RepoMirrors
/
mpv
mirror of https://github.com/mpv-player/mpv
1
0
Fork 0
Code Issues Releases Wiki Activity
mpv/audio/out/internal.h

154 lines
5.6 KiB
C
Raw Normal View History

audio/out: make ao struct opaque We want to move the AO to its own thread. There's no technical reason for making the ao struct opaque to do this. But it helps us sleep at night, because we can control access to shared state better.
2014-03-07 14:24:32 +00:00
/*
* This file is part of MPlayer.
*
* MPlayer is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 2 of the License, or
* (at your option) any later version.
*
* MPlayer is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License along
* with MPlayer; if not, write to the Free Software Foundation, Inc.,
* 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
*/
#ifndef MP_AO_INTERNAL_H_
#define MP_AO_INTERNAL_H_
audio/out: feed AOs from a separate thread This has 2 goals: - Ensure that AOs have always enough data, even if the device buffers are very small. - Reduce complexity in some AOs, which do their own buffering. One disadvantage is that performance is slightly reduced due to more copying. Implementation-wise, we don't change ao.c much, and instead "redirect" the driver's callback to an API wrapper in push.c. Additionally, we add code for dealing with AOs that have a pull API. These AOs usually do their own buffering (jack, coreaudio, portaudio), and adding a thread is basically a waste. The code in pull.c manages a ringbuffer, and allows callback-based AOs to read data directly.
2014-03-08 23:04:37 +00:00
#include <stdbool.h>
#include "audio/chmap.h"
#include "audio/chmap_sel.h"
audio/out: reduce amount of audio buffering Since the addition of the AO feed thread, 200ms of latency (MIN_BUFFER) was added to all push-based AOs. This is not so nice, because even AOs with relatively small buffering (e.g. ao_alsa on my system with ~170ms of buffer size), the additional latency becomes noticable when e.g. toggling mute with softvol. Fix this by trying to keep not only 200ms minimum buffer, but also 200ms maximum buffer. In other words, never buffer beyond 200ms in total. Do this by estimating the AO's buffer fill status using get_space and the initially known AO buffer size (the get_space return value on initialization, before any audio was played). We limit the maximum amount of data written to the soft buffer so that soft buffer size and audio buffer size equal to 200ms (MIN_BUFFER). To avoid weird problems with weird AOs, we buffer beyond MIN_BUFFER if the AO's get_space requests more data than that, and as long as the soft buffer is large enough. Note that this is just a hack to improve the latency. When the audio chain gains the ability to refilter data, this won't be needed anymore, and instead we can introduce some sort of buffer replacement function in order to update data in the soft buffer.
2014-03-10 00:13:40 +00:00
// Minimum buffer size in seconds.
#define MIN_BUFFER 0.2
player: unrangle one aspect of audio EOF handling For some reason, the buffered_audio variable was used to "cache" the ao_get_delay() result. But I can't really see any reason why this should be done, and it just seems to complicate everything. One reason might be that the value should be checked only if the AO buffers have been recently filled (as otherwise the delay could go low and trigger an accidental EOF condition), but this didn't work anyway, since buffered_audio is set from ao_get_delay() anyway at a later point if it was unset. And in both cases, the value is used _after_ filling the audio buffers anyway. Simplify it. Also, move the audio EOF condition to a separate function. (Note that ao_eof_reached() probably could/should whether the last ao_play() call had AOPLAY_FINAL_CHUNK set to avoid accidental EOF on underflows, but for now let's keep the code equivalent.)
2014-04-17 21:48:09 +00:00
// If ao_get_delay() reaches this value after ao_play() was called with the
// AOPLAY_FINAL_CHUNK flag set, the playback core expects that the audio has
// all been played.
#define AO_EOF_DELAY 0.05
audio/out: make ao struct opaque We want to move the AO to its own thread. There's no technical reason for making the ao struct opaque to do this. But it helps us sleep at night, because we can control access to shared state better.
2014-03-07 14:24:32 +00:00
/* global data used by ao.c and ao drivers */
struct ao {
int samplerate;
struct mp_chmap channels;
int format; // one of AF_FORMAT_...
audio/out: feed AOs from a separate thread This has 2 goals: - Ensure that AOs have always enough data, even if the device buffers are very small. - Reduce complexity in some AOs, which do their own buffering. One disadvantage is that performance is slightly reduced due to more copying. Implementation-wise, we don't change ao.c much, and instead "redirect" the driver's callback to an API wrapper in push.c. Additionally, we add code for dealing with AOs that have a pull API. These AOs usually do their own buffering (jack, coreaudio, portaudio), and adding a thread is basically a waste. The code in pull.c manages a ringbuffer, and allows callback-based AOs to read data directly.
2014-03-08 23:04:37 +00:00
int bps; // bytes per second (per plane)
audio/out: make ao struct opaque We want to move the AO to its own thread. There's no technical reason for making the ao struct opaque to do this. But it helps us sleep at night, because we can control access to shared state better.
2014-03-07 14:24:32 +00:00
int sstride; // size of a sample on each plane
// (format_size*num_channels/num_planes)
audio/out: feed AOs from a separate thread This has 2 goals: - Ensure that AOs have always enough data, even if the device buffers are very small. - Reduce complexity in some AOs, which do their own buffering. One disadvantage is that performance is slightly reduced due to more copying. Implementation-wise, we don't change ao.c much, and instead "redirect" the driver's callback to an API wrapper in push.c. Additionally, we add code for dealing with AOs that have a pull API. These AOs usually do their own buffering (jack, coreaudio, portaudio), and adding a thread is basically a waste. The code in pull.c manages a ringbuffer, and allows callback-based AOs to read data directly.
2014-03-08 23:04:37 +00:00
int num_planes;
audio/out: make ao struct opaque We want to move the AO to its own thread. There's no technical reason for making the ao struct opaque to do this. But it helps us sleep at night, because we can control access to shared state better.
2014-03-07 14:24:32 +00:00
bool probing; // if true, don't fail loudly on init
bool untimed; // don't assume realtime playback
bool no_persistent_volume; // the AO does the equivalent of af_volume
bool per_application_mixer; // like above, but volume persists (per app)
audio/out: feed AOs from a separate thread This has 2 goals: - Ensure that AOs have always enough data, even if the device buffers are very small. - Reduce complexity in some AOs, which do their own buffering. One disadvantage is that performance is slightly reduced due to more copying. Implementation-wise, we don't change ao.c much, and instead "redirect" the driver's callback to an API wrapper in push.c. Additionally, we add code for dealing with AOs that have a pull API. These AOs usually do their own buffering (jack, coreaudio, portaudio), and adding a thread is basically a waste. The code in pull.c manages a ringbuffer, and allows callback-based AOs to read data directly.
2014-03-08 23:04:37 +00:00
int device_buffer; // device buffer in samples (guessed by
// common init code if not set by driver)
const struct ao_driver *api; // entrypoints to the wrapper (push.c/pull.c)
audio/out: make ao struct opaque We want to move the AO to its own thread. There's no technical reason for making the ao struct opaque to do this. But it helps us sleep at night, because we can control access to shared state better.
2014-03-07 14:24:32 +00:00
const struct ao_driver *driver;
void *priv;
struct encode_lavc_context *encode_lavc_ctx;
struct input_ctx *input_ctx;
struct mp_log *log; // Using e.g. "[ao/coreaudio]" as prefix
audio/out: feed AOs from a separate thread This has 2 goals: - Ensure that AOs have always enough data, even if the device buffers are very small. - Reduce complexity in some AOs, which do their own buffering. One disadvantage is that performance is slightly reduced due to more copying. Implementation-wise, we don't change ao.c much, and instead "redirect" the driver's callback to an API wrapper in push.c. Additionally, we add code for dealing with AOs that have a pull API. These AOs usually do their own buffering (jack, coreaudio, portaudio), and adding a thread is basically a waste. The code in pull.c manages a ringbuffer, and allows callback-based AOs to read data directly.
2014-03-08 23:04:37 +00:00
int buffer;
void *api_priv;
audio/out: make ao struct opaque We want to move the AO to its own thread. There's no technical reason for making the ao struct opaque to do this. But it helps us sleep at night, because we can control access to shared state better.
2014-03-07 14:24:32 +00:00
};
audio/out: feed AOs from a separate thread This has 2 goals: - Ensure that AOs have always enough data, even if the device buffers are very small. - Reduce complexity in some AOs, which do their own buffering. One disadvantage is that performance is slightly reduced due to more copying. Implementation-wise, we don't change ao.c much, and instead "redirect" the driver's callback to an API wrapper in push.c. Additionally, we add code for dealing with AOs that have a pull API. These AOs usually do their own buffering (jack, coreaudio, portaudio), and adding a thread is basically a waste. The code in pull.c manages a ringbuffer, and allows callback-based AOs to read data directly.
2014-03-08 23:04:37 +00:00
extern const struct ao_driver ao_api_push;
extern const struct ao_driver ao_api_pull;
/* Note:
*
* In general, there are two types of audio drivers:
* a) push based (the user queues data that should be played)
* b) pull callback based (the audio API calls a callback to get audio)
*
* The ao.c code can handle both. It basically implements two audio paths
* and provides a uniform API for them. If ao_driver->play is NULL, it assumes
* that the driver uses a callback based audio API, otherwise push based.
*
* Requirements:
audio/out: adjust documentation comments
2014-05-29 21:56:44 +00:00
* a) ->play is called to queue audio. push.c creates a thread to regularly
* refill audio device buffers with ->play, but all driver functions are
* always called under an exclusive lock.
audio/out: feed AOs from a separate thread This has 2 goals: - Ensure that AOs have always enough data, even if the device buffers are very small. - Reduce complexity in some AOs, which do their own buffering. One disadvantage is that performance is slightly reduced due to more copying. Implementation-wise, we don't change ao.c much, and instead "redirect" the driver's callback to an API wrapper in push.c. Additionally, we add code for dealing with AOs that have a pull API. These AOs usually do their own buffering (jack, coreaudio, portaudio), and adding a thread is basically a waste. The code in pull.c manages a ringbuffer, and allows callback-based AOs to read data directly.
2014-03-08 23:04:37 +00:00
* Mandatory:
* init
* uninit
* reset
* get_space
* play
* get_delay
* pause
* resume
audio/out/pull: remove race conditions There were subtle and minor race conditions in the pull.c code, and AOs using it (jack, portaudio, sdl, wasapi). Attempt to remove these. There was at least a race condition in the ao_reset() implementation: mp_ring_reset() was called concurrently to the audio callback. While the ringbuffer uses atomics to allow concurrent access, the reset function wasn't concurrency-safe (and can't easily be made to). Fix this by stopping the audio callback before doing a reset. After that, we can do anything without needing synchronization. The callback is resumed when resuming playback at a later point. Don't call driver->pause, and make driver->resume and driver->reset start/stop the audio callback. In the initial state, the audio callback must be disabled. JackAudio of course is different. Maybe there is no way to suspend the audio callback without "disconnecting" it (what jack_deactivate() would do), so I'm not trying my luck, and implemented a really bad hack doing active waiting until we get the audio callback into a state where it won't interfere. Once the callback goes from AO_STATE_WAIT to NONE, we can be sure that the callback doesn't access the ringbuffer or anything else anymore. Since both sched_yield() and pthread_yield() apparently are not always available, use mp_sleep_us(1) to avoid burning CPU during active waiting. The ao_jack.c change also removes a race condition: apparently we didn't initialize _all_ ao fields before starting the audio callback. In ao_wasapi.c, I'm not sure whether reset really waits for the audio callback to return. Kovensky says it's not guaranteed, so disable the reset callback - for now the behavior of ao_wasapi.c is like with ao_jack.c, and active waiting is used to deal with the audio callback.
2014-05-29 00:24:17 +00:00
* Optional:
* control
* drain
audio/out: adjust documentation comments
2014-05-29 21:56:44 +00:00
* b) ->play must be NULL. ->resume must be provided, and should make the
* audio API start calling the audio callback. Your audio callback should
* in turn call ao_read_data() to get audio data. Most functions are
* optional and will be emulated if missing (e.g. pausing is emulated as
* silence). ->get_delay and ->get_space are never called.
audio/out: feed AOs from a separate thread This has 2 goals: - Ensure that AOs have always enough data, even if the device buffers are very small. - Reduce complexity in some AOs, which do their own buffering. One disadvantage is that performance is slightly reduced due to more copying. Implementation-wise, we don't change ao.c much, and instead "redirect" the driver's callback to an API wrapper in push.c. Additionally, we add code for dealing with AOs that have a pull API. These AOs usually do their own buffering (jack, coreaudio, portaudio), and adding a thread is basically a waste. The code in pull.c manages a ringbuffer, and allows callback-based AOs to read data directly.
2014-03-08 23:04:37 +00:00
* Mandatory:
* init
* uninit
audio/out/pull: remove race conditions There were subtle and minor race conditions in the pull.c code, and AOs using it (jack, portaudio, sdl, wasapi). Attempt to remove these. There was at least a race condition in the ao_reset() implementation: mp_ring_reset() was called concurrently to the audio callback. While the ringbuffer uses atomics to allow concurrent access, the reset function wasn't concurrency-safe (and can't easily be made to). Fix this by stopping the audio callback before doing a reset. After that, we can do anything without needing synchronization. The callback is resumed when resuming playback at a later point. Don't call driver->pause, and make driver->resume and driver->reset start/stop the audio callback. In the initial state, the audio callback must be disabled. JackAudio of course is different. Maybe there is no way to suspend the audio callback without "disconnecting" it (what jack_deactivate() would do), so I'm not trying my luck, and implemented a really bad hack doing active waiting until we get the audio callback into a state where it won't interfere. Once the callback goes from AO_STATE_WAIT to NONE, we can be sure that the callback doesn't access the ringbuffer or anything else anymore. Since both sched_yield() and pthread_yield() apparently are not always available, use mp_sleep_us(1) to avoid burning CPU during active waiting. The ao_jack.c change also removes a race condition: apparently we didn't initialize _all_ ao fields before starting the audio callback. In ao_wasapi.c, I'm not sure whether reset really waits for the audio callback to return. Kovensky says it's not guaranteed, so disable the reset callback - for now the behavior of ao_wasapi.c is like with ao_jack.c, and active waiting is used to deal with the audio callback.
2014-05-29 00:24:17 +00:00
* resume (starts the audio callback)
* Also, the following optional callbacks can be provided:
* reset (stops the audio callback, resume() restarts it)
* control
audio/out: feed AOs from a separate thread This has 2 goals: - Ensure that AOs have always enough data, even if the device buffers are very small. - Reduce complexity in some AOs, which do their own buffering. One disadvantage is that performance is slightly reduced due to more copying. Implementation-wise, we don't change ao.c much, and instead "redirect" the driver's callback to an API wrapper in push.c. Additionally, we add code for dealing with AOs that have a pull API. These AOs usually do their own buffering (jack, coreaudio, portaudio), and adding a thread is basically a waste. The code in pull.c manages a ringbuffer, and allows callback-based AOs to read data directly.
2014-03-08 23:04:37 +00:00
*/
audio/out: make ao struct opaque We want to move the AO to its own thread. There's no technical reason for making the ao struct opaque to do this. But it helps us sleep at night, because we can control access to shared state better.
2014-03-07 14:24:32 +00:00
struct ao_driver {
// If true, use with encoding only.
bool encode;
// Name used for --ao.
const char *name;
// Description shown with --ao=help.
const char *description;
// Init the device using ao->format/ao->channels/ao->samplerate. If the
// device doesn't accept these parameters, you can attempt to negotiate
// fallback parameters, and set the ao format fields accordingly.
int (*init)(struct ao *ao);
audio/out: feed AOs from a separate thread This has 2 goals: - Ensure that AOs have always enough data, even if the device buffers are very small. - Reduce complexity in some AOs, which do their own buffering. One disadvantage is that performance is slightly reduced due to more copying. Implementation-wise, we don't change ao.c much, and instead "redirect" the driver's callback to an API wrapper in push.c. Additionally, we add code for dealing with AOs that have a pull API. These AOs usually do their own buffering (jack, coreaudio, portaudio), and adding a thread is basically a waste. The code in pull.c manages a ringbuffer, and allows callback-based AOs to read data directly.
2014-03-08 23:04:37 +00:00
// Optional. See ao_control() etc. in ao.c
audio/out: make ao struct opaque We want to move the AO to its own thread. There's no technical reason for making the ao struct opaque to do this. But it helps us sleep at night, because we can control access to shared state better.
2014-03-07 14:24:32 +00:00
int (*control)(struct ao *ao, enum aocontrol cmd, void *arg);
audio/out: make draining a separate operation Until now, this was always conflated with uninit. This was ugly, and also many AOs emulated this manually (or just ignored it). Make draining an explicit operation, so AOs which support it can provide it, and for all others generic code will emulate it. For ao_wasapi, we keep it simple and basically disable the internal draining implementation (maybe it should be restored later). Tested on Linux only.
2014-03-08 23:49:39 +00:00
void (*uninit)(struct ao *ao);
audio/out: adjust documentation comments
2014-05-29 21:56:44 +00:00
// push based: see ao_reset()
// pull based: stop the audio callback
audio/out: make ao struct opaque We want to move the AO to its own thread. There's no technical reason for making the ao struct opaque to do this. But it helps us sleep at night, because we can control access to shared state better.
2014-03-07 14:24:32 +00:00
void (*reset)(struct ao*ao);
audio/out: adjust documentation comments
2014-05-29 21:56:44 +00:00
// push based: see ao_pause()
void (*pause)(struct ao *ao);
// push based: see ao_resume()
// pull based: start the audio callback
void (*resume)(struct ao *ao);
// push based: see ao_play()
audio/out: make ao struct opaque We want to move the AO to its own thread. There's no technical reason for making the ao struct opaque to do this. But it helps us sleep at night, because we can control access to shared state better.
2014-03-07 14:24:32 +00:00
int (*get_space)(struct ao *ao);
audio/out: adjust documentation comments
2014-05-29 21:56:44 +00:00
// push based: see ao_play()
audio/out: make ao struct opaque We want to move the AO to its own thread. There's no technical reason for making the ao struct opaque to do this. But it helps us sleep at night, because we can control access to shared state better.
2014-03-07 14:24:32 +00:00
int (*play)(struct ao *ao, void **data, int samples, int flags);
audio/out: adjust documentation comments
2014-05-29 21:56:44 +00:00
// push based: see ao_get_delay()
audio/out: make ao struct opaque We want to move the AO to its own thread. There's no technical reason for making the ao struct opaque to do this. But it helps us sleep at night, because we can control access to shared state better.
2014-03-07 14:24:32 +00:00
float (*get_delay)(struct ao *ao);
audio/out: adjust documentation comments
2014-05-29 21:56:44 +00:00
// push based: block until all queued audio is played (optional)
audio/out: make draining a separate operation Until now, this was always conflated with uninit. This was ugly, and also many AOs emulated this manually (or just ignored it). Make draining an explicit operation, so AOs which support it can provide it, and for all others generic code will emulate it. For ao_wasapi, we keep it simple and basically disable the internal draining implementation (maybe it should be restored later). Tested on Linux only.
2014-03-08 23:49:39 +00:00
void (*drain)(struct ao *ao);
audio/out: make ao struct opaque We want to move the AO to its own thread. There's no technical reason for making the ao struct opaque to do this. But it helps us sleep at night, because we can control access to shared state better.
2014-03-07 14:24:32 +00:00
// For option parsing (see vo.h)
int priv_size;
const void *priv_defaults;
const struct m_option *options;
};
audio/out: feed AOs from a separate thread This has 2 goals: - Ensure that AOs have always enough data, even if the device buffers are very small. - Reduce complexity in some AOs, which do their own buffering. One disadvantage is that performance is slightly reduced due to more copying. Implementation-wise, we don't change ao.c much, and instead "redirect" the driver's callback to an API wrapper in push.c. Additionally, we add code for dealing with AOs that have a pull API. These AOs usually do their own buffering (jack, coreaudio, portaudio), and adding a thread is basically a waste. The code in pull.c manages a ringbuffer, and allows callback-based AOs to read data directly.
2014-03-08 23:04:37 +00:00
// These functions can be called by AOs.
audio/out: make ao struct opaque We want to move the AO to its own thread. There's no technical reason for making the ao struct opaque to do this. But it helps us sleep at night, because we can control access to shared state better.
2014-03-07 14:24:32 +00:00
int ao_play_silence(struct ao *ao, int samples);
audio: wake up the core when audio buffer is running low And also add a function ao_need_data(), which AO drivers can call if their audio buffer runs low. This change intends to make it easier for the playback thread: instead of making the playback thread calculate a timeout at which the audio buffer should be refilled, make the push.c audio thread wakeup the core instead. ao_need_data() is going to be used by ao_pulse, and we need to workaround a stupid situation with pulseaudio causing a deadlock because its callback still holds the internal pulseaudio lock. For AOs that don't call ao_need_data(), the deadline is calculated by the buffer fill status and latency, as before.
2014-04-15 20:38:16 +00:00
void ao_need_data(struct ao *ao);
audio/out: feed AOs from a separate thread This has 2 goals: - Ensure that AOs have always enough data, even if the device buffers are very small. - Reduce complexity in some AOs, which do their own buffering. One disadvantage is that performance is slightly reduced due to more copying. Implementation-wise, we don't change ao.c much, and instead "redirect" the driver's callback to an API wrapper in push.c. Additionally, we add code for dealing with AOs that have a pull API. These AOs usually do their own buffering (jack, coreaudio, portaudio), and adding a thread is basically a waste. The code in pull.c manages a ringbuffer, and allows callback-based AOs to read data directly.
2014-03-08 23:04:37 +00:00
void ao_wait_drain(struct ao *ao);
int ao_read_data(struct ao *ao, void **data, int samples, int64_t out_time_us);
audio/out: make ao struct opaque We want to move the AO to its own thread. There's no technical reason for making the ao struct opaque to do this. But it helps us sleep at night, because we can control access to shared state better.
2014-03-07 14:24:32 +00:00
bool ao_chmap_sel_adjust(struct ao *ao, const struct mp_chmap_sel *s,
struct mp_chmap *map);
bool ao_chmap_sel_get_def(struct ao *ao, const struct mp_chmap_sel *s,
struct mp_chmap *map, int num);
#endif