1
Fork 0
mirror of https://github.com/RGBCube/serenity synced 2025-05-14 08:14:58 +00:00
serenity/Userland/Services/AudioServer/Mixer.cpp
kleines Filmröllchen b4fbd30b70 AudioServer+Userland: Decouple client sample rates from device rate
This change was a long time in the making ever since we obtained sample
rate awareness in the system. Now, each client has its own sample rate,
accessible via new IPC APIs, and the device sample rate is only
accessible via the management interface. AudioServer takes care of
resampling client streams into the device sample rate. Therefore, the
main improvement introduced with this commit is full responsiveness to
sample rate changes; all open audio programs will continue to play at
correct speed with the audio resampled to the new device rate.

The immediate benefits are manifold:
- Gets rid of the legacy hardware sample rate IPC message in the
  non-managing client
- Removes duplicate resampling and sample index rescaling code
  everywhere
- Avoids potential sample index scaling bugs in SoundPlayer (which have
  happened many times before) and fixes a sample index scaling bug in
  aplay
- Removes several FIXMEs
- Reduces amount of sample copying in all applications (especially
  Piano, where this is critical), improving performance
- Reduces number of resampling users, making future API changes (which
  will need to happen for correct resampling to be implemented) easier

I also threw in a simple race condition fix for Piano's audio player
loop.
2023-07-01 23:27:24 +01:00

198 lines
6.6 KiB
C++

/*
* Copyright (c) 2018-2020, Andreas Kling <kling@serenityos.org>
* Copyright (c) 2021, kleines Filmröllchen <filmroellchen@serenityos.org>
*
* SPDX-License-Identifier: BSD-2-Clause
*/
#include "Mixer.h"
#include <AK/Array.h>
#include <AK/Format.h>
#include <AK/MemoryStream.h>
#include <AK/NumericLimits.h>
#include <AudioServer/ConnectionFromClient.h>
#include <AudioServer/ConnectionFromManagerClient.h>
#include <AudioServer/Mixer.h>
#include <LibCore/ConfigFile.h>
#include <LibCore/Timer.h>
#include <pthread.h>
#include <sys/ioctl.h>
namespace AudioServer {
Mixer::Mixer(NonnullRefPtr<Core::ConfigFile> config, NonnullOwnPtr<Core::File> device)
: m_device(move(device))
, m_sound_thread(Threading::Thread::construct(
[this] {
mix();
return 0;
},
"AudioServer[mixer]"sv))
, m_config(move(config))
{
m_muted = m_config->read_bool_entry("Master", "Mute", false);
m_main_volume = static_cast<double>(m_config->read_num_entry("Master", "Volume", 100)) / 100.0;
m_sound_thread->start();
}
NonnullRefPtr<ClientAudioStream> Mixer::create_queue(ConnectionFromClient& client)
{
auto queue = adopt_ref(*new ClientAudioStream(client));
queue->set_sample_rate(audiodevice_get_sample_rate());
{
Threading::MutexLocker const locker(m_pending_mutex);
m_pending_mixing.append(*queue);
}
// Signal the mixer thread to start back up, in case nobody was connected before.
m_mixing_necessary.signal();
return queue;
}
void Mixer::mix()
{
decltype(m_pending_mixing) active_mix_queues;
for (;;) {
{
Threading::MutexLocker const locker(m_pending_mutex);
// While we have nothing to mix, wait on the condition.
// HACK: HDA is currently broken when we don't constantly feed it a buffer stream.
// Commenting out this line makes it "just work" for the time being. Please add this line back once the issue is fixed.
// See:
// m_mixing_necessary.wait_while([this, &active_mix_queues]() { return m_pending_mixing.is_empty() && active_mix_queues.is_empty(); });
if (!m_pending_mixing.is_empty()) {
active_mix_queues.extend(move(m_pending_mixing));
m_pending_mixing.clear();
}
}
active_mix_queues.remove_all_matching([&](auto& entry) { return !entry->is_connected(); });
Array<Audio::Sample, HARDWARE_BUFFER_SIZE> mixed_buffer;
m_main_volume.advance_time();
// Mix the buffers together into the output
for (auto& queue : active_mix_queues) {
if (!queue->client()) {
queue->clear();
continue;
}
queue->volume().advance_time();
for (auto& mixed_sample : mixed_buffer) {
Audio::Sample sample;
if (!queue->get_next_sample(sample, audiodevice_get_sample_rate()))
break;
if (queue->is_muted())
continue;
sample.log_multiply(SAMPLE_HEADROOM);
sample.log_multiply(static_cast<float>(queue->volume()));
mixed_sample += sample;
}
}
// Even though it's not realistic, the user expects no sound at 0%.
if (m_muted || m_main_volume < 0.01) {
m_device->write_until_depleted(m_zero_filled_buffer).release_value_but_fixme_should_propagate_errors();
} else {
FixedMemoryStream stream { m_stream_buffer.span() };
for (auto& mixed_sample : mixed_buffer) {
mixed_sample.log_multiply(static_cast<float>(m_main_volume));
mixed_sample.clip();
LittleEndian<i16> out_sample;
out_sample = static_cast<i16>(mixed_sample.left * NumericLimits<i16>::max());
MUST(stream.write_value(out_sample));
out_sample = static_cast<i16>(mixed_sample.right * NumericLimits<i16>::max());
MUST(stream.write_value(out_sample));
}
auto buffered_bytes = MUST(stream.tell());
VERIFY(buffered_bytes == m_stream_buffer.size());
m_device->write_until_depleted({ m_stream_buffer.data(), buffered_bytes })
.release_value_but_fixme_should_propagate_errors();
}
}
}
void Mixer::set_main_volume(double volume)
{
if (volume < 0)
m_main_volume = 0;
else if (volume > 2)
m_main_volume = 2;
else
m_main_volume = volume;
m_config->write_num_entry("Master", "Volume", static_cast<int>(volume * 100));
request_setting_sync();
ConnectionFromManagerClient::for_each([&](auto& client) {
client.did_change_main_mix_volume({}, main_volume());
});
}
void Mixer::set_muted(bool muted)
{
if (m_muted == muted)
return;
m_muted = muted;
m_config->write_bool_entry("Master", "Mute", m_muted);
request_setting_sync();
ConnectionFromManagerClient::for_each([muted](auto& client) {
client.did_change_main_mix_muted_state({}, muted);
});
}
int Mixer::audiodevice_set_sample_rate(u32 sample_rate)
{
int code = ioctl(m_device->fd(), SOUNDCARD_IOCTL_SET_SAMPLE_RATE, sample_rate);
if (code != 0)
dbgln("Error while setting sample rate to {}: ioctl error: {}", sample_rate, strerror(errno));
// Note that the effective sample rate may be different depending on device restrictions.
// Therefore, we delete our cache, but for efficency don't immediately read the sample rate back.
m_cached_sample_rate = {};
return code;
}
u32 Mixer::audiodevice_get_sample_rate() const
{
if (m_cached_sample_rate.has_value())
return m_cached_sample_rate.value();
u32 sample_rate = 0;
int code = ioctl(m_device->fd(), SOUNDCARD_IOCTL_GET_SAMPLE_RATE, &sample_rate);
if (code != 0)
dbgln("Error while getting sample rate: ioctl error: {}", strerror(errno));
else
m_cached_sample_rate = sample_rate;
return sample_rate;
}
void Mixer::request_setting_sync()
{
if (m_config_write_timer.is_null() || !m_config_write_timer->is_active()) {
m_config_write_timer = Core::Timer::create_single_shot(
AUDIO_CONFIG_WRITE_INTERVAL,
[this] {
if (auto result = m_config->sync(); result.is_error())
dbgln("Failed to write audio mixer config: {}", result.error());
},
this)
.release_value_but_fixme_should_propagate_errors();
m_config_write_timer->start();
}
}
ClientAudioStream::ClientAudioStream(ConnectionFromClient& client)
: m_client(client)
{
}
}