Mechanism for utilizing voice path DMA in packetized voice communication system to decrease latency and processor overhead

摘要

A voice path direct memory access (DMA)-based packet generation mechanism writes digitally encoded voice samples directly into prescribed subportions of a preallocated portion of random access memory, to avoid interrupting a main processor for the purpose. A pointer to a respective buffer space subportion is presented to a protocol stack, so that one or more overhead bytes for the stored voice samples can be generated and written into adjacent address space of the preallocated portion of random access memory. The contents of the preallocated memory space are then serialized out for transmission to a destination receiver.