Using additional intermediate buffer queues to identify interleaved media data to be read together

Abstract

Techniques are disclosed for identifying multiple sections from one or more tracks of a media file and reading them together in a consumption-driven pipeline process. A render pipeline may comprise a sample generator, a sample buffer, and a destination buffer. Multiple render pipelines may be used for parsing multiple tracks of the media file. An I/O manager may determine that a destination buffer requires new data. The I/O manager may schedule a memory read for a data element from the sample buffer corresponding to the destination buffer and may determine if any of the sample buffers have data elements with memory locations close to the scheduled read. If so, the I/O manager may also schedule those memory locations to be read. After reading, the filled data elements corresponding to the read memory may then be sent to their corresponding destination buffers to be consumed and added to their corresponding tracks.