hi, all ,
i am setting up RTP streaming environment between a X86 server and a embedded system client, host app is VLC, and target/client is gst-launch, commands are following:
HOST:
# vlc -vvv big_buck_bunny_480p_h264.mov --sout '#rtp{dst=<dst ip>, port=5004,sdp=rtsp://<server ip>:8080/test.sdp}'
Target/client:
# gst-launch udpsrc multicast-group=<server ip> caps="application/x-rtp,media=(string)video,clock-rate=(int)90000,encoding-name=(string)H264" port=5004 ! rtph264depay ! my-codec-hw ! my-render-hw"
the command works well, then i hope playbin2 creates hardware pipeline automatically, command line here:
# gst-launch playbin2 uri=rtsp://<server ip>:8080/test.sdp
it fails to play. btw : playbin2 works well with my optimized codec and render, i test it with command. (gst-launch playbin2 uri=file:///big_buck.mov ), it can find the right elements, such as "my-codec-hw" and "my-render-hw".
then i dump the log of gst-launch , and find playbin2 not perform preroll for live streams, which cause full or real pipeline is not ready before getting GstSystemClock.
in fact, my optimized render can only use the specified clock provided with *_sink_provide_clock(), not GstSystemClock. so the pipeline hang even it links the optimized elements.
my question is that how can i tell playbin2 to use my provided clock for live pipeline?
if any misunderstanding , pls correct me.
--
It's not the things you do in life that you regret ,
but the things that you do not do
------------------------------------------------------------------------------
Start uncovering the many advantages of virtual appliances
and start using them to simplify application deployment and
accelerate your shift to cloud computing.
http://p.sf.net/sfu/novell-sfdev2dev_______________________________________________
gstreamer-devel mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gstreamer-devel