处理加速解码后视频数据copy的方法

首先是Intel的方式intel 官方文档

其次是可以使用intel指令集 sse4

再然后可以使用ffmpeg的：

void av_image_copy_uc_from     (     uint8_t *      dst_data[4],
            const ptrdiff_t      dst_linesizes[4],
            const uint8_t *      src_data[4],
            const ptrdiff_t      src_linesizes[4],
            enum AVPixelFormat      pix_fmt,
            int      width,
            int      height
      )           

Copy image data located in uncacheable (e.g.

GPU mapped) memory. Where available, this function will use special functionality for reading from such memory, which may result in greatly improved performance compared to plain av_image_copy().

The data pointers and the linesizes must be aligned to the maximum required by the CPU architecture.

Note
    The linesize parameters have the type ptrdiff_t here, while they are int for av_image_copy().
    On x86, the linesizes currently need to be aligned to the cacheline size (i.e. 64) to get improved performance.