转自:http://docs.gstreamer.com/display/GstSDK/Playback+tutorial+8%3A+Hardware-accelerated+video+decoding
GStreamer needs to keep track of where these “hardware buffers” are though, so conventional buffers still travel from element to element, but their only content is a hardware buffer ID, or handler. If retrieved with an appsink, for example, hardware buffers make no sense, since they are meant to be handled only by the plugin that generated them.
To indicate this, these buffers have special Caps, like video/x-vdpau-output or video/x-fluendo-va. In this way, the auto-plugging mechanism of GStreamer will not try to feed hardware buffers to conventional elements, as they would not understand the received buffers. Moreover, using these Caps, the auto-plugger is able to automatically build pipelines that use hardware acceleration, since, after a VAAPI decoder, a VAAPI sink is the only element that fits.
本文介绍GStreamer中如何使用硬件加速进行视频解码。文章解释了硬件缓冲区在管道中的工作方式及其特殊能力,如video/x-vdpau-output和video/x-fluendo-va。这些特性确保了GStreamer能够自动构建利用硬件加速的播放管线。
5675

被折叠的 条评论
为什么被折叠?



