[FFmpeg-devel] [PATCH] HWAccel infrastructure (take 3)

Stephen Warren swarren
Fri Feb 20 17:23:43 CET 2009


Gwenole Beauchesne wrote:
> 
> On Thu, 19 Feb 2009, Stephen Warren wrote:
> 
> > Of course, perhaps this could be added later, if and when needed, rather
> > than right now, and for performance reasons could be optional based on
> some
> > accelerator-provided flag. Presumably though, since ffmpeg searches the
> > data for start codes to find the next picture, recording the slice start
> > positions would be almost zero extra work anyway.
> 
> That's what was done initially with my buf+offset approach (as an
> optimization hint if buf is the frame start) but Michael preferred
> buf+size instead. So, you will now have to compute your offsets yourself
> too. ;-)

Just so I'm clear (because I'm ill and possibly not thinking straight), the
API will be presented with one buf+size per slice that the ffmpeg parser
found, so all the "compute" is is adding up the sizes of previous slices?
That's no big deal. Just so long as the ffmpeg->VDPAU-specific structure
conversion doesn't have to rescan the entire bitstream looking for start
codes.

-- 
nvpublic





More information about the ffmpeg-devel mailing list