在看AudioTrack的write函数的时候，了解到，音频数据最终都写到了audio_track_cblk_t的结构体中。
这个结构体是在AudioFlinger中创建的。
AudioFlinger是如何来使用这些数据的呢？
今天就来学习学习。


我们写数据的时候，调用了audio_track_cblk_t::framesAvailable_l函数，来判断是否有可用的空间，以供写用。
类audio_track_cblk_t中还有另外一个函数framesReady，看名字，应该是告诉我们已经准备好了多少东东。
看样子，AudioFlinger在使用音频数据的时候，应该是先调用了framesReady函数，来看看我们已经写进去多少音频数据了，然后再使用这些数据。


*****************************************源码************************************************* uint32_t audio_track_cblk_t::framesReady(){ uint64_t u = this->user; uint64_t s = this->server; if (flags & CBLK_DIRECTION_MSK) { if (u < loopEnd) { return u - s; } else { Mutex::Autolock _l(lock); if (loopCount >= 0) { return (loopEnd - loopStart)*loopCount + u - s; } else { return UINT_MAX; } } } else { return s - u; }}

**********************************************************************************************
源码路径：
frameworks\base\media\libmedia\AudioTrack.cpp


#######################说明################################

/*之前看代码都是顺藤摸瓜，今天要顺瓜摸藤了。看看什么地方调用了函数framesReady。搜了一下，调用的地方还真不少。不过，很多地方只是用调用结果来作判断，只有在函数AudioFlinger::PlaybackThread::Track::getNextBuffer中，把返回值保存了下来。以前看代码知道，写数据的时候，framesAvailable返回值是被保存起来，并使用的。类推一下，读数据的时候，framesReady的返回值也应该被保存使用。于是就来到了函数AudioFlinger::PlaybackThread::Track::getNextBuffer中：*/// ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++status_t AudioFlinger::PlaybackThread::Track::getNextBuffer(AudioBufferProvider::Buffer* buffer){ audio_track_cblk_t* cblk = this->cblk(); uint32_t framesReady; uint32_t framesReq = buffer->frameCount; // Check if last stepServer failed, try to step now if (mFlags & TrackBase::STEPSERVER_FAILED) { if (!step()) goto getNextBuffer_exit; LOGV("stepServer recovered"); mFlags &= ~TrackBase::STEPSERVER_FAILED; } framesReady = cblk->framesReady(); if (LIKELY(framesReady)) { uint64_t s = cblk->server; uint64_t bufferEnd = cblk->serverBase + cblk->frameCount; bufferEnd = (cblk->loopEnd < bufferEnd) ? cblk->loopEnd : bufferEnd; if (framesReq > framesReady) { framesReq = framesReady; } if (s + framesReq > bufferEnd) { framesReq = bufferEnd - s; } buffer->raw = getBuffer(s, framesReq); if (buffer->raw == 0) goto getNextBuffer_exit;// ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++void* AudioFlinger::ThreadBase::TrackBase::getBuffer(uint32_t offset, uint32_t frames) const { audio_track_cblk_t* cblk = this->cblk(); int8_t *bufferStart = (int8_t *)mBuffer + (offset-cblk->serverBase)*cblk->frameSize; int8_t *bufferEnd = bufferStart + frames * cblk->frameSize; // Check validity of returned pointer in case the track control block would have been corrupted. if (bufferStart < mBuffer || bufferStart > bufferEnd || bufferEnd > mBufferEnd || ((unsigned long)bufferStart & (unsigned long)(cblk->frameSize - 1))) { LOGE("TrackBase::getBuffer buffer out of range:\n start: %p, end %p , mBuffer %p mBufferEnd %p\n \ server %lld, serverBase %lld, user %lld, userBase %lld, channelCount %d", bufferStart, bufferEnd, mBuffer, mBufferEnd, cblk->server, cblk->serverBase, cblk->user, cblk->userBase, cblk->channelCount); return 0; } return bufferStart;}// ---------------------------------------------------------------- buffer->frameCount = framesReq; return NO_ERROR; }getNextBuffer_exit: buffer->raw = 0; buffer->frameCount = 0; LOGV("getNextBuffer() no more data for track %d on thread %p", mName, mThread.unsafe_get()); return NOT_ENOUGH_DATA;}// ----------------------------------------------------------------/*下面看看哪个地方调用了函数AudioFlinger::PlaybackThread::Track::getNextBuffer。搜索发现，AudioFlinger中只有函数AudioFlinger::DirectOutputThread::threadLoop调用了，而我们说的AudioTrack是播放音乐用的，所以，肯定不是这儿。另外，发现AudioMixer中有几个地方调用了函数getNextBuffer，不过调用代码如下：t.bufferProvider->getNextBuffer(&t.buffer);bufferProvider和AudioFlinger::PlaybackThread::Track又存在怎样的关系呢？bufferProvider的定义代码：AudioBufferProvider* bufferProvider;存在以下继承关系：class Track : public TrackBaseclass TrackBase : public AudioBufferProvider, public RefBase可见，bufferProvider最终指向的是AudioFlinger::PlaybackThread::Track对象，所以，t.bufferProvider->getNextBuffer(&t.buffer)其实是调用的函数：AudioFlinger::PlaybackThread::Track::getNextBuffer函数AudioFlinger::MixerThread::prepareTracks_l调用了函数AudioMixer::setBufferProvider。函数AudioMixer::setBufferProvider中给bufferProvider进行赋值。函数AudioFlinger::MixerThread::threadLoop和函数AudioFlinger::DuplicatingThread::threadLoop中有调用函数AudioFlinger::MixerThread::prepareTracks_l。先看看AudioMixer中调用函数AudioFlinger::PlaybackThread::Track::getNextBuffer的地方吧。AudioMixer中的以下几个函数都调用了函数getNextBuffer：process__nop函数process__genericNoResampling函数process__genericResampling函数process__OneTrack16BitsStereoNoResampling函数process__TwoTracks16BitsStereoNoResampling函数这儿取其中的process__OneTrack16BitsStereoNoResampling函数来看看。*/// ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++// one track, 16 bits stereo without resampling is the most common casevoid AudioMixer::process__OneTrack16BitsStereoNoResampling(state_t* state){ const int i = 31 - __builtin_clz(state->enabledTracks); const track_t& t = state->tracks[i]; AudioBufferProvider::Buffer& b(t.buffer); int32_t* out = t.mainBuffer; size_t numFrames = state->frameCount; const int16_t vl = t.volume[0]; const int16_t vr = t.volume[1]; const uint32_t vrl = t.volumeRL; while (numFrames) { b.frameCount = numFrames; t.bufferProvider->getNextBuffer(&b); int16_t const *in = b.i16; // in == NULL can happen if the track was flushed just after having // been enabled for mixing. if (in == NULL || ((unsigned long)in & 3)) { memset(out, 0, numFrames*MAX_NUM_CHANNELS*sizeof(int16_t)); LOGE_IF(((unsigned long)in & 3), "process stereo track: input buffer alignment pb: buffer %p track %d, channels %d, needs %08x", in, i, t.channelCount, t.needs); return; } size_t outFrames = b.frameCount; if (UNLIKELY(uint32_t(vl) > UNITY_GAIN || uint32_t(vr) > UNITY_GAIN)) { // volume is boosted, so we might need to clamp even though // we process only one track. do { uint32_t rl = *reinterpret_cast<uint32_t const *>(in); in += 2; int32_t l = mulRL(1, rl, vrl) >> 12; int32_t r = mulRL(0, rl, vrl) >> 12; // clamping... l = clamp16(l); r = clamp16(r); *out++ = (r<<16) | (l & 0xFFFF); } while (--outFrames); } else { do { uint32_t rl = *reinterpret_cast<uint32_t const *>(in); in += 2; int32_t l = mulRL(1, rl, vrl) >> 12; int32_t r = mulRL(0, rl, vrl) >> 12; *out++ = (r<<16) | (l & 0xFFFF); } while (--outFrames); } numFrames -= b.frameCount; t.bufferProvider->releaseBuffer(&b); }}// ----------------------------------------------------------------/*可见，函数的主要功能是将数据从audio_track_cblk_t中copy到track_t的mainBuffer中。track_t是个什么东东呢？是state_t的一个成员。state_t又是个什么东东呢？是传入的函数参数。现在有两个问题需要探讨：1、process__OneTrack16BitsStereoNoResampling是如何被调用的？以及传入的参数是从哪儿来的？2、放入track_t的mainBuffer中的数据，是如何被使用的？*/// ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++// 先看看process__OneTrack16BitsStereoNoResampling是怎么被调用的。// 函数AudioMixer::process__validate中有使用process__OneTrack16BitsStereoNoResampling函数。// ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++void AudioMixer::process__validate(state_t* state){ LOGW_IF(!state->needsChanged, "in process__validate() but nothing's invalid"); uint32_t changed = state->needsChanged; state->needsChanged = 0; // clear the validation flag // recompute which tracks are enabled / disabled uint32_t enabled = 0; uint32_t disabled = 0; while (changed) { const int i = 31 - __builtin_clz(changed); const uint32_t mask = 1<<i; changed &= ~mask; track_t& t = state->tracks[i]; (t.enabled ? enabled : disabled) |= mask; } state->enabledTracks &= ~disabled; state->enabledTracks |= enabled; // compute everything we need... int countActiveTracks = 0; int all16BitsStereoNoResample = 1; int resampling = 0; int volumeRamp = 0; uint32_t en = state->enabledTracks; while (en) { const int i = 31 - __builtin_clz(en); en &= ~(1<<i); countActiveTracks++; track_t& t = state->tracks[i]; uint32_t n = 0; n |= NEEDS_CHANNEL_1 + t.channelCount - 1; n |= NEEDS_FORMAT_16; n |= t.doesResample() ? NEEDS_RESAMPLE_ENABLED : NEEDS_RESAMPLE_DISABLED; if (t.auxLevel != 0 && t.auxBuffer != NULL) { n |= NEEDS_AUX_ENABLED; } if (t.volumeInc[0]|t.volumeInc[1]) { volumeRamp = 1; } else if (!t.doesResample() && t.volumeRL == 0) { n |= NEEDS_MUTE_ENABLED; } t.needs = n; if ((n & NEEDS_MUTE__MASK) == NEEDS_MUTE_ENABLED) { t.hook = track__nop; } else { if ((n & NEEDS_AUX__MASK) == NEEDS_AUX_ENABLED) { all16BitsStereoNoResample = 0; } if ((n & NEEDS_RESAMPLE__MASK) == NEEDS_RESAMPLE_ENABLED) { all16BitsStereoNoResample = 0; resampling = 1; t.hook = track__genericResample; } else { if ((n & NEEDS_CHANNEL_COUNT__MASK) == NEEDS_CHANNEL_1){ t.hook = track__16BitsMono; all16BitsStereoNoResample = 0; } if ((n & NEEDS_CHANNEL_COUNT__MASK) == NEEDS_CHANNEL_2){ t.hook = track__16BitsStereo; } } } } // select the processing hooks state->hook = process__nop; if (countActiveTracks) { if (resampling) { if (!state->outputTemp) { state->outputTemp = new int32_t[MAX_NUM_CHANNELS * state->frameCount]; } if (!state->resampleTemp) { state->resampleTemp = new int32_t[MAX_NUM_CHANNELS * state->frameCount]; } state->hook = process__genericResampling; } else { if (state->outputTemp) { delete [] state->outputTemp; state->outputTemp = 0; } if (state->resampleTemp) { delete [] state->resampleTemp; state->resampleTemp = 0; } state->hook = process__genericNoResampling; if (all16BitsStereoNoResample && !volumeRamp) { if (countActiveTracks == 1) { state->hook = process__OneTrack16BitsStereoNoResampling; } } } } LOGV("mixer configuration change: %d activeTracks (%08x) " "all16BitsStereoNoResample=%d, resampling=%d, volumeRamp=%d", countActiveTracks, state->enabledTracks, all16BitsStereoNoResample, resampling, volumeRamp); state->hook(state); // Now that the volume ramp has been done, set optimal state and // track hooks for subsequent mixer process if (countActiveTracks) { int allMuted = 1; uint32_t en = state->enabledTracks; while (en) { const int i = 31 - __builtin_clz(en); en &= ~(1<<i); track_t& t = state->tracks[i]; if (!t.doesResample() && t.volumeRL == 0) { t.needs |= NEEDS_MUTE_ENABLED; t.hook = track__nop; } else { allMuted = 0; } } if (allMuted) { state->hook = process__nop; } else if (all16BitsStereoNoResample) { if (countActiveTracks == 1) { state->hook = process__OneTrack16BitsStereoNoResampling; } } }}// ----------------------------------------------------------------/* 把其赋值给了state的hook。state是传入的参数。hook是函数指针。又引出来两个问题：1、既然hook是函数指针，那么它何时被调用的？2、函数process__validate有是何时被调用的？第一个问题比较简单，函数process__validate中调用了hook。来看看哪儿有使用函数process__validate。函数AudioMixer::invalidateState中将process__validate赋值给了mState的hook。mState是AudioMixer的成员变量。hook是个函数指针。*/// ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ void AudioMixer::invalidateState(uint32_t mask) { if (mask) { mState.needsChanged |= mask; mState.hook = process__validate; } }// ----------------------------------------------------------------/*与上面类似，又出来两个问题：1、mState的hook被调用的地方。2、函数invalidateState被调用的地方。*/// ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++/* 先看第一个问题。函数AudioMixer::process中调用了mState的hook。void AudioMixer::process(){ mState.hook(&mState);}可见，函数process__OneTrack16BitsStereoNoResampling的参数state就是此处的mState：mState.hook(&mState);mState.hook = process__validate;state->hook = process__OneTrack16BitsStereoNoResampling;state->hook(state);*/// ----------------------------------------------------------------// ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++/* 再看第二个问题：函数invalidateState被调用的地方。函数invalidateState被调用的地方还不是，不过，我们只关心以下两个函数：AudioMixer::enableAudioMixer::setParameter这两个函数中函数AudioFlinger::MixerThread::prepareTracks_l中都有没调到。而函数AudioFlinger::MixerThread::prepareTracks_l在AudioFlinger::MixerThread::threadLoop中被调到。*/// 一个一个来。// ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++status_t AudioMixer::enable(int name){ switch (name) { case MIXING: { if (mState.tracks[ mActiveTrack ].enabled != 1) { mState.tracks[ mActiveTrack ].enabled = 1; LOGV("enable(%d)", mActiveTrack); invalidateState(1<<mActiveTrack); } } break; default: return NAME_NOT_FOUND; } return NO_ERROR;}// ----------------------------------------------------------------// ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++status_t AudioMixer::setParameter(int target, int name, void *value){ int valueInt = (int)value; int32_t *valueBuf = (int32_t *)value; switch (target) { case TRACK: if (name == CHANNEL_COUNT) { if ((uint32_t(valueInt) <= MAX_NUM_CHANNELS) && (valueInt)) { if (mState.tracks[ mActiveTrack ].channelCount != valueInt) { mState.tracks[ mActiveTrack ].channelCount = valueInt; LOGV("setParameter(TRACK, CHANNEL_COUNT, %d)", valueInt); invalidateState(1<<mActiveTrack); } return NO_ERROR; } } if (name == MAIN_BUFFER) { if (mState.tracks[ mActiveTrack ].mainBuffer != valueBuf) { mState.tracks[ mActiveTrack ].mainBuffer = valueBuf; LOGV("setParameter(TRACK, MAIN_BUFFER, %p)", valueBuf); invalidateState(1<<mActiveTrack); } return NO_ERROR; } if (name == AUX_BUFFER) { if (mState.tracks[ mActiveTrack ].auxBuffer != valueBuf) { mState.tracks[ mActiveTrack ].auxBuffer = valueBuf; LOGV("setParameter(TRACK, AUX_BUFFER, %p)", valueBuf); invalidateState(1<<mActiveTrack); } return NO_ERROR; } break; case RESAMPLE: if (name == SAMPLE_RATE) { if (valueInt > 0) { track_t& track = mState.tracks[ mActiveTrack ]; if (track.setResampler(uint32_t(valueInt), mSampleRate)) { LOGV("setParameter(RESAMPLE, SAMPLE_RATE, %u)", uint32_t(valueInt)); invalidateState(1<<mActiveTrack); } return NO_ERROR; } } break; case RAMP_VOLUME: case VOLUME: if ((uint32_t(name-VOLUME0) < MAX_NUM_CHANNELS)) { track_t& track = mState.tracks[ mActiveTrack ]; if (track.volume[name-VOLUME0] != valueInt) { LOGV("setParameter(VOLUME, VOLUME0/1: %04x)", valueInt); track.prevVolume[name-VOLUME0] = track.volume[name-VOLUME0] << 16; track.volume[name-VOLUME0] = valueInt; if (target == VOLUME) { track.prevVolume[name-VOLUME0] = valueInt << 16; track.volumeInc[name-VOLUME0] = 0; } else { int32_t d = (valueInt<<16) - track.prevVolume[name-VOLUME0]; int32_t volInc = d / int32_t(mState.frameCount); track.volumeInc[name-VOLUME0] = volInc; if (volInc == 0) { track.prevVolume[name-VOLUME0] = valueInt << 16; } } invalidateState(1<<mActiveTrack); } return NO_ERROR; } else if (name == AUXLEVEL) { track_t& track = mState.tracks[ mActiveTrack ]; if (track.auxLevel != valueInt) { LOGV("setParameter(VOLUME, AUXLEVEL: %04x)", valueInt); track.prevAuxLevel = track.auxLevel << 16; track.auxLevel = valueInt; if (target == VOLUME) { track.prevAuxLevel = valueInt << 16; track.auxInc = 0; } else { int32_t d = (valueInt<<16) - track.prevAuxLevel; int32_t volInc = d / int32_t(mState.frameCount); track.auxInc = volInc; if (volInc == 0) { track.prevAuxLevel = valueInt << 16; } } invalidateState(1<<mActiveTrack); } return NO_ERROR; } break; } return BAD_VALUE;}// ----------------------------------------------------------------// ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++// prepareTracks_l() must be called with ThreadBase::mLock helduint32_t AudioFlinger::MixerThread::prepareTracks_l(const SortedVector< wp<Track> >& activeTracks, Vector< sp<Track> > *tracksToRemove){ uint32_t mixerStatus = MIXER_IDLE; // find out which tracks need to be processed size_t count = activeTracks.size(); size_t mixedTracks = 0; size_t tracksWithEffect = 0; float masterVolume = mMasterVolume; bool masterMute = mMasterMute; if (masterMute) { masterVolume = 0; }#ifdef LVMX bool tracksConnectedChanged = false; bool stateChanged = false; int audioOutputType = LifeVibes::getMixerType(mId, mType); if (LifeVibes::audioOutputTypeIsLifeVibes(audioOutputType)) { int activeTypes = 0; for (size_t i=0 ; i<count ; i++) { sp<Track> t = activeTracks[i].promote(); if (t == 0) continue; Track* const track = t.get(); int iTracktype=track->type(); activeTypes |= 1<<track->type(); } LifeVibes::computeVolumes(audioOutputType, activeTypes, tracksConnectedChanged, stateChanged, masterVolume, masterMute); }#endif // Delegate master volume control to effect in output mix effect chain if needed sp<EffectChain> chain = getEffectChain_l(AudioSystem::SESSION_OUTPUT_MIX); if (chain != 0) { uint32_t v = (uint32_t)(masterVolume * (1 << 24)); chain->setVolume_l(&v, &v); masterVolume = (float)((v + (1 << 23)) >> 24); chain.clear(); } for (size_t i=0 ; i<count ; i++) { sp<Track> t = activeTracks[i].promote(); if (t == 0) continue; Track* const track = t.get(); audio_track_cblk_t* cblk = track->cblk(); // The first time a track is added we wait // for all its buffers to be filled before processing it mAudioMixer->setActiveTrack(track->name());// ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++status_t AudioMixer::setActiveTrack(int track){ if (uint32_t(track-TRACK0) >= MAX_NUM_TRACKS) { return BAD_VALUE; } mActiveTrack = track - TRACK0; return NO_ERROR;}// ---------------------------------------------------------------- if (cblk->framesReady() && track->isReady() && !track->isPaused() && !track->isTerminated()) { //LOGV("track %d u=%08x, s=%08x [OK] on thread %p", track->name(), cblk->user, cblk->server, this); mixedTracks++; // track->mainBuffer() != mMixBuffer means there is an effect chain // connected to the track chain.clear(); if (track->mainBuffer() != mMixBuffer) { chain = getEffectChain_l(track->sessionId()); // Delegate volume control to effect in track effect chain if needed if (chain != 0) { tracksWithEffect++; } else { LOGW("prepareTracks_l(): track %08x attached to effect but no chain found on session %d", track->name(), track->sessionId()); } } int param = AudioMixer::VOLUME; if (track->mFillingUpStatus == Track::FS_FILLED) { // no ramp for the first volume setting track->mFillingUpStatus = Track::FS_ACTIVE; if (track->mState == TrackBase::RESUMING) { track->mState = TrackBase::ACTIVE; param = AudioMixer::RAMP_VOLUME; } } else if (cblk->server != 0) { // If the track is stopped before the first frame was mixed, // do not apply ramp param = AudioMixer::RAMP_VOLUME; } // compute volume for this track uint32_t vl, vr, va; if (track->isMuted() || track->isPausing() || mStreamTypes[track->type()].mute) { vl = vr = va = 0; if (track->isPausing()) { track->setPaused(); } } else { // read original volumes with volume control float typeVolume = mStreamTypes[track->type()].volume;#ifdef LVMX bool streamMute=false; // read the volume from the LivesVibes audio engine. if (LifeVibes::audioOutputTypeIsLifeVibes(audioOutputType)) { LifeVibes::getStreamVolumes(audioOutputType, track->type(), &typeVolume, &streamMute); if (streamMute) { typeVolume = 0; } }#endif float v = masterVolume * typeVolume; vl = (uint32_t)(v * cblk->volume[0]) << 12; vr = (uint32_t)(v * cblk->volume[1]) << 12; va = (uint32_t)(v * cblk->sendLevel); } // Delegate volume control to effect in track effect chain if needed if (chain != 0 && chain->setVolume_l(&vl, &vr)) { // Do not ramp volume if volume is controlled by effect param = AudioMixer::VOLUME; track->mHasVolumeController = true; } else { // force no volume ramp when volume controller was just disabled or removed // from effect chain to avoid volume spike if (track->mHasVolumeController) { param = AudioMixer::VOLUME; } track->mHasVolumeController = false; } // Convert volumes from 8.24 to 4.12 format int16_t left, right, aux; uint32_t v_clamped = (vl + (1 << 11)) >> 12; if (v_clamped > MAX_GAIN_INT) v_clamped = MAX_GAIN_INT; left = int16_t(v_clamped); v_clamped = (vr + (1 << 11)) >> 12; if (v_clamped > MAX_GAIN_INT) v_clamped = MAX_GAIN_INT; right = int16_t(v_clamped); if (va > MAX_GAIN_INT) va = MAX_GAIN_INT; aux = int16_t(va);#ifdef LVMX if ( tracksConnectedChanged || stateChanged ) { // only do the ramp when the volume is changed by the user / application param = AudioMixer::VOLUME; }#endif // XXX: these things DON'T need to be done each time// 前面有说到，调用getNextBuffer函数的代码为：t.bufferProvider->getNextBuffer(&b);// 猜到没错的话，此处应该就是给bufferProvider赋值的地方 mAudioMixer->setBufferProvider(track);// ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++status_t AudioMixer::setBufferProvider(AudioBufferProvider* buffer){ mState.tracks[ mActiveTrack ].bufferProvider = buffer; return NO_ERROR;}// ----------------------------------------------------------------// 这个就是前面说的AudioMixer::enable函数 mAudioMixer->enable(AudioMixer::MIXING);// 这个是前面说的AudioMixer::setParameter函数 mAudioMixer->setParameter(param, AudioMixer::VOLUME0, (void *)left); mAudioMixer->setParameter(param, AudioMixer::VOLUME1, (void *)right); mAudioMixer->setParameter(param, AudioMixer::AUXLEVEL, (void *)aux); mAudioMixer->setParameter( AudioMixer::TRACK, AudioMixer::FORMAT, (void *)track->format()); mAudioMixer->setParameter( AudioMixer::TRACK, AudioMixer::CHANNEL_COUNT, (void *)track->channelCount()); mAudioMixer->setParameter( AudioMixer::RESAMPLE, AudioMixer::SAMPLE_RATE, (void *)(cblk->sampleRate)); mAudioMixer->setParameter( AudioMixer::TRACK, AudioMixer::MAIN_BUFFER, (void *)track->mainBuffer()); mAudioMixer->setParameter( AudioMixer::TRACK, AudioMixer::AUX_BUFFER, (void *)track->auxBuffer()); // reset retry count track->mRetryCount = kMaxTrackRetries; mixerStatus = MIXER_TRACKS_READY; } else { //LOGV("track %d u=%08x, s=%08x [NOT READY] on thread %p", track->name(), cblk->user, cblk->server, this); if (track->isStopped()) { track->reset(); } if (track->isTerminated() || track->isStopped() || track->isPaused()) { // We have consumed all the buffers of this track. // Remove it from the list of active tracks. tracksToRemove->add(track); } else { // No buffers for this track. Give it a few chances to // fill a buffer, then remove it from active list.// 如果track中没有数据，给它个机会往里填数据。// 在mRetryCount次之后还没数据，就把它从active list中咔嚓掉。 if (--(track->mRetryCount) <= 0) { LOGV("BUFFER TIMEOUT: remove(%d) from active list on thread %p", track->name(), this); tracksToRemove->add(track); // indicate to client process that the track was disabled because of underrun cblk->flags |= CBLK_DISABLED_ON; } else if (mixerStatus != MIXER_TRACKS_READY) { mixerStatus = MIXER_TRACKS_ENABLED; } } mAudioMixer->disable(AudioMixer::MIXING);// ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++status_t AudioMixer::disable(int name){ switch (name) { case MIXING: { if (mState.tracks[ mActiveTrack ].enabled != 0) { mState.tracks[ mActiveTrack ].enabled = 0; LOGV("disable(%d)", mActiveTrack); invalidateState(1<<mActiveTrack); } } break; default: return NAME_NOT_FOUND; } return NO_ERROR;}// 此处也有对函数disable的调用。// ---------------------------------------------------------------- } }// 干掉那些需要被干掉的track // remove all the tracks that need to be... count = tracksToRemove->size(); if (UNLIKELY(count)) { for (size_t i=0 ; i<count ; i++) { const sp<Track>& track = tracksToRemove->itemAt(i); mActiveTracks.remove(track); if (track->mainBuffer() != mMixBuffer) { chain = getEffectChain_l(track->sessionId()); if (chain != 0) { LOGV("stopping track on chain %p for session Id: %d", chain.get(), track->sessionId()); chain->stopTrack(); } } if (track->isTerminated()) { mTracks.remove(track); deleteTrackName_l(track->mName); } } } // mix buffer must be cleared if all tracks are connected to an // effect chain as in this case the mixer will not write to // mix buffer and track effects will accumulate into it if (mixedTracks != 0 && mixedTracks == tracksWithEffect) { memset(mMixBuffer, 0, mFrameCount * mChannelCount * sizeof(int16_t)); } return mixerStatus;}// ----------------------------------------------------------------// ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++bool AudioFlinger::MixerThread::threadLoop(){ Vector< sp<Track> > tracksToRemove; uint32_t mixerStatus = MIXER_IDLE; nsecs_t standbyTime = systemTime(); size_t mixBufferSize = mFrameCount * mFrameSize; // FIXME: Relaxed timing because of a certain device that can't meet latency // Should be reduced to 2x after the vendor fixes the driver issue nsecs_t maxPeriod = seconds(mFrameCount) / mSampleRate * 3; nsecs_t lastWarning = 0; bool longStandbyExit = false; uint32_t activeSleepTime = activeSleepTimeUs(); uint32_t idleSleepTime = idleSleepTimeUs(); uint32_t sleepTime = idleSleepTime; Vector< sp<EffectChain> > effectChains; while (!exitPending()) { processConfigEvents();// ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++void AudioFlinger::ThreadBase::processConfigEvents(){ mLock.lock(); while(!mConfigEvents.isEmpty()) { LOGV("processConfigEvents() remaining events %d", mConfigEvents.size()); ConfigEvent *configEvent = mConfigEvents[0]; mConfigEvents.removeAt(0); // release mLock before locking AudioFlinger mLock: lock order is always // AudioFlinger then ThreadBase to avoid cross deadlock mLock.unlock(); mAudioFlinger->mLock.lock(); audioConfigChanged_l(configEvent->mEvent, configEvent->mParam);// ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++// audioConfigChanged_l() must be called with AudioFlinger::mLock heldvoid AudioFlinger::audioConfigChanged_l(int event, int ioHandle, void *param2){ size_t size = mNotificationClients.size(); for (size_t i = 0; i < size; i++) { mNotificationClients.valueAt(i)->client()->ioConfigChanged(event, ioHandle, param2); }}// ---------------------------------------------------------------- mAudioFlinger->mLock.unlock(); delete configEvent; mLock.lock(); } mLock.unlock();}// ---------------------------------------------------------------- mixerStatus = MIXER_IDLE; { // scope for mLock Mutex::Autolock _l(mLock); if (checkForNewParameters_l()) { mixBufferSize = mFrameCount * mFrameSize; // FIXME: Relaxed timing because of a certain device that can't meet latency // Should be reduced to 2x after the vendor fixes the driver issue maxPeriod = seconds(mFrameCount) / mSampleRate * 3; activeSleepTime = activeSleepTimeUs(); idleSleepTime = idleSleepTimeUs(); } const SortedVector< wp<Track> >& activeTracks = mActiveTracks; // put audio hardware into standby after short delay if UNLIKELY((!activeTracks.size() && systemTime() > standbyTime) || mSuspended) { if (!mStandby) { LOGV("Audio hardware entering standby, mixer %p, mSuspended %d\n", this, mSuspended); mOutput->standby(); mStandby = true; mBytesWritten = 0; } if (!activeTracks.size() && mConfigEvents.isEmpty()) { // we're about to wait, flush the binder command buffer IPCThreadState::self()->flushCommands(); if (exitPending()) break; // wait until we have something to do... LOGV("MixerThread %p TID %d going to sleep\n", this, gettid()); mWaitWorkCV.wait(mLock); LOGV("MixerThread %p TID %d waking up\n", this, gettid()); if (mMasterMute == false) { char value[PROPERTY_VALUE_MAX]; property_get("ro.audio.silent", value, "0"); if (atoi(value)) { LOGD("Silence is golden"); setMasterMute(true); } } standbyTime = systemTime() + kStandbyTimeInNsecs; sleepTime = idleSleepTime; continue; } }// 此处调用了函数prepareTracks_l mixerStatus = prepareTracks_l(activeTracks, &tracksToRemove); // prevent any changes in effect chain list and in each effect chain // during mixing and effect process as the audio buffers could be deleted // or modified if an effect is created or deleted lockEffectChains_l(effectChains); } if (LIKELY(mixerStatus == MIXER_TRACKS_READY)) {// 这儿调用了前面介绍的AudioMixer::process函数 // mix buffers... mAudioMixer->process(); sleepTime = 0; standbyTime = systemTime() + kStandbyTimeInNsecs; //TODO: delay standby when effects have a tail } else { // If no tracks are ready, sleep once for the duration of an output // buffer size, then write 0s to the output if (sleepTime == 0) { if (mixerStatus == MIXER_TRACKS_ENABLED) { sleepTime = activeSleepTime; } else { sleepTime = idleSleepTime; } } else if (mBytesWritten != 0 || (mixerStatus == MIXER_TRACKS_ENABLED && longStandbyExit)) { memset (mMixBuffer, 0, mixBufferSize); sleepTime = 0; LOGV_IF((mBytesWritten == 0 && (mixerStatus == MIXER_TRACKS_ENABLED && longStandbyExit)), "anticipated start"); } // TODO add standby time extension fct of effect tail } if (mSuspended) { sleepTime = suspendSleepTimeUs(); }// 往硬件中写数据 // sleepTime == 0 means we must write to audio hardware if (sleepTime == 0) { for (size_t i = 0; i < effectChains.size(); i ++) { effectChains[i]->process_l(); } // enable changes in effect chain unlockEffectChains(effectChains);#ifdef LVMX int audioOutputType = LifeVibes::getMixerType(mId, mType); if (LifeVibes::audioOutputTypeIsLifeVibes(audioOutputType)) { LifeVibes::process(audioOutputType, mMixBuffer, mixBufferSize); }#endif mLastWriteTime = systemTime(); mInWrite = true; mBytesWritten += mixBufferSize;// 是这儿写的。 int bytesWritten = (int)mOutput->write(mMixBuffer, mixBufferSize); if (bytesWritten < 0) mBytesWritten -= mixBufferSize; mNumWrites++; mInWrite = false; nsecs_t now = systemTime(); nsecs_t delta = now - mLastWriteTime; if (delta > maxPeriod) { mNumDelayedWrites++; if ((now - lastWarning) > kWarningThrottle) { LOGW("write blocked for %llu msecs, %d delayed writes, thread %p", ns2ms(delta), mNumDelayedWrites, this); lastWarning = now; } if (mStandby) { longStandbyExit = true; } } mStandby = false; } else { // enable changes in effect chain unlockEffectChains(effectChains); usleep(sleepTime); } // finally let go of all our tracks, without the lock held // since we can't guarantee the destructors won't acquire that // same lock. tracksToRemove.clear(); // Effect chains will be actually deleted here if they were removed from // mEffectChains list during mixing or effects processing effectChains.clear(); } if (!mStandby) { mOutput->standby(); } LOGV("MixerThread %p exiting", this); return false;}// ----------------------------------------------------------------// ----------------------------------------------------------------// ----------------------------------------------------------------/*在前面，看函数AudioMixer::process__OneTrack16BitsStereoNoResampling的时候，留下了一个问题：放入track_t的mainBuffer中的数据，是如何被使用的？函数函数AudioMixer::process__OneTrack16BitsStereoNoResampling的主要功能就是将数据从audio_track_cblk_t中copy到track_t的mainBuffer中。先看看mainBuffer的来历。*/// ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++// mainBuffer是track_t的成员。// 函数AudioMixer::setParameter中有对mainBuffer赋值。// ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ if (name == MAIN_BUFFER) { if (mState.tracks[ mActiveTrack ].mainBuffer != valueBuf) { mState.tracks[ mActiveTrack ].mainBuffer = valueBuf; LOGV("setParameter(TRACK, MAIN_BUFFER, %p)", valueBuf); invalidateState(1<<mActiveTrack); } return NO_ERROR; }// ----------------------------------------------------------------// 函数AudioFlinger::MixerThread::prepareTracks_l中调用了函数AudioMixer::setParameter。// ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ mAudioMixer->setParameter( AudioMixer::TRACK, AudioMixer::MAIN_BUFFER, (void *)track->mainBuffer());// 这个 track之前也被使用过：// mAudioMixer->setBufferProvider(track);// track的来历：// Track* const track = t.get();// t的来历：// sp<Track> t = activeTracks[i].promote();// activeTracks是函数AudioFlinger::MixerThread::prepareTracks_l的参数。// ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++// 函数函数AudioFlinger::MixerThread::prepareTracks_l在函数AudioFlinger::MixerThread::threadLoop中被调用：mixerStatus = prepareTracks_l(activeTracks, &tracksToRemove);/*activeTracks的来历：const SortedVector< wp<Track> >& activeTracks = mActiveTracks;mActiveTracks是PlaybackThread的成员变量。函数AudioFlinger::PlaybackThread::addTrack_l，往mActiveTracks中添加成员。*/// ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++// addTrack_l() must be called with ThreadBase::mLock heldstatus_t AudioFlinger::PlaybackThread::addTrack_l(const sp<Track>& track){ status_t status = ALREADY_EXISTS; // set retry count for buffer fill track->mRetryCount = kMaxTrackStartupRetries; if (mActiveTracks.indexOf(track) < 0) { // the track is newly added, make sure it fills up all its // buffers before playing. This is to ensure the client will // effectively get the latency it requested. track->mFillingUpStatus = Track::FS_FILLING; track->mResetDone = false; mActiveTracks.add(track); if (track->mainBuffer() != mMixBuffer) { sp<EffectChain> chain = getEffectChain_l(track->sessionId()); if (chain != 0) { LOGV("addTrack_l() starting track on chain %p for session %d", chain.get(), track->sessionId()); chain->startTrack(); } } status = NO_ERROR; } LOGV("mWaitWorkCV.broadcast"); mWaitWorkCV.broadcast(); return status;}// ----------------------------------------------------------------// 函数AudioFlinger::PlaybackThread::Track::start调用了函数AudioFlinger::PlaybackThread::addTrack_l。// ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++status_t AudioFlinger::PlaybackThread::Track::start(){ status_t status = NO_ERROR; LOGV("start(%d), calling thread %d session %d", mName, IPCThreadState::self()->getCallingPid(), mSessionId); sp<ThreadBase> thread = mThread.promote(); if (thread != 0) { Mutex::Autolock _l(thread->mLock); int state = mState; // here the track could be either new, or restarted // in both cases "unstop" the track if (mState == PAUSED) { mState = TrackBase::RESUMING; LOGV("PAUSED => RESUMING (%d) on thread %p", mName, this); } else { mState = TrackBase::ACTIVE; LOGV("? => ACTIVE (%d) on thread %p", mName, this); } if (!isOutputTrack() && state != ACTIVE && state != RESUMING) { thread->mLock.unlock(); status = AudioSystem::startOutput(thread->id(), (AudioSystem::stream_type)mStreamType, mSessionId); thread->mLock.lock(); } if (status == NO_ERROR) { PlaybackThread *playbackThread = (PlaybackThread *)thread.get(); playbackThread->addTrack_l(this); } else { mState = state; } } else { status = BAD_VALUE; } return status;}// ----------------------------------------------------------------函数AudioFlinger::TrackHandle::start调用了函数AudioFlinger::PlaybackThread::Track::start。// ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++status_t AudioFlinger::TrackHandle::start() { return mTrack->start();}// ----------------------------------------------------------------// 再往上拱不是很好拱，还是跳到上面往下钻吧。/*从java的测试代码中可知，调用完write写完数据后，一般会调用play函数。如在函数testSetStereoVolumeMid中： track.write(data, 0, data.length); track.play();*/// play函数的实现：// ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ /** * Starts playing an AudioTrack. * @throws IllegalStateException */ public void play() throws IllegalStateException { if (mState != STATE_INITIALIZED) { throw(new IllegalStateException("play() called on uninitialized AudioTrack.")); } synchronized(mPlayStateLock) {// 调到了native中的start函数。 native_start();// ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++static voidandroid_media_AudioTrack_start(JNIEnv *env, jobject thiz){// 此处的AudioTrack对象是创建AudioTrack的时候保存过去的 AudioTrack *lpTrack = (AudioTrack *)env->GetIntField( thiz, javaAudioTrackFields.nativeTrackInJavaObj); if (lpTrack == NULL ) { jniThrowException(env, "java/lang/IllegalStateException", "Unable to retrieve AudioTrack pointer for start()"); return; } lpTrack->start();// ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++void AudioTrack::start(){ sp<AudioTrackThread> t = mAudioTrackThread; status_t status; LOGV("start %p", this); if (t != 0) { if (t->exitPending()) { if (t->requestExitAndWait() == WOULD_BLOCK) { LOGE("AudioTrack::start called from thread"); return; } } t->mLock.lock(); } if (android_atomic_or(1, &mActive) == 0) { mNewPosition = mCblk->server + mUpdatePeriod; mCblk->bufferTimeoutMs = MAX_STARTUP_TIMEOUT_MS; mCblk->waitTimeMs = 0; mCblk->flags &= ~CBLK_DISABLED_ON; if (t != 0) { t->run("AudioTrackThread", THREAD_PRIORITY_AUDIO_CLIENT); } else { setpriority(PRIO_PROCESS, 0, THREAD_PRIORITY_AUDIO_CLIENT); } if (mCblk->flags & CBLK_INVALID_MSK) { LOGW("start() track %p invalidated, creating a new one", this); // no need to clear the invalid flag as this cblk will not be used anymore // force new track creation status = DEAD_OBJECT; } else {// 此处的mAudioTrack是在函数AudioTrack::createTrack中创建并赋值的// sp<IAudioTrack> track = audioFlinger->createTrack()// mAudioTrack = track;// 函数AudioFlinger::createTrack返回的是一个TrackHandle对象// trackHandle = new TrackHandle(track);// return trackHandle;// 也就是说，此处调用的是函数AudioFlinger::TrackHandle::start status = mAudioTrack->start();// ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++status_t AudioFlinger::TrackHandle::start() {// TrackHandle的构造函数中传入了一个track// 该track的创建：track = thread->createTrack_l()// 函数AudioFlinger::PlaybackThread::createTrack_l返回的是一个AudioFlinger::PlaybackThread::Track对象// 因此，此处调用的是函数AudioFlinger::PlaybackThread::Track::start// ！！！！接上头了// 也就是说，函数AudioFlinger::MixerThread::prepareTracks_l调用函数AudioMixer::setParameter时使用的track对象，其实就是我们创建AudioTrack时创建的那个东东// 也就是说函数函数AudioMixer::process__OneTrack16BitsStereoNoResampling，其实是把数据copy到了AudioTrack对象的mian buffer中 return mTrack->start();}---------------------------------------------------------------- } if (status == DEAD_OBJECT) { LOGV("start() dead IAudioTrack: creating a new one"); status = createTrack(mStreamType, mCblk->sampleRate, mFormat, mChannelCount, mFrameCount, mFlags, mSharedBuffer, getOutput(), false); if (status == NO_ERROR) { status = mAudioTrack->start(); if (status == NO_ERROR) { mNewPosition = mCblk->server + mUpdatePeriod; } } } if (status != NO_ERROR) { LOGV("start() failed"); android_atomic_and(~1, &mActive); if (t != 0) { t->requestExit(); } else { setpriority(PRIO_PROCESS, 0, ANDROID_PRIORITY_NORMAL); } } } if (t != 0) { t->mLock.unlock(); }}// ----------------------------------------------------------------}// ---------------------------------------------------------------- mPlayState = PLAYSTATE_PLAYING; } }// ----------------------------------------------------------------// ----------------------------------------------------------------// ----------------------------------------------------------------// ----------------------------------------------------------------/*还有一个遗留问题，函数AudioFlinger::MixerThread::threadLoop往硬件写数据使用的是以下代码：int bytesWritten = (int)mOutput->write(mMixBuffer, mixBufferSize);mMixBuffer是MixerThread的成员变量。而函数函数AudioMixer::process__OneTrack16BitsStereoNoResampling是把数据copy到AudioTrack对象的mainBuffer中。这两个东东之间是如何联系起来的呢？下面的这个链条，貌似将它们关联了起来：1、函数AudioFlinger::createTrack在调用thread->createTrack_l成功之后，会根据条件调用函数AudioFlinger::moveEffectChain_l。*/// ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ track = thread->createTrack_l(client, streamType, sampleRate, format, channelCount, frameCount, sharedBuffer, lSessionId, &lStatus); // move effect chain to this output thread if an effect on same session was waiting // for a track to be created if (lStatus == NO_ERROR && effectThread != NULL) { Mutex::Autolock _dl(thread->mLock); Mutex::Autolock _sl(effectThread->mLock); moveEffectChain_l(lSessionId, effectThread, thread, true); }// ----------------------------------------------------------------// 2、函数函数AudioFlinger::moveEffectChain_l调用了函数AudioFlinger::PlaybackThread::removeEffectChain_l。// ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ // remove chain first. This is useful only if reconfiguring effect chain on same output thread, // so that a new chain is created with correct parameters when first effect is added. This is // otherwise unecessary as removeEffect_l() will remove the chain when last effect is // removed. srcThread->removeEffectChain_l(chain);// ----------------------------------------------------------------// 3、函数AudioFlinger::PlaybackThread::removeEffectChain_l将PlaybackThread的mMixBuffer设置到了AudioTrack对象的mian buffer。// ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++size_t AudioFlinger::PlaybackThread::removeEffectChain_l(const sp<EffectChain>& chain){ int session = chain->sessionId(); LOGV("removeEffectChain_l() %p from thread %p for session %d", chain.get(), this, session); for (size_t i = 0; i < mEffectChains.size(); i++) { if (chain == mEffectChains[i]) { mEffectChains.removeAt(i); // detach all tracks with same session ID from this chain for (size_t i = 0; i < mTracks.size(); ++i) { sp<Track> track = mTracks[i]; if (session == track->sessionId()) { track->setMainBuffer(mMixBuffer); } } break; } } return mEffectChains.size();}// ----------------------------------------------------------------/*函数AudioFlinger::openOutput中在调用HAL层的openOutputStream成功打开一个output之后会创建一个MixerThread对象。并将其添加到成员变量mPlaybackThreads中： thread = new MixerThread(this, output, id, *pDevices); mPlaybackThreads.add(id, thread);*/

###########################################################


&&&&&&&&&&&&&&&&&&&&&&&总结&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&
MixerThread的threadloop函数，会先检查数据是否已经通过AudioTrack对象写到了audio_track_cblk_t中。
如果已经写好了，就将数据copy到AudioTrack对象的main buffer中。
在创建AudioTrack对象的时候，已经将AudioTrack对象的mian buffer和PlaybackThread的mix buffer进行了关联。
MixerThread的threadloop函数之后就会调用HAL层的write函数将数据写到硬件。
&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&

更多相关文章

1. SpringBoot 2.0 中 HikariCP 数据库连接池原理解析
2. 一句话锁定MySQL数据占用元凶
3. android 实现服务器连接获取数据和传递数据（1）
4. Bitmap recycle()
5. Android基础(八)：文件存储
6. android 之 PhoneGap+jQuery Mobile+Rest 访问远程数据
7. (转 Android(安卓)UI学习 - Menu)
8. Android(安卓)自定义 ExpandableListView
9. android graphic(3)—surfaceflinger的启动流程

随机推荐

1. android dialog 不变暗
2. Android Desigin Library
3. android设置属性
4. 【30篇突击 android】源码统计七
5. BlueStacks将Android程序运行在Windows上
6. android 逐帧动画
7. Android开发学习之Gallery和GridView浅析
8. 网络请求框架AsyncHttpclient的简单使用
9. RadioGroup+RadioButton嵌套实现多行单选
10. android TextToSpeech 源码分析