Commit graph

209 commits

Author SHA1 Message Date
Stephen Birarda
30d2e9fd23 add unsafeEachNode to iterate nodes when read lock held elsewhere 2017-06-12 17:54:16 -07:00
Stephen Birarda
4688fe4c39 fix comment for change from replicant to downstream 2017-06-12 16:29:20 -07:00
Stephen Birarda
03a8d7b8c8 only replicate packets for agents being replicated 2017-06-12 16:29:20 -07:00
Stephen Birarda
1868971cfc fix debug for manual codec in upstream agents 2017-06-12 16:29:20 -07:00
Stephen Birarda
91c25d4270 don't continously set codec for upstream agent 2017-06-12 16:29:20 -07:00
Stephen Birarda
88af8b5842 use new downstream/upstream nomeclature 2017-06-12 16:29:20 -07:00
Stephen Birarda
a4aa9689a6 don't construct a packet of exact size for replicated audio packets 2017-06-12 16:29:20 -07:00
Stephen Birarda
fe668b1bb1 make codec negotiation stateless for replicated agents 2017-06-12 16:29:20 -07:00
Stephen Birarda
c7b3b79a23 use replicant nodes for audio packet replication 2017-06-12 16:29:19 -07:00
Stephen Birarda
8ebf1db9a8 switch to replicator/replicant and add new node types 2017-06-12 10:58:34 -07:00
Stephen Birarda
b5c26b596e seek past source UUID in mirrored packets 2017-06-12 02:39:54 -07:00
Stephen Birarda
e629881665 mirror audio format negotiation 2017-06-12 02:18:27 -07:00
Stephen Birarda
0ee0082052 handle new packet types in AudioMixerClientData::processPackets 2017-06-12 02:05:40 -07:00
Stephen Birarda
f726c5bc3a handle mirrored packets in AudioMixerClientData 2017-06-12 01:16:51 -07:00
Stephen Birarda
45bb323723 experimental support for mirrored audio packets to other mixers 2017-06-12 00:19:24 -07:00
Zach Pomerantz
07e97a0cab unqueue from front for audio packets 2017-02-16 21:40:53 +00:00
Zach Pomerantz
2e9217fb47 loosen audio memoization precondition 2017-02-15 13:36:03 -05:00
Zach Pomerantz
06c8fba35d mv Node out of packet queue 2017-02-14 21:10:30 +00:00
Zach Pomerantz
a7386392a1 fix codec parsing 2017-02-14 20:46:27 +00:00
Zach Pomerantz
7d8a947e6d add node-isolated packet parsing to client data 2017-02-14 20:46:27 +00:00
Zach Pomerantz
da5a89f582 move codec negotiation to client data 2017-02-14 20:46:27 +00:00
Zach Pomerantz
481cd0ca6a move audio packets to client data 2017-02-14 20:46:27 +00:00
Zach Pomerantz
2f457ae891 add packet queue to AudioMixerClientData 2017-02-14 20:46:27 +00:00
Zach Pomerantz
5e9fb17949 use tbb::atomic for cp 2017-02-07 15:01:23 -05:00
Zach Pomerantz
ad7c01e86e modularize audio ignore zone computations 2017-02-07 15:01:22 -05:00
Zach Pomerantz
308e3cab71 fix memoized ignoreZone assertion for initialization 2017-02-07 15:01:22 -05:00
Zach Pomerantz
093f748d7d make auto stream explicit 2017-02-07 15:01:22 -05:00
Zach Pomerantz
e7cf84324b abstract audio ignore caching 2017-02-07 15:01:22 -05:00
Zach Pomerantz
207d2e78f0 fix should ignore opts 2017-02-07 15:01:21 -05:00
Zach Pomerantz
310c8b18ee add caching over symmetric nodes for audio shouldIgnore 2017-02-07 15:01:21 -05:00
Zach Pomerantz
8a42755e8f mv shouldIgnore from AudioMixerSlave to ClientData 2017-02-07 15:01:21 -05:00
Zach Pomerantz
3c1cf504d0 rename getIgnoreBox to getIgnoreZone to prevent confusion 2017-02-07 15:01:21 -05:00
Zach Pomerantz
9bcc5c95b4 reduce stream lock usage in AudioMixerClientData::getIgnoreBox 2017-02-07 15:01:21 -05:00
Zach Pomerantz
569ae113a0 memoize audio ignore box in AudioMixerClientData 2017-02-07 15:01:21 -05:00
Brad Hefta-Gaub
fe709f51de personal space feature 2016-11-19 16:30:15 -08:00
David Kelly
ec0382daf7 PR feedback
Completely deleting hrtfs incorrectly.  This seems to be better, we
only want to delete HRTF for the QUuid() stream for the killed node in each
node's list of HRTFs.
2016-11-01 18:29:31 -07:00
David Kelly
0b80610e65 Needed to properly remove hrtfs for avatar audio 2016-11-01 15:09:14 -07:00
David Kelly
80e38c395e Send KillAvatar to AudioMixer too
So it can remove the listening agent avatar audio streams.  Otherwise
they stick around forever, and eventually swamp the audio mixer.
2016-10-31 10:35:16 -07:00
David Kelly
e3a22618d1 CR feedback 2016-10-27 17:49:43 -07:00
David Kelly
52a14bf5c3 Merge branch 'master' of github.com:highfidelity/hifi into dk/moreACAudioArtifacts 2016-10-27 16:52:25 -07:00
Howard Stearns
d2b83b47ba Merge pull request #8936 from zzmp/clean/audio-mixer
clean audio mixer
2016-10-27 15:57:16 -07:00
David Kelly
5e4f539144 AudioMixer flushes now
And protected agent from case with no codec
2016-10-27 14:46:04 -07:00
Zach Pomerantz
c14a928db5 convert size_t to int 2016-10-27 14:34:10 -07:00
Zach Pomerantz
32040a9047 add stat for streams in audio mixer 2016-10-27 13:55:17 -07:00
Zach Pomerantz
9e9f8666ce clean up handleMismatchAudioFormat comment 2016-10-27 13:46:53 -07:00
David Kelly
6a61762659 First try
AudioMixer has same issue transitioning from silence to encoded
audio and back, it seems.  This makes that go away, but I don't
like it.  Better to support a flush function in the encoder but
for now lets do this and be sure it is the entire issue.
2016-10-26 17:41:34 -07:00
Clément Brisset
b63eb6ba9f Merge pull request #8665 from zzmp/feat/audio-stats
move audio stats to QML/JS, with graphs
2016-10-20 16:43:49 -07:00
Zach Pomerantz
055fabf57b fix audio AppendFlag semantics 2016-09-26 18:51:51 -07:00
Zach Pomerantz
d3132d8746 denote terminal audio stats packet 2016-09-26 16:50:51 -07:00
Zach Pomerantz
b9c4018b8e expose AudioStats to qml/js 2016-09-26 15:35:29 -07:00