回复：Re: how does clutter-gst/totem work with dmabuf buffers from v4l2 decoders?

[HalleyZhao]

Hi Junyan:

do you mean the following?

video frame maps to cpu memory address space, 

then a deep copy is required to upload the data to gpu memory address space by clutter_gst_yv12_upload().

it looks inefficient, usually we can bind texture from dmabuf fd.

--------------------------------

----- 原始邮件 -----
发件人：Junyan He <[hidden email]>
收件人："[hidden email]" <[hidden email]>, Discussion of the development of and with GStreamer <[hidden email]>
主题：Re: how does clutter-gst/totem work with dmabuf buffers from v4l2 decoders?
日期：2019年09月10日 14点55分

vaapi buffer can be mapped by the GstVideoMeta->map function.
On 二, 2019-09-10 at 14:24 +0800, HalleyZhao wrote:

> Hi experts:
>
> I notice that totem uses clutter-gst as videosink, then I look into
> clutter-gst (https://github.com/GNOME/clutter-gst).
> however, I haven't found how it work with hardware accelerated
> decoder (usually generate dma_buf buffer for a video frame).
> anyone can give some hints? thanks.
>
>
> --------------------------------
>
> _______________________________________________
> gstreamer-devel mailing list
> [hidden email]
> https://lists.freedesktop.org/mailman/listinfo/gstreamer-devel

_______________________________________________
gstreamer-devel mailing list
[hidden email]
https://lists.freedesktop.org/mailman/listinfo/gstreamer-devel

_______________________________________________
gstreamer-devel mailing list
[hidden email]
https://lists.freedesktop.org/mailman/listinfo/gstreamer-devel