AudioRecord#
Initialization#
public AudioRecord(int audioSource, int sampleRateInHz, int channelConfig, int audioFormat,int bufferSizeInBytes)
Regarding the sample rate, all devices are guaranteed to support 44100Hz, but other frequencies may not be supported. In this case, you can use the parameter SAMPLE_RATE_UNSPECIFIED to let the device decide the sample rate to use. channelConfig is the sampling channel; currently, all devices guarantee support for mono, while multi-channel may not be supported. bufferSizeInBytes is the size of the buffer used to store the collected data, which is generally directly set to the return value of the getMinBufferSize interface.
getMinBufferSize#
//frameworks/base/media/java/android/media/AudioRecord.java
static public int getMinBufferSize(int sampleRateInHz, int channelConfig, int audioFormat) {
int channelCount = 0;
switch (channelConfig) {
case AudioFormat.CHANNEL_IN_DEFAULT: // AudioFormat.CHANNEL_CONFIGURATION_DEFAULT
case AudioFormat.CHANNEL_IN_MONO:
case AudioFormat.CHANNEL_CONFIGURATION_MONO:
channelCount = 1;
break;
case AudioFormat.CHANNEL_IN_STEREO:
case AudioFormat.CHANNEL_CONFIGURATION_STEREO:
case (AudioFormat.CHANNEL_IN_FRONT | AudioFormat.CHANNEL_IN_BACK):
channelCount = 2;
break;
case AudioFormat.CHANNEL_INVALID:
default:
loge("getMinBufferSize(): Invalid channel configuration.");
return ERROR_BAD_VALUE;
}
int size = native_get_min_buff_size(sampleRateInHz, channelCount, audioFormat);
if (size == 0) {
return ERROR_BAD_VALUE;
}
else if (size == -1) {
return ERROR;
}
else {
return size;
}
}
//frameworks/base/core/jni/android_media_AudioRecord.cpp
static jint android_media_AudioRecord_get_min_buff_size(JNIEnv *env, jobject thiz,
jint sampleRateInHertz, jint channelCount, jint audioFormat) {
ALOGV(">> android_media_AudioRecord_get_min_buff_size(%d, %d, %d)",
sampleRateInHertz, channelCount, audioFormat);
size_t frameCount = 0;
audio_format_t format = audioFormatToNative(audioFormat); // Convert Java format to native
status_t result = AudioRecord::getMinFrameCount(&frameCount, // Get minimum frame count
sampleRateInHertz,
format,
audio_channel_in_mask_from_count(channelCount));
if (result == BAD_VALUE) {
return 0;
}
if (result != NO_ERROR) {
return -1;
}
return frameCount * audio_bytes_per_frame(channelCount, format);
}
//frameworks/av/media/libaudioclient/AudioRecord.cpp
status_t AudioRecord::getMinFrameCount(
size_t* frameCount,
uint32_t sampleRate,
audio_format_t format,
audio_channel_mask_t channelMask)
{
if (frameCount == NULL) {
return BAD_VALUE;
}
size_t size;
status_t status = AudioSystem::getInputBufferSize(sampleRate, format, channelMask, &size);
if (status != NO_ERROR) {
ALOGE("%s(): AudioSystem could not query the input buffer size for"
" sampleRate %u, format %#x, channelMask %#x; status %d",
__func__, sampleRate, format, channelMask, status);
return status;
}
// We double the size of input buffer for ping pong use of record buffer.
const auto frameSize = audio_bytes_per_frame(
audio_channel_count_from_in_mask(channelMask), format);
if (frameSize == 0 || ((*frameCount = (size * 2) / frameSize) == 0)) {
ALOGE("%s(): Unsupported configuration: sampleRate %u, format %#x, channelMask %#x",
__func__, sampleRate, format, channelMask);
return BAD_VALUE;
}
return NO_ERROR;
}
//frameworks/av/media/libaudioclient/AudioSystem.cpp
status_t AudioSystem::getInputBufferSize(uint32_t sampleRate, audio_format_t format,
audio_channel_mask_t channelMask, size_t* buffSize) {
const sp<AudioFlingerClient> afc = getAudioFlingerClient();
if (afc == 0) {
return NO_INIT;
}
return afc->getInputBufferSize(sampleRate, format, channelMask, buffSize);
}
status_t AudioSystem::AudioFlingerClient::getInputBufferSize(
uint32_t sampleRate, audio_format_t format,
audio_channel_mask_t channelMask, size_t* buffSize) {
const sp<IAudioFlinger> af = get_audio_flinger();
if (af == 0) {
return PERMISSION_DENIED;
}
std::lock_guard _l(mMutex);
// Do we have a stale mInBuffSize or are we requesting the input buffer size for new values
if ((mInBuffSize == 0) || (sampleRate != mInSamplingRate) || (format != mInFormat)
|| (channelMask != mInChannelMask)) {
size_t inBuffSize = af->getInputBufferSize(sampleRate, format, channelMask);
if (inBuffSize == 0) {
ALOGE("AudioSystem::getInputBufferSize failed sampleRate %d format %#x channelMask %#x",
sampleRate, format, channelMask);
return BAD_VALUE;
}
// A benign race is possible here: we could overwrite a fresher cache entry
// save the request params
mInSamplingRate = sampleRate;
mInFormat = format;
mInChannelMask = channelMask;
mInBuffSize = inBuffSize;
}
*buffSize = mInBuffSize;
return NO_ERROR;
}
//frameworks/av/services/audioflinger/AudioFlinger.cpp
size_t AudioFlinger::getInputBufferSize(uint32_t sampleRate, audio_format_t format,
audio_channel_mask_t channelMask) const
{
status_t ret = initCheck();
if (ret != NO_ERROR) {
return 0;
}
if ((sampleRate == 0) ||
!audio_is_valid_format(format) ||
!audio_is_input_channel(channelMask)) {
return 0;
}
audio_utils::lock_guard lock(hardwareMutex());
if (mPrimaryHardwareDev == nullptr) {
return 0;
}
if (mInputBufferSizeOrderedDevs.empty()) {
return 0;
}
mHardwareStatus = AUDIO_HW_GET_INPUT_BUFFER_SIZE;
std::vector<audio_channel_mask_t> channelMasks = {channelMask};
if (channelMask != AUDIO_CHANNEL_IN_MONO) {
channelMasks.push_back(AUDIO_CHANNEL_IN_MONO);
}
if (channelMask != AUDIO_CHANNEL_IN_STEREO) {
channelMasks.push_back(AUDIO_CHANNEL_IN_STEREO);
}
std::vector<audio_format_t> formats = {format};
if (format != AUDIO_FORMAT_PCM_16_BIT) {
// For compressed format, buffer size may be queried using PCM. Allow this for compatibility
// in cases the primary hw dev does not support the format.
// TODO: replace with a table of formats and nominal buffer sizes (based on nominal bitrate
// and codec frame size).
formats.push_back(AUDIO_FORMAT_PCM_16_BIT);
}
std::vector<uint32_t> sampleRates = {sampleRate};
static const uint32_t SR_44100 = 44100;
static const uint32_t SR_48000 = 48000;
if (sampleRate != SR_48000) {
sampleRates.push_back(SR_48000);
}
if (sampleRate != SR_44100) {
sampleRates.push_back(SR_44100);
}
mHardwareStatus = AUDIO_HW_IDLE;
auto getInputBufferSize = [](const sp<DeviceHalInterface>& dev, audio_config_t config,
size_t* bytes) -> status_t {
if (!dev) {
return BAD_VALUE;
}
status_t result = dev->getInputBufferSize(&config, bytes);
if (result == BAD_VALUE) {
// Retry with the config suggested by the HAL.
result = dev->getInputBufferSize(&config, bytes);
}
if (result != OK || *bytes == 0) {
return BAD_VALUE;
}
return result;
};
// Change parameters of the configuration each iteration until we find a
// configuration that the device will support, or HAL suggests what it supports.
audio_config_t config = AUDIO_CONFIG_INITIALIZER;
for (auto testChannelMask : channelMasks) {
config.channel_mask = testChannelMask;
for (auto testFormat : formats) {
config.format = testFormat;
for (auto testSampleRate : sampleRates) {
config.sample_rate = testSampleRate;
size_t bytes = 0;
ret = BAD_VALUE;
for (const AudioHwDevice* dev : mInputBufferSizeOrderedDevs) {
ret = getInputBufferSize(dev->hwDevice(), config, &bytes);
if (ret == OK) {
break;
}
}
if (ret == BAD_VALUE) continue;
if (config.sample_rate != sampleRate || config.channel_mask != channelMask ||
config.format != format) {
uint32_t dstChannelCount = audio_channel_count_from_in_mask(channelMask);
uint32_t srcChannelCount =
audio_channel_count_from_in_mask(config.channel_mask);
size_t srcFrames =
bytes / audio_bytes_per_frame(srcChannelCount, config.format);
size_t dstFrames = destinationFramesPossible(
srcFrames, config.sample_rate, sampleRate);
bytes = dstFrames * audio_bytes_per_frame(dstChannelCount, format);
}
return bytes;
}
}
}
ALOGW("getInputBufferSize failed with minimum buffer size sampleRate %u, "
"format %#x, channelMask %#x",sampleRate, format, channelMask);
return 0;
}
AudioRecord Initialization#
//frameworks/base/media/java/android/media/AudioRecord.java
private AudioRecord(AudioAttributes attributes, AudioFormat format, int bufferSizeInBytes,
int sessionId, @Nullable Context context,
int maxSharedAudioHistoryMs, int halInputFlags) throws IllegalArgumentException {
mRecordingState = RECORDSTATE_STOPPED;
mHalInputFlags = halInputFlags;
if (attributes == null) {
throw new IllegalArgumentException("Illegal null AudioAttributes");
}
if (format == null) {
throw new IllegalArgumentException("Illegal null AudioFormat");
}
// remember which looper is associated with the AudioRecord instantiation
if ((mInitializationLooper = Looper.myLooper()) == null) {
mInitializationLooper = Looper.getMainLooper();
}
// is this AudioRecord using REMOTE_SUBMIX at full volume?
if (attributes.getCapturePreset() == MediaRecorder.AudioSource.REMOTE_SUBMIX) {
final AudioAttributes.Builder ab =
new AudioAttributes.Builder(attributes);
HashSet<String> filteredTags = new HashSet<String>();
final Iterator<String> tagsIter = attributes.getTags().iterator();
while (tagsIter.hasNext()) {
final String tag = tagsIter.next();
if (tag.equalsIgnoreCase(SUBMIX_FIXED_VOLUME)) {
mIsSubmixFullVolume = true;
Log.v(TAG, "Will record from REMOTE_SUBMIX at full fixed volume");
} else { // SUBMIX_FIXED_VOLUME: is not to be propagated to the native layers
filteredTags.add(tag);
}
}
ab.replaceTags(filteredTags);
attributes = ab.build();
}
mAudioAttributes = attributes;
int rate = format.getSampleRate();
if (rate == AudioFormat.SAMPLE_RATE_UNSPECIFIED) {
rate = 0;
}
int encoding = AudioFormat.ENCODING_DEFAULT;
if ((format.getPropertySetMask() & AudioFormat.AUDIO_FORMAT_HAS_PROPERTY_ENCODING) != 0)
{
encoding = format.getEncoding();
}
audioParamCheck(mAudioAttributes.getCapturePreset(), rate, encoding);
if ((format.getPropertySetMask()
& AudioFormat.AUDIO_FORMAT_HAS_PROPERTY_CHANNEL_INDEX_MASK) != 0) {
mChannelIndexMask = format.getChannelIndexMask();
mChannelCount = format.getChannelCount();
}
if ((format.getPropertySetMask()
& AudioFormat.AUDIO_FORMAT_HAS_PROPERTY_CHANNEL_MASK) != 0) {
mChannelMask = getChannelMaskFromLegacyConfig(format.getChannelMask(), false);
mChannelCount = format.getChannelCount();
} else if (mChannelIndexMask == 0) {
mChannelMask = getChannelMaskFromLegacyConfig(AudioFormat.CHANNEL_IN_DEFAULT, false);
mChannelCount = AudioFormat.channelCountFromInChannelMask(mChannelMask);
}
audioBuffSizeCheck(bufferSizeInBytes);
AttributionSource attributionSource = (context != null)
? context.getAttributionSource() : AttributionSource.myAttributionSource();
if (attributionSource.getPackageName() == null) {
// Command line utility
attributionSource = attributionSource.withPackageName("uid:" + Binder.getCallingUid());
}
int[] sampleRate = new int[] {mSampleRate};
int[] session = new int[1];
session[0] = resolveSessionId(context, sessionId);
//TODO: update native initialization when information about hardware init failure
// due to capture device already open is available.
try (ScopedParcelState attributionSourceState = attributionSource.asScopedParcelState()) {
int initResult = native_setup(new WeakReference<AudioRecord>(this), mAudioAttributes,
sampleRate, mChannelMask, mChannelIndexMask, mAudioFormat,
mNativeBufferSizeInBytes, session, attributionSourceState.getParcel(),
0 /*nativeRecordInJavaObj*/, maxSharedAudioHistoryMs, mHalInputFlags);
if (initResult != SUCCESS) {
loge("Error code " + initResult + " when initializing native AudioRecord object.");
return; // with mState == STATE_UNINITIALIZED
}
}
mSampleRate = sampleRate[0];
mSessionId = session[0];
mState = STATE_INITIALIZED;
}
//frameworks/base/core/jni/android_media_AudioRecord.cpp
static jint android_media_AudioRecord_setup(JNIEnv *env, jobject thiz, jobject weak_this,
jobject jaa, jintArray jSampleRate, jint channelMask,
jint channelIndexMask, jint audioFormat,
jint buffSizeInBytes, jintArray jSession,
jobject jAttributionSource, jlong nativeRecordInJavaObj,
jint sharedAudioHistoryMs,
jint halFlags) {
//ALOGV(">> Entering android_media_AudioRecord_setup");
//ALOGV("sampleRate=%d, audioFormat=%d, channel mask=%x, buffSizeInBytes=%d "
// "nativeRecordInJavaObj=0x%llX",
// sampleRateInHertz, audioFormat, channelMask, buffSizeInBytes, nativeRecordInJavaObj);
audio_channel_mask_t localChanMask = inChannelMaskToNative(channelMask);
if (jSession == NULL) {
ALOGE("Error creating AudioRecord: invalid session ID pointer");
return (jint) AUDIO_JAVA_ERROR;
}
jint* nSession = env->GetIntArrayElements(jSession, nullptr /* isCopy */);
if (nSession == NULL) {
ALOGE("Error creating AudioRecord: Error retrieving session id pointer");
return (jint) AUDIO_JAVA_ERROR;
}
audio_session_t sessionId = (audio_session_t) nSession[0];
env->ReleaseIntArrayElements(jSession, nSession, 0 /* mode */);
nSession = NULL;
sp<AudioRecord> lpRecorder;
sp<AudioRecordJNIStorage> callbackData;
jclass clazz = env->GetObjectClass(thiz);
if (clazz == NULL) {
ALOGE("Can't find %s when setting up callback.", kClassPathName);
return (jint) AUDIORECORD_ERROR_SETUP_NATIVEINITFAILED;
}
// if we pass in an existing *Native* AudioRecord, we don't need to create/initialize one.
if (nativeRecordInJavaObj == 0) { // native audio record has not been created yet, so we need to create it
if (jaa == 0) {
ALOGE("Error creating AudioRecord: invalid audio attributes");
return (jint) AUDIO_JAVA_ERROR;
}
if (jSampleRate == 0) {
ALOGE("Error creating AudioRecord: invalid sample rates");
return (jint) AUDIO_JAVA_ERROR;
}
jint elements[1];
env->GetIntArrayRegion(jSampleRate, 0, 1, elements);
int sampleRateInHertz = elements[0];
// channel index mask takes priority over channel position masks.
if (channelIndexMask) {
// Java channel index masks need the representation bits set.
localChanMask = audio_channel_mask_from_representation_and_bits(
AUDIO_CHANNEL_REPRESENTATION_INDEX,
channelIndexMask);
}
// Java channel position masks map directly to the native definition
if (!audio_is_input_channel(localChanMask)) {
ALOGE("Error creating AudioRecord: channel mask %#x is not valid.", localChanMask);
return (jint) AUDIORECORD_ERROR_SETUP_INVALIDCHANNELMASK;
}
uint32_t channelCount = audio_channel_count_from_in_mask(localChanMask);
// compare the format against the Java constants
audio_format_t format = audioFormatToNative(audioFormat);
if (format == AUDIO_FORMAT_INVALID) {
ALOGE("Error creating AudioRecord: unsupported audio format %d.", audioFormat);
return (jint) AUDIORECORD_ERROR_SETUP_INVALIDFORMAT;
}
if (buffSizeInBytes == 0) {
ALOGE("Error creating AudioRecord: frameCount is 0.");
return (jint) AUDIORECORD_ERROR_SETUP_ZEROFRAMECOUNT;
}
size_t frameCount = buffSizeInBytes / audio_bytes_per_frame(channelCount, format);
// create an uninitialized AudioRecord object
Parcel* parcel = parcelForJavaObject(env, jAttributionSource);
android::content::AttributionSourceState attributionSource;
attributionSource.readFromParcel(parcel);
lpRecorder = new AudioRecord(attributionSource);
// read the AudioAttributes values
auto paa = JNIAudioAttributeHelper::makeUnique();
jint jStatus = JNIAudioAttributeHelper::nativeFromJava(env, jaa, paa.get());
if (jStatus != (jint)AUDIO_JAVA_SUCCESS) {
return jStatus;
}
ALOGV("AudioRecord_setup for source=%d tags=%s flags=%08x", paa->source, paa->tags, paa->flags);
const auto flags = static_cast<audio_input_flags_t>(halFlags);
// create the callback information:
// this data will be passed with every AudioRecord callback
// we use a weak reference so the AudioRecord object can be garbage collected.
callbackData = sp<AudioRecordJNIStorage>::make(clazz, weak_this);
const status_t status =
lpRecorder->set(paa->source, sampleRateInHertz, // set configuration
format, // word length, PCM
localChanMask, frameCount,
callbackData, // callback
0, // notificationFrames,
true, // threadCanCallJava
sessionId, AudioRecord::TRANSFER_DEFAULT, flags, -1,
-1, // default uid, pid
paa.get(), AUDIO_PORT_HANDLE_NONE, MIC_DIRECTION_UNSPECIFIED,
MIC_FIELD_DIMENSION_DEFAULT, sharedAudioHistoryMs);
if (status != NO_ERROR) {
ALOGE("Error creating AudioRecord instance: initialization check failed with status %d.",
status);
goto native_init_failure;
}
// Set caller name so it can be logged in destructor.
// MediaMetricsConstants.h: AMEDIAMETRICS_PROP_CALLERNAME_VALUE_JAVA
lpRecorder->setCallerName("java");
} else { // end if nativeRecordInJavaObj == 0) // native AudioRecord has already been created
lpRecorder = (AudioRecord*)nativeRecordInJavaObj;
// TODO: We need to find out which members of the Java AudioRecord might need to be
// initialized from the Native AudioRecord
// these are directly returned from getters:
// mSampleRate
// mRecordSource
// mAudioFormat
// mChannelMask
// mChannelCount
// mState (?)
// mRecordingState (?)
// mPreferredDevice
// create the callback information:
// this data will be passed with every AudioRecord callback
// This next line makes little sense
// callbackData = sp<AudioRecordJNIStorage>::make(clazz, weak_this);
}
nSession = env->GetIntArrayElements(jSession, nullptr /* isCopy */);
if (nSession == NULL) {
ALOGE("Error creating AudioRecord: Error retrieving session id pointer");
goto native_init_failure;
}
// read the audio session ID back from AudioRecord in case a new session was created during set()
nSession[0] = lpRecorder->getSessionId();
env->ReleaseIntArrayElements(jSession, nSession, 0 /* mode */);
nSession = NULL;
{
const jint elements[1] = { (jint) lpRecorder->getSampleRate() };
env->SetIntArrayRegion(jSampleRate, 0, 1, elements);
}
// save our newly created C++ AudioRecord in the "nativeRecorderInJavaObj" field
// of the Java object
// associate Java object
setFieldSp(env, thiz, lpRecorder, javaAudioRecordFields.nativeRecorderInJavaObj);
// save our newly created callback information in the "jniData" field
// of the Java object (in mNativeJNIDataHandle) so we can free the memory in finalize()
setFieldSp(env, thiz, callbackData, javaAudioRecordFields.jniData);
return (jint) AUDIO_JAVA_SUCCESS;
// failure:
native_init_failure:
setFieldSp(env, thiz, sp<AudioRecord>{}, javaAudioRecordFields.nativeRecorderInJavaObj);
setFieldSp(env, thiz, sp<AudioRecordJNIStorage>{}, javaAudioRecordFields.jniData);
// lpRecorder goes out of scope, so reference count drops to zero
return (jint) AUDIORECORD_ERROR_SETUP_NATIVEINITFAILED;
}
//frameworks/av/media/libaudioclient/AudioRecord.cpp
status_t AudioRecord::set(
audio_source_t inputSource,
uint32_t sampleRate,
audio_format_t format,
audio_channel_mask_t channelMask,
size_t frameCount,
const wp<IAudioRecordCallback>& callback,
uint32_t notificationFrames,
bool threadCanCallJava,
audio_session_t sessionId,
transfer_type transferType,
audio_input_flags_t flags,
uid_t uid,
pid_t pid,
const audio_attributes_t* pAttributes,
audio_port_handle_t selectedDeviceId,
audio_microphone_direction_t selectedMicDirection,
float microphoneFieldDimension,
int32_t maxSharedAudioHistoryMs)
{
status_t status = NO_ERROR;
LOG_ALWAYS_FATAL_IF(mInitialized, "%s: should not be called twice", __func__);
mInitialized = true;
// Note mPortId is not valid until the track is created, so omit mPortId in ALOG for set.
ALOGV("%s(): inputSource %d, sampleRate %u, format %#x, channelMask %#x, frameCount %zu, "
"notificationFrames %u, sessionId %d, transferType %d, flags %#x, attributionSource %s"
"uid %d, pid %d",
__func__,
inputSource, sampleRate, format, channelMask, frameCount, notificationFrames,
sessionId, transferType, flags, mClientAttributionSource.toString().c_str(), uid, pid);
// TODO b/182392553: refactor or remove
pid_t callingPid = IPCThreadState::self()->getCallingPid();
pid_t myPid = getpid();
pid_t adjPid = pid;
if (pid == -1 || (callingPid != myPid)) {
adjPid = callingPid;
}
auto clientAttributionSourcePid = legacy2aidl_pid_t_int32_t(adjPid);
if (!clientAttributionSourcePid.ok()) {
return logIfErrorAndReturnStatus(BAD_VALUE,
StringPrintf("%s: received invalid client attribution "
"source pid, pid: %d, sessionId: %d",
__func__, pid, sessionId),
__func__);
}
mClientAttributionSource.pid = clientAttributionSourcePid.value();
uid_t adjUid = uid;
if (uid == -1 || (callingPid != myPid)) {
adjUid = IPCThreadState::self()->getCallingUid();
}
auto clientAttributionSourceUid = legacy2aidl_uid_t_int32_t(adjUid);
if (!clientAttributionSourceUid.ok()) {
return logIfErrorAndReturnStatus(BAD_VALUE,
StringPrintf("%s: received invalid client attribution "
"source uid, pid: %d, session id: %d",
__func__, pid, sessionId),
__func__);
}
mClientAttributionSource.uid = clientAttributionSourceUid.value();
mTracker.reset(new RecordingActivityTracker());
mSelectedDeviceId = selectedDeviceId;
mSelectedMicDirection = selectedMicDirection;
mSelectedMicFieldDimension = microphoneFieldDimension;
mMaxSharedAudioHistoryMs = maxSharedAudioHistoryMs;
// Copy the state variables early so they are available for error reporting.
if (pAttributes == nullptr) {
mAttributes = AUDIO_ATTRIBUTES_INITIALIZER;
mAttributes.source = inputSource;
if (inputSource == AUDIO_SOURCE_VOICE_COMMUNICATION
|| inputSource == AUDIO_SOURCE_CAMCORDER) {
mAttributes.flags = static_cast<audio_flags_mask_t>(
mAttributes.flags | AUDIO_FLAG_CAPTURE_PRIVATE);
}
} else {
// stream type shouldn't be looked at, this track has audio attributes
memcpy(&mAttributes, pAttributes, sizeof(audio_attributes_t));
ALOGV("%s: Building AudioRecord with attributes: source=%d flags=0x%x tags=[%s]",
__func__, mAttributes.source, mAttributes.flags, mAttributes.tags);
}
mSampleRate = sampleRate;
if (format == AUDIO_FORMAT_DEFAULT) {
format = AUDIO_FORMAT_PCM_16_BIT;
}
if (!audio_is_linear_pcm(format)) {
// Compressed capture requires direct
flags = (audio_input_flags_t) (flags | AUDIO_INPUT_FLAG_DIRECT);
ALOGI("%s(): Format %#x is not linear pcm. Setting DIRECT, using flags %#x", __func__,
format, flags);
}
mFormat = format;
mChannelMask = channelMask;
mSessionId = sessionId;
ALOGV("%s: mSessionId %d", __func__, mSessionId);
mOrigFlags = mFlags = flags;
mTransfer = transferType;
switch (mTransfer) { // The incoming transfer is default; the other cases can be ignored for now
case TRANSFER_DEFAULT:
if (callback == nullptr || threadCanCallJava) {
mTransfer = TRANSFER_SYNC;
} else {
mTransfer = TRANSFER_CALLBACK;
}
break;
case TRANSFER_CALLBACK:
if (callback == nullptr) {
return logIfErrorAndReturnStatus(
BAD_VALUE,
StringPrintf("%s: Transfer type TRANSFER_CALLBACK but callback == nullptr, "
"pid: %d, session id: %d",
__func__, pid, sessionId),
__func__);
}
break;
case TRANSFER_OBTAIN:
case TRANSFER_SYNC:
break;
default:
return logIfErrorAndReturnStatus(
BAD_VALUE,
StringPrintf("%s: Invalid transfer type %d, pid: %d, session id: %d", __func__,
mTransfer, pid, sessionId),
__func__);
}
// invariant that mAudioRecord != 0 is true only after set() returns successfully
if (mAudioRecord != 0) {
return logIfErrorAndReturnStatus(
INVALID_OPERATION,
StringPrintf("%s: Track already in use, pid: %d, session id: %d", __func__, pid,
sessionId),
__func__);
}
if (!audio_is_valid_format(mFormat)) {
return logIfErrorAndReturnStatus(
BAD_VALUE,
StringPrintf("%s: Format %#x is not valid, pid: %d, session id: %d", __func__,
mFormat, pid, sessionId),
__func__);
}
if (!audio_is_input_channel(mChannelMask)) {
return logIfErrorAndReturnStatus(
BAD_VALUE,
StringPrintf("%s: Invalid channel mask %#x, pid: %d, session id: %d", __func__,
mChannelMask, pid, sessionId),
__func__);
}
mChannelCount = audio_channel_count_from_in_mask(mChannelMask);
mFrameSize = audio_bytes_per_frame(mChannelCount, mFormat);
// mFrameCount is initialized in createRecord_l
mReqFrameCount = frameCount;
mNotificationFramesReq = notificationFrames;
// mNotificationFramesAct is initialized in createRecord_l
mCallback = callback;
if (mCallback != nullptr) {
mAudioRecordThread = new AudioRecordThread(*this); // Callback thread, so that after data is collected, it can actively call back the caller
mAudioRecordThread->run("AudioRecord", ANDROID_PRIORITY_AUDIO);
// thread begins in paused state, and will not reference us until start()
}
// create the IAudioRecord
// 关键
{
AutoMutex lock(mLock);
status = createRecord_l(0 /*epoch*/);
}
ALOGV("%s(%d): status %d", __func__, mPortId, status);
if (status != NO_ERROR) {
if (mAudioRecordThread != 0) {
mAudioRecordThread->requestExit(); // see comment in AudioRecord.h
mAudioRecordThread->requestExitAndWait();
mAudioRecordThread.clear();
}
// bypass error message to avoid logging twice (createRecord_l logs the error).
mStatus = status;
return mStatus;
}
// TODO: add audio hardware input latency here
mLatency = (1000LL * mFrameCount) / mSampleRate;
mMarkerPosition = 0;
mMarkerReached = false;
mNewPosition = 0;
mUpdatePeriod = 0;
AudioSystem::acquireAudioSessionId(mSessionId, adjPid, adjUid);
mSequence = 1;
mObservedSequence = mSequence;
mInOverrun = false;
mFramesRead = 0;
mFramesReadServerOffset = 0;
return logIfErrorAndReturnStatus(status, "", __func__);
}
status_t AudioRecord::createRecord_l(const Modulo<uint32_t> &epoch)
{
const int64_t beginNs = systemTime();
const sp<IAudioFlinger>& audioFlinger = AudioSystem::get_audio_flinger();
IAudioFlinger::CreateRecordInput input;
IAudioFlinger::CreateRecordOutput output;
[[maybe_unused]] audio_session_t originalSessionId;
void *iMemPointer;
audio_track_cblk_t* cblk;
status_t status;
static const int32_t kMaxCreateAttempts = 3;
int32_t remainingAttempts = kMaxCreateAttempts;
if (audioFlinger == 0) {
return logIfErrorAndReturnStatus(
NO_INIT, StringPrintf("%s(%d): Could not get audioflinger", __func__, mPortId), "");
}
// mFlags (not mOrigFlags) is modified depending on whether fast request is accepted.
// After fast request is denied, we will request again if IAudioRecord is re-created.
// Now that we have a reference to an I/O handle and have not yet handed it off to AudioFlinger,
// we must release it ourselves if anything goes wrong.
// Client can only express a preference for FAST. Server will perform additional tests.
if (mFlags & AUDIO_INPUT_FLAG_FAST) {
bool useCaseAllowed =
// any of these use cases:
// use case 1: callback transfer mode
(mTransfer == TRANSFER_CALLBACK) ||
// use case 2: blocking read mode
// The default buffer capacity at 48 kHz is 2048 frames, or ~42.6 ms.
// That's enough for double-buffering with our standard 20 ms rule of thumb for
// the minimum period of a non-SCHED_FIFO thread.
// This is needed so that AAudio apps can do a low latency non-blocking read from a
// callback running with SCHED_FIFO.
(mTransfer == TRANSFER_SYNC) ||
// use case 3: obtain/release mode
(mTransfer == TRANSFER_OBTAIN);
if (!useCaseAllowed) {
ALOGD("%s(%d): AUDIO_INPUT_FLAG_FAST denied, incompatible transfer = %s",
__func__, mPortId,
convertTransferToText(mTransfer));
mFlags = (audio_input_flags_t) (mFlags & ~(AUDIO_INPUT_FLAG_FAST |
AUDIO_INPUT_FLAG_RAW));
}
}
input.attr = mAttributes;
input.config.sample_rate = mSampleRate;
input.config.channel_mask = mChannelMask;
input.config.format = mFormat;
input.clientInfo.attributionSource = mClientAttributionSource;
input.clientInfo.clientTid = -1;
if (mFlags & AUDIO_INPUT_FLAG_FAST) {
if (mAudioRecordThread != 0) {
input.clientInfo.clientTid = mAudioRecordThread->getTid();
}
}
input.riid = mTracker->getRiid();
input.flags = mFlags;
// The notification frame count is the period between callbacks, as suggested by the client
// but moderated by the server. For record, the calculations are done entirely on server side.
input.frameCount = mReqFrameCount;
input.notificationFrameCount = mNotificationFramesReq;
input.selectedDeviceId = mSelectedDeviceId;
input.sessionId = mSessionId;
originalSessionId = mSessionId;
input.maxSharedAudioHistoryMs = mMaxSharedAudioHistoryMs;
do {
media::CreateRecordResponse response;
auto aidlInput = input.toAidl();
if (!aidlInput.ok()) {
return logIfErrorAndReturnStatus(
BAD_VALUE,
StringPrintf("%s(%d): Could not create record due to invalid input", __func__,
mPortId),
"");
}
status = audioFlinger->createRecord(aidlInput.value(), response); // Create record
auto recordOutput = IAudioFlinger::CreateRecordOutput::fromAidl(response);
if (!recordOutput.ok()) {
return logIfErrorAndReturnStatus(
BAD_VALUE,
StringPrintf("%s(%d): Could not create record output due to invalid response",
__func__, mPortId),
"");
}
output = recordOutput.value();
if (status == NO_ERROR) {
break;
}
if (status != FAILED_TRANSACTION || --remainingAttempts <= 0) {
return logIfErrorAndReturnStatus(
status,
StringPrintf("%s(%d): AudioFlinger could not create record track, status: %d",
__func__, mPortId, status),
"");
}
// FAILED_TRANSACTION happens under very specific conditions causing a state mismatch
// between audio policy manager and audio flinger during the input stream open sequence
// and can be recovered by retrying.
// Leave time for race condition to clear before retrying and randomize delay
// to reduce the probability of concurrent retries in locked steps.
usleep((20 + rand() % 30) * 10000);
} while (1);
ALOG_ASSERT(output.audioRecord != 0);
// AudioFlinger now owns the reference to the I/O handle,
// so we are no longer responsible for releasing it.
mAwaitBoost = false;
if (output.flags & AUDIO_INPUT_FLAG_FAST) {
ALOGI("%s(%d): AUDIO_INPUT_FLAG_FAST successful; frameCount %zu -> %zu",
__func__, mPortId,
mReqFrameCount, output.frameCount);
mAwaitBoost = true;
}
mFlags = output.flags;
mRoutedDeviceId = output.selectedDeviceId;
mSessionId = output.sessionId;
mSampleRate = output.sampleRate;
mServerConfig = output.serverConfig;
mServerFrameSize = audio_bytes_per_frame(
audio_channel_count_from_in_mask(mServerConfig.channel_mask), mServerConfig.format);
mServerSampleSize = audio_bytes_per_sample(mServerConfig.format);
mHalSampleRate = output.halConfig.sample_rate;
mHalChannelCount = audio_channel_count_from_in_mask(output.halConfig.channel_mask);
mHalFormat = output.halConfig.format;
if (output.cblk == 0) {
return logIfErrorAndReturnStatus(
NO_INIT, StringPrintf("%s(%d): Could not get control block", __func__, mPortId),
"");
}
// TODO: Using unsecurePointer() has some associated security pitfalls
// (see declaration for details).
// Either document why it is safe in this case or address the
// issue (e.g. by copying).
iMemPointer = output.cblk ->unsecurePointer(); // Get shared memory
if (iMemPointer == NULL) {
return logIfErrorAndReturnStatus(
NO_INIT,
StringPrintf("%s(%d): Could not get control block pointer", __func__, mPortId), "");
}
cblk = static_cast<audio_track_cblk_t*>(iMemPointer);
// Starting address of buffers in shared memory.
// The buffers are either immediately after the control block,
// or in a separate area at the discretion of the server.
void *buffers;
if (output.buffers == 0) {
buffers = cblk + 1;
} else {
// TODO: Using unsecurePointer() has some associated security pitfalls
// (see declaration for details).
// Either document why it is safe in this case or address the
// issue (e.g. by copying).
buffers = output.buffers->unsecurePointer();
if (buffers == NULL) {
return logIfErrorAndReturnStatus(
NO_INIT,
StringPrintf("%s(%d): Could not get buffer pointer", __func__, mPortId), "");
}
}
// invariant that mAudioRecord != 0 is true only after set() returns successfully
if (mAudioRecord != 0) {
IInterface::asBinder(mAudioRecord)->unlinkToDeath(mDeathNotifier, this);
mDeathNotifier.clear();
}
mAudioRecord = output.audioRecord;
mCblkMemory = output.cblk;
mBufferMemory = output.buffers;
IPCThreadState::self()->flushCommands();
mCblk = cblk;
// note that output.frameCount is the (possibly revised) value of mReqFrameCount
if (output.frameCount < mReqFrameCount || (mReqFrameCount == 0 && output.frameCount == 0)) {
ALOGW("%s(%d): Requested frameCount %zu but received frameCount %zu",
__func__, output.portId,
mReqFrameCount, output.frameCount);
}
// Make sure that the application is notified with sufficient margin before overrun.
// The computation is done on the server side.
if (mNotificationFramesReq > 0 && output.notificationFrameCount != mNotificationFramesReq) {
ALOGW("%s(%d): Server adjusted notificationFrames from %u to %zu for frameCount %zu",
__func__, output.portId,
mNotificationFramesReq, output.notificationFrameCount, output.frameCount);
}
mNotificationFramesAct = (uint32_t)output.notificationFrameCount;
if (mServerConfig.format != mFormat && mCallback != nullptr) {
mFormatConversionBufRaw = std::make_unique<uint8_t[]>(mNotificationFramesAct * mFrameSize);
mFormatConversionBuffer.raw = mFormatConversionBufRaw.get();
}
//mInput != input includes the case where mInput == AUDIO_IO_HANDLE_NONE for first creation
if (mDeviceCallback != 0) {
if (mInput != AUDIO_IO_HANDLE_NONE) {
AudioSystem::removeAudioDeviceCallback(this, mInput, mPortId);
}
AudioSystem::addAudioDeviceCallback(this, output.inputId, output.portId);
}
if (!mSharedAudioPackageName.empty()) {
mAudioRecord->shareAudioHistory(mSharedAudioPackageName, mSharedAudioStartMs);
}
mPortId = output.portId;
// We retain a copy of the I/O handle, but don't own the reference
mInput = output.inputId;
mRefreshRemaining = true;
mFrameCount = output.frameCount;
// If IAudioRecord is re-created, don't let the requested frameCount
// decrease. This can confuse clients that cache frameCount().
if (mFrameCount > mReqFrameCount) {
mReqFrameCount = mFrameCount;
}
// update proxy
mProxy = new AudioRecordClientProxy(cblk, buffers, mFrameCount, mServerFrameSize);
mProxy->setEpoch(epoch);
mProxy->setMinimum(mNotificationFramesAct);
mDeathNotifier = new DeathNotifier(this);
IInterface::asBinder(mAudioRecord)->linkToDeath(mDeathNotifier, this);
mMetricsId = std::string(AMEDIAMETRICS_KEY_PREFIX_AUDIO_RECORD) + std::to_string(mPortId);
mediametrics::LogItem(mMetricsId)
.set(AMEDIAMETRICS_PROP_EVENT, AMEDIAMETRICS_PROP_EVENT_VALUE_CREATE)
.set(AMEDIAMETRICS_PROP_EXECUTIONTIMENS, (int64_t)(systemTime() - beginNs))
// the following are immutable (at least until restore)
.set(AMEDIAMETRICS_PROP_FLAGS, toString(mFlags).c_str())
.set(AMEDIAMETRICS_PROP_ORIGINALFLAGS, toString(mOrigFlags).c_str())
.set(AMEDIAMETRICS_PROP_SESSIONID, (int32_t)mSessionId)
.set(AMEDIAMETRICS_PROP_TRACKID, mPortId)
.set(AMEDIAMETRICS_PROP_LOGSESSIONID, mLogSessionId)
.set(AMEDIAMETRICS_PROP_SOURCE, toString(mAttributes.source).c_str())
.set(AMEDIAMETRICS_PROP_THREADID, (int32_t)output.inputId)
.set(AMEDIAMETRICS_PROP_SELECTEDDEVICEID, (int32_t)mSelectedDeviceId)
.set(AMEDIAMETRICS_PROP_ROUTEDDEVICEID, (int32_t)mRoutedDeviceId)
.set(AMEDIAMETRICS_PROP_ENCODING, toString(mFormat).c_str())
.set(AMEDIAMETRICS_PROP_CHANNELMASK, (int32_t)mChannelMask)
.set(AMEDIAMETRICS_PROP_FRAMECOUNT, (int32_t)mFrameCount)
.set(AMEDIAMETRICS_PROP_SAMPLERATE, (int32_t)mSampleRate)
// the following are NOT immutable
.set(AMEDIAMETRICS_PROP_STATE, stateToString(mActive))
.set(AMEDIAMETRICS_PROP_STATUS, (int32_t)status)
.set(AMEDIAMETRICS_PROP_SELECTEDMICDIRECTION, (int32_t)mSelectedMicDirection)
.set(AMEDIAMETRICS_PROP_SELECTEDMICFIELDDIRECTION, (double)mSelectedMicFieldDimension)
.record();
// sp<IAudioTrack> track destructor will cause releaseOutput() to be called by AudioFlinger
return logIfErrorAndReturnStatus(status, "", "");
}
// frameworks/av/services/audioflinger/AudioFlinger.cpp
status_t AudioFlinger::createRecord(const media::CreateRecordRequest& _input,
media::CreateRecordResponse& _output)
{
CreateRecordInput input = VALUE_OR_RETURN_STATUS(CreateRecordInput::fromAidl(_input));
CreateRecordOutput output;
sp<IAfRecordTrack> recordTrack;
sp<Client> client;
status_t lStatus;
audio_session_t sessionId = input.sessionId;
audio_port_handle_t portId = AUDIO_PORT_HANDLE_NONE;
output.cblk.clear();
output.buffers.clear();
output.inputId = AUDIO_IO_HANDLE_NONE;
// TODO b/182392553: refactor or clean up
AttributionSourceState adjAttributionSource = input.clientInfo.attributionSource;
bool updatePid = (adjAttributionSource.pid == -1);
const uid_t callingUid = IPCThreadState::self()->getCallingUid();
const uid_t currentUid = VALUE_OR_RETURN_STATUS(legacy2aidl_uid_t_int32_t(
adjAttributionSource.uid));
if (!isAudioServerOrMediaServerOrSystemServerOrRootUid(callingUid)) {
ALOGW_IF(currentUid != callingUid,
"%s uid %d tried to pass itself off as %d",
__FUNCTION__, callingUid, currentUid);
adjAttributionSource.uid = VALUE_OR_RETURN_STATUS(legacy2aidl_uid_t_int32_t(callingUid));
updatePid = true;
}
const pid_t callingPid = IPCThreadState::self()->getCallingPid();
const pid_t currentPid = VALUE_OR_RETURN_STATUS(aidl2legacy_int32_t_pid_t(
adjAttributionSource.pid));
if (updatePid) {
ALOGW_IF(currentPid != (pid_t)-1 && currentPid != callingPid,
"%s uid %d pid %d tried to pass itself off as pid %d",
__func__, callingUid, callingPid, currentPid);
adjAttributionSource.pid = VALUE_OR_RETURN_STATUS(legacy2aidl_pid_t_int32_t(callingPid));
}
adjAttributionSource = afutils::checkAttributionSourcePackage(
adjAttributionSource);
// further format checks are performed by createRecordTrack_l()
if (!audio_is_valid_format(input.config.format)) {
ALOGE("createRecord() invalid format %#x", input.config.format);
lStatus = BAD_VALUE;
goto Exit;
}
// further channel mask checks are performed by createRecordTrack_l()
if (!audio_is_input_channel(input.config.channel_mask)) {
ALOGE("createRecord() invalid channel mask %#x", input.config.channel_mask);
lStatus = BAD_VALUE;
goto Exit;
}
if (sessionId == AUDIO_SESSION_ALLOCATE) {
sessionId = (audio_session_t) newAudioUniqueId(AUDIO_UNIQUE_ID_USE_SESSION);
} else if (audio_unique_id_get_use(sessionId) != AUDIO_UNIQUE_ID_USE_SESSION) {
lStatus = BAD_VALUE;
goto Exit;
}
output.sessionId = sessionId;
output.selectedDeviceId = input.selectedDeviceId;
output.flags = input.flags;
client = registerPid(VALUE_OR_FATAL(aidl2legacy_int32_t_pid_t(adjAttributionSource.pid))); // Create shared memory
// Not a conventional loop, but a retry loop for at most two iterations total.
// Try first maybe with FAST flag then try again without FAST flag if that fails.
// Exits loop via break on no error or goes to exit on error
// The sp<> references will be dropped when re-entering scope.
// The lack of indentation is deliberate, to reduce code churn and ease merges.
for (;;) {
// release previously opened input if retrying.
if (output.inputId != AUDIO_IO_HANDLE_NONE) {
recordTrack.clear();
AudioSystem::releaseInput(portId);
output.inputId = AUDIO_IO_HANDLE_NONE;
portId = AUDIO_PORT_HANDLE_NONE;
}
lStatus = AudioSystem::getInputForAttr(&input.attr, &output.inputId, // Call HAL to get attribute information
input.riid,
sessionId,
// FIXME compare to AudioTrack
adjAttributionSource,
&input.config,
output.flags, &output.selectedDeviceId, &portId);
if (lStatus != NO_ERROR) {
ALOGE("createRecord() getInputForAttr return error %d", lStatus);
goto Exit;
}
{
audio_utils::lock_guard _l(mutex());
IAfRecordThread* const thread = checkRecordThread_l(output.inputId); // Get capture thread
if (thread == NULL) {
ALOGW("createRecord() checkRecordThread_l failed, input handle %d", output.inputId);
lStatus = FAILED_TRANSACTION;
goto Exit;
}
ALOGV("createRecord() lSessionId: %d input %d", sessionId, output.inputId);
output.sampleRate = input.config.sample_rate;
output.frameCount = input.frameCount;
output.notificationFrameCount = input.notificationFrameCount;
recordTrack = thread->createRecordTrack_l(client, input.attr, &output.sampleRate, // Create Record in capture thread
input.config.format, input.config.channel_mask,
&output.frameCount, sessionId,
&output.notificationFrameCount,
callingPid, adjAttributionSource, &output.flags,
input.clientInfo.clientTid,
&lStatus, portId, input.maxSharedAudioHistoryMs);
LOG_ALWAYS_FATAL_IF((lStatus == NO_ERROR) && (recordTrack == 0));
// lStatus == BAD_TYPE means FAST flag was rejected: request a new input from
// audio policy manager without FAST constraint
if (lStatus == BAD_TYPE) {
continue;
}
if (lStatus != NO_ERROR) {
goto Exit;
}
if (recordTrack->isFastTrack()) {
output.serverConfig = {
thread->sampleRate(),
thread->channelMask(),
thread->format()
};
} else {
output.serverConfig = {
recordTrack->sampleRate(),
recordTrack->channelMask(),
recordTrack->format()
};
}
output.halConfig = {
thread->sampleRate(),
thread->channelMask(),
thread->format()
};
// Check if one effect chain was awaiting for an AudioRecord to be created on this
// session and move it to this thread.
sp<IAfEffectChain> chain = getOrphanEffectChain_l(sessionId);
if (chain != 0) {
audio_utils::lock_guard _l2(thread->mutex());
thread->addEffectChain_l(chain);
}
break;
}
// End of retry loop.
// The lack of indentation is deliberate, to reduce code churn and ease merges.
}
output.cblk = recordTrack->getCblk();
output.buffers = recordTrack->getBuffers(); // Return anonymous shared memory operator
output.portId = portId;
output.audioRecord = IAfRecordTrack::createIAudioRecordAdapter(recordTrack);
_output = VALUE_OR_FATAL(output.toAidl()); // Wrap into binder
Exit:
if (lStatus != NO_ERROR) {
// remove local strong reference to Client before deleting the RecordTrack so that the
// Client destructor is called by the TrackBase destructor with clientMutex() held
// Don't hold clientMutex() when releasing the reference on the track as the
// destructor will acquire it.
{
audio_utils::lock_guard _cl(clientMutex());
client.clear();
}
recordTrack.clear();
if (output.inputId != AUDIO_IO_HANDLE_NONE) {
AudioSystem::releaseInput(portId);
}
}
return lStatus;
}
// frameworks/av/services/audioflinger/Threads.cpp
// RecordThread::createRecordTrack_l() must be called with AudioFlinger::mutex() held
sp<IAfRecordTrack> RecordThread::createRecordTrack_l(
const sp<Client>& client,
const audio_attributes_t& attr,
uint32_t *pSampleRate,
audio_format_t format,
audio_channel_mask_t channelMask,
size_t *pFrameCount,
audio_session_t sessionId,
size_t *pNotificationFrameCount,
pid_t creatorPid,
const AttributionSourceState& attributionSource,
audio_input_flags_t *flags,
pid_t tid,
status_t *status,
audio_port_handle_t portId,
int32_t maxSharedAudioHistoryMs)
{
size_t frameCount = *pFrameCount;
size_t notificationFrameCount = *pNotificationFrameCount;
sp<IAfRecordTrack> track;
status_t lStatus;
audio_input_flags_t inputFlags = mInput->flags;
audio_input_flags_t requestedFlags = *flags;
uint32_t sampleRate;
lStatus = initCheck();
if (lStatus != NO_ERROR) {
ALOGE("createRecordTrack_l() audio driver not initialized");
goto Exit;
}
if (!audio_is_linear_pcm(mFormat) && (*flags & AUDIO_INPUT_FLAG_DIRECT) == 0) {
ALOGE("createRecordTrack_l() on an encoded stream requires AUDIO_INPUT_FLAG_DIRECT");
lStatus = BAD_VALUE;
goto Exit;
}
if (maxSharedAudioHistoryMs != 0) {
if (!captureHotwordAllowed(attributionSource)) {
lStatus = PERMISSION_DENIED;
goto Exit;
}
if (maxSharedAudioHistoryMs < 0
|| maxSharedAudioHistoryMs > kMaxSharedAudioHistoryMs) {
lStatus = BAD_VALUE;
goto Exit;
}
}
if (*pSampleRate == 0) {
*pSampleRate = mSampleRate;
}
sampleRate = *pSampleRate;
// special case for FAST flag considered OK if fast capture is present and access to
// audio history is not required
if (hasFastCapture() && mMaxSharedAudioHistoryMs == 0) {
inputFlags = (audio_input_flags_t)(inputFlags | AUDIO_INPUT_FLAG_FAST);
}
// Check if requested flags are compatible with input stream flags
if ((*flags & inputFlags) != *flags) {
ALOGW("createRecordTrack_l(): mismatch between requested flags (%08x) and"
" input flags (%08x)",
*flags, inputFlags);
*flags = (audio_input_flags_t)(*flags & inputFlags);
}
// client expresses a preference for FAST and no access to audio history,
// but we get the final say
if (*flags & AUDIO_INPUT_FLAG_FAST && maxSharedAudioHistoryMs == 0) {
if (
// we formerly checked for a callback handler (non-0 tid),
// but that is no longer required for TRANSFER_OBTAIN mode
// No need to match hardware format, format conversion will be done in client side.
//
// Frame count is not specified (0), or is less than or equal the pipe depth.
// It is OK to provide a higher capacity than requested.
// We will force it to mPipeFramesP2 below.
(frameCount <= mPipeFramesP2) &&
// PCM data
audio_is_linear_pcm(format) &&
// hardware channel mask
(channelMask == mChannelMask) &&
// hardware sample rate
(sampleRate == mSampleRate) &&
// record thread has an associated fast capture
hasFastCapture() &&
// there are sufficient fast track slots available
mFastTrackAvail
) {
// check compatibility with audio effects.
audio_utils::lock_guard _l(mutex());
// Do not accept FAST flag if the session has software effects
sp<IAfEffectChain> chain = getEffectChain_l(sessionId);
if (chain != 0) {
audio_input_flags_t old = *flags;
chain->checkInputFlagCompatibility(flags);
if (old != *flags) {
ALOGV("%p AUDIO_INPUT_FLAGS denied by effect old=%#x new=%#x",
this, (int)old, (int)*flags);
}
}
ALOGV_IF((*flags & AUDIO_INPUT_FLAG_FAST) != 0,
"%p AUDIO_INPUT_FLAG_FAST accepted: frameCount=%zu mFrameCount=%zu",
this, frameCount, mFrameCount);
} else {
ALOGV("%p AUDIO_INPUT_FLAG_FAST denied: frameCount=%zu mFrameCount=%zu mPipeFramesP2=%zu "
"format=%#x isLinear=%d mFormat=%#x channelMask=%#x sampleRate=%u mSampleRate=%u "
"hasFastCapture=%d tid=%d mFastTrackAvail=%d",
this, frameCount, mFrameCount, mPipeFramesP2,
format, audio_is_linear_pcm(format), mFormat, channelMask, sampleRate, mSampleRate,
hasFastCapture(), tid, mFastTrackAvail);
*flags = (audio_input_flags_t)(*flags & ~AUDIO_INPUT_FLAG_FAST);
}
}
// If FAST or RAW flags were corrected, ask caller to request new input from audio policy
if ((*flags & AUDIO_INPUT_FLAG_FAST) !=
(requestedFlags & AUDIO_INPUT_FLAG_FAST)) {
*flags = (audio_input_flags_t) (*flags & ~(AUDIO_INPUT_FLAG_FAST | AUDIO_INPUT_FLAG_RAW));
lStatus = BAD_TYPE;
goto Exit;
}
// compute track buffer size in frames, and suggest the notification frame count
if (*flags & AUDIO_INPUT_FLAG_FAST) {
// fast track: frame count is exactly the pipe depth
frameCount = mPipeFramesP2;
// ignore requested notificationFrames, and always notify exactly once every HAL buffer
notificationFrameCount = mFrameCount;
} else {
// not fast track: max notification period is resampled equivalent of one HAL buffer time
// or 20 ms if there is a fast capture
// TODO This could be a roundupRatio inline, and const
size_t maxNotificationFrames = ((int64_t) (hasFastCapture() ? mSampleRate/50 : mFrameCount)
* sampleRate + mSampleRate - 1) / mSampleRate;
// minimum number of notification periods is at least kMinNotifications,
// and at least kMinMs rounded up to a whole notification period (minNotificationsByMs)
static const size_t kMinNotifications = 3;
static const uint32_t kMinMs = 30;
// TODO This could be a roundupRatio inline
const size_t minFramesByMs = (sampleRate * kMinMs + 1000 - 1) / 1000;
// TODO This could be a roundupRatio inline
const size_t minNotificationsByMs = (minFramesByMs + maxNotificationFrames - 1) /
maxNotificationFrames;
const size_t minFrameCount = maxNotificationFrames *
max(kMinNotifications, minNotificationsByMs);
frameCount = max(frameCount, minFrameCount);
if (notificationFrameCount == 0 || notificationFrameCount > maxNotificationFrames) {
notificationFrameCount = maxNotificationFrames;
}
}
*pFrameCount = frameCount;
*pNotificationFrameCount = notificationFrameCount;
{ // scope for mutex()
audio_utils::lock_guard _l(mutex());
int32_t startFrames = -1;
if (!mSharedAudioPackageName.empty()
&& mSharedAudioPackageName == attributionSource.packageName
&& mSharedAudioSessionId == sessionId
&& captureHotwordAllowed(attributionSource)) {
startFrames = mSharedAudioStartFrames;
}
track = IAfRecordTrack::create(this, client, attr, sampleRate, // Create Record
format, channelMask, frameCount,
nullptr /* buffer */, (size_t)0 /* bufferSize */, sessionId, creatorPid,
attributionSource, *flags, IAfTrackBase::TYPE_DEFAULT, portId,
startFrames);
lStatus = track->initCheck();
if (lStatus != NO_ERROR) {
ALOGE("createRecordTrack_l() initCheck failed %d; no control block?", lStatus);
// track must be cleared from the caller as the caller has the AF lock
goto Exit;
}
mTracks.add(track); // Add Record to the list
if ((*flags & AUDIO_INPUT_FLAG_FAST) && (tid != -1)) {
pid_t callingPid = IPCThreadState::self()->getCallingPid();
// we don't have CAP_SYS_NICE, nor do we want to have it as it's too powerful,
// so ask activity manager to do this on our behalf
sendPrioConfigEvent_l(callingPid, tid, kPriorityAudioApp, true /*forApp*/);
}
if (maxSharedAudioHistoryMs != 0) {
sendResizeBufferConfigEvent_l(maxSharedAudioHistoryMs);
}
}
lStatus = NO_ERROR;
Exit:
*status = lStatus;
return track;
}
// frameworks/av/services/audioflinger/Threads.cpp
bool RecordThread::threadLoop()
{
nsecs_t lastWarning = 0;
inputStandBy();
reacquire_wakelock:
{
audio_utils::lock_guard _l(mutex());
acquireWakeLock_l();
}
// used to request a deferred sleep, to be executed later while mutex is unlocked
uint32_t sleepUs = 0;
// timestamp correction enable is determined under lock, used in processing step.
bool timestampCorrectionEnabled = false;
int64_t lastLoopCountRead = -2; // never matches "previous" loop, when loopCount = 0.
// loop while there is work to do
for (int64_t loopCount = 0;; ++loopCount) { // loopCount used for statistics tracking
// Note: these sp<> are released at the end of the for loop outside of the mutex() lock.
sp<IAfRecordTrack> activeTrack;
std::vector<sp<IAfRecordTrack>> oldActiveTracks;
Vector<sp<IAfEffectChain>> effectChains;
// activeTracks accumulates a copy of a subset of mActiveTracks
Vector<sp<IAfRecordTrack>> activeTracks;
// reference to the (first and only) active fast track
sp<IAfRecordTrack> fastTrack;
// reference to a fast track which is about to be removed
sp<IAfRecordTrack> fastTrackToRemove;
bool silenceFastCapture = false;
{ // scope for mutex()
audio_utils::unique_lock _l(mutex());
processConfigEvents_l();
// check exitPending here because checkForNewParameters_l() and
// checkForNewParameters_l() can temporarily release mutex()
if (exitPending()) {
break;
}
// sleep with mutex unlocked
if (sleepUs > 0) {
ATRACE_BEGIN("sleepC");
(void)mWaitWorkCV.wait_for(_l, std::chrono::microseconds(sleepUs));
ATRACE_END();
sleepUs = 0;
continue;
}
// if no active track(s), then standby and release wakelock
size_t size = mActiveTracks.size();
if (size == 0) {
standbyIfNotAlreadyInStandby();
// exitPending() can't become true here
releaseWakeLock_l();
ALOGV("RecordThread: loop stopping");
// go to sleep
mWaitWorkCV.wait(_l);
ALOGV("RecordThread: loop starting");
goto reacquire_wakelock;
}
bool doBroadcast = false;
bool allStopped = true;
for (size_t i = 0; i < size; ) {
if (activeTrack) { // ensure track release is outside lock.
oldActiveTracks.emplace_back(std::move(activeTrack));
}
activeTrack = mActiveTracks[i];
if (activeTrack->isTerminated()) {
if (activeTrack->isFastTrack()) {
ALOG_ASSERT(fastTrackToRemove == 0);
fastTrackToRemove = activeTrack;
}
removeTrack_l(activeTrack);
mActiveTracks.remove(activeTrack);
size--;
continue;
}
IAfTrackBase::track_state activeTrackState = activeTrack->state();
switch (activeTrackState) {
case IAfTrackBase::PAUSING:
mActiveTracks.remove(activeTrack);
activeTrack->setState(IAfTrackBase::PAUSED);
if (activeTrack->isFastTrack()) {
ALOGV("%s fast track is paused, thus removed from active list", __func__);
// Keep a ref on fast track to wait for FastCapture thread to get updated
// state before potential track removal
fastTrackToRemove = activeTrack;
}
doBroadcast = true;
size--;
continue;
case IAfTrackBase::STARTING_1:
sleepUs = 10000;
i++;
allStopped = false;
continue;
case IAfTrackBase::STARTING_2:
doBroadcast = true;
if (mStandby) {
mThreadMetrics.logBeginInterval();
mThreadSnapshot.onBegin();
mStandby = false;
}
activeTrack->setState(IAfTrackBase::ACTIVE);
allStopped = false;
break;
case IAfTrackBase::ACTIVE:
allStopped = false;
break;
case IAfTrackBase::IDLE: // cannot be on ActiveTracks if idle
case IAfTrackBase::PAUSED: // cannot be on ActiveTracks if paused
case IAfTrackBase::STOPPED: // cannot be on ActiveTracks if destroyed/terminated
default:
LOG_ALWAYS_FATAL("%s: Unexpected active track state:%d, id:%d, tracks:%zu",
__func__, activeTrackState, activeTrack->id(), size);
}
if (activeTrack->isFastTrack()) {
ALOG_ASSERT(!mFastTrackAvail);
ALOG_ASSERT(fastTrack == 0);
// if the active fast track is silenced either:
// 1) silence the whole capture from fast capture buffer if this is
// the only active track
// 2) invalidate this track: this will cause the client to reconnect and possibly
// be invalidated again until unsilenced
bool invalidate = false;
if (activeTrack->isSilenced()) {
if (size > 1) {
invalidate = true;
} else {
silenceFastCapture = true;
}
}
// Invalidate fast tracks if access to audio history is required as this is not
// possible with fast tracks. Once the fast track has been invalidated, no new
// fast track will be created until mMaxSharedAudioHistoryMs is cleared.
if (mMaxSharedAudioHistoryMs != 0) {
invalidate = true;
}
if (invalidate) {
activeTrack->invalidate();
ALOG_ASSERT(fastTrackToRemove == 0);
fastTrackToRemove = activeTrack;
removeTrack_l(activeTrack);
mActiveTracks.remove(activeTrack);
size--;
continue;
}
fastTrack = activeTrack;
}
activeTracks.add(activeTrack);
i++;
}
mActiveTracks.updatePowerState_l(this);
updateMetadata_l();
if (allStopped) {
standbyIfNotAlreadyInStandby();
}
if (doBroadcast) {
mStartStopCV.notify_all();
}
// sleep if there are no active tracks to process
if (activeTracks.isEmpty()) {
if (sleepUs == 0) {
sleepUs = kRecordThreadSleepUs;
}
continue;
}
sleepUs = 0;
timestampCorrectionEnabled = isTimestampCorrectionEnabled_l();
lockEffectChains_l(effectChains);
}
// thread mutex is now unlocked, mActiveTracks unknown, activeTracks.size() > 0
size_t size = effectChains.size();
for (size_t i = 0; i < size; i++) {
// thread mutex is not locked, but effect chain is locked
effectChains[i]->process_l();
}
// Push a new fast capture state if fast capture is not already running, or cblk change
if (mFastCapture != 0) {
FastCaptureStateQueue *sq = mFastCapture->sq();
FastCaptureState *state = sq->begin();
bool didModify = false;
FastCaptureStateQueue::block_t block = FastCaptureStateQueue::BLOCK_UNTIL_PUSHED;
if (state->mCommand != FastCaptureState::READ_WRITE /* FIXME &&
(kUseFastMixer != FastMixer_Dynamic || state->mTrackMask > 1)*/) {
if (state->mCommand == FastCaptureState::COLD_IDLE) {
int32_t old = android_atomic_inc(&mFastCaptureFutex);
if (old == -1) {
(void) syscall(__NR_futex, &mFastCaptureFutex, FUTEX_WAKE_PRIVATE, 1);
}
}
state->mCommand = FastCaptureState::READ_WRITE;
#if 0 // FIXME
mFastCaptureDumpState.increaseSamplingN(mAfThreadCallback->isLowRamDevice() ?
FastThreadDumpState::kSamplingNforLowRamDevice :
FastThreadDumpState::kSamplingN);
#endif
didModify = true;
}
audio_track_cblk_t *cblkOld = state->mCblk;
audio_track_cblk_t *cblkNew = fastTrack != 0 ? fastTrack->cblk() : NULL;
if (cblkNew != cblkOld) {
state->mCblk = cblkNew;
// block until acked if removing a fast track
if (cblkOld != NULL) {
block = FastCaptureStateQueue::BLOCK_UNTIL_ACKED;
}
didModify = true;
}
AudioBufferProvider* abp = (fastTrack != 0 && fastTrack->isPatchTrack()) ?
reinterpret_cast<AudioBufferProvider*>(fastTrack.get()) : nullptr;
if (state->mFastPatchRecordBufferProvider != abp) {
state->mFastPatchRecordBufferProvider = abp;
state->mFastPatchRecordFormat = fastTrack == 0 ?
AUDIO_FORMAT_INVALID : fastTrack->format();
didModify = true;
}
if (state->mSilenceCapture != silenceFastCapture) {
state->mSilenceCapture = silenceFastCapture;
didModify = true;
}
sq->end(didModify);
if (didModify) {
sq->push(block);
#if 0
if (kUseFastCapture == FastCapture_Dynamic) {
mNormalSource = mPipeSource;
}
#endif
}
}
// now run the fast track destructor with thread mutex unlocked
fastTrackToRemove.clear();
// Read from HAL to keep up with fastest client if multiple active tracks, not slowest one.
// Only the client(s) that are too slow will overrun. But if even the fastest client is too
// slow, then this RecordThread will overrun by not calling HAL read often enough.
// If destination is non-contiguous, first read past the nominal end of buffer, then
// copy to the right place. Permitted because mRsmpInBuffer was over-allocated.
int32_t rear = mRsmpInRear & (mRsmpInFramesP2 - 1);
ssize_t framesRead = 0; // not needed, remove clang-tidy warning.
const int64_t lastIoBeginNs = systemTime(); // start IO timing
// If an NBAIO source is present, use it to read the normal capture's data
// Next, we start collecting
if (mPipeSource != 0) {
size_t framesToRead = min(mRsmpInFramesOA - rear, mPipeFramesP2 / 2);
// The audio fifo read() returns OVERRUN on overflow, and advances the read pointer
// to the full buffer point (clearing the overflow condition). Upon OVERRUN error,
// we immediately retry the read() to get data and prevent another overflow.
for (int retries = 0; retries <= 2; ++retries) {
ALOGW_IF(retries > 0, "overrun on read from pipe, retry #%d", retries);
framesRead = mPipeSource->read((uint8_t*)mRsmpInBuffer + rear * mFrameSize, // Collect through HAL
framesToRead);
if (framesRead != OVERRUN) break;
}
const ssize_t availableToRead = mPipeSource->availableToRead();
if (availableToRead >= 0) {
mMonopipePipeDepthStats.add(availableToRead);
// PipeSource is the primary clock. It is up to the AudioRecord client to keep up.
LOG_ALWAYS_FATAL_IF((size_t)availableToRead > mPipeFramesP2,
"more frames to read than fifo size, %zd > %zu",
availableToRead, mPipeFramesP2);
const size_t pipeFramesFree = mPipeFramesP2 - availableToRead;
const size_t sleepFrames = min(pipeFramesFree, mRsmpInFramesP2) / 2;
ALOGVV("mPipeFramesP2:%zu mRsmpInFramesP2:%zu sleepFrames:%zu availableToRead:%zd",
mPipeFramesP2, mRsmpInFramesP2, sleepFrames, availableToRead);
sleepUs = (sleepFrames * 1000000LL) / mSampleRate;
}
if (framesRead < 0) {
status_t status = (status_t) framesRead;
switch (status) {
case OVERRUN:
ALOGW("overrun on read from pipe");
framesRead = 0;
break;
case NEGOTIATE:
ALOGE("re-negotiation is needed");
framesRead = -1; // Will cause an attempt to recover.
break;
default:
ALOGE("unknown error %d on read from pipe", status);
break;
}
}
// otherwise use the HAL / AudioStreamIn directly