Android学习（九）AudioTrack(1)

摘要: AudioTrack

AudioTrack（1）

AudioTrack简介（简介部分转发）

在Android中播放声音可以用MediaPlayer和AudioTrack两种方案的，但是两种方案是有很大区别的，MediaPlayer可以播放多种格式的声音文件，例如MP3，AAC，WAV，OGG，MIDI等。而AudioTrack只能播放PCM数据流。
事实上，两种本质上是没啥区别的，MediaPlayer在播放音频时，在framework层还是会创建AudioTrack，把解码后的PCM数流传递给AudioTrack，最后由AudioFlinger进行混音，传递音频给硬件播放出来。利用AudioTrack播放只是跳过Mediaplayer的解码部分而已。Mediaplayer的解码核心部分是基于OpenCORE 来实现的，支持通用的音视频和图像格式，codec使用的是OpenMAX接口来进行扩展。因此使用audiotrack播放mp3文件的话，要自己加入一个音频解码器，如libmad。否则只能播放PCM数据，如大多数WAV格式的音频文件。
常规的AudioTrack播放流程如下：

mAudioMinBufSize = AudioTrack.getMinBufferSize(samplerate, AudioFormat.CHANNEL_CONFIGURATION_STEREO, AudioFormat.ENCODING_PCM_16BIT); mAudioTrack = new AudioTrack(AudioManager.STREAM_MUSIC, // 指定在流的类型 // STREAM_ALARM：警告声 // STREAM_MUSCI：音乐声，例如music等 // STREAM_RING：铃声 // STREAM_SYSTEM：系统声音 // STREAM_VOCIE_CALL：电话声音 samplerate,// 设置音频数据的采样率 AudioFormat.CHANNEL_CONFIGURATION_STEREO,// 设置输出声道为双声道立体声 AudioFormat.ENCODING_PCM_16BIT,// 设置音频数据块是8位还是16位 mAudioMinBufSize, AudioTrack.MODE_STREAM);// 设置模式类型，在这里设置为流类型， AudioTrack中有MODE_STATIC和MODE_STREAM两种分类。STREAM方式表示由用户通过write方式把数据一次一次得写到audiotrack中。这种方式的缺点就是JAVA层和Native层不断地交换数据，效率损失较大。而STATIC方式表示是一开始创建的时候，就把音频数据放到一个固定的buffer，然后直接传给audiotrack，后续就不用一次次得write了。AudioTrack会自己播放这个buffer中的数据。这种方法对于铃声等体积较小的文件比较合适。 mAudioTrack.play();//开始播放 mAudioTrack.write(audioBuffer, 0, mAudioMinBufSize);//往track中写数据 mAudioTrack.stop();//停止播放 mAudioTrack.release();// 关闭并释放资源

源码解析

首先看mAudioMinBufSize = AudioTrack.getMinBufferSize(samplerate,

AudioFormat.CHANNEL_CONFIGURATION_STEREO,

AudioFormat.ENCODING_PCM_16BIT);
源码：AudioTrack.java

static public int getMinBufferSize(int sampleRateInHz, int channelConfig, int audioFormat) { int channelCount = 0; switch(channelConfig) { case AudioFormat.CHANNEL_OUT_MONO: case AudioFormat.CHANNEL_CONFIGURATION_MONO: channelCount = 1; break; case AudioFormat.CHANNEL_OUT_STEREO: case AudioFormat.CHANNEL_CONFIGURATION_STEREO: channelCount = 2;à最多支持双声道 break; default: loge("getMinBufferSize(): Invalid channel configuration."); return AudioTrack.ERROR_BAD_VALUE; } if ((audioFormat != AudioFormat.ENCODING_PCM_16BIT) && (audioFormat != AudioFormat.ENCODING_PCM_8BIT)) { loge("getMinBufferSize(): Invalid audio format."); return AudioTrack.ERROR_BAD_VALUE;à只支持PCM8和PCM16音频数据 } if ( (sampleRateInHz < 4000) || (sampleRateInHz > 48000) ) { loge("getMinBufferSize(): " + sampleRateInHz +"Hz is not a supported sample rate.");à采用频率区间 return AudioTrack.ERROR_BAD_VALUE; } int size = native_get_min_buff_size(sampleRateInHz, channelCount, audioFormat);à调用native函数，对硬件做查询，是否支持相关参数 if ((size == -1) || (size == 0)) { loge("getMinBufferSize(): error querying hardware"); return AudioTrack.ERROR; } else { return size; } }

native_get_min_buff_size(sampleRateInHz, channelCount, audioFormat);在android_media_AudioTrack.cpp中：

static jint android_media_AudioTrack_get_min_buff_size(JNIEnv *env, jobject thiz, jint sampleRateInHertz, jint nbChannels, jint audioFormat) { int frameCount = 0; if (AudioTrack::getMinFrameCount(&frameCount, AudioSystem::DEFAULT, sampleRateInHertz) != NO_ERROR) { return -1; } return frameCount * nbChannels * (audioFormat == javaAudioTrackFields.PCM16 ? 2 : 1); }

一个单位的帧（Frame）等于1个采样点的字节数*声道数（如PCM16，双声道的一个帧=2*2=4字节），在声卡驱动中，内部缓冲区也采用Frame为单位来分配和管理。
再看AudioTrack的构造函数：

public AudioTrack(int streamType, int sampleRateInHz, int channelConfig, int audioFormat, int bufferSizeInBytes, int mode, int sessionId) throws IllegalArgumentException { mState = STATE_UNINITIALIZED; // remember which looper is associated with the AudioTrack instanciation if ((mInitializationLooper = Looper.myLooper()) == null) { mInitializationLooper = Looper.getMainLooper(); } audioParamCheck(streamType, sampleRateInHz, channelConfig, audioFormat, mode); audioBuffSizeCheck(bufferSizeInBytes); if (sessionId < 0) { throw (new IllegalArgumentException("Invalid audio session ID: "+sessionId)); } int[] session = new int[1]; session[0] = sessionId; // native initialization int initResult = native_setup(new WeakReference<AudioTrack>(this), mStreamType, mSampleRate, mChannels, mAudioFormat, mNativeBufferSizeInBytes, mDataLoadMode, session); if (initResult != SUCCESS) { loge("Error code "+initResult+" when initializing AudioTrack."); return; // with mState == STATE_UNINITIALIZED } mSessionId = session[0]; if (mDataLoadMode == MODE_STATIC) { mState = STATE_NO_STATIC_DATA; } else { mState = STATE_INITIALIZED; } }

在native层做setup：android_media_AudioTrack.cpp

static int android_media_AudioTrack_native_setup(JNIEnv *env, jobject thiz, jobject weak_this, jint streamType, jint sampleRateInHertz, jint channels, jint audioFormat, jint buffSizeInBytes, jint memoryMode, jintArray jSession) { LOGV("sampleRate=%d, audioFormat(from Java)=%d, channels=%x, buffSize=%d", sampleRateInHertz, audioFormat, channels, buffSizeInBytes); int afSampleRate; int afFrameCount; if (AudioSystem::getOutputFrameCount(&afFrameCount, streamType) != NO_ERROR) { LOGE("Error creating AudioTrack: Could not get AudioSystem frame count."); return AUDIOTRACK_ERROR_SETUP_AUDIOSYSTEM; } if (AudioSystem::getOutputSamplingRate(&afSampleRate, streamType) != NO_ERROR) { LOGE("Error creating AudioTrack: Could not get AudioSystem sampling rate."); return AUDIOTRACK_ERROR_SETUP_AUDIOSYSTEM; } if (!AudioSystem::isOutputChannel(channels)) { LOGE("Error creating AudioTrack: invalid channel mask."); return AUDIOTRACK_ERROR_SETUP_INVALIDCHANNELMASK; } int nbChannels = AudioSystem::popCount(channels);à统计channels中多少位为1 // check the stream type AudioSystem::stream_type atStreamType; if (streamType == javaAudioTrackFields.STREAM_VOICE_CALL) { atStreamType = AudioSystem::VOICE_CALL; } … } else if (streamType == javaAudioTrackFields.STREAM_MUSIC) { atStreamType = AudioSystem::MUSIC;àjava层值和jni层数值转换 … } // check the format. // This function was called from Java, so we compare the format against the Java constants if ((audioFormat != javaAudioTrackFields.PCM16) && (audioFormat != javaAudioTrackFields.PCM8)) { LOGE("Error creating AudioTrack: unsupported audio format."); return AUDIOTRACK_ERROR_SETUP_INVALIDFORMAT; } … // compute the frame count int bytesPerSample = audioFormat == javaAudioTrackFields.PCM16 ? 2 : 1; int format = audioFormat == javaAudioTrackFields.PCM16 ? AudioSystem::PCM_16_BIT : AudioSystem::PCM_8_BIT; int frameCount = buffSizeInBytes / (nbChannels * bytesPerSample); AudioTrackJniStorage* lpJniStorage = new AudioTrackJniStorage(); ……. // create the native AudioTrack object AudioTrack* lpTrack = new AudioTrack();à创建native的AudioTrack if (lpTrack == NULL) { LOGE("Error creating uninitialized AudioTrack"); goto native_track_failure; } // initialize the native AudioTrack object if (memoryMode == javaAudioTrackFields.MODE_STREAM) { lpTrack->set( atStreamType,// stream type sampleRateInHertz, format,// word length, PCM channels, frameCount, 0,// flags audioCallback, &(lpJniStorage->mCallbackData),//callback, callback data (user) 0,// notificationFrames == 0 since not using EVENT_MORE_DATA to feed the AudioTrack 0,// shared mem true,// thread can call Java sessionId);// audio session ID } else if (memoryMode == javaAudioTrackFields.MODE_STATIC) { // AudioTrack is using shared memory if (!lpJniStorage->allocSharedMem(buffSizeInBytes)) { LOGE("Error creating AudioTrack in static mode: error creating mem heap base"); goto native_init_failure; } lpTrack->set( atStreamType,// stream type sampleRateInHertz, format,// word length, PCM channels, frameCount, 0,// flags audioCallback, &(lpJniStorage->mCallbackData),//callback, callback data (user)); 0,// notificationFrames == 0 since not using EVENT_MORE_DATA to feed the AudioTrack lpJniStorage->mMemBase,// shared mem true,// thread can call Java sessionId);// audio session ID } if (lpTrack->initCheck() != NO_ERROR) { LOGE("Error initializing AudioTrack"); goto native_init_failure; } nSession = (jint *) env->GetPrimitiveArrayCritical(jSession, NULL); if (nSession == NULL) { LOGE("Error creating AudioTrack: Error retrieving session id pointer"); goto native_init_failure; } // read the audio session ID back from AudioTrack in case we create a new session nSession[0] = lpTrack->getSessionId(); env->ReleasePrimitiveArrayCritical(jSession, nSession, 0); nSession = NULL; // save our newly created C++ AudioTrack in the "nativeTrackInJavaObj" field // of the Java object (in mNativeTrackInJavaObj) env->SetIntField(thiz, javaAudioTrackFields.nativeTrackInJavaObj, (int)lpTrack); à把jni层中new出来的AudioTrack对象指针保存到java对象的一个变量中，这样就把jni层的AudioTrack对象和java层的AudioTrack对象关联起来 // save the JNI resources so we can free them later //LOGV("storing lpJniStorage: %x\n", (int)lpJniStorage); env->SetIntField(thiz, javaAudioTrackFields.jniData, (int)lpJniStorage); à将lpJniStorage对象指针也保存到java层中 return AUDIOTRACK_SUCCESS; // failures: native_init_failure: delete lpTrack; env->SetIntField(thiz, javaAudioTrackFields.nativeTrackInJavaObj, 0); native_track_failure: if (nSession != NULL) { env->ReleasePrimitiveArrayCritical(jSession, nSession, 0); } env->DeleteGlobalRef(lpJniStorage->mCallbackData.audioTrack_class); env->DeleteGlobalRef(lpJniStorage->mCallbackData.audioTrack_ref); delete lpJniStorage; env->SetIntField(thiz, javaAudioTrackFields.jniData, 0); return AUDIOTRACK_ERROR_SETUP_NATIVEINITFAILED; }

查看AudioTrackJniStorage* lpJniStorage = new AudioTrackJniStorage();

struct audiotrack_callback_cookie { jclass audioTrack_class; jobject audioTrack_ref; }; // ---------------------------------------------------------------------------- class AudioTrackJniStorage { public: sp<MemoryHeapBase> mMemHeap; sp<MemoryBase> mMemBase; audiotrack_callback_cookie mCallbackData; int mStreamType; AudioTrackJniStorage() { mCallbackData.audioTrack_class = 0; mCallbackData.audioTrack_ref = 0; mStreamType = AudioSystem::DEFAULT; } ~AudioTrackJniStorage() { mMemBase.clear(); mMemHeap.clear(); } bool allocSharedMem(int sizeInBytes) { mMemHeap = new MemoryHeapBase(sizeInBytes, 0, "AudioTrack Heap Base"); if (mMemHeap->getHeapID() < 0) { return false; } mMemBase = new MemoryBase(mMemHeap, 0, sizeInBytes); return true; } };

查看MemoryHeapBase的构造函数：

MemoryHeapBase::MemoryHeapBase(size_t size, uint32_t flags, char const * name) : mFD(-1), mSize(0), mBase(MAP_FAILED), mFlags(flags), mDevice(0), mNeedUnmap(false) { const size_t pagesize = getpagesize();à获取系统中的内存页大小，一般为4kB size = ((size + pagesize-1) & ~(pagesize-1)); int fd = ashmem_create_region(name == NULL ? "MemoryHeapBase" : name, size);à ashmem_create_region来创建共享内存，在真实设备上将打开/dev/ashmem设备得到一个文件描述符 LOGE_IF(fd<0, "error creating ashmem region: %s", strerror(errno)); if (fd >= 0) { if (mapfd(fd, size) == NO_ERROR) {à通过mmap方式得到内存地址 if (flags & READ_ONLY) { ashmem_set_prot_region(fd, PROT_READ); } } } }

MemoryHeapBase构造完后得到几个结果：

mBase变量指向共享内存的起始地址

mSize是所要求分配的内存大小

mFD是ashmem_create_region返回的文件描述符

查看MemoryBase：

class MemoryBase : public BnMemory { public: MemoryBase(const sp<IMemoryHeap>& heap, ssize_t offset, size_t size); virtual ~MemoryBase(); virtual sp<IMemoryHeap> getMemory(ssize_t* offset, size_t* size) const; protected: size_t getSize() const { return mSize; } ssize_t getOffset() const { return mOffset; } const sp<IMemoryHeap>& getHeap() const { return mHeap; } private: size_t mSize; ssize_t mOffset; sp<IMemoryHeap> mHeap; };

这里暂时只看到分配了一块共享内存，这样两个进程就可以共享这块内存，但是没有跨进程的同步对象。
分析mAudioTrack.play();：
play的jni层：

static void android_media_AudioTrack_start(JNIEnv *env, jobject thiz) { AudioTrack *lpTrack = (AudioTrack *)env->GetIntField( thiz, javaAudioTrackFields.nativeTrackInJavaObj);à从java的AudioTrack对象中获取native层的AudioTrack对象指针，从int型转换指针 if (lpTrack == NULL ) { jniThrowException(env, "java/lang/IllegalStateException", "Unable to retrieve AudioTrack pointer for start()"); return; } lpTrack->start(); }

write的jni层，最终都会调用writeToTrack()函数：

jint writeToTrack(AudioTrack* pTrack, jint audioFormat, jbyte* data, jint offsetInBytes, jint sizeInBytes) { ssize_t written = 0; à如果是STATIC模式sharedBuffer()返回不为0，如果为STREAM模式返回为0 if (pTrack->sharedBuffer() == 0) { written = pTrack->write(data + offsetInBytes, sizeInBytes); } else { if (audioFormat == javaAudioTrackFields.PCM16) { if ((size_t)sizeInBytes > pTrack->sharedBuffer()->size()) { sizeInBytes = pTrack->sharedBuffer()->size(); } memcpy(pTrack->sharedBuffer()->pointer(), data + offsetInBytes, sizeInBytes);à STATIC直接把数据memcpy到共享内存 written = sizeInBytes; } else if (audioFormat == javaAudioTrackFields.PCM8) { … } return written; }

到这里只分析了AudioTrack的java层和jni层的交互，开辟了一份共享内存。