#807: RTP modules do not allow the null sink to suspend when monitoring it using
lots of CPU
--------------------------+-------------------------------------------------
Reporter: psihozefir | Owner: lennart
Type: enhancement | Status: new
Milestone: | Component: module-rtp-*
Resolution: | Keywords: CPU frequency, drain battery
--------------------------+-------------------------------------------------
Comment(by psihozefir):
I was reading your post over and over again, and I'm trying to understand
as much as I can from it. Since the RTP transmision involves a client and
a server, I guess "the server" is the one listening from connections on
the network and that outputs the audio stream to the physical device not
the one serving the audio stream over the network, right? In this sense,
my laptop is a client and my desktop is a server.
htop or top on my laptop shows the CPU stays below 10% at all times,
except when I'm compiling something, when it can be seen the CPU's usage
is 100%. The problem is, the kernel's conservative CPU frequency governor
decides to "clock" the CPU at its highest available freqency when the RTP
client and the RTP monitor are running (that is, without an increase in
CPU's usage as shown by top or other CPU monitoring program).
So... "using lots of CPU" is not quite correct in my opinion. This would
be correct if you would get the same results as when running the following
bash script:
while true; do echo >/dev/null; done
--
Ticket URL: <http://pulseaudio.org/ticket/807#comment:7>
PulseAudio <http://pulseaudio.org/>
The PulseAudio Sound Server
_______________________________________________
pulseaudio-tickets mailing list
pulseaudio-tickets@mail.0pointer.de
https://tango.0pointer.de/mailman/listinfo/pulseaudio-tickets
