13
0
Commit Graph

8970 Commits

Author SHA1 Message Date
3f5c01e4eb Move Windows timer utility functions from PA backend into libpbd 2015-09-16 11:22:16 +10:00
4ffe8ffc0f Put Windows timer functions into PBD namespace in preparation for moving them to libpbd
Add functions for get/set the Multimedia timer resolution, although we are
really only interested in the minimum, this will facilitate testing

Put timer utility functions inside nested namespaces as they are platform
specific
2015-09-16 11:22:16 +10:00
4330db1aa7 Update timestamp types in CycleTimer class to be signed 2015-09-16 11:22:16 +10:00
08e18a0cb4 Move implementation for DSPLoadCalculator back into header
It can be inline now that it is much simpler
2015-09-16 11:22:16 +10:00
f5e7aa11f9 Use same algorithm as CA and ALSA backends in DSPLoadCalculation
Stop using an averaging of the values until we can establish whether it is
really necessary.
2015-09-16 11:22:16 +10:00
cf88bbc472 Add check for invalid timer values from the DummyBackend
Needed for systems where the Windows QPC timer returns erratic values
2015-09-16 11:22:16 +10:00
b2a7393553 Change types in DSPLoadCalculator to signed integer after change in utils::get_microsecond 2015-09-16 11:22:16 +10:00
a974659504 Check return value of windows timer functions and handle errors
Change return type of utils::get_microseconds to signed int and return -1 when
failing to read timer
2015-09-16 11:22:16 +10:00
50a6487e77 move speed quietning code into Delivery, where it belongs.
Note: I find the -12dB reduction too much of a reduction.
2015-09-15 17:01:12 -04:00
601a34521c add an Amp to Delivery, remove it from Send, make use of this in various ::run() methods
Delivery::_amp now will handle monitor-related delicks assuming the Session::config.get_use_monitor_fades() is
true.
2015-09-15 16:35:37 -04:00
fc74894def another note-off fix.
For nearly coincident note-on the sequence does not matter,
but note-off must be sent before a new note-on in strict
order (could be the same note).
2015-09-15 17:50:27 +02:00
4b20e719a8 prepare displaying peakfile queue-len 2015-09-15 17:50:27 +02:00
bed85f98d1 logic rearrangement to avoid unnecessary call to Route::monitoring_state() in instrument/generator tracks 2015-09-15 09:08:38 -04:00
392836a7d2 NO-OP: comment clarification 2015-09-15 09:06:16 -04:00
952443a4b7 NO-OP: parenthesis and brace style 2015-09-15 09:05:08 -04:00
2b1a30ca96 only apply transport declicking if requested to do so (default is to do so) 2015-09-15 09:00:50 -04:00
563d9dfb65 add 2 new session properties related to implicit fades/gain ramps 2015-09-15 09:00:50 -04:00
d92fceabb0 use an explicit black list of route names that require numbering.
This replaces an attempt to check port names which fails for a number of reasons.
2015-09-15 09:00:50 -04:00
04e7cfabbe prepare peak-file cleanup/recreate. 2015-09-15 14:44:05 +02:00
551e20b926 Remove ifstream / ofstream when accessing session template files (or creating a new session from a template file)
On Windows, the above classes don't support UTF-8 so let's get rid of them and use g_fopen / stringstream / fwrite() etc.
2015-09-15 13:17:22 +01:00
8a4bf386a4 amend previous commit. 2015-09-14 21:25:57 +02:00
9a533c925b fix short x-fade rendering 2015-09-14 21:18:56 +02:00
e63c3d028a note-off ordering - fixes #6340
Evoral::Beats::operator>() rounds to (1.0 / PPQN), hardcoded 1/1920.0.

If the time difference between two events is smaller than 1/PPQN,
Beats::operator>()  and  Beats::operator<()  produce ambiguous results.
The same pair of values is both "less than" and "greater than" depending
which operator is used.

While it's fine for some cases to ignore the order of nearly concurent
events, the std::priority_queue must be strictly ordered.
2015-09-14 04:55:01 +02:00
d83889079b hack around a bug in cppunit/mingw/windows. 2015-09-13 20:23:12 +02:00
e45db26d97 adjust precision to specified value, avoid "0". 2015-09-13 20:12:10 +02:00
8e8f8d92c2 allow to run unit-test under wine from srcdir. 2015-09-13 15:30:00 +02:00
69ff6d0507 fix mismatched delete/free 2015-09-13 04:09:24 +02:00
6f6dde4f06 Fix typo in DSPLoadCalculator causing assertions on Windows debug builds 2015-09-13 09:01:26 +10:00
eb998b3264 plug some memory leaks in libs 2015-09-12 20:05:25 +02:00
cb6c975417 remove cruft 2015-09-12 17:55:04 +02:00
a5e3371b80 Event Pool usage debugging, see also 6ade16b38 2015-09-12 02:18:38 +02:00
ec06f2c49f spelling error fixes (notably "overriden" => "overidden") from IOhannes m zmölnig 2015-09-11 09:23:43 -04:00
9fcc48d140 fail to create Monitor section if port-names are not unique
We should still prevent a user from naming a route "Monitor", but
since "Monitor" is a translatable string, this is an i18n safeguard. eg.
 - create a session with a mon-section and a track Монитор.
 - send the session to a friend in Russia.
2015-09-11 13:14:27 +02:00
def273ea71 queue disk buffer overwrite when playlist layering changes.
This is a first approximation. Doing this 100% efficiently is quite hard because it would require comparing the range
currently in the disk buffer and the range impacted by the layer change. I suspect this will work fine for the
foreseeable future
2015-09-10 16:49:15 -04:00
bfd66b2ea0 use PortManager::port_name_prefix_is_unique to check for new route names 2015-09-10 16:41:34 -04:00
c7e755b25c add PortManager::port_name_prefix_is_unique() 2015-09-10 16:41:34 -04:00
8d54a2472d fix recent session loading.
std::stringstream::operator<< calls strlen(), but the string in the
temp buffer may not [yet] be NULL terminated.
2015-09-10 17:39:31 +02:00
0ef8175b5a when removing routes, don't do potentially expensive work once for each route.
We can update solo state and tell interested parties about the removal once the actual removal is done
2015-09-10 11:35:57 -04:00
cb85080ca8 fix deadlock when removing routes and using JACK1.
graph reorder callback needed an early exit if we were deleting routes, is all.
2015-09-10 11:35:57 -04:00
a86aa9e5bc use FileSource::within_session() for peak-files 2015-09-10 16:47:26 +02:00
b3e8495662 remove duplicate enum registration 2015-09-10 08:15:11 -04:00
1a6c08dcfc Make sure that the DSPLoadCalculator class is exportable 2015-09-10 10:24:04 +01:00
d2e083397c Add support for newly introduced class ARDOUR::DSPLoadCalculator (when building with MSVC) 2015-09-10 10:22:38 +01:00
28803fface Remove DSPLoadCalculator class in portaudio backend source 2015-09-10 12:12:56 +10:00
c2e5bdedff Use ARDOUR::DSPLoadCalculator in PortAudioBackend 2015-09-10 12:12:56 +10:00
8625362cac Add DSPLoadCalculator class to libardour
This is similar to the class in the PortaudioBackend but uses an average of the
values if raw load is under 80%
2015-09-10 12:12:56 +10:00
09caf8336f allow LV2 plugins to query current block-size.
This is akin to VST2's audioMasterGetBlockSize.
It returns the current nominal block size (think jack-buffersize).

It's not the only block size that may be used when calling run(), it's
just the normal one. The actual block sizes used may be larger or smaller
and may vary between successive calls of run().

This change became neccesary after 53e969e9. Some plugins expected 
maxBlockLength to be the /current/ buffer-size and not all-time maxiumum.
Those plugins can now use nominalBlockLength.
2015-09-10 00:55:59 +02:00
390ea007c5 rework peakfile handling:
- copy old peak-files to new (do not require re-calc)
- keep old peak-files (for now, backwards compat)
- fix cleanup-sources to remove *new* peak-file
- include channel-number in hash (like it was done before)

see also 624f76b

TODO: add Session > Cleanup > remove/re-create peaks
2015-09-10 00:55:58 +02:00
78ff3c05a3 remove Ardour broken peak-file support
Ardour-2.0 put peak-files in a "broken" location for several months.
Since then Ardour renamed those files. No more. When loading ancient
sessions peak-files are now re-created (in the background).
2015-09-10 00:55:58 +02:00
48579d42b0 update libltc to v1.1.4-4-gb034a23 (endianess issue) 2015-09-09 16:54:26 +02:00