13
0
Commit Graph

839 Commits

Author SHA1 Message Date
cc7b8b1bc5
gcc-11 compat, volatile atomic variables (1/2)
"While 'atomic' has a volatile qualifier, this is a historical
artifact and the pointer passed to it should not be volatile."

Furthermore "It is very important that all accesses to a
particular integer or pointer be performed using only this API"
(from https://developer.gnome.org/glib/2.68/glib-Atomic-Operations.html)

Hence initialization of atomic variables is changed to also use
this API, instead of directly initializing the value.

This also fixes a few cases where atomic variables were
accessed directly.

see also libs/pbd/pbd/g_atomic_compat.h
2021-03-22 15:30:07 +01:00
John Emmas
a2b1482cad Fix a problem with ASIO buffer sizes on Windows
PortAudio uses what it calls 'default suggested latencies' but in callback streaming mode, they can result in wildly inaccurate buffer sizing (e.g. the user requests a buffer size of 128 but PortAudio actually instructs ASIO to use a much bigger size).

What we do now is to improve PortAudio's suggested latency calculation by basing it on the actual buffer size requested by the user.
2021-03-18 16:12:28 +00:00
291de1d277
Fix OSX and non GCC builds 2021-03-01 16:20:38 +01:00
7195ca20e0
Update weak_libjack - https://github.com/x42/weakjack/
This fixes an issue with recent MSVC causing a compile-error
due to __attribute__((constructor))
2021-02-28 16:27:22 +01:00
ace11bd243
Consistency: replace "x-run" with "xrun" (1/2) 2021-02-26 22:22:08 +01:00
John Emmas
d2cb383965 MSVC changes to support the new code modules (Tabbable API stuff) 2021-02-24 14:32:23 +00:00
3f4e2fa348
ALSA: retain port-connections of slave devices
When the engine is re-started, ports should only be reconnected
when all slave devices are running. Previously those connections
were lost.
2021-02-19 14:03:37 +01:00
634d325e5d
Prevent deadlock when disconnecting
The backend holds `_port_callback_mutex` while disconnecting ports.
In some cases disconnecting a port can drop the last reference
resulting in a port-deletion from the connection handler.
This in turn will eventually aquire the `_port_callback_mutex`
and deadlock.

This is now circumvented by using atomic operations instead of
taking a lock to set the `_port_change_flag`.

The flag is also used to trigger a latency update in some cases,
atomic is preferable to taking a lock to set this flag.

--

Full bt: https://paste.debian.net/1184056/
Short:

#1  in pthread_mutex_lock ()
#2  in ARDOUR::PortEngineSharedImpl::port_connect_add_remove_callback()
#3  in ARDOUR::BackendPort::~BackendPort()
#4  in ARDOUR::DummyPort::~DummyPort()
#6  in ARDOUR::DummyAudioPort::~DummyAudioPort()
#7  in boost::checked_delete<ARDOUR::BackendPort>(ARDOUR::BackendPort*)
#12 in boost::shared_ptr<ARDOUR::ProtoPort>::reset()
#13 in ARDOUR::Port::drop()
#14 in ARDOUR::Port::~Port()
#15 in ARDOUR::AudioPort::~AudioPort()
#17 in ARDOUR::AudioEngine::add_pending_port_deletion(ARDOUR::Port*)
#20 in boost::detail::sp_counted_base::release()
#37 in ARDOUR::PortManager::connect_callback() at libs/ardour/port_manager.cc:788
#38 in ARDOUR::DummyAudioBackend::main_process_thread() at libs/backends/dummy/dummy_audiobackend.cc:1018
2021-02-04 21:06:20 +01:00
f0a89c4f1d
ALSA: allow to select clock-source when using multiple devices 2021-01-24 18:23:37 +01:00
8426bb3aaf
ALSA: explicitly list millisecond buffer-sizes 2021-01-22 16:07:33 +01:00
0b97175d49
ALSA: fall back to nearest available buffer-size
The warning "samples per period does not match." never triggered.
Previously not being able to set the requested buffersize was a
fatal error.

This adds support for soundcards that only support msec.
e.g. recent HDA Intel via SOF (Sound Open Firmware)
2021-01-22 16:07:29 +01:00
cff58bd9da
ALSA: Fix slave device lookup 2021-01-22 16:07:24 +01:00
c5b986f420
ALSA: set pretty port names for audio I/O 2021-01-18 03:18:21 +01:00
21e66216be
Separate user-set pretty-names from hardware I/O names
This allow to restore original engine port-names as set
by the backend. ALSA MIDI, CoreAudio, CoreMIDI and PortAudio
drivers can provide human readable physical port names for
some devices.
2021-01-18 03:18:20 +01:00
0c81ba33d2
JACK latency compensation for newly created tracks #8472
Ardour native backends trigger a latency-callback when ports_changed.
This was lost for JACK when the shared port engine was implemented.
2021-01-14 03:22:25 +01:00
5e76d63c06
Resolve some Xcode12 C++11, sigc++ issues 2020-12-12 04:46:22 +01:00
60901d26b7
ALSA: improve debug output 2020-12-03 21:29:21 +01:00
80e3c191e4
Pulseaudio: allow steam being suspended during export
When exporting long sessions with freewheeling, pulseaudio
may meanwhile suspend the corked audio device. The "FAIL_ON_SUSPEND"
option then prevents ardour to uncork it after export, and the
audio-backend is halted.
2020-11-22 21:41:47 +01:00
81862f8c30
Allow to move pulse stream #8467
This potentially breaks various assumptions (e.g. no resampling,
fixed buffersize) when the stream is moved to a different device.
Then again it's pulseaudio, which is unsuitable for pro-audio to
begin with.
2020-11-11 15:59:26 +01:00
bb468ee4de
Update jackd.exe registry key
This ensures compatibility with upcoming jack > 1.9.16,
current jack 1.9.16, as well as older versions.
2020-11-09 20:06:49 +01:00
c87ef15e82
ALSA: configure capture settings before playback
This fixes an issue with some soundcards e.g. "AxeFx III".
Device configuration fails unless set_hwpar() is performed
for the capture device before configuring the playack
device (half duplex is fine, too).
2020-11-04 14:46:49 +01:00
cf0904f752
Handle vfork/exec failure
Don't call atexit, exit without cleanup.
2020-10-31 22:33:10 +01:00
a9bd7b2848
Remove unused compile-time option to not use vfork 2020-10-31 22:30:45 +01:00
7896c30508
Fix cmdline parameter escape 2020-10-31 13:42:14 +01:00
John Emmas
40ed19767a In a Windows build, support launching both older and newer versions of Jack (Stephane's + Filipe's) 2020-10-31 08:01:55 +00:00
John Emmas
5c06fdf1de In an MSVC build, modify our VC project to use weak Jack
(needed to support Jack v1.9.16)
2020-10-31 07:59:09 +00:00
f7cc5e1f38
Search for JACK on windows using the registry 2020-10-24 04:01:54 +02:00
17f0b87648
Fix Apple Carbon #define interference 2020-10-11 21:40:31 +02:00
472ce90fab
JACK backend allows to calibrate latency
However, if jackd is running, we should test if jackd is using
a full-duplex device to establish if latency calibration is
possible -- fix 92f54b3e98.
2020-09-20 17:16:01 +02:00
216935f16e
Remove ALSA debug message in optimized builds 2020-09-18 15:40:44 +02:00
802ddf344a
Remove unused mutex (amend a80064981e) 2020-09-06 05:43:03 +02:00
a80064981e
NO-OP: Consolidate PortConnect code into shared parent class 2020-09-05 23:05:57 +02:00
481cf37052
RT-safe jack transport time-master
Prefer a try-lock, instead of locking the tempo-map.
2020-08-19 19:07:41 +02:00
Stefan Westerfeld
ad8eb4d05b
Set JACK transport bar_start_tick field. 2020-08-19 19:04:11 +02:00
92f54b3e98
Add API to query if systemic latency can be measured 2020-06-30 03:34:26 +02:00
09aa0a3d1a
Consolidate code using pthread_attr_setstacksize
This also adds some stack constraint to rt and fallback threads
that didn't have those before (ALSA MIDI for example)
2020-06-06 18:35:44 +02:00
ae3c8b19c6
Tweak PulseAudio re-initialization
This fixes an issue with pulse disconnecting after freewheeling,
and possibly also address #8193
2020-06-05 04:24:09 +02:00
b46a7d43c2
Dummy Backend: add a MIDI Clock generator 2020-05-27 23:40:06 +02:00
feaf87d200
Dummy Backend: remove hardcoded MTC sequence 2020-05-12 01:49:45 +02:00
7c15ad099f
Dummy Backend: add a MTC generator 2020-05-12 01:49:41 +02:00
88fc77b4c0
Double dummy MTC sequence duration 2020-05-11 21:35:22 +02:00
4ff6fbe6b8
ALSA: Use same period for in+out #8075
Since the input is never flushed, capture latency is still
only one cycle.
2020-05-10 22:28:40 +02:00
abe23b5885
ALSA backend: ignore idempotent latency settings 2020-05-09 01:22:18 +02:00
c75b2355c1
Fix potential malloc(0) issue 2020-05-05 17:13:18 +02:00
27f26f5cfd
Cont'd work on ALSA nperiods (amend 2326fb163) 2020-05-03 00:19:56 +02:00
2326fb1638
ALSA: only read/clear user-requested nperiods before snd_pcm_start
This should fix excessive latency with the RayDay that
has a fixed buffersize of 16384 samples.
2020-05-01 23:17:31 +02:00
8fa0049d28
ALSA: set start threshold to zero 2020-04-30 04:16:17 +02:00
11f68a02cd
Another step towards nperiod fixes for the RayDay 2020-04-30 02:05:14 +02:00
3a6a9cfa6c
Tweak ALSA nperiod setting, fallback to nearest
This is mainly for RME RayDAT that has a fixed buffersize of 16k:

  dev_name : hw:HDSPMxc2f6c5,0
  channels : 36
  min_rate : 32000
  max_rate : 192000
  min_bufz : 16384
  max_bufz : 16384
  min_nper : 4
  max_nper : 512

However nperiod configuration determines the effective latency
regardless.

This is similar to https://github.com/jackaudio/jack1/blob/master/drivers/alsa/alsa_driver.c#L476-L486
2020-04-29 23:56:24 +02:00
47bb331bfa
Allow to refresh ALSA device list - #8051 2020-04-29 02:19:38 +02:00