音量的调节控制:
按下音量的加减键,对应的keyevent是:KEYCODE_VOLUME_DOWN,KEYCODE_VOLUME_UP。
主要看两个地方对这一事件的处理,一是:handleKeyDown()@AudioManager.java,一是:
interceptKeyBeforeQueueing ()@PhoneWindowManager.java。
先看AudioManager.java类中的处理:
frameworks/base/media/java/android/media/
public void handleKeyDown(KeyEvent event, int stream) @AudioManager.java{
int keyCode = event.getKeyCode();
switch (keyCode) {
case KeyEvent.KEYCODE_VOLUME_UP:
case KeyEvent.KEYCODE_VOLUME_DOWN:
//在key-down事件时,调整音量,要更灵敏的相应用户,这里的flags是FLAG_SHOW_UI | FLAG_VIBRATE,会显示进度条,有震动,
adjustSuggestedStreamVolume(
keyCode == KeyEvent.KEYCODE_VOLUME_UP ? ADJUST_RAISE: ADJUST_LOWER,
stream, FLAG_SHOW_UI | FLAG_VIBRATE);
break;
}
}
public void handleKeyUp(KeyEvent event, int stream) @AudioManager.java {
nt keyCode = event.getKeyCode();
switch (keyCode) {
case KeyEvent.KEYCODE_VOLUME_UP:
case KeyEvent.KEYCODE_VOLUME_DOWN:
//在key-up时,播放调节音量的声音。
adjustSuggestedStreamVolume(ADJUST_SAME, stream,
FLAG_PLAY_SOUND);
}
}
根据音量加减键,进一步调用:adjustSuggestedStreamVolume。
public void adjustSuggestedStreamVolume(int direction, int suggestedStreamType, int flags) {
MediaSessionLegacyHelper helper = MediaSessionLegacyHelper.getHelper(getContext());
helper.sendAdjustVolumeBy(suggestedStreamType, direction, flags);
往下的过程,会通过ISessionManager这个Binder client 句柄,把调节音量的请求传到AudioService。
这个调用,接下来会走到MediaSessionService.java中,调用SessionManagerImpl中的方法:dispatchAdjustVolumeLocked (),SessionManagerImpl是ISessionManager的服务端的实现类。
class SessionManagerImpl extends ISessionManager.Stub;
}
MediaSessionService.java
private void dispatchAdjustVolumeLocked(int suggestedStream, int direction, int flags,
MediaSessionRecord session) @SessionManagerImpl{
String packageName = getContext().getOpPackageName();
//这个mAudioService就是AUDIO_SERVICE = "audio"对应的Binder服务。所以请求就到了AudioService的server端。
mAudioService.adjustSuggestedStreamVolume(direction, suggestedStream,
flags, packageName, TAG);
}
private void adjustSuggestedStreamVolume(int direction, int suggestedStreamType, int flags,
String callingPackage, String caller, int uid) {
//声音分为STREAM_MUSIC,STREAM_VOICE_CALL,等多种流类型,当前要调节的是哪一种流类型的音量,先通过getActiveStreamType获取当前活跃的流类型。
先判断是不是在通话,如果是通话,还要分是蓝牙STREAM_BLUETOOTH_SCO,还是STREAM_VOICE_CALL;然后是不是播放音乐,
int streamType;
streamType = getActiveStreamType(suggestedStreamType);
//执行调整音量
adjustStreamVolume(streamType, direction, flags, callingPackage, caller, uid);
}
adjustStreamVolume稍后分析,
下面看下PhoneWindowManager.java中对音量键的处理:
public int interceptKeyBeforeQueueing(KeyEvent event, int policyFlags)
@ PhoneWindowManager.java {
final int keyCode = event.getKeyCode();
switch (keyCode) {
case KeyEvent.KEYCODE_VOLUME_DOWN:
case KeyEvent.KEYCODE_VOLUME_UP:
//对一些物理按键,系统有优先拦截的权利,做些预处理,这里会根据当前的具体情况,如是不是做截屏处理,是不是在通话等,然后决定需要调整的的stream类型。通过mediasession进一步处理按键事件,跟audiomanager类似。
MediaSessionLegacyHelper.getHelper(mContext)
.sendVolumeKeyEvent(event, true);
}
}
public void sendVolumeKeyEvent(KeyEvent keyEvent, boolean musicOnly)
@ MediaSessionLegacyHelper.java {
//接下来的处理,就是通过MediaSessionManager.,SessionManagerImpl(位于MediaSessionService.java中),把请求传到audioservice那边。
mSessionManager.dispatchAdjustVolume(
AudioManager.USE_DEFAULT_STREAM_TYPE, direction, flags);
}
接着看audioservice,如何处理音量调节的。
private void adjustStreamVolume(int streamType, int direction, int flags,
String callingPackage, String caller, int uid) @ AudioService.java {
//step1,为streamType查找alias归类,使用流类型别名是为了让有相同别名的流,有相同的行为。
int streamTypeAlias = mStreamVolumeAlias[streamType];
VolumeStreamState streamState = mStreamStates[streamTypeAlias];
//step2,为streamtypealias寻找匹配的device。
final int device = getDeviceForStream(streamTypeAlias);
//step3,获取对应的aliasindex
int aliasIndex = streamState.getIndex(device);
//step4,调节index,把UI上的加1、减1的步骤,转成流别名上内部单元中的数字。
step = rescaleIndex(10, streamType, streamTypeAlias);
//step5,对音量模式的影响,
if (((flags & AudioManager.FLAG_ALLOW_RINGER_MODES) != 0) ||
(streamTypeAlias == getUiSoundsStreamType())) {
int ringerMode = getRingerModeInternal();
//如果已经是震动模式,就不要震动了
if (ringerMode == AudioManager.RINGER_MODE_VIBRATE) {
flags &= ~AudioManager.FLAG_VIBRATE;
}
//检查这个调节是否会改变铃声模式,而不仅仅是调节音量,如果是这样,就要设置合适的铃声模式,在流状态上音量指数。
final int result = checkForRingerModeChange(aliasIndex, direction, step,
streamState.mIsMuted, callingPackage, flags);
adjustVolume = (result & FLAG_ADJUST_VOLUME) != 0;
}
//step6,将调节音量的事件发给下一个处理者。
int oldIndex = mStreamStates[streamType].getIndex(device);
sendMsg(mAudioHandler, MSG_SET_DEVICE_VOLUME, SENDMSG_QUEUE,
device, 0, streamState,0);
int index = mStreamStates[streamType].getIndex(device);
sendVolumeUpdate(streamType, oldIndex, index, flags);
}
Step1,流类型别名,虽然stream有多种类型,但他们在某些方面的行为是一样的,这就是stream-alias存在的意义。
private final int[] STREAM_VOLUME_ALIAS_DEFAULT = new int[] {
AudioSystem.STREAM_VOICE_CALL, // STREAM_VOICE_CALL
AudioSystem.STREAM_RING, // STREAM_SYSTEM
AudioSystem.STREAM_RING, // STREAM_RING
AudioSystem.STREAM_MUSIC, // STREAM_MUSIC
AudioSystem.STREAM_ALARM, // STREAM_ALARM
AudioSystem.STREAM_RING, // STREAM_NOTIFICATION
AudioSystem.STREAM_BLUETOOTH_SCO, // STREAM_BLUETOOTH_SCO
AudioSystem.STREAM_RING, // STREAM_SYSTEM_ENFORCED
AudioSystem.STREAM_RING, // STREAM_DTMF
AudioSystem.STREAM_MUSIC // STREAM_TTS
};
public int getIndex(int device) {
synchronized (VolumeStreamState.class) {
int index = mIndexMap.get(device, -1);
if (index == -1) {
// there is always an entry for AudioSystem.DEVICE_OUT_DEFAULT
index = mIndexMap.get(AudioSystem.DEVICE_OUT_DEFAULT);
}
return index;
}
}
mIndexMap实际是个数组,不含映射的,记录了device对应的index值。
Step6,sendMsg把消息投到队列,有mAudioHandler处理,一方面把音量值保存到系统设置文件,另一方面调用audiopolicyservice接口调整音频设备的音量。
private void persistVolume(VolumeStreamState streamState, int device)
@ AudioHandler$ AudioService.java {
//把音量值写到系统文件。
System.putIntForUser(mContentResolver,
streamState.getSettingNameForDevice(device),
(streamState.getIndex(device) + 5)/ 10,
UserHandle.USER_CURRENT);
}
调用audiopolicyservice接口的流程:
public void applyDeviceVolume_syncVSS(int device)@ VolumeStreamState $ AudioService.java {
AudioSystem.setStreamVolumeIndex(mStreamType, index, device);
}
setStreamVolumeIndex是native方法。
status_t AudioSystem::setStreamVolumeIndex(audio_stream_type_t stream,…)
@ AudioSystem.cpp {
const sp& aps = AudioSystem::get_audio_policy_service();
return aps->setStreamVolumeIndex(stream, index, device);
}
IAudioPolicyService对应的服务端的实现:
status_t AudioPolicyService::setStreamVolumeIndex(audio_stream_type_t stream, int index,
audio_devices_t device)@AudioPolicyInterfaceImplLegacy.cpp{
return mpAudioPolicy->set_stream_volume_index(mpAudioPolicy, stream, index);
}
根据前面的分析,mpAudioPolicy的默认实现是legacy_audio_policy 结构体中的audio_policy 。相关实现(create_legacy_ap()@audio_policy_hal.cpp);
static int ap_set_stream_volume_index(struct audio_policy *pol, audio_stream_type_t stream,
int index)@ audio_policy_hal.cpp {
struct legacy_audio_policy *lap = to_lap(pol);
return lap->apm->setStreamVolumeIndex((AudioSystem::stream_type)stream,
index, AUDIO_DEVICE_OUT_DEFAULT);
}
setStreamVolumeIndex的最终实现调用了AudioPolicyManagerBase.cpp中接口。
status_t AudioPolicyManagerBase::setStreamVolumeIndex(AudioSystem::stream_type stream,
int index, audio_devices_t device)@ AudioPolicyManagerBase.cpp {
//根据连接的设备,在所有的输出上计算、应用流音量,
for (size_t i = 0; i < mOutputs.size(); i++) {
audio_devices_t curDevice = getDeviceForVolume(mOutputs.valueAt(i)->device());
if ((device == AUDIO_DEVICE_OUT_DEFAULT) || (device == curDevice)) {
status_t volStatus = checkAndSetVolume(stream, index, mOutputs.keyAt(i),
curDevice);
status = volStatus;
}
}
return status;
}
系统中有多个output,每个output又支持多个device,所以checkAndSetVolume函数,先计算音量,最终会调用Audioflinger的接口完成音量的设置。
主要看跳转到Audioflinger的流程:
status_t AudioPolicyManagerBase::checkAndSetVolume(…)@ AudioPolicyManagerBase.cpp{
mpClientInterface->setStreamVolume((AudioSystem::stream_type)stream, volume, output,
delayMs);
}
AudioPolicyClientInterface*mpClientInterface;是在AudioPolicyManagerBase构造函数中赋值的。调用AudioPolicyManagerBase构造函数的根在audio_policy_hal.cpp
static int create_legacy_ap(struct audio_policy_service_ops *aps_ops,…)@ audio_policy_hal.cpp {
lap->service_client = new AudioPolicyCompatClient(aps_ops, service);
//这个函数最终创建了AudioPolicyManagerBase实例。
lap->apm = createAudioPolicyManager(lap->service_client);
}
status_t AudioPolicyCompatClient::setStreamVolume(…){
// mServiceOps即是函数create_legacy_ap中的aps_ops。想知道aps_ops是怎么来的,就是要知道谁调用了create_legacy_ap。
return mServiceOps->set_stream_volume(mService, (audio_stream_type_t)stream,
volume, output, delayMs);
}
想知道create_legacy_ap函数中的参数aps_ops是怎么来的,就是要知道谁调用了create_legacy_ap。
audio_policy_hal.cpp
//在打开一个hal层设备时,会执行这个函数:
static int legacy_ap_dev_open(…)@ audio_policy_hal.cpp{
dev->device.create_audio_policy = create_legacy_ap;
}
函数指针create_audio_policy指向了create_legacy_ap。
AudioPolicyService.cpp
在AudioPolicyService实例第一次被引用时:
void AudioPolicyService::onFirstRef()@AudioPolicyService.cpp{
//create_legacy_ap函数中的参数aps_op是在这里赋值的(aps_ops)
rc = mpAudioPolicyDev->create_audio_policy(mpAudioPolicyDev, &aps_ops, this,
&mpAudioPolicy);
}
根据aps_ops 结构体的定义,set_stream_volume实现指向了aps_set_stream_volume。
struct audio_policy_service_ops aps_ops = {
.set_stream_volume = aps_set_stream_volume,
}
有函数aps_set_stream_volume,经过AudioPolicyService.cpp,经过AudioCommandThread的SET_VOLUME命令,在经过AudioSystem.cpp,最终实现是在Audioflinger。
status_t AudioSystem::setStreamVolume(audio_stream_type_t stream, float value,
audio_io_handle_t output)@ AudioSystem.cpp {
af->setStreamVolume(stream, value, output);
const sp& af = AudioSystem::get_audio_flinger();
}
status_t AudioFlinger::setStreamVolume(audio_stream_type_t stream, float value,
audio_io_handle_t output)@ AudioFlinger.cpp {
PlaybackThread *thread = NULL;
//根据output找到对应的playbackthread,
thread = checkPlaybackThread_l(output);
//把新的值记录到mStreamTypes中。
mStreamTypes[stream].volume = value;
thread->setStreamVolume(stream, value);
}
然后playbackthread对音量值做下记录,在混音线程准备音频数据时,会把playbackthread保存的值提取出来。
Frameworks/av/services/audiofligner/
AudioFlinger::PlaybackThread::mixer_state AudioFlinger::MixerThread::prepareTracks_l(
Vector< sp
读出之前这种的音量值后,和主音量综合运算,把计算后的值,通过setParameter传给AudioMixer。