*****************************************源码*************************************************
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; 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; } if ( (sampleRateInHz < 4000) || (sampleRateInHz > 48000) ) { loge("getMinBufferSize(): " + sampleRateInHz +"Hz is not a supported sample rate."); return AudioTrack.ERROR_BAD_VALUE; } int size = native_get_min_buff_size(sampleRateInHz, channelCount, audioFormat); if ((size == -1) || (size == 0)) { loge("getMinBufferSize(): error querying hardware"); return AudioTrack.ERROR; } else { return size; } }

***********************************************************************************************
源码路径：
frameworks\base\media\java\android\media\AudioTrack.java

###########################################说明##############################################################
先把自带的注释拿来看看吧：
/** * Returns the minimum buffer size required for the successful creation of an AudioTrack * object to be created in the {@link #MODE_STREAM} mode. Note that this size doesn't * guarantee a smooth playback under load, and higher values should be chosen according to * the expected frequency at which the buffer will be refilled with additional data to play. * @param sampleRateInHz the sample rate expressed in Hertz. * @param channelConfig describes the configuration of the audio channels. * See {@link AudioFormat#CHANNEL_OUT_MONO} and * {@link AudioFormat#CHANNEL_OUT_STEREO} * @param audioFormat the format in which the audio data is represented. * See {@link AudioFormat#ENCODING_PCM_16BIT} and * {@link AudioFormat#ENCODING_PCM_8BIT} * @return {@link #ERROR_BAD_VALUE} if an invalid parameter was passed, * or {@link #ERROR} if the implementation was unable to query the hardware for its output * properties, * or the minimum buffer size expressed in bytes. */

从注释可以看出，通过该函数获取的最小buffer size，只是保证在MODE_STREAM模式下成功地创建一个AudioTrack对象。
并不能保证流畅地播放。

1、参数就不说了，可以参考上面注释，上一篇文章中也有说。
2、定义了一个内部变量：
     int channelCount = 0;
   用来记录声道数量。
   调用native函数native_get_min_buff_size时会用。
   可见buffer size也是由native层来决定的。
3、接下来根据Channel类型，计算声道数量：
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; break; default: loge("getMinBufferSize(): Invalid channel configuration."); return AudioTrack.ERROR_BAD_VALUE; }

MONO都是1，Stereo的都是2。
   不过，我们之前看过，Channel类型不止这几种。有以下一堆呢：
public static final int CHANNEL_OUT_FRONT_LEFT = 0x4; public static final int CHANNEL_OUT_FRONT_RIGHT = 0x8; public static final int CHANNEL_OUT_FRONT_CENTER = 0x10; public static final int CHANNEL_OUT_LOW_FREQUENCY = 0x20; public static final int CHANNEL_OUT_BACK_LEFT = 0x40; public static final int CHANNEL_OUT_BACK_RIGHT = 0x80; public static final int CHANNEL_OUT_FRONT_LEFT_OF_CENTER = 0x100; public static final int CHANNEL_OUT_FRONT_RIGHT_OF_CENTER = 0x200; public static final int CHANNEL_OUT_BACK_CENTER = 0x400; public static final int CHANNEL_OUT_MONO = CHANNEL_OUT_FRONT_LEFT; public static final int CHANNEL_OUT_STEREO = (CHANNEL_OUT_FRONT_LEFT | CHANNEL_OUT_FRONT_RIGHT); public static final int CHANNEL_OUT_QUAD = (CHANNEL_OUT_FRONT_LEFT | CHANNEL_OUT_FRONT_RIGHT | CHANNEL_OUT_BACK_LEFT | CHANNEL_OUT_BACK_RIGHT); public static final int CHANNEL_OUT_SURROUND = (CHANNEL_OUT_FRONT_LEFT | CHANNEL_OUT_FRONT_RIGHT | CHANNEL_OUT_FRONT_CENTER | CHANNEL_OUT_BACK_CENTER); public static final int CHANNEL_OUT_5POINT1 = (CHANNEL_OUT_FRONT_LEFT | CHANNEL_OUT_FRONT_RIGHT | CHANNEL_OUT_FRONT_CENTER | CHANNEL_OUT_LOW_FREQUENCY | CHANNEL_OUT_BACK_LEFT | CHANNEL_OUT_BACK_RIGHT); public static final int CHANNEL_OUT_7POINT1 = (CHANNEL_OUT_FRONT_LEFT | CHANNEL_OUT_FRONT_RIGHT | CHANNEL_OUT_FRONT_CENTER | CHANNEL_OUT_LOW_FREQUENCY | CHANNEL_OUT_BACK_LEFT | CHANNEL_OUT_BACK_RIGHT | CHANNEL_OUT_FRONT_LEFT_OF_CENTER | CHANNEL_OUT_FRONT_RIGHT_OF_CENTER);

并且，AudioFormat.CHANNEL_CONFIGURATION_MONO和AudioFormat.CHANNEL_CONFIGURATION_STEREO的定义还不包含在这一堆之中，而是在它们之前定义：
/** Mono audio configuration */ /** @deprecated use CHANNEL_OUT_MONO or CHANNEL_IN_MONO instead */ @Deprecated public static final int CHANNEL_CONFIGURATION_MONO = 2; /** Stereo (2 channel) audio configuration */ /** @deprecated use CHANNEL_OUT_STEREO or CHANNEL_IN_STEREO instead */ @Deprecated public static final int CHANNEL_CONFIGURATION_STEREO = 3;

难道其他的Channel类型都不需要获取这个min buffer size？？？
   还是说，目前只支持单声道和双声道？？？

4、下面判断音频格式，即采样点数据所占的bit数：
if ((audioFormat != AudioFormat.ENCODING_PCM_16BIT) && (audioFormat != AudioFormat.ENCODING_PCM_8BIT)) { loge("getMinBufferSize(): Invalid audio format."); return AudioTrack.ERROR_BAD_VALUE; }

可见，只支持16bit和8bit两种。

5、判断采用率：
if ( (sampleRateInHz < 4000) || (sampleRateInHz > 48000) ) { loge("getMinBufferSize(): " + sampleRateInHz +"Hz is not a supported sample rate."); return AudioTrack.ERROR_BAD_VALUE; }

只支持4000Hz到48000Hz之间。

6、接下来调到native中去：
int size = native_get_min_buff_size(sampleRateInHz, channelCount, audioFormat); if ((size == -1) || (size == 0)) { loge("getMinBufferSize(): error querying hardware"); return AudioTrack.ERROR; } else { return size; }

可见，真正干活的是在native中，java层中只是做些辅助操作。

通过前文中JNI的函数对照表，可知native_get_min_buff_size函数对应的是native中的android_media_AudioTrack_get_min_buff_size函数。
   路径：frameworks\base\core\jni\android_media_AudioTrack.cpp

函数android_media_AudioTrack_get_min_buff_size的实现：
// returns the minimum required size for the successful creation of a streaming AudioTrack // returns -1 if there was an error querying the hardware. 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); }

可见，最小buffer size是frameCoun乘以声道个数，在根据音频格式乘以1或2得到。
   声道个数和音频格式都是传入的，不再说。
   frameCount是调用函数AudioTrack::getMinFrameCount取得的。从函数名可知，此处取得的应该是最小frame数。
   传入的三个参数：
     &frameCount是用来保存frame计数的。
     sampleRateInHertz是采样率。
     AudioSystem::DEFAULT是写死的。其定义在类AudioSystem中，其他的定义如下：
enum stream_type { DEFAULT =-1, VOICE_CALL = 0, SYSTEM = 1, RING = 2, MUSIC = 3, ALARM = 4, NOTIFICATION = 5, BLUETOOTH_SCO = 6, ENFORCED_AUDIBLE = 7, // Sounds that cannot be muted by user and must be routed to speaker DTMF = 8, TTS = 9, NUM_STREAM_TYPES };

原来是stream的类型。
   为什么不在调用getMinBufferSize的时候传入stream类型，而在此处使用DEFAULT呢？？？

先放放，继续看函数AudioTrack::getMinFrameCount。

函数AudioTrack::getMinFrameCount的实现：
status_t AudioTrack::getMinFrameCount( int* frameCount, int streamType, uint32_t sampleRate) { int afSampleRate; if (AudioSystem::getOutputSamplingRate(&afSampleRate, streamType) != NO_ERROR) { return NO_INIT; } int afFrameCount; if (AudioSystem::getOutputFrameCount(&afFrameCount, streamType) != NO_ERROR) { return NO_INIT; } uint32_t afLatency; if (AudioSystem::getOutputLatency(&afLatency, streamType) != NO_ERROR) { return NO_INIT; } // Ensure that buffer depth covers at least audio hardware latency uint32_t minBufCount = afLatency / ((1000 * afFrameCount) / afSampleRate); if (minBufCount < 2) minBufCount = 2; *frameCount = (sampleRate == 0) ? afFrameCount * minBufCount : afFrameCount * minBufCount * sampleRate / afSampleRate; return NO_ERROR; }

开始，调用了三个AudioSystem的函数，似曾谋面，不过当时被无视了，今天看看吧。

函数AudioSystem::getOutputSamplingRate的实现：
status_t AudioSystem::getOutputSamplingRate(int* samplingRate, int streamType) { OutputDescriptor *outputDesc; audio_io_handle_t output; if (streamType == DEFAULT) { streamType = MUSIC; } output = getOutput((stream_type)streamType); if (output == 0) { return PERMISSION_DENIED; } gLock.lock(); outputDesc = AudioSystem::gOutputs.valueFor(output); if (outputDesc == 0) { LOGV("getOutputSamplingRate() no output descriptor for output %d in gOutputs", output); gLock.unlock(); const sp<IAudioFlinger>& af = AudioSystem::get_audio_flinger(); if (af == 0) return PERMISSION_DENIED; *samplingRate = af->sampleRate(output); } else { LOGV("getOutputSamplingRate() reading from output desc"); *samplingRate = outputDesc->samplingRate; gLock.unlock(); } LOGV("getOutputSamplingRate() streamType %d, output %d, sampling rate %d", streamType, output, *samplingRate); return NO_ERROR; }

判断流的类型，如果是DEFAULT，将其设置为MUSIC！
   纳炉嚎啕！！！
   DEFAULT的流类型原来是这么用的。

接下来根据stream type获取output。
   然后获取output的描述。

若获取成功，则output描述中的采样率就是要获取的采样率。
   否则，尝试从AudioFlinger中获取采样率。

函数AudioSystem::getOutputFrameCount，AudioSystem::getOutputLatency，与函数AudioSystem::getOutputSamplingRate的处理类似。

至此，采样率，frameCount和延迟都取得了。
   
   接下来计算minBufCount：
// Ensure that buffer depth covers at least audio hardware latency uint32_t minBufCount = afLatency / ((1000 * afFrameCount) / afSampleRate); if (minBufCount < 2) minBufCount = 2;

从注释可知，buff大小应至少能覆盖audio 硬件的延迟。
   公式不太明白。
   先看看从链接：http://blog.csdn.net/innost/article/details/6125779
   中摘过来的frame的说明：
     一个frame就是1个采样点的字节数*声道。为啥搞个frame出来？因为对于多声道的话，用1个采样点的字节数表示不全，
     因为播放的时候肯定是多个声道的数据都要播出来才行。所以为了方便，就说1秒钟有多少个frame，这样就能抛开声道数，把意思表示全了。

还不是很明白。先放放。
   猜了半天也猜不出来。哪位大侠指点指点。

下面计算frameCount：
*frameCount = (sampleRate == 0) ? afFrameCount * minBufCount : afFrameCount * minBufCount * sampleRate / afSampleRate;

我们的sampleRate肯定不为0，所以最后的计算应该为：afFrameCount * minBufCount * sampleRate / afSampleRate