13
0
Commit Graph

31842 Commits

Author SHA1 Message Date
dcc0f1cb17
Fix playlist edit after edit race
After an edit option the playlist is thawed and triggers
the butler thread (SessionEvent::Overwrite) to re-read the
playlist.

If another edit operation is started the butler may reads a
region using the region's new position, but the playlist's
old range.

See also 4db1c02bd1
2021-05-12 19:24:00 +02:00
bdaca1d783
Skip redundant disk buffer changes
This speeds up initial session loading or sessions with many regions.
2021-05-12 19:10:10 +02:00
3a7304ca01
Fix session loading when aborting latency measurement #8691
Stop the engine if it was started for latency measurement
but measurement was aborted.

Previously the engine was kept running and the "Start' button
was not available.
If this happened during initial session loading, a user
would have to manually restart the engine to proceed.
2021-05-12 00:11:43 +02:00
0fc50b9bf2 Mackie Control: set select state correctly during initial setup 2021-05-11 15:55:04 -06:00
0567c143f1
Tweak export-report to fit on smaller screens #8698
When the screen-height is insufficient, the loudness-graph is
not included in the dialog. It is still visible in the image
saved with the export (save-export-analysis-image).

For tiny screens the conformity analysis may also be skipped.
2021-05-11 23:12:31 +02:00
0033929e7f
Update split-benchmark script (fix autoconnect) 2021-05-11 20:30:53 +02:00
58830d857c
Lua API for sends, sidechains and monitoring 2021-05-11 20:30:53 +02:00
ca1ff5afaf
Remove cruft 2021-05-11 20:30:53 +02:00
19c9622a30 NOOP: comment fix, there is no deadlock! 2021-05-11 08:35:14 -06:00
99022a6157 waveview: add some extensive notes about threading design 2021-05-11 07:57:31 -06:00
1a9df476c2 waveview: now that _quit is protected by a mutex, it doesn't need to be atomic
No reason to have two memory fences when we only need one
2021-05-11 07:57:31 -06:00
John Emmas
952416c596 Accommodate some changes to 'ssize_t' in VS2019
In more recent versions of MSVC, ssize_t equates to 'long' (for a 32-bit build) or '__int64' (for 64-bit)
2021-05-11 14:06:25 +01:00
166ac63924 waveview: redesign thread exit strategy
The previous design had a race condition. When WaveViewThreads::stop_threads() was called, it would
first acquire the mutex, then set _quit, then call condition.broadcast(). But worker threads would
check _quit without holding the mutex. It was therefore for a thread to be delayed in its
own lock acquisition by the ::stop_threads() caller, then end up back in cond.wait() AFTER
the cond.broadcast() was done. Such a thread would sleep forever and never wake up.

This new design removes WaveViewDrawRequestQueue, which was a clean encapsulation of the
queueing aspects of WaveViewThreads, but unfortunately made correct mutex acquisition
and condition signalling/waiting needlessly complex. THe mutex, condition variable
and actual queue were moved into WaveViewThreads, and all worker threads execute a method
of the class which gives the appropriate code easy access to the mutex and condition var,
which must always be used together.
2021-05-10 21:41:26 -06:00
e64e140809
Update Russian translation 2021-05-11 02:21:15 +02:00
5343d00382 when adding locations with empty names, generate one (related to 1c753a7f38) 2021-05-09 17:52:29 -06:00
c45be2266e support SMF markers without labels 2021-05-09 17:52:29 -06:00
bfb5bfcd2d
Fix builds (float, double ambiguity)
```
gtk2_ardour/ardour_ui.cc:2060: error: ISO C++ says that these are ambiguous, even though the worst conversion for the first is better than the worst conversion for the second:
/usr/include/architecture/i386/math.h:343: note: candidate 1: double pow(double, double)
/usr/include/c++/4.2.1/cmath:357: note: candidate 2: float std::pow(float, float)

```
2021-05-09 05:59:29 +02:00
292547b264
Use RWLock for Locations
This replaces a Mutex and adds additional read-locks.

This is needed to address some threading issues with rt-threads
calling auto_loop_location() while the GUI changes locations.

Since locations are C-pointers this is still not entirely safe!
Locations::remove() may delete a location while a pointer
to it is being used in another thread.
2021-05-09 03:20:14 +02:00
600e959db5 set a boolean flag as needed to break out of MIDI tempo map import 2021-05-08 18:38:39 -06:00
78d56daf0a fix position of imported MIDI markers
Note: this fix may imply that the same change needed when importing a tempo map
2021-05-08 18:38:19 -06:00
8b1df3e083 fix bad transition problem in transportFSM when slowing to a stop that is then reversed 2021-05-08 17:58:46 -06:00
66bc15f8c5 add stderr output for TransportFSM bad transitions, and remove stacktrace 2021-05-08 17:58:46 -06:00
ad14077587 rearrange transportFSM code to be more explicit about a speed-reset-after-stop 2021-05-08 17:58:46 -06:00
e588fe2057 avoid crash when stopping with "abort" set to true, but after _last_roll_location was reset to -1 2021-05-08 17:58:46 -06:00
05ef7f2647 come to a complete stop if rewind/ffwd steps past -40 semitones slower 2021-05-08 17:58:46 -06:00
8803b55f45
Remove invalid assertion
There are some valid cases to call no-roll with non-zero
transport speed. e.g. export pre_roll/fail_roll when
exporting multiple timespans.
2021-05-08 23:46:10 +02:00
79bf025862
Fix export race condition (and crash)
Previously the freewheel export thread directly called
Session::butler_transport_work(). The butler thread
may concurrently call the same function. This can lead
double free or memory corruption (see below)

Now export thread summons the butler and does nothing
until it completed its work.

```
Export Thread:
3   XMLNode::~XMLNode
4   ARDOUR::AutomationList::snapshot_history
5   ARDOUR::AutomationList::start_write_pass
6   ARDOUR::Automatable::non_realtime_locate
7   ARDOUR::Route::non_realtime_locate
8   ARDOUR::Session::non_realtime_locate
9   ARDOUR::Session::butler_transport_work
10  ARDOUR::Session::process_export_fw

Butler thread:
7   XMLNode::~XMLNode
8   ARDOUR::AutomationList::snapshot_history
9   ARDOUR::AutomationList::start_write_pass
10  ARDOUR::Automatable::non_realtime_locate
11  ARDOUR::Route::non_realtime_locate
12  ARDOUR::Session::non_realtime_locate
13  ARDOUR::Session::butler_transport_work
14  ARDOUR::Butler::thread_work
15  ARDOUR::Butler::_thread_work
```
2021-05-08 23:29:49 +02:00
cc83cdd9df
Config isn't editable 2021-05-08 19:19:16 +02:00
1c47f0adbe use 4 semitone steps whenever rewind/ffwd is adjusting in the range below 1 octave slower 2021-05-08 09:43:04 -06:00
57b3df8b22 copy over (boo!) transport control changes to BasicUI (control surfaces) 2021-05-08 09:20:21 -06:00
6b98e56da2 const-ification tweak 2021-05-08 09:19:50 -06:00
6ca18bfcfd tweak gradual rewind/ffwd action to move in 4 semitone steps once the speed is below "one octave down"
This avoids taking forever to get to -40st, where we reverse direction
2021-05-08 09:13:32 -06:00
9d7ded8bfe allow rewind/ffwd to incrementally move between speeds of zero and 1/-1
It's a little slow in getting all the way to zero, so this may need some tweaking
2021-05-08 08:16:28 -06:00
d88e0c6673
Fix windows builds (declare class binding)
see also 2ff8f5a5de
2021-05-08 14:38:43 +02:00
08092d22b4
Prevent invalid frame/child size allocation 2021-05-08 14:36:43 +02:00
ce1e05fc3d
Update DnD min. movement threshold -- #8686
Previously x-axis threshold was in samples. This is
useless unless zoomed in to the max. So in most cases
the first-move was always in x-direction, making constrained
y-axis drags near impossible.

Furthermore the threshold for copy-drags was increased
as per #8686
2021-05-08 04:28:36 +02:00
4d815554dd Various Control Surfaces: Fix one off in calculation of fader position
MIDI pitch bend holds 14 bits (16384) which gives values of 0 to 16383.
2021-05-07 18:02:44 -05:00
Carlo Stemberger
67cce59f4e
Avoid regex convention in script documentation 2021-05-08 00:35:31 +02:00
Edgar Aichinger
a103a89f4f
also install desktop icon files 2021-05-08 00:27:33 +02:00
Edgar Aichinger
433d7dc8a0
freedesktop: install (not only build) desktop, mimeinfo and appdata files 2021-05-08 00:27:33 +02:00
Tiefseetauchner
720a7eed77 Update compile.sh
cd in line 4 would not work if path has blank as it would have two arguments
./compile.sh: line 4: cd: too many arguments
2021-05-07 16:25:21 -06:00
e2a28ec776 avoid playhead jumping while we locate from a control surface 2021-05-07 16:24:53 -06:00
b0d4f77d92
Optimize RegionList redisplay for batch changes 2021-05-07 23:29:25 +02:00
f6c011cb65
Remove old RegionPropertyChanged signal (3/3) 2021-05-07 23:29:25 +02:00
283cade057
Update GUI Region[s]PropertyChanged signal handlers (2/3) 2021-05-07 23:29:25 +02:00
2ff8f5a5de
Consolidate RegionsPropertyChanged signal emission (1/3)
This prepares to combine multiple RegionPropertyChanged signals
(potentially thousands) into a single signal.
2021-05-07 23:29:25 +02:00
5af480083c
MIDI List Editor, subscribe to region's signal (not all regions) 2021-05-07 23:29:25 +02:00
bb7f919a07 fix import file Lua script to follow new API to PublicEditor::do_import() 2021-05-07 14:58:16 -06:00
3e47057609 import process now allows optional import of MIDI markers/cues 2021-05-07 14:58:16 -06:00
f689e9ecab Evoral: add methods to SMF to load and access markers/cues in SMF files 2021-05-07 14:58:16 -06:00