audioserver 启动流程
本篇文章目标
- 了解 audioserver 启动流程
- 了解 AudioPolicyService 启动流程
- 了解 AudioFlinger 启动流程
Android 系统的启动流程
在了解audioserver的启动之前先看Android系统的启动流程
通过上图可以看到Android系统的启动整体流程，而audioserver的启动就是由init进程拉起的，因此audioserver有单独的进程号。
在Android系统中，音频主要有两个子服务，AudioFlinger和AudioPolicy，这两个服务都是运行在audioserver进程中的。
AudioFlinger负责和底层audio hal进行交互，是audio系统的实现者。
AudioPolicy负责音频策略的定制，是audio系统的指挥者。
Audio Server 的启动流程
加载 audioserver.rc
audioserver.rc是由init进程拉起的，如下图
init进程对应的main函数
system/core/init/init.cpp
int SecondStageMain(int argc, char** argv) {
// ......
LoadBootScripts(am, sm);
// ......
}
// 看LoadBootScripts()函数
static void LoadBootScripts(ActionManager& action_manager, ServiceList& service_list) {
Parser parser = CreateParser(action_manager, service_list);
std::string bootscript = GetProperty("ro.boot.init_rc", "");
if (bootscript.empty()) {
parser.ParseConfig("/system/etc/init/hw/init.rc");
if (!parser.ParseConfig("/system/etc/init")) {
late_import_paths.emplace_back("/system/etc/init");
}
// late_import is available only in Q and earlier release. As we don't
// have system_ext in those versions, skip late_import for system_ext.
parser.ParseConfig("/system_ext/etc/init");
if (!parser.ParseConfig("/vendor/etc/init")) {
late_import_paths.emplace_back("/vendor/etc/init");
}
if (!parser.ParseConfig("/odm/etc/init")) {
late_import_paths.emplace_back("/odm/etc/init");
}
if (!parser.ParseConfig("/product/etc/init")) {
late_import_paths.emplace_back("/product/etc/init");
}
} else {
parser.ParseConfig(bootscript);
}
}
LoadBootScrips()函数中，会首先加载init.rc文件
然后会加载/system/etc/init/、/product/etc/init/、/odm/etc/init/、/vendor/etc/init目录下的rc文件
而audioserver.rc位于/system/etc/init 目录下。
启动 audioserver 进程
看下audioserver.rc文件，init进程加载audioserver.rc文件时就会启动audioserver进程。
frameworks/av/media/audioserver/audioserver.rc
service audioserver /system/bin/audioserver
class core
user audioserver # 设置所有者
# media gid needed for /dev/fm (radio) and for /data/misc/media (tee)
group audio camera drmrpc media mediadrm net_bt net_bt_admin net_bw_acct wakelock # 设置所属组
capabilities BLOCK_SUSPEND
ioprio rt 4
task_profiles ProcessCapacityHigh HighPerformance
onrestart restart vendor.audio-hal
onrestart restart vendor.audio-hal-4-0-msd
onrestart restart audio_proxy_service
# Keep the original service names for backward compatibility
onrestart restart vendor.audio-hal-2-0
onrestart restart audio-hal-2-0
on property:vts.native_server.on=1
stop audioserver
on property:vts.native_server.on=0
start audioserver
audioserver以及audio hal的进程启动/关闭都是在该文件下执行的。
audioserver 加载 AudioFlinger 以及 AudioPolicy 的流程
audioserver是在main_audioserver.cpp里面去创建audioflinger和audiopolicy的实例的
frameworks\av\media\audioserver\main_audioserver.cpp
int main(int argc __unused, char **argv) {
// ......
// 如果doLog为true则创建子进程
if (doLog && (childPid = fork()) != 0) {
// ......
} else {
// all other services
if (doLog) {
prctl(PR_SET_PDEATHSIG, SIGKILL); // if parent media.log dies before me, kill me also
setpgid(0, 0); // but if I die first, don't kill my parent
}
android::hardware::configureRpcThreadpool(4, false /*callerWillJoin*/);
sp<ProcessState> proc(ProcessState::self());
sp<IServiceManager> sm = defaultServiceManager();
ALOGI("ServiceManager: %p", sm.get());
AudioFlinger::instantiate();
AudioPolicyService::instantiate();
// AAudioService should only be used in OC-MR1 and later.
// And only enable the AAudioService if the system MMAP policy explicitly allows it.
// This prevents a client from misusing AAudioService when it is not supported.
aaudio_policy_t mmapPolicy = property_get_int32(AAUDIO_PROP_MMAP_POLICY,
AAUDIO_POLICY_NEVER);
if (mmapPolicy == AAUDIO_POLICY_AUTO || mmapPolicy == AAUDIO_POLICY_ALWAYS) {
AAudioService::instantiate();
}
ProcessState::self()->startThreadPool();
IPCThreadState::self()->joinThreadPool();
}
}
在main_audioserver.cpp里面主要的逻辑：
① 启动AudioFlinger
② 启动AudioPolicyService
③ 如果支持AAudio，则启动AAudioService
④ 创建线程池并将当前线程加入线程池
audioserver启动的完整流程如下：
经过上面的流程audioserver已经启动处于待命状态，如果有应用需要播放则会通过服务最终选择合适的硬件将声音播出，接下来按照上面的流程进行进一步的细分。
AudioFlinger 初始化流程
AudioFlinger::instantiate()，在Android S版本之前，在AudioFlinger并没有实现instantiate函数，直接在其父类找到对应函数。
frameworks/native/libs/binder/include/binder/BinderService.h
class BinderService
{
public:
static status_t publish(bool allowIsolated = false,
int dumpFlags = IServiceManager::DUMP_FLAG_PRIORITY_DEFAULT) {
sp<IServiceManager> sm(defaultServiceManager());
return sm->addService(String16(SERVICE::getServiceName()), new SERVICE(), allowIsolated,
dumpFlags);
}
static void instantiate() { publish(); }
}
在BinderService.h中主要的业务：
① 获取IServiceManager
② new SERVICE()，这里对应的就是new AudioFlinger()
③ 将AudioFlinger注册到ServiceManager里面，方便后续别的进程使用该服务
Android S版本里面将instantiate放在了AudioFlinger里面，代码如下：
AudioFlinger.cpp
void AudioFlinger::instantiate() {
sp<IServiceManager> sm(defaultServiceManager());
sm->addService(String16(IAudioFlinger::DEFAULT_SERVICE_NAME),
new AudioFlingerServerAdapter(new AudioFlinger()), false,
IServiceManager::DUMP_FLAG_PRIORITY_DEFAULT);
}
由于sm是带sp的(sp
接下来先看AudioFlinger的构造函数：
AudioFlinger::AudioFlinger()
// 默认全局变量的初始化辅助
......
{
// ......
// 实例化打开设备的接口，通过它可以获取到hal层audio硬件的接口
mDevicesFactoryHal = DevicesFactoryHalInterface::create();
// 实例化音效的接口，通过它可以获取到hal层音效的接口
mEffectsFactoryHal = EffectsFactoryHalInterface::create();
}
看AudioFlinger::onFirstRef()的代码：
void AudioFlinger::onFirstRef()
{
Mutex::Autolock _l(mLock);
// ......
mMode = AUDIO_MODE_NORMAL;
gAudioFlinger = this; // we are already refcounted, store into atomic pointer.
// 实例化和hal的回调并设置回调函数
mDevicesFactoryHalCallback = new DevicesFactoryHalCallbackImpl;
mDevicesFactoryHal->setCallbackOnce(mDevicesFactoryHalCallback);
}
至此AudioFlinger就实例化完成了
AudioPolicyService 的初始化流程
AudioPolicyService也是BinderService子类，其instantiate过程和AudioFlinger是类似的，创建AudioPolicyService，注册到ServiceManager中。直接看AudioPolicy的初始化
先看构造函数：做一些全局变量初始化赋值操作
AudioPolicyService::AudioPolicyService()
: BnAudioPolicyService(),
mAudioPolicyManager(NULL),
mAudioPolicyClient(NULL),
mPhoneState(AUDIO_MODE_INVALID),
mCaptureStateNotifier(false),
mCreateAudioPolicyManager(createAudioPolicyManager),
mDestroyAudioPolicyManager(destroyAudioPolicyManager) {
}
和AudioFlinger一样，在初始化的时候会调用AudioPolicyService::onFirstRef()函数，接下来看AudioPolicyService::onFirstRef()函数：
void AudioPolicyService::onFirstRef()
{
{
Mutex::Autolock _l(mLock);
// start audio commands thread
// ①创建AudioCommandThread线程，用于执行audio命令
mAudioCommandThread = new AudioCommandThread(String8("ApmAudio"), this);
// start output activity command thread
// ②用于执行audio 输出命令
mOutputCommandThread = new AudioCommandThread(String8("ApmOutput"), this);
// ③实例化AudioPolicyClient对象，该对象是AudioFlinger和AudioPolicyService通信的中间
mAudioPolicyClient = new AudioPolicyClient(this);
// ④ Android S版本新增，用于加载自定义AudioPolicyManager的支持，如果自定义库存在则加载，如果不存在则恢复默认
loadAudioPolicyManager();
// ⑤实例化AudioPolicyManager对象
mAudioPolicyManager = mCreateAudioPolicyManager(mAudioPolicyClient);
}
// load audio processing modules
// ⑥初始化音效相关的内容
sp<AudioPolicyEffects> audioPolicyEffects = new AudioPolicyEffects();
sp<UidPolicy> uidPolicy = new UidPolicy(this);
sp<SensorPrivacyPolicy> sensorPrivacyPolicy = new SensorPrivacyPolicy(this);
{
Mutex::Autolock _l(mLock);
mAudioPolicyEffects = audioPolicyEffects;
mUidPolicy = uidPolicy;
mSensorPrivacyPolicy = sensorPrivacyPolicy;
}
uidPolicy->registerSelf();
sensorPrivacyPolicy->registerSelf();
}
而AudioPolicy的管理主要在AudioPolicyManager里面，接下来看AudioPolicyManager
AudioPolicyManager的构造
AudioPolicyManager::AudioPolicyManager(AudioPolicyClientInterface *clientInterface)
: AudioPolicyManager(clientInterface, false /*forTesting*/)
{
loadConfig();
}
Andorid S版本是在AudioPolicyService.cpp的createAudioPolicyManager()函数中去调用initialize()函数的
static AudioPolicyInterface* createAudioPolicyManager(AudioPolicyClientInterface *clientInterface)
{
// #ifndef VENDOR_EDIT
// pengzhao.chen@tcl.com, 2022/04/25, [FAT-9999] Modified for RTK patch
AudioPolicyManager *apm = nullptr;
apm = new TclAudioPolicyManager(clientInterface);
/*
AudioPolicyManager *apm = new AudioPolicyManager(clientInterface);
*/
// #endif /* VENDOR_EDIT */
status_t status = apm->initialize();
if (status != NO_ERROR) {
delete apm;
apm = nullptr;
}
return apm;
}
再看一下initialize()函数的实现：
AudioPolicyManager::initialize()
status_t AudioPolicyManager::initialize() {
{
// ① 加载音频engine库
auto engLib = EngineLibrary::load(
"libaudiopolicyengine" + getConfig().getEngineLibraryNameSuffix() + ".so");
if (!engLib) {
ALOGE("%s: Failed to load the engine library", __FUNCTION__);
return NO_INIT;
}
// ② 根据前面加载的库调用该函数创建Engine对象
mEngine = engLib->createEngine();
if (mEngine == nullptr) {
ALOGE("%s: Failed to instantiate the APM engine", __FUNCTION__);
return NO_INIT;
}
}
// ③ 设置APM为监听者
mEngine->setObserver(this);
// ④ 初始化检查
status_t status = mEngine->initCheck();
if (status != NO_ERROR) {
LOG_FATAL("Policy engine not initialized(err=%d)", status);
return status;
}
// 获取productStrategy
mCommunnicationStrategy = mEngine->getProductStrategyForAttributes(
mEngine->getAttributesForStreamType(AUDIO_STREAM_VOICE_CALL));
// ⑤ 输入输出设备的打开和线程创建
onNewAudioModulesAvailableInt(nullptr /*newDevices*/);
// make sure default device is reachable
if (mDefaultOutputDevice == 0 || !mAvailableOutputDevices.contains(mDefaultOutputDevice)) {
ALOGE_IF(mDefaultOutputDevice != 0, "Default device %s is unreachable",
mDefaultOutputDevice->toString().c_str());
status = NO_INIT;
}
// If microphones address is empty, set it according to device type
for (size_t i = 0; i < mAvailableInputDevices.size(); i++) {
if (mAvailableInputDevices[i]->address().empty()) {
if (mAvailableInputDevices[i]->type() == AUDIO_DEVICE_IN_BUILTIN_MIC) {
mAvailableInputDevices[i]->setAddress(AUDIO_BOTTOM_MICROPHONE_ADDRESS);
} else if (mAvailableInputDevices[i]->type() == AUDIO_DEVICE_IN_BACK_MIC) {
mAvailableInputDevices[i]->setAddress(AUDIO_BACK_MICROPHONE_ADDRESS);
}
}
}
ALOGW_IF(mPrimaryOutput == nullptr, "The policy configuration does not declare a primary output");
// Silence ALOGV statements
property_set("log.tag." LOG_TAG, "D");
// ⑥ 更新设备
updateDevicesAndOutputs();
return status;
}
AudioPolicyManager初始化主要依赖于loadConfig()和initialize()函数：
loadConfig()用来加载音频配置文件（audio_policy_configuration.xml）
initialize()主要用于：
①加载音频engine库
②根据前面加载的库调用该函数创建Engine对象
③设置APM为监听者
④初始化检查
⑤依次加载 HwModule 并打开其所含 profile 的 outputStream 及 inputStream（audio module模块包括：primary、A2DP、USB、remote_submix、hearing_aid）
最后看下AudioPolicyService的整体加载流程图：
总结
audioserver启动会加载AudioFlinger和AudioPolicyService两个服务，这两个服务会加载配置文件，解析音频输入、输出硬件信息，并且加载HIDL，建立与hardware的通信连接，至此audioserver启动完成。