On 19.02.2018 15:48, Tanu Kaskinen wrote: > This adds a pa_suspend_cause_t parameter to the sink/source_set_state() > functions, and moves part of the work that pa_sink/source_suspend() does > to sink/source_set_state(). The reason for this code shuffling is that I > plan to make all suspend cause changes available to modules through the > state change callbacks. This is the first step towards that. > > Additionally, pa_source_sync_suspend() is changed to also update the > suspend cause of the monitor source when the suspend cause of the > monitored sink changes. That probably doesn't have much effect on > anything, but I think it makes sense to mirror the sink suspend cause in > the monitor source. > > pa_source_sync_suspend() has also a bug fix: previously it was probably > possible that a sink might get suspended while in the passthrough mode. > When the sink then resumed (while still in the passthrough mode), > pa_source_sync_suspend() would resume also the monitor source, even > though the monitor source should be kept suspended when the sink is in > the passthrough mode. Now the monitor source won't be resumed in this > situation. > --- > src/pulsecore/sink.c | 125 ++++++++++++++++++++++++++--------------------- > src/pulsecore/source.c | 128 ++++++++++++++++++++++++++++++------------------- > 2 files changed, 150 insertions(+), 103 deletions(-) > LGTM
