectively use a memory.A pattern generation section 11 is provided in a DMA controller 17 and generates data of a predetermined pattern, such as a zero matrix or unit matrix, in the DMA controller when data is transferred from an external shared memory 14 to an internal memory 15. Further, transfer data read out from the external shared memory is temporarily held in a queuing section 13 for queuing. At this time, switching between the transfer data from the queuing section and predetermined pattern data is made based on the number of the transfer data.