The patent badge is an abbreviated version of the USPTO patent document. The patent badge does contain a link to the full patent document.
The patent badge is an abbreviated version of the USPTO patent document. The patent badge covers the following: Patent number, Date patent was issued, Date patent was filed, Title of the patent, Applicant, Inventor, Assignee, Attorney firm, Primary examiner, Assistant examiner, CPCs, and Abstract. The patent badge does contain a link to the full patent document (in Adobe Acrobat format, aka pdf). To download or print any patent click here.
Patent No.:
Date of Patent:
Oct. 29, 2002
Filed:
Dec. 15, 1999
Method and system for concurrent processing of slices of a bitstream in a multiprocessor (mp) system
Ekman Tsang, Vancouver, CA;
Silicon Magic Corporation, Sunnyvale, CA (US);
Abstract
A method and system for concurrent processing of slices of a bitstream in a multiprocessor (MP) system is disclosed. The MP system includes a number of identical processors and a common memory. The memory is for receiving a plurality of bitstreams (preferably MPEG2 bitstreams) as a plurality of slices. The method and system comprises accessing a semaphore register by one of the plurality of processors and searching for an associated slice within the memory by the one processor. The method and system further comprises processing the associated slice by the one processor. Finally, the method and system comprises updating a memory location which holds the last address of the associated slice by the one processor; wherein subsequent processors search for each of the plurality of slices from the updated last address in the register. A system and method in accordance with the present invention provides for intercommunication between the plurality of processors within a multiprocessing system. By determining within a semaphore the most recent location of the decoding task, the time and effort spent on searching for new starting location of new task is minimized. Accordingly, the decoding process is parallel, and in most cases averages out the decoding demand on the processors. Accordingly, each of the processors within the multiprocessor system, only needs to search from the address of the bitstream through the remainder of the bitstream to obtain the slice. This process is repeated for each of the processors until the bitstream is processed. Therefore, in a system and method in accordance with the present invention there is no requirement that a processor search from the beginning of the bitstream to the point where the slice originates because the address pointers are updated as at the point where the last slice has been operated on. This provides for a more efficient system for processing bitstreams in a parallel fashion.