summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorJan Schmidt <jan@centricular.com>2016-04-16 02:17:26 +1000
committerJan Schmidt <jan@centricular.com>2016-04-16 22:49:37 +1000
commit51e1949879e5b21dec9d5ff14a7016a12a4f5430 (patch)
tree03c944e95fa90bacd1fe251e7fb5eda0132586cd
parent1c65828968f4b0c7c344bf46606af3404698d444 (diff)
Revert "pulsesink: uncork if needed upon commit"
This reverts commit 0dd46accf6d282ff07065852bd91c85c78af3394. With some audiosinks, starting the ringbuffer on the first commit causes audio glitches at startup by starting to output segments from the ringbuffer before it has been filled / fully prerolled. This doesn't usually happen with pulsesink because we map the pulseaudio ringbuffer directly, but we should keep things consistent with other sinks with regards to startup latency, plus it gives more headway to avoid glitching, should the initial 2nd segment take more than 10ms to generate. https://bugzilla.gnome.org/show_bug.cgi?id=657076
-rw-r--r--ext/pulse/pulsesink.c6
1 files changed, 0 insertions, 6 deletions
diff --git a/ext/pulse/pulsesink.c b/ext/pulse/pulsesink.c
index f7685f71f..c88fa965b 100644
--- a/ext/pulse/pulsesink.c
+++ b/ext/pulse/pulsesink.c
@@ -1516,12 +1516,6 @@ gst_pulseringbuffer_commit (GstAudioRingBuffer * buf, guint64 * sample,
if (pbuf->paused)
goto was_paused;
- /* ensure running clock for whatever out there */
- if (pbuf->corked) {
- if (!gst_pulsering_set_corked (pbuf, FALSE, FALSE))
- goto uncork_failed;
- }
-
/* offset is in bytes */
offset = *sample * bpf;