From 2120ea4f083228dde0d7307203391a4ec8f57e2d Mon Sep 17 00:00:00 2001 From: "Tim E. Real" Date: Fri, 9 Sep 2011 10:04:11 +0000 Subject: Mostly engine fixes/corrections. Please see ChangeLog. --- muse2/ChangeLog | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) (limited to 'muse2/ChangeLog') diff --git a/muse2/ChangeLog b/muse2/ChangeLog index bdc4cb57..b0bc5121 100644 --- a/muse2/ChangeLog +++ b/muse2/ChangeLog @@ -1,3 +1,19 @@ +09.09.2011: + - Removed sending of SEQM_ADD_TRACK, SEQM_REMOVE_TRACK, SEQM_CHANGE_TRACK, SEQM_REMOVE_PART, and + SEQM_CHANGE_PART to ALSA midi thread (which waits) from inside Audio::processMsg. + Replaced with hand-offs to Song::processMsg. TODO: SEQM_SET_TRACK_OUT_CHAN etc. (Tim) + - MidiDevice play and stuck event lists no longer directly accessible from outside - + replaced with ::addScheduledEvent and ::addStuckNote. (Tim) + - Installed a ring buffer for ALSA MidiDevice play/stuck notes event lists, filled by ::addScheduledEvent + and ::addStuckNote, because ALSA midi is processed in a thread other than audio. (Tim) + TODO: processMidi could do stuck notes better by bypassing their insertion in the play event list, + also applies to seek/stop handler block, but I'm not sure about the time comparisons. + - Some more separation of ALSA and Jack Midi, one step closer to configurable choice of either/or. (Tim) + - Massive cleanup of audio, midi, mididev, alsamidi, jackmidi modules. Moved some stuff from Audio::processMidi + into MidiDevice. (Tim) + - Test OK here so far, with stress files and complete pre-existing songs, ALSA, Jack, and Synth midi. (Tim) + TODO: Restoration of sustain upon play seems broken for a while - try putEvent instead of addScheduledEvent + in Audio::startRolling(). 08.09.2011: - Finished namespace MusEWidget. Added namespaces MusEUtil, MusEDialog. (Orcan) 07.09.2011: -- cgit v1.2.3