Mon May 12 19:36:15 CEST 2008
I'm interested in knowing whether ffmpeg was designed
with parallelization in mind or was it transformed later on
to enable parallelization.
I'm collecting benchmarks for a parallelization study and
I'm interested in looking at what type of clean up would
be required before the code can be made parallel.
I noticed that early revisions (-r 2000) for example,
lack (there could be others) thread entry point encode_thread. However,
since I'm not that familiar with the code I'm not sure
whether some clean up has already been performed by then.
More information about the ffmpeg-devel