On 07/05/18 21:52, Timo Rothenpieler wrote: >> Nack. Implementation-specific details go in the implementation-specific >> structure (AVHWFramesContext.hwctx). >> >> What are you actually thining of using this for? If you want to add flags >> which are in common between multiple different implementations then maybe it >> would be suitable to put it here, but something implementation-specific >> really shouldn't be. > > I want to add a mapped-frame/dummy mode to the CUDA frame allocator, where it > does everything a normal hwframes ctx does, except allocating memory, because > the backing memory comes mapped from the cuvid frame. > > Which is a simple flag to set, but there is no way to tell it right now. > > The structures in the struct you mentioned are private to libavutil, so I > can't set them from a decoder.
They are public, it's just the frames one for CUDA doesn't currently exist. You want to make a new AVCUDAFramesContext structure in libavutil/hwcontext_cuda.h. - Mark _______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
