Android 音频子系统(2) ---- AudioServer

---

前言

本系列文章基于Android14源码进行分析解读，部分框图直接使用了原作者的图片，侵权必删。
我们既生成文章，也是各路文章的搬运工。

---

一、What&Why AudioFlinger

        AudioHAL(AHAL)提供隔离Vendor硬件差异的抽象，通过tinyalsa实现对底层硬件声卡节点的利用。AHAL作为一个独立的service，在Android系统中肯定有client通过binder与其通信，实现数据流与控制流的交互，完成对声卡节点数据的访问。
        AudioTrack.java和AudioRecord.java是应用层的API，透过GNI访问了Native层的AudioTrack.cpp和AudioRecord.cpp，AudioTrack和AudioRecord又是如何把数据写进和读出底层声卡的呢？
        Linux声卡设备节点是一个互斥的节点，每次只能被一个应用打开，其他应用尝试open一个已经打开的节点会返回device busy的错误。但是在实际Android使用过程中又存在多个音频数据流同时播放的场景，这是如何做到的呢？
        所有这些疑问都会在AudioFlinger服务中解决。AudioFlinger向下访问AudioHardware，实现输出音频数据，控制音频参数；AudioFlinger向上通过IAudioFinger接口对AudioTrack和AudioRecord供服务。AudioFlinger在Android的音频系统框架中起着承上启下的作用，地位相当告急。
   AudioFlinger创建PlaybackThread/RecordThread服务线程，实现对AHAL层音频节点的独占利用。多个AudioTrack/AudioRecord作为Clinet端同PlaybackThread/RecordThread服务线程交互，Thread实现音频流的mix/snoop，来实现整个音频框架。
  下面通过分析代码，完成对AudioFlinger的深入明白。
二、AudioFlinger初始化

AudioFlinger在audioserver中启动，代码在：
frameworks/av/media/audioserver/main_audioserver.cpp

1. #define LOG_TAG "audioserver"
3. int main(int argc __unused, char **argv)
4. {
5.      ......
6.     这里修改了AudioFling和AudioPolicyService的初始化方式，以解决独立初始化模式下引起的
7.     服务启动时间差，这个时间差会导致部分服务时间检测异常而强行退出。
8.     // Instantiating AudioFlinger (making it public, e.g. through ::initialize())
9.     // and then instantiating AudioPolicy (and making it public)
10.     // leads to situations where AudioFlinger is accessed remotely before
11.     // AudioPolicy is initialized.  Not only might this
12.     // cause inaccurate results, but if AudioPolicy has slow audio HAL
13.     // initialization, it can cause a TimeCheck abort to occur on an AudioFlinger
14.     // call which tries to access AudioPolicy.
15.     //
16.     // We create AudioFlinger and AudioPolicy locally then make it public to ServiceManager.
17.     // This requires both AudioFlinger and AudioPolicy to be in-proc.
18.     //
20.     /* 1、创建AudioFling实例，必须在AudioPolicyService之前，因为aps会访问af中的接口 */
21.     const auto af = sp<AudioFlinger>::make();
22.     const auto afAdapter = sp<AudioFlingerServerAdapter>::make(af);
23.    
24.     /* 2、创建AudioPolicyService实例 */
25.     const auto aps = sp<AudioPolicyService>::make();
27.     /* 3、添加AudioFlinger和AudioPolicyService服务到ServiceManager */   
28.     // Add AudioFlinger and AudioPolicy to ServiceManager.
29.     sp<IServiceManager> sm = defaultServiceManager();
30.     sm->addService(String16(IAudioFlinger::DEFAULT_SERVICE_NAME), afAdapter,
31.             false /* allowIsolated */, IServiceManager::DUMP_FLAG_PRIORITY_DEFAULT);
32.     sm->addService(String16(AudioPolicyService::getServiceName()), aps,
33.             false /* allowIsolated */, IServiceManager::DUMP_FLAG_PRIORITY_DEFAULT);
34.    
35.     // AAudioService should only be used in OC-MR1 and later.
36.     // And only enable the AAudioService if the system MMAP policy explicitly allows it.
37.     std::vector<AudioMMapPolicyInfo> policyInfos;
38.     status_t status = af->getMmapPolicyInfos(
39.             AudioMMapPolicyType::DEFAULT, &policyInfos);
40.     // Initialize aaudio service when querying mmap policy succeeds and
41.     // any of the policy supports MMAP.
42.     if (status == NO_ERROR &&
43.         std::any_of(policyInfos.begin(), policyInfos.end(), [](const auto& info) {
44.                 return info.mmapPolicy == AudioMMapPolicy::AUTO ||
45.                        info.mmapPolicy == AudioMMapPolicy::ALWAYS;
46.         })) {
47.         
48.         /* 4、创建AAudioService实例 */
49.         AAudioService::instantiate();
50.     }
52.     ......
54.     IPCThreadState::self()->joinThreadPool();
55.     }
56. }

复制代码

        主函数里面没有做太多的任务，创建了AudioFlinger和AudioPolicyService两个实例，并将服务注册到servicemanager中，除此之外还检测并创建了AAudioService。

---

1、AudioFlinger实例构建

1. AudioFlinger::AudioFlinger() {
2.     ......
4.     // reset battery stats.
5.     // if the audio service has crashed, battery stats could be left
6.     // in bad state, reset the state upon service start.
7.     BatteryNotifier::getInstance().noteResetAudio();
9.     /* 构建音频device和音效effect实例 */
10.     mDevicesFactoryHal = DevicesFactoryHalInterface::create();
11.     mEffectsFactoryHal = EffectsFactoryHalInterface::create();
12. }

复制代码

AudioFlinger的实例没有做太多的工作：

• 完成电池状态的复位；
  
• 构建音频设备Device工厂实例；
  
• 构建音效Effect工厂实例；
  

1. sp<DevicesFactoryHalInterface> DevicesFactoryHalInterface::create() {
2.       return createPreferredImpl<DevicesFactoryHalInterface>(true /* isDevice */);
3. }
6. void *createPreferredImpl(bool isDevice) {
7.     ......
8.     if (...) {
9.        .....
10.         /*  1、创建AudioHal服务  */
11.         if (createHalService(std::max(*ifaceVersionIt, *siblingVersionIt), isDevice,
12.                              &rawInterface)) {
13.             return rawInterface;
14.         }
15.     return nullptr;
16. }
19. bool createHalService(const AudioHalVersionInfo& version, bool isDevice, void** rawInterface) {
20.     const std::string libName = "libaudiohal@" + version.toVersionString() + ".so";
21.     const std::string factoryFunctionName =
22.             isDevice ? "createIDevicesFactory" : "createIEffectsFactory";
23.     constexpr int dlMode = RTLD_LAZY;
24.     void* handle = nullptr;
25.     dlerror(); // clear
26.     /*  2、打开 libaudiohal@version.so 动态库 */
27.     handle = dlopen(libName.c_str(), dlMode);
28.     ......
29.     void* (*factoryFunction)();
30.     /*3、函数进来后首先初始化factoryFunctionName为createIDevicesFactory， 所以这里获取到的factoryFunction句柄为createIDevicesFactory函数 */
31.     *(void **)(&factoryFunction) = dlsym(handle, factoryFunctionName.c_str());
32.     if (!factoryFunction) {
33.        ......
34.     }
35.     /* 4、调用libaudiohal动态库的createIDevicesFactory函数，完成AHAL句柄的创建 */
36.     *rawInterface = (*factoryFunction)();
37.     return true;
38. }
41. extern "C" __attribute__((visibility("default"))) void* createIDevicesFactory() {
42.     return createIDevicesFactoryImpl();
43. }
45. extern "C" __attribute__((visibility("default"))) void* createIDevicesFactoryImpl() {
46.     auto service = hardware::audio::CPP_VERSION::IDevicesFactory::getService();
47.     return service ? new DevicesFactoryHalHidl(service) : nullptr;
48. }

复制代码

1. DevicesFactoryHalEntry.cpp::createIDevicesFactory() ->
2.     DevicesFactoryHalHidl.cpp::createIDevicesFactoryImpl() ->         
3.          DevicesFactoryHalHidl.cpp::DevicesFactoryHalHidl()

复制代码

具体逻辑这里不做深入的分析了，后面有机会再展开。
        整个AudioFlinger完成初始化之后，并没有做太多的任务。而是作为一个服务运行在系统中，等候客户端与其进行通信：这也符合AF的功能，我只做一个执行者，不做决策。那谁优势决策者呢？别忘了我们还没有分析AudioPolicyService呢。
2、AudioPolicyService实例构建

1. AudioPolicyService::AudioPolicyService()
2.     : BnAudioPolicyService(),
3.       mAudioPolicyManager(NULL),
4.       mAudioPolicyClient(NULL),
5.       mPhoneState(AUDIO_MODE_INVALID),
6.       mCaptureStateNotifier(false),
7.      /* 1 - 初始化mCreateAudioPolicyManager为createAudioPolicyManager */
8.       mCreateAudioPolicyManager(createAudioPolicyManager),
9.       mDestroyAudioPolicyManager(destroyAudioPolicyManager) {
10.       setMinSchedulerPolicy(SCHED_NORMAL, ANDROID_PRIORITY_AUDIO);
11. }
13. /* 在完成实例后，首先调用的函数就是onFirstRef，具体原因请搜索其他文章 */
14. void AudioPolicyService::onFirstRef()
15. {
16.      {
17.         Mutex::Autolock _l(mLock);
19.         /* 2 - 创建两个命令线程，用于接收并处理其他服务发送的命令 */
20.         // start audio commands thread
21.         mAudioCommandThread = new AudioCommandThread(String8("ApmAudio"), this);
22.         // start output activity command thread
23.         mOutputCommandThread = new AudioCommandThread(String8("ApmOutput"), this);
24.         
25.         /* 3 - 构造AudioPolicyClient实例：APC是AF的clinet端，很多任务都是通过跨进程调用到AF中 */
26.         mAudioPolicyClient = new AudioPolicyClient(this);
28.         /* 4 - 加载用户自定义的的AudioPolicyManager实现：libaudiopolicymanagercustom.so */
29.         loadAudioPolicyManager();
31.         /* 5 - 调用createAudioPolicyManager接口 */
32.         mAudioPolicyManager = mCreateAudioPolicyManager(mAudioPolicyClient);
33.     }

复制代码

2.1、createAudioPolicyManager

1. static AudioPolicyInterface* createAudioPolicyManager(AudioPolicyClientInterface *clientInterface)
2. {
3.     AudioPolicyManager *apm = nullptr;
4.     media::AudioPolicyConfig apmConfig;
5.     /* 1、与AHAL交互获取modules信息，以确定是否存在用户自定义的configurable AudioPolicyConfig */
6.     if (status_t status = clientInterface->getAudioPolicyConfig(&apmConfig); status == OK) {
7.         auto config = AudioPolicyConfig::loadFromApmAidlConfigWithFallback(apmConfig);
8.         LOG_ALWAYS_FATAL_IF(config->getEngineLibraryNameSuffix() !=
9.                 AudioPolicyConfig::kDefaultEngineLibraryNameSuffix,
10.                 "Only default engine is currently supported with the AIDL HAL");
11.         apm = new AudioPolicyManager(config,
12.                 loadApmEngineLibraryAndCreateEngine(
13.                         config->getEngineLibraryNameSuffix(), apmConfig.engineConfig),
14.                 clientInterface);
15.     } else {
16.         /* 2、加载系统预定义的audio_policy_configuration配置 */
17.         auto config = AudioPolicyConfig::loadFromApmXmlConfigWithFallback();
18.          /* 3、config->getEngineLibraryNameSuffix()返回的
19.              字符串为“default”(kDefaultEngineLibraryNameSuffix)，
20.              加载libaudiopolicyengine.default.so并完成AudioPolicyManager实例构造 */
21.         apm = new AudioPolicyManager(config,
22.                 loadApmEngineLibraryAndCreateEngine(config->getEngineLibraryNameSuffix()),
23.                 clientInterface);
24.     }
25.     /* 4、初始化AudioPolicyManager */
26.     status_t status = apm->initialize();
27.     if (status != NO_ERROR) {
28.         delete apm;
29.         apm = nullptr;
30.     }
31.     return apm;
32. }
34. // static
35. sp<const AudioPolicyConfig> AudioPolicyConfig::loadFromApmXmlConfigWithFallback(
36.         const std::string& xmlFilePath) {
37.     /* 2.1 - xmlFilePath为空，audio_get_audio_policy_config_file获取到的路径为/vendor/etc/audio_policy_configuration.xml */
38.     const std::string filePath =
39.             xmlFilePath.empty() ? audio_get_audio_policy_config_file() : xmlFilePath;
40.     auto config = sp<AudioPolicyConfig>::make();
41.     /* 2.2 - 解析audio_policy_configuration.xml 文件并获得音频策略配置信息 */
42.     if (status_t status = config->loadFromXml(filePath, false /*forVts*/); status == NO_ERROR) {
43.         return config;
44.     }
45.      /* 2.3 - 不会运行到这里 */
46.     return createDefault();
47. }
50. status_t AudioPolicyConfig::loadFromXml(const std::string& xmlFilePath, bool forVts) {
51.     ...
52.     status_t status = deserializeAudioPolicyFile(xmlFilePath.c_str(), this);
53.     ...
54. }
56. status_t deserializeAudioPolicyFile(const char *fileName, AudioPolicyConfig *config)
57. {
58.     PolicySerializer serializer;
59.     return status = serializer.deserialize(fileName, config);
60. }
62. status_t PolicySerializer::deserialize(const char *configFile, AudioPolicyConfig *config,
63.                                        bool ignoreVendorExtensions)
64. {
65.     /* 完成audio_policy_configuration.xml的解析*/
66.     ......
67. }

复制代码

以Android的car/emulator的audio_policy_configuration.xml文件为例，下图是xml文件剖析后得到的结果，整个剖析按照mixPort，devicePort和route的属性剖析并终极形成HwModule，条理比力简单，读者自行撸代码也能看懂。
 

得益于Android丰富的工具，可以通过 dumpsys media.audio_policy 命令查看终极配置的路由计谋。

1. dumpsys media.audio_policy
2. Supported System Usages:
3.   AUDIO_USAGE_CALL_ASSISTANT, AUDIO_USAGE_EMERGENCY, AUDIO_USAGE_SAFETY, AUDIO_USAGE_VEHICLE_STATUS, AUDIO_USAGE_ANNOUNCEMENT
4. UID Policy:
5.         mObserverRegistered=True
6.         Assistants UIDs:
7.                 No UIDs present.
8.         Active Assistants UIDs:
9.                 No UIDs present.
10.         Accessibility UIDs:
11.                 No UIDs present.
12.         Input Method Service UID=1010067
13.         Is RTT Enabled: False
14. AudioCommandThread: 0xb400007416624090
15. - Commands:
16.    Command Time        Wait pParam
17.   Last Command
18.    05      092141.687  0    0xb40000739666f2e0
19. OutputCommandThread: 0xb400007416625950
20. - Commands:
21.    Command Time        Wait pParam
22.   Last Command
23.    09      092141.966  0    0xb400007306620340
25. AudioPolicyManager Dump: 0xb4000074466246b0
26. Primary Output I/O handle: 13
27. Phone state: AUDIO_MODE_NORMAL
28. Force use for communications: 0
29. Force use for media: 10
30. Force use for record: 0
31. Force use for dock: 9
32. Force use for system: 0
33. Force use for HDMI system audio: 0
34. Force use for encoded surround output: 0
35. Force use for vibrate ringing: 0
36. TTS output not available
37. Master mono: off
38. Communication Strategy id: 14
39. Config source: /vendor/etc/audio_policy_configuration.xml
41. Available output devices (17):
42.   1. Port ID: 50; "bus1000_mirror_device"; {AUDIO_DEVICE_OUT_BUS, @:bus1000_mirror_device}
43.      Encapsulation modes: 0, metadata types: 0
44.    - Profiles (1):
45.       1. ""; AUDIO_FORMAT_PCM_16_BIT (0x1)
46.          sampling rates: 48000
47.          channel masks: 0x0003
48.          AUDIO_ENCAPSULATION_TYPE_NONE
49.      - gains:
50.        Gain 1:
51.        - mode: 00000001
52.        - channel_mask: 00000000
53.        - min_value: -3200 mB
54.        - max_value: 600 mB
55.        - default_value: 0 mB
56.        - step_value: 100 mB
57.        - min_ramp_ms: 0 ms
58.        - max_ramp_ms: 0 ms
59.   2. Port ID: 47; "bus211_audio_zone_2"; {AUDIO_DEVICE_OUT_BUS, @:bus211_audio_zone_2}
60.      Encapsulation modes: 0, metadata types: 0
61.    - Profiles (1):
62.       1. ""; AUDIO_FORMAT_PCM_16_BIT (0x1)
63.          sampling rates: 48000
64.          channel masks: 0x0003
65.          AUDIO_ENCAPSULATION_TYPE_NONE
66.      - gains:
67.        Gain 1:
68.        - mode: 00000001
69.        - channel_mask: 00000000
70.        - min_value: -3200 mB
71.        - max_value: 600 mB
72.        - default_value: 0 mB
73.        - step_value: 100 mB
74.        - min_ramp_ms: 0 ms
75.        - max_ramp_ms: 0 ms
76.   ...
77.   6. Port ID: 35; "bus111_audio_zone_1"; {AUDIO_DEVICE_OUT_BUS, @:bus111_audio_zone_1}
78.      Encapsulation modes: 0, metadata types: 0
79.    - Profiles (1):
80.       1. ""; AUDIO_FORMAT_PCM_16_BIT (0x1)
81.          sampling rates: 48000
82.          channel masks: 0x0003
83.          AUDIO_ENCAPSULATION_TYPE_NONE
84.      - gains:
85.        Gain 1:
86.        - mode: 00000001
87.        - channel_mask: 00000000
88.        - min_value: -3200 mB
89.        - max_value: 600 mB
90.        - default_value: 0 mB
91.        - step_value: 100 mB
92.        - min_ramp_ms: 0 ms
93.        - max_ramp_ms: 0 ms
94.    ...
95.   10. Port ID: 23; "bus7_system_sound_out"; {AUDIO_DEVICE_OUT_BUS, @:bus7_system_sound_out}
96.       Encapsulation modes: 0, metadata types: 0
97.     - Profiles (1):
98.        1. ""; AUDIO_FORMAT_PCM_16_BIT (0x1)
99.           sampling rates: 48000
100.           channel masks: 0x0003
101.           AUDIO_ENCAPSULATION_TYPE_NONE
102.       - gains:
103.         Gain 1:
104.         - mode: 00000001
105.         - channel_mask: 00000000
106.         - min_value: -3200 mB
107.         - max_value: 600 mB
108.         - default_value: 0 mB
109.         - step_value: 100 mB
110.         - min_ramp_ms: 0 ms
111.         - max_ramp_ms: 0 ms
112.   17. Port ID: 2; "bus0_media_out"; {AUDIO_DEVICE_OUT_BUS, @:bus0_media_out}
113.       Encapsulation modes: 0, metadata types: 0
114.     - Profiles (1):
115.        1. ""; AUDIO_FORMAT_PCM_16_BIT (0x1)
116.           sampling rates: 48000
117.           channel masks: 0x0003
118.           AUDIO_ENCAPSULATION_TYPE_NONE
119.       - gains:
120.         Gain 1:
121.         - mode: 00000001
122.         - channel_mask: 00000000
123.         - min_value: -3200 mB
124.         - max_value: 600 mB
125.         - default_value: 0 mB
126.         - step_value: 100 mB
127.         - min_ramp_ms: 0 ms
128.         - max_ramp_ms: 0 ms
130. Available input devices (7):
131.   1. Port ID: 53; "Built-In Mic"; {AUDIO_DEVICE_IN_BUILTIN_MIC, @:Built-In Mic}
132.      Encapsulation modes: 0, metadata types: 0
133.    - Profiles (1):
134.       1. ""; AUDIO_FORMAT_PCM_16_BIT (0x1)
135.          sampling rates: 8000, 11025, 12000, 16000, 22050, 24000, 32000, 44100, 48000
136.          channel masks: 0x000c, 0x0010, 0x0030
137.          AUDIO_ENCAPSULATION_TYPE_NONE
138.   2. Port ID: 54; "Built-In Back Mic"; {AUDIO_DEVICE_IN_BACK_MIC, @:Built-In Back Mic}
139.      Encapsulation modes: 0, metadata types: 0
140.    - Profiles (1):
141.       1. ""; AUDIO_FORMAT_PCM_16_BIT (0x1)
142.          sampling rates: 8000, 11025, 12000, 16000, 22050, 24000, 32000, 44100, 48000
143.          channel masks: 0x000c, 0x0010, 0x0030
144.          AUDIO_ENCAPSULATION_TYPE_NONE
145.      ...
147. Hardware modules (3):
148.   1. Handle: 10; "primary"
149.    - Output MixPorts (17):
150.       1. "mixport_bus0_media_out"; 0x0002 (AUDIO_OUTPUT_FLAG_PRIMARY)
151.        - Profiles (1):
152.           1. ""; AUDIO_FORMAT_PCM_16_BIT (0x1)
153.              sampling rates: 48000
154.              channel masks: 0x0003
155.              AUDIO_ENCAPSULATION_TYPE_NONE
156.        - Supported devices (1):
157.         1. Port ID: 2; "bus0_media_out"; {AUDIO_DEVICE_OUT_BUS, @:bus0_media_out}
158.            Encapsulation modes: 0, metadata types: 0
159.        - maxOpenCount: 1; curOpenCount: 1
160.        - maxActiveCount: 1; curActiveCount: 0
161.        - recommendedMuteDurationMs: 0 ms
162.       2. "mixport_bus1_navigation_out"; 0x0000 (AUDIO_OUTPUT_FLAG_NONE)
163.        - Profiles (1):
164.           1. ""; AUDIO_FORMAT_PCM_16_BIT (0x1)
165.              sampling rates: 48000
166.              channel masks: 0x0003
167.              AUDIO_ENCAPSULATION_TYPE_NONE
168.        - Supported devices (1):
169.         1. Port ID: 5; "bus1_navigation_out"; {AUDIO_DEVICE_OUT_BUS, @:bus1_navigation_out}
170.            Encapsulation modes: 0, metadata types: 0
171.        - maxOpenCount: 1; curOpenCount: 1
172.        - maxActiveCount: 1; curActiveCount: 0
173.        - recommendedMuteDurationMs: 0 ms
174.       ...
175.    - Input MixPorts (4):
176.       1. "primary input"; 0x0000 (AUDIO_INPUT_FLAG_NONE)
177.        - Profiles (1):
178.           1. ""; AUDIO_FORMAT_PCM_16_BIT (0x1)
179.              sampling rates: 8000, 11025, 12000, 16000, 22050, 24000, 32000, 44100, 48000
180.              channel masks: 0x000c, 0x0010, 0x0030
181.              AUDIO_ENCAPSULATION_TYPE_NONE
182.        - Supported devices (3):
183.         1. Port ID: 53; "Built-In Mic"; {AUDIO_DEVICE_IN_BUILTIN_MIC, @:Built-In Mic}
184.            Encapsulation modes: 0, metadata types: 0
185.         2. Port ID: 54; "Built-In Back Mic"; {AUDIO_DEVICE_IN_BACK_MIC, @:Built-In Back Mic}
186.            Encapsulation modes: 0, metadata types: 0
187.         3. Port ID: 55; "Echo-Reference Mic"; {AUDIO_DEVICE_IN_ECHO_REFERENCE, @:Echo-Reference Mic}
188.            Encapsulation modes: 0, metadata types: 0
189.        - maxOpenCount: 1; curOpenCount: 0
190.        - maxActiveCount: 1; curActiveCount: 0
191.        - recommendedMuteDurationMs: 0 ms
192.       ...
193.    - Declared devices (23):
194.     1. Port ID: 50; "bus1000_mirror_device"; {AUDIO_DEVICE_OUT_BUS, @:bus1000_mirror_device}
195.        Encapsulation modes: 0, metadata types: 0
196.      - Profiles (1):
197.         1. ""; AUDIO_FORMAT_PCM_16_BIT (0x1)
198.            sampling rates: 48000
199.            channel masks: 0x0003
200.            AUDIO_ENCAPSULATION_TYPE_NONE
201.        - gains:
202.          Gain 1:
203.          - mode: 00000001
204.          - channel_mask: 00000000
205.          - min_value: -3200 mB
206.          - max_value: 600 mB
207.          - default_value: 0 mB
208.          - step_value: 100 mB
209.          - min_ramp_ms: 0 ms
210.          - max_ramp_ms: 0 ms
211.     ...
212.    - Audio Routes (21):
213.       1. Mix; Sink: "bus0_media_out"
214.          Sources: "mixport_bus0_media_out"
215.       2. Mix; Sink: "bus1_navigation_out"
216.          Sources: "mixport_bus1_navigation_out"
217.       3. Mix; Sink: "bus2_voice_command_out"
218.          Sources: "mixport_bus2_voice_command_out"
219.      ...
220.   

复制代码

2.2 AudioPolicyManager实例构建 

1. static AudioPolicyInterface* createAudioPolicyManager(AudioPolicyClientInterface *clientInterface)
2. {
3.     ...
4.         apm = new AudioPolicyManager(config,
5.                 loadApmEngineLibraryAndCreateEngine(config->getEngineLibraryNameSuffix()),
6.                 clientInterface);
7. }

复制代码

2.2.1  loadApmEngineLibraryAndCreateEngine 

在2.1章的第3步中config->getEngineLibraryNameSuffix()返回的字符串为“default”(kDefaultEngineLibraryNameSuffix)，以是loadApmEngineLibraryAndCreateEngine传入的参数为（"default”,null)。

1. EngineInstance loadApmEngineLibraryAndCreateEngine(const std::string& librarySuffix,
2.         const std::string& configXmlFilePath)
3. {
4.     /* 1 - 加载libaudiopolicyengine.default.so，获取createEngineInstance句柄并赋值给mCreateEngineInstance */
5.     auto engLib = EngineLibrary::load(librarySuffix);
6.     if (!engLib) {
7.         ALOGE("%s: Failed to load the engine library, suffix "%s"",
8.                 __func__, librarySuffix.c_str());
9.         return nullptr;
10.     }
12.     /* 2 - 创建ape，这里的configXmlFilePath为空 */
13.     auto engine = engLib->createEngineUsingXmlConfig(configXmlFilePath);
14.     if (engine == nullptr) {
15.         ALOGE("%s: Failed to instantiate the APM engine", __func__);
16.         return nullptr;
17.     }
18.     return engine;
19. }
22. EngineInstance EngineLibrary::createEngineUsingXmlConfig(const std::string& xmlFilePath)
23. {
24.     auto instance = createEngine();
25.     if (instance != nullptr) {
26.         /* 2.1 - xmlFilePath为空，加载engine 配置文件*/
27.         if (status_t status = instance->loadFromXmlConfigWithFallback(xmlFilePath);
28.                 status == OK) {
29.             return instance;
30.         } else {
31.             ALOGE("%s: loading of the engine config with XML configuration file "%s" failed: %d",
32.                     __func__, xmlFilePath.empty() ? "default" : xmlFilePath.c_str(), status);
33.         }
34.     }
35.     return nullptr;
36. }
38. status_t Engine::loadFromXmlConfigWithFallback(const std::string& xmlFilePath) {
39.     return loadWithFallback(xmlFilePath);
40. }
42. template<typename T>
43. status_t Engine::loadWithFallback(const T& configSource) {
44.     /* 2.1.1 - 加载engineConfig::DEFAULT_PATH("/vendor/etc/audio_policy_engine_configuration.xml") 下的ape配置文件 */
45.     auto result = EngineBase::loadAudioPolicyEngineConfig(configSource);
46.     auto legacyStrategy = getLegacyStrategy();
47.     for (const auto &strategy : legacyStrategy) {
48.         mLegacyStrategyMap[getProductStrategyByName(strategy.name)] = strategy.id;
49.     }
51.     return OK;
52. }

复制代码

        APE的调用逻辑比力繁琐，但是原理很清楚：起首确定是否存在用户自定义的configurable engine，如果不存在则使用default engine；确认之后渐渐调用并剖析相干的APE配置文件。譬如车机的APE配置如下：/frameworks/av/services/audiopolicy/engineconfigurable/config/example/automotive/audio_policy_engine_configuration.xml

1. <configuration version="1.0" xmlns:xi="http://www.w3.org/2001/XInclude">
3.     <xi:include href="audio_policy_engine_product_strategies.xml"/>
4.     <xi:include href="audio_policy_engine_criterion_types.xml"/>
5.     <xi:include href="audio_policy_engine_criteria.xml"/>
6.     <xi:include href="audio_policy_engine_volumes.xml"/>
8. </configuration>

复制代码

具体内容在后续使用的过程中分析。
2.3、AudioPolicyManager::initialize

        回到2.1的第4步，也是最后一步：AudioPolicyManager的initialize。

2. status_t AudioPolicyManager::initialize() {
3.     /* 1 - 这里的mEngine就是上面分析的APE，肯定非空了 */
4.     if (mEngine == nullptr) {
5.         return NO_INIT;
6.     }
8.     /* 2 - 将APM设置为Engine的观察器（mApmObserver=APM），
9.            允许Engine检索Devices、Streams、HwModules的信息：
10.            Engine通过APM获取/设置相关信息 */
11.     mEngine->setObserver(this);
12.    
13.     /* 3 - 检测Engine是否被正确初始化：mApmObserver!=null, 这不是废话吗？
14.            前面刚设置完成*/
15.     status_t status = mEngine->initCheck();
16.     if (status != NO_ERROR) {
17.         LOG_FATAL("Policy engine not initialized(err=%d)", status);
18.         return status;
19.     }
21.     // The actual device selection cache will be updated when calling `updateDevicesAndOutputs`
22.     // at the end of this function.
23.     /* 4 - 使用产品策略初始化设备，这些设备的策略会在音频模块初始化后更新。*/
24.     mEngine->initializeDeviceSelectionCache();
25.     mCommunnicationStrategy = mEngine->getProductStrategyForAttributes(
26.         mEngine->getAttributesForStreamType(AUDIO_STREAM_VOICE_CALL));
28.      /* 5 - 有音频模块可用（实际上是首次初始化），完成模块的初始化。*/
29.     // after parsing the config, mConfig contain all known devices;
30.     // open all output streams needed to access attached devices
31.     onNewAudioModulesAvailableInt(nullptr /*newDevices*/);
33.     // make sure default device is reachable
34.     if (const auto defaultOutputDevice = mConfig->getDefaultOutputDevice();
35.             defaultOutputDevice == nullptr ||
36.             !mAvailableOutputDevices.contains(defaultOutputDevice)) {
37.         ALOGE_IF(defaultOutputDevice != nullptr, "Default device %s is unreachable",
38.                  defaultOutputDevice->toString().c_str());
39.         status = NO_INIT;
40.     }
41.       
42.     /* 6 - 第5步中完成了整个音频模块的初始化，跟新相关信息到Cache缓存中。
43.        在音频系统稳定后，为了加快处理速度，引擎会缓存基于AudioAttribute/Streams的设备选择。当一个设备连接时，Android的模式会发生变化，引擎会收到通知并更新缓存。如果某个输出上音频流的启动/停止可能影响通知时，引擎需要通过updateDevicesAndOutputs函数更新缓存。*/
44.     updateDevicesAndOutputs();
46.     return status;
47. }

复制代码

2.3.1、AudioPolicyManager:nNewAudioModulesAvailableInt

        经过2.2章节的分析，我们已经清楚系统已经完成xml配置文件的剖析并形成HwModule的多维数据，该函数就是根据先前分析的数据完成音频系统的最告急的一步：调用AF->AHAL完成设备节
点的open，并在AF中创建数据流播放/录制线程，为后面AudioTrack做好预备

1. void AudioPolicyManager::onNewAudioModulesAvailableInt(DeviceVector *newDevices)
2. {
3.     ALOGD("AudioAnt: onNewAudioModulesAvailableInt enter ......");
5.     /* 1. 遍历audio_policy_configuration.xml中配置的所有模块 */
6.     for (const auto& hwModule : mConfig->getHwModules()) {
7.         if (std::find(mHwModules.begin(), mHwModules.end(), hwModule) != mHwModules.end()) {
8.             continue;
9.         }
10.         if (hwModule->getHandle() == AUDIO_MODULE_HANDLE_NONE) {
11.             ALOGD("AudioAnt: loadHwModule with module name %s", hwModule->getName());
12.             /* 2. 加载相关module，获取module的句柄handle */
13.             if (audio_module_handle_t handle = mpClientInterface->loadHwModule(hwModule->getName());
14.                     handle != AUDIO_MODULE_HANDLE_NONE) {
15.                 hwModule->setHandle(handle);
16.             } else {
17.                 ALOGW("could not load HW module %s", hwModule->getName());
18.                 continue;
19.             }
20.         }
22.         /* 3. 如果xml中配置的module存在，则记录到mHwModules容器中以便后面访问使用 */
23.         mHwModules.push_back(hwModule);
25.         /* 4. 遍历当前模块支持的profiles */
26.         for (const auto& outProfile : hwModule->getOutputProfiles()) {
27.             ALOGD("AudioAnt: outProfile with name: %s", outProfile->getTagName().c_str());
28.             
29.             /* 5. 获取当前profile支持的设备 */
30.             const DeviceVector &supportedDevices = outProfile->getSupportedDevices();
31.             for(auto& device: supportedDevices) {
32.                 ALOGD("AudioAnt: supportedDevices with name: %s", device->getTagName().c_str());
33.             }
35.             DeviceVector availProfileDevices = supportedDevices.filter(mConfig->getOutputDevices());
36.             sp<DeviceDescriptor> supportedDevice = 0;
37.             if (supportedDevices.contains(mConfig->getDefaultOutputDevice())) {
38.                 supportedDevice = mConfig->getDefaultOutputDevice();
39.             } else {
40.                 // choose first device present in profile's SupportedDevices also part of
41.                 // mAvailableOutputDevices.
42.                 if (availProfileDevices.isEmpty()) {
43.                     continue;
44.                 }
45.                 supportedDevice = availProfileDevices.itemAt(0);
46.             }
47.             if (!mConfig->getOutputDevices().contains(supportedDevice)) {
48.                 continue;
49.             }
50.             sp<SwAudioOutputDescriptor> outputDesc = new SwAudioOutputDescriptor(outProfile, mpClientInterface);
51.             audio_io_handle_t output = AUDIO_IO_HANDLE_NONE;
52.             
53.             /* 6. 打开设备，涉及AHAL的open操作，并在AF中创建相关播放/录制线程 */
54.             status_t status = outputDesc->open(nullptr /* halConfig */, nullptr /* mixerConfig */,
55.                                                DeviceVector(supportedDevice),
56.                                                AUDIO_STREAM_DEFAULT,
57.                                                AUDIO_OUTPUT_FLAG_NONE, &output);
58.             if (status != NO_ERROR) {
59.                 ALOGW("Cannot open output stream for devices %s on hw module %s",
60.                       supportedDevice->toString().c_str(), hwModule->getName());
61.                 continue;
62.             }
63.             for (const auto &device : availProfileDevices) {
64.                 ALOGD("AudioAnt: availProfileDevices with name: %s", device->getTagName().c_str());
65.                 // give a valid ID to an attached device once confirmed it is reachable
66.                 if (!device->isAttached()) {
67.                     device->attach(hwModule);
68.                     mAvailableOutputDevices.add(device);
69.                     device->setEncapsulationInfoFromHal(mpClientInterface);
70.                     if (newDevices) newDevices->add(device);
71.                     setEngineDeviceConnectionState(device, AUDIO_POLICY_DEVICE_STATE_AVAILABLE);
72.                 }
73.             }
74.             if (mPrimaryOutput == nullptr &&
75.                     outProfile->getFlags() & AUDIO_OUTPUT_FLAG_PRIMARY) {
76.                 mPrimaryOutput = outputDesc;
77.             }
78.             if ((outProfile->getFlags() & AUDIO_OUTPUT_FLAG_DIRECT) != 0) {
79.                 outputDesc->close();
80.             } else {
81.                 addOutput(output, outputDesc);
82.                 setOutputDevices(outputDesc,
83.                                  DeviceVector(supportedDevice),
84.                                  true,
85.                                  0,
86.                                  NULL);
87.             }
88.         }
89.         
90.         /* 7. 同输出设置一样的操作逻辑，不赘述 */
91.         // open input streams needed to access attached devices to validate
92.         // mAvailableInputDevices list
93.         for (const auto& inProfile : hwModule->getInputProfiles()) {
94.         }
95.     }
97.     /* 8. 如果有声场相关输出，在使用之前先关闭它 */
98.     // Check if spatializer outputs can be closed until used.
99.     // mOutputs vector never contains duplicated outputs at this point.
100.     std::vector<audio_io_handle_t> outputsClosed;
101.     for (size_t i = 0; i < mOutputs.size(); i++) {
102.         sp<SwAudioOutputDescriptor> desc = mOutputs.valueAt(i);
103.         if ((desc->mFlags & AUDIO_OUTPUT_FLAG_SPATIALIZER) != 0
104.                 && !isOutputOnlyAvailableRouteToSomeDevice(desc)) {
105.             outputsClosed.push_back(desc->mIoHandle);
106.             desc->close();
107.         }
108.     }
109.     for (auto output : outputsClosed) {
110.         removeOutput(output);
111.     }
112. }

复制代码

        终极的利用输出如下：

1. 06-06 21:51:06.077 16876 16876 D APM_AudioPolicyManager: AudioAnt: onNewAudioModulesAvailableInt enter ......
2. 06-06 21:51:06.077 16876 16876 D APM_AudioPolicyManager: AudioAnt: loadHwModule with module name primary
3. 06-06 21:51:09.092 16876 16876 D APM_AudioPolicyManager: AudioAnt: outProfile with name: mixport_bus0_media_out
4. 06-06 21:51:09.092 16876 16876 D APM_AudioPolicyManager: AudioAnt: supportedDevices with name: bus0_media_out
5. 06-06 21:51:09.096 16876 16876 I AudioFlinger: AudioAnt: openOutput() this 0xb40000738661f010, module 10 Device AUDIO_DEVICE_OUT_BUS, @:bus0_media_out, SamplingRate
6. 48000, Format 0x000001, Channels 0x3, flags 0x2
7. 06-06 21:51:09.266 16876 16876 D APM_AudioPolicyManager: AudioAnt: availProfileDevices with name: bus0_media_out
8. 06-06 21:51:09.316 16876 16876 D APM_AudioPolicyManager: AudioAnt: outProfile with name: mixport_bus1_navigation_out
9. 06-06 21:51:09.316 16876 16876 D APM_AudioPolicyManager: AudioAnt: supportedDevices with name: bus1_navigation_out
10. 06-06 21:51:09.316 16876 16876 I AudioFlinger: AudioAnt: openOutput() this 0xb40000738661f010, module 10 Device AUDIO_DEVICE_OUT_BUS, @:bus1_navigation_out, Samplin
11. gRate 48000, Format 0x000001, Channels 0x3, flags 0
12. 06-06 21:51:09.327 16876 16876 D APM_AudioPolicyManager: AudioAnt: availProfileDevices with name: bus1_navigation_out
13. 06-06 21:51:09.354 16876 16876 D APM_AudioPolicyManager: AudioAnt: outProfile with name: mixport_bus2_voice_command_out
14. 06-06 21:51:09.354 16876 16876 D APM_AudioPolicyManager: AudioAnt: supportedDevices with name: bus2_voice_command_out
15. 06-06 21:51:09.354 16876 16876 I AudioFlinger: AudioAnt: openOutput() this 0xb40000738661f010, module 10 Device AUDIO_DEVICE_OUT_BUS, @:bus2_voice_command_out, Samp
16. lingRate 48000, Format 0x000001, Channels 0x3, flags 0
17. 06-06 21:51:09.365 16876 16876 D APM_AudioPolicyManager: AudioAnt: availProfileDevices with name: bus2_voice_command_out
18. 06-06 21:51:09.391 16876 16876 D APM_AudioPolicyManager: AudioAnt: outProfile with name: mixport_bus3_call_ring_out
19. 06-06 21:51:09.392 16876 16876 D APM_AudioPolicyManager: AudioAnt: supportedDevices with name: bus3_call_ring_out
20. 06-06 21:51:09.392 16876 16876 I AudioFlinger: AudioAnt: openOutput() this 0xb40000738661f010, module 10 Device AUDIO_DEVICE_OUT_BUS, @:bus3_call_ring_out, Sampling
21. Rate 48000, Format 0x000001, Channels 0x3, flags 0
22. 06-06 21:51:09.403 16876 16876 D APM_AudioPolicyManager: AudioAnt: availProfileDevices with name: bus3_call_ring_out
23. 06-06 21:51:09.416 16876 16876 D APM_AudioPolicyManager: AudioAnt: outProfile with name: mixport_bus4_call_out
24. 06-06 21:51:09.416 16876 16876 D APM_AudioPolicyManager: AudioAnt: supportedDevices with name: bus4_call_out
25. 06-06 21:51:09.416 16876 16876 I AudioFlinger: AudioAnt: openOutput() this 0xb40000738661f010, module 10 Device AUDIO_DEVICE_OUT_BUS, @:bus4_call_out, SamplingRate
26. 48000, Format 0x000001, Channels 0x3, flags 0
27. 06-06 21:51:09.423 16876 16876 D APM_AudioPolicyManager: AudioAnt: availProfileDevices with name: bus4_call_out
28. 06-06 21:51:09.441 16876 16876 D APM_AudioPolicyManager: AudioAnt: outProfile with name: mixport_bus5_alarm_out
29. 06-06 21:51:09.442 16876 16876 D APM_AudioPolicyManager: AudioAnt: supportedDevices with name: bus5_alarm_out
30. 06-06 21:51:09.442 16876 16876 I AudioFlinger: AudioAnt: openOutput() this 0xb40000738661f010, module 10 Device AUDIO_DEVICE_OUT_BUS, @:bus5_alarm_out, SamplingRate
31. 48000, Format 0x000001, Channels 0x3, flags 0
32. 06-06 21:51:09.451 16876 16876 D APM_AudioPolicyManager: AudioAnt: availProfileDevices with name: bus5_alarm_out
33. 06-06 21:51:09.463 16876 16876 D APM_AudioPolicyManager: AudioAnt: outProfile with name: mixport_bus6_notification_out
34. 06-06 21:51:09.463 16876 16876 D APM_AudioPolicyManager: AudioAnt: supportedDevices with name: bus6_notification_out
35. 06-06 21:51:09.463 16876 16876 I AudioFlinger: AudioAnt: openOutput() this 0xb40000738661f010, module 10 Device AUDIO_DEVICE_OUT_BUS, @:bus6_notification_out, Sampl
36. ingRate 48000, Format 0x000001, Channels 0x3, flags 0
37. 06-06 21:51:09.471 16876 16876 D APM_AudioPolicyManager: AudioAnt: availProfileDevices with name: bus6_notification_out
38. 06-06 21:51:09.491 16876 16876 D APM_AudioPolicyManager: AudioAnt: outProfile with name: mixport_bus7_system_sound_out
39. 06-06 21:51:09.491 16876 16876 D APM_AudioPolicyManager: AudioAnt: supportedDevices with name: bus7_system_sound_out
40. 06-06 21:51:09.491 16876 16876 I AudioFlinger: AudioAnt: openOutput() this 0xb40000738661f010, module 10 Device AUDIO_DEVICE_OUT_BUS, @:bus7_system_sound_out, Sampl
41. ingRate 48000, Format 0x000001, Channels 0x3, flags 0
42. 06-06 21:51:09.499 16876 16876 D APM_AudioPolicyManager: AudioAnt: availProfileDevices with name: bus7_system_sound_out
43. 06-06 21:51:09.519 16876 16876 D APM_AudioPolicyManager: AudioAnt: outProfile with name: mixport_bus100_audio_zone_1
44. 06-06 21:51:09.519 16876 16876 D APM_AudioPolicyManager: AudioAnt: supportedDevices with name: bus100_audio_zone_1
45. 06-06 21:51:09.520 16876 16876 I AudioFlinger: AudioAnt: openOutput() this 0xb40000738661f010, module 10 Device AUDIO_DEVICE_OUT_BUS, @:bus100_audio_zone_1, Samplin
46. gRate 48000, Format 0x000001, Channels 0x3, flags 0
47. 06-06 21:51:09.530 16876 16876 D APM_AudioPolicyManager: AudioAnt: availProfileDevices with name: bus100_audio_zone_1
48. 06-06 21:51:09.557 16876 16876 D APM_AudioPolicyManager: AudioAnt: outProfile with name: mixport_bus101_audio_zone_1
49. 06-06 21:51:09.558 16876 16876 D APM_AudioPolicyManager: AudioAnt: supportedDevices with name: bus101_audio_zone_1
50. 06-06 21:51:09.558 16876 16876 I AudioFlinger: AudioAnt: openOutput() this 0xb40000738661f010, module 10 Device AUDIO_DEVICE_OUT_BUS, @:bus101_audio_zone_1, Samplin
51. gRate 48000, Format 0x000001, Channels 0x3, flags 0
52. 06-06 21:51:09.569 16876 16876 D APM_AudioPolicyManager: AudioAnt: availProfileDevices with name: bus101_audio_zone_1
53. 06-06 21:51:09.598 16876 16876 D APM_AudioPolicyManager: AudioAnt: outProfile with name: mixport_bus110_audio_zone_1
54. 06-06 21:51:09.598 16876 16876 D APM_AudioPolicyManager: AudioAnt: supportedDevices with name: bus110_audio_zone_1
55. 06-06 21:51:09.598 16876 16876 I AudioFlinger: AudioAnt: openOutput() this 0xb40000738661f010, module 10 Device AUDIO_DEVICE_OUT_BUS, @:bus110_audio_zone_1, Samplin
56. gRate 48000, Format 0x000001, Channels 0x3, flags 0
57. 06-06 21:51:09.604 16876 16876 D APM_AudioPolicyManager: AudioAnt: availProfileDevices with name: bus110_audio_zone_1
58. 06-06 21:51:09.630 16876 16876 D APM_AudioPolicyManager: AudioAnt: outProfile with name: mixport_bus111_audio_zone_1
59. 06-06 21:51:09.630 16876 16876 D APM_AudioPolicyManager: AudioAnt: supportedDevices with name: bus111_audio_zone_1
60. 06-06 21:51:09.631 16876 16876 I AudioFlinger: AudioAnt: openOutput() this 0xb40000738661f010, module 10 Device AUDIO_DEVICE_OUT_BUS, @:bus111_audio_zone_1, Samplin
61. gRate 48000, Format 0x000001, Channels 0x3, flags 0
62. 06-06 21:51:09.642 16876 16876 D APM_AudioPolicyManager: AudioAnt: availProfileDevices with name: bus111_audio_zone_1
63. 06-06 21:51:09.678 16876 16876 D APM_AudioPolicyManager: AudioAnt: outProfile with name: mixport_bus200_audio_zone_2
64. 06-06 21:51:09.679 16876 16876 D APM_AudioPolicyManager: AudioAnt: supportedDevices with name: bus200_audio_zone_2
65. 06-06 21:51:09.679 16876 16876 I AudioFlinger: AudioAnt: openOutput() this 0xb40000738661f010, module 10 Device AUDIO_DEVICE_OUT_BUS, @:bus200_audio_zone_2, Samplin
66. gRate 48000, Format 0x000001, Channels 0x3, flags 0
67. 06-06 21:51:09.690 16876 16876 D APM_AudioPolicyManager: AudioAnt: availProfileDevices with name: bus200_audio_zone_2
68. 06-06 21:51:09.713 16876 16876 D APM_AudioPolicyManager: AudioAnt: outProfile with name: mixport_bus201_audio_zone_2
69. 06-06 21:51:09.713 16876 16876 D APM_AudioPolicyManager: AudioAnt: supportedDevices with name: bus201_audio_zone_2
70. 06-06 21:51:09.713 16876 16876 I AudioFlinger: AudioAnt: openOutput() this 0xb40000738661f010, module 10 Device AUDIO_DEVICE_OUT_BUS, @:bus201_audio_zone_2, Samplin
71. gRate 48000, Format 0x000001, Channels 0x3, flags 0
72. 06-06 21:51:09.720 16876 16876 D APM_AudioPolicyManager: AudioAnt: availProfileDevices with name: bus201_audio_zone_2
73. 06-06 21:51:09.742 16876 16876 D APM_AudioPolicyManager: AudioAnt: outProfile with name: mixport_bus210_audio_zone_2
74. 06-06 21:51:09.742 16876 16876 D APM_AudioPolicyManager: AudioAnt: supportedDevices with name: bus210_audio_zone_2
75. 06-06 21:51:09.742 16876 16876 I AudioFlinger: AudioAnt: openOutput() this 0xb40000738661f010, module 10 Device AUDIO_DEVICE_OUT_BUS, @:bus210_audio_zone_2, Samplin
76. gRate 48000, Format 0x000001, Channels 0x3, flags 0
77. 06-06 21:51:09.747 16876 16876 D APM_AudioPolicyManager: AudioAnt: availProfileDevices with name: bus210_audio_zone_2
78. 06-06 21:51:09.773 16876 16876 D APM_AudioPolicyManager: AudioAnt: outProfile with name: mixport_bus211_audio_zone_2
79. 06-06 21:51:09.773 16876 16876 D APM_AudioPolicyManager: AudioAnt: supportedDevices with name: bus211_audio_zone_2
80. 06-06 21:51:09.773 16876 16876 I AudioFlinger: AudioAnt: openOutput() this 0xb40000738661f010, module 10 Device AUDIO_DEVICE_OUT_BUS, @:bus211_audio_zone_2, Samplin
81. gRate 48000, Format 0x000001, Channels 0x3, flags 0
82. 06-06 21:51:09.779 16876 16876 D APM_AudioPolicyManager: AudioAnt: availProfileDevices with name: bus211_audio_zone_2
83. 06-06 21:51:09.810 16876 16876 D APM_AudioPolicyManager: AudioAnt: outProfile with name: mixport_bus1000_mirror_device
84. 06-06 21:51:09.810 16876 16876 D APM_AudioPolicyManager: AudioAnt: supportedDevices with name: bus1000_mirror_device
85. 06-06 21:51:09.810 16876 16876 I AudioFlinger: AudioAnt: openOutput() this 0xb40000738661f010, module 10 Device AUDIO_DEVICE_OUT_BUS, @:bus1000_mirror_device, Sampl
86. ingRate 48000, Format 0x000001, Channels 0x3, flags 0
87. 06-06 21:51:09.817 16876 16876 D APM_AudioPolicyManager: AudioAnt: availProfileDevices with name: bus1000_mirror_device
88. 06-06 21:51:09.881 16876 16876 D APM_AudioPolicyManager: AudioAnt: loadHwModule with module name a2dp
89. 06-06 21:51:09.882 16876 16876 D APM_AudioPolicyManager: AudioAnt: loadHwModule with module name usb
90. 06-06 21:51:09.884 16876 16876 D APM_AudioPolicyManager: AudioAnt: outProfile with name: usb_accessory output
91. 06-06 21:51:09.884 16876 16876 D APM_AudioPolicyManager: AudioAnt: supportedDevices with name: USB Host Out
92. 06-06 21:51:09.884 16876 16876 D APM_AudioPolicyManager: AudioAnt: outProfile with name: usb_device output
93. 06-06 21:51:09.884 16876 16876 D APM_AudioPolicyManager: AudioAnt: supportedDevices with name: USB Device Out
94. 06-06 21:51:09.885 16876 16876 D APM_AudioPolicyManager: AudioAnt: supportedDevices with name: USB Headset Out
95. 06-06 21:51:09.885 16876 16876 D APM_AudioPolicyManager: AudioAnt: loadHwModule with module name r_submix
96. 06-06 21:51:09.888 16876 16876 D APM_AudioPolicyManager: AudioAnt: outProfile with name: r_submix output
97. 06-06 21:51:09.888 16876 16876 D APM_AudioPolicyManager: AudioAnt: supportedDevices with name: Remote Submix Out

复制代码

        可以看到此时系统已经创建了多个AudioOut线程(主历程为audioservice)，随后如果应用创建了AudioTrack，就可以根据strategy绑定到相干线程，触发或者加入到播放/录制线程。

1. auto8678p1_64_bsp_vm:/ # ps -ef | grep audioserver
2. audioserver  16876     1 0 21:51:04 ?     00:00:02 audioserver
4. auto8678p1_64_bsp_vm:/ # ps -T -p 16876
5. USER           PID   TID  PPID        VSZ    RSS WCHAN            ADDR S CMD
6. audioserver  16876 16876     1   12552792  50996 binder_th+          0 S audioserver
7. audioserver  16876 16884     1   12552792  50996 binder_th+          0 S binder:16876_1
8. audioserver  16876 16885     1   12552792  50996 futex_wai+          0 S AudioFlinger_Pa
9. audioserver  16876 16886     1   12552792  50996 binder_th+          0 S binder:16876_2
10. audioserver  16876 16887     1   12552792  50996 binder_th+          0 S HwBinder:16876_
11. audioserver  16876 16888     1   12552792  50996 binder_th+          0 S HwBinder:16876_
12. audioserver  16876 16889     1   12552792  50996 binder_th+          0 S HwBinder:16876_
13. audioserver  16876 16891     1   12552792  50996 futex_wai+          0 S AudioUtilThread
14. audioserver  16876 16892     1   12552792  50996 futex_wai+          0 S ApmAudio
15. audioserver  16876 16893     1   12552792  50996 futex_wai+          0 S ApmOutput
16. audioserver  16876 16913     1   12552792  50996 futex_wai+          0 S TimerThread
17. audioserver  16876 16914     1   12552792  50996 futex_wai+          0 S AudioOut_D
18. audioserver  16876 16915     1   12552792  50996 futex_wai+          0 S AudioOut_15
19. audioserver  16876 16916     1   12552792  50996 futex_wai+          0 S AudioOut_1D
20. audioserver  16876 16917     1   12552792  50996 futex_wai+          0 S AudioOut_25
21. audioserver  16876 16918     1   12552792  50996 futex_wai+          0 S AudioOut_2D
22. audioserver  16876 16919     1   12552792  50996 futex_wai+          0 S AudioOut_35
23. audioserver  16876 16920     1   12552792  50996 futex_wai+          0 S AudioOut_3D
24. audioserver  16876 16921     1   12552792  50996 futex_wai+          0 S AudioOut_45
25. audioserver  16876 16922     1   12552792  50996 futex_wai+          0 S AudioOut_4D
26. audioserver  16876 16923     1   12552792  50996 futex_wai+          0 S AudioOut_55
27. audioserver  16876 16924     1   12552792  50996 futex_wai+          0 S AudioOut_5D
28. audioserver  16876 16925     1   12552792  50996 futex_wai+          0 S AudioOut_65
29. audioserver  16876 16926     1   12552792  50996 futex_wai+          0 S AudioOut_6D
30. audioserver  16876 16927     1   12552792  50996 futex_wai+          0 S AudioOut_75
31. audioserver  16876 16928     1   12552792  50996 futex_wai+          0 S AudioOut_7D
32. audioserver  16876 16929     1   12552792  50996 futex_wai+          0 S AudioOut_85
33. audioserver  16876 16930     1   12552792  50996 futex_wai+          0 S AudioOut_8D
34. audioserver  16876 16945     1   12552792  50996 binder_th+          0 S binder:16876_3
35. audioserver  16876 16948     1   12552792  50996 binder_th+          0 S binder:16876_4
36. audioserver  16876 17743     1   12552792  50996 binder_th+          0 S binder:16876_5

复制代码

总结

团体音频链路的预备工作暂时画上一个句号，复杂但是容易明白。音频服务端已经预备停当，等候音频客户端(AudioTrack/AudioRecord)的加入并开始奥妙的音频之旅。

免责声明：如果侵犯了您的权益，请联系站长，我们会及时删除侵权内容，谢谢合作！更多信息从访问主页：qidao123.com:ToB企服之家，中国第一个企服评测及商务社交产业平台。