Hi, the preferred way of adding new hardware-accelerated decoding wrappers is implementing the hwcontext API in libavutil and the hwaccel hooks in libavcodec decoders. Handles to hardware frames are exported to the caller to avoid a possibly-unnecessary memory copy. See e.g. vaapi_decode for an example. Deviating from this pattern requires a sufficiently strong justification.
Beyond that, we already have two NVidia hardware decoding APIs. Why is there a need for a third one? -- Anton Khirnov _______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-devel To unsubscribe, visit link above, or email ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".
