On 25/08/17 16:35, wm4 wrote: > That looks generally OK. Is there any chance a hwaccel approach would > be possible instead? If I've learned anything about hardware decoding, > then that hwaccel is vastly superior to vendor-implemented full stream > decoders.
Unfortunately I think the whole point of the V4L2 M2M push is to define a common way for vendors to implement full-stream decoders in the kernel without it affecting userspace at all. It's pretty much OpenMAX IL, except with all of the code pushed into the kernel so that users don't need to deal with extra userspace libraries. So, I don't think there is any chance of it working like this at all, though there might be something said for parsing the headers externally to extract all the metadata correctly before passing it to the kernel. - Mark _______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
