On Friday, March 29, 2013 02:47:20 PM Helmar Sch?tz wrote: > On Fr 29 M?r 2013 08:54:16 CET, Tanu Kaskinen wrote: > > I wouldn't mind having a glance at a log with maximum verbosity, with > > realtime-scheduling enabled. > > http://mayoneso.de/pulseaudio/log_pa_rt_enabled.tar.gz I didn't see anything suspicious (the ConsoleKit activity at the end of the second log looks strange, but it shouldn't have any effect on the realtime threads, and it's not in the other logs). > > Getting killed due to excessive CPU use suggests that PulseAudio enters a > > busy-loop. Another possibility is that PulseAudio does some heavy > > processing, but that shouldn't be the case with the default > > configuration. Have you loaded some filter modules in PulseAudio? > > No, PulseAudio runs with the default configuration. > > > If you disable realtime-scheduling, does pulseaudio consume a lot of CPU > > time? > > It uses about 2-6 % CPU load and ~1184 seconds CPU time. No problem with that then. > > I guess you don't have rtkit installed. rtkit should be able to > > dynamically remove the realtime scheduling from PulseAudio in case > > PulseAudio doesn't behave nicely, before PulseAudio gets killed. Of > > course, if there's a busy- loop involved, there would still be excessive > > CPU load... > > rtkit is installed. So it seems, it's visible in the logs too. I'm not particularly familiar with rtkit, so I don't know if it's misbehaving or are you expected to do some configuration to prevent the killing. I don't really have much ideas for debugging this. gdb is useless, because you won't get any backtraces when the process is killed. -- Tanu
