오디오 믹서수정

02_day/test-mixer/Android.mk

@@ -1,24 +1,18 @@
 LOCAL_PATH:= $(call my-dir)
-
 include $(CLEAR_VARS)
 
-LOCAL_SRC_FILES:= \
+LOCAL_SRC_FILES:=  test-mixer.cpp  AudioMixer.cpp.arm BufferProviders.cpp
-	test-mixer.cpp \
-	AudioMixer.cpp.arm \
 
 LOCAL_C_INCLUDES := \
-	bionic \
-	bionic/libstdc++/include \
-	external/stlport/stlport \
 	$(call include-path-for, audio-effects) \
 	$(call include-path-for, audio-utils) \
-	frameworks/av/services/audioflinger
+	frameworks/av/services/audioflinger \
+	external/sonic
 
 LOCAL_STATIC_LIBRARIES := \
 	libsndfile
 
 LOCAL_SHARED_LIBRARIES := \
-	libstlport \
 	libeffects \
 	libnbaio \
 	libcommon_time_client \
@@ -27,10 +21,13 @@ LOCAL_SHARED_LIBRARIES := \
 	libdl \
 	libcutils \
 	libutils \
-	liblog
+	liblog \
+	libsonic
 
 LOCAL_MODULE:= test-mixer
 
 LOCAL_MODULE_TAGS := optional
 
+LOCAL_CXX_STL := libc++
+
 include $(BUILD_EXECUTABLE)

02_day/test-mixer/AudioMixer.cpp

@@ -39,9 +39,6 @@
 #include <common_time/local_clock.h>
 #include <common_time/cc_helper.h>
 
-#include <media/EffectsFactoryApi.h>
-#include <audio_effects/effect_downmix.h>
-
 #include "AudioMixerOps.h"
 #include "AudioMixer.h"
 
@@ -69,9 +66,16 @@
 #define ARRAY_SIZE(x) (sizeof(x)/sizeof((x)[0]))
 #endif
 
-// Set kUseNewMixer to true to use the new mixer engine. Otherwise the
+// TODO: Move these macro/inlines to a header file.
-// original code will be used.  This is false for now.
+template <typename T>
-static const bool kUseNewMixer = false;
+static inline
+T max(const T& x, const T& y) {
+    return x > y ? x : y;
+}
+
+// Set kUseNewMixer to true to use the new mixer engine always. Otherwise the
+// original code will be used for stereo sinks, the new mixer for multichannel.
+static const bool kUseNewMixer = true;
 
 // Set kUseFloat to true to allow floating input into the mixer engine.
 // If kUseNewMixer is false, this is ignored or may be overridden internally
@@ -91,288 +95,6 @@ T min(const T& a, const T& b)
     return a < b ? a : b;
 }
 
-AudioMixer::CopyBufferProvider::CopyBufferProvider(size_t inputFrameSize,
-        size_t outputFrameSize, size_t bufferFrameCount) :
-        mInputFrameSize(inputFrameSize),
-        mOutputFrameSize(outputFrameSize),
-        mLocalBufferFrameCount(bufferFrameCount),
-        mLocalBufferData(NULL),
-        mConsumed(0)
-{
-    ALOGV("CopyBufferProvider(%p)(%zu, %zu, %zu)", this,
-            inputFrameSize, outputFrameSize, bufferFrameCount);
-    LOG_ALWAYS_FATAL_IF(inputFrameSize < outputFrameSize && bufferFrameCount == 0,
-            "Requires local buffer if inputFrameSize(%zu) < outputFrameSize(%zu)",
-            inputFrameSize, outputFrameSize);
-    if (mLocalBufferFrameCount) {
-        (void)posix_memalign(&mLocalBufferData, 32, mLocalBufferFrameCount * mOutputFrameSize);
-    }
-    mBuffer.frameCount = 0;
-}
-
-AudioMixer::CopyBufferProvider::~CopyBufferProvider()
-{
-    ALOGV("~CopyBufferProvider(%p)", this);
-    if (mBuffer.frameCount != 0) {
-        mTrackBufferProvider->releaseBuffer(&mBuffer);
-    }
-    free(mLocalBufferData);
-}
-
-status_t AudioMixer::CopyBufferProvider::getNextBuffer(AudioBufferProvider::Buffer *pBuffer,
-        int64_t pts)
-{
-    //ALOGV("CopyBufferProvider(%p)::getNextBuffer(%p (%zu), %lld)",
-    //        this, pBuffer, pBuffer->frameCount, pts);
-    if (mLocalBufferFrameCount == 0) {
-        status_t res = mTrackBufferProvider->getNextBuffer(pBuffer, pts);
-        if (res == OK) {
-            copyFrames(pBuffer->raw, pBuffer->raw, pBuffer->frameCount);
-        }
-        return res;
-    }
-    if (mBuffer.frameCount == 0) {
-        mBuffer.frameCount = pBuffer->frameCount;
-        status_t res = mTrackBufferProvider->getNextBuffer(&mBuffer, pts);
-        // At one time an upstream buffer provider had
-        // res == OK and mBuffer.frameCount == 0, doesn't seem to happen now 7/18/2014.
-        //
-        // By API spec, if res != OK, then mBuffer.frameCount == 0.
-        // but there may be improper implementations.
-        ALOG_ASSERT(res == OK || mBuffer.frameCount == 0);
-        if (res != OK || mBuffer.frameCount == 0) { // not needed by API spec, but to be safe.
-            pBuffer->raw = NULL;
-            pBuffer->frameCount = 0;
-            return res;
-        }
-        mConsumed = 0;
-    }
-    ALOG_ASSERT(mConsumed < mBuffer.frameCount);
-    size_t count = min(mLocalBufferFrameCount, mBuffer.frameCount - mConsumed);
-    count = min(count, pBuffer->frameCount);
-    pBuffer->raw = mLocalBufferData;
-    pBuffer->frameCount = count;
-    copyFrames(pBuffer->raw, (uint8_t*)mBuffer.raw + mConsumed * mInputFrameSize,
-            pBuffer->frameCount);
-    return OK;
-}
-
-void AudioMixer::CopyBufferProvider::releaseBuffer(AudioBufferProvider::Buffer *pBuffer)
-{
-    //ALOGV("CopyBufferProvider(%p)::releaseBuffer(%p(%zu))",
-    //        this, pBuffer, pBuffer->frameCount);
-    if (mLocalBufferFrameCount == 0) {
-        mTrackBufferProvider->releaseBuffer(pBuffer);
-        return;
-    }
-    // LOG_ALWAYS_FATAL_IF(pBuffer->frameCount == 0, "Invalid framecount");
-    mConsumed += pBuffer->frameCount; // TODO: update for efficiency to reuse existing content
-    if (mConsumed != 0 && mConsumed >= mBuffer.frameCount) {
-        mTrackBufferProvider->releaseBuffer(&mBuffer);
-        ALOG_ASSERT(mBuffer.frameCount == 0);
-    }
-    pBuffer->raw = NULL;
-    pBuffer->frameCount = 0;
-}
-
-void AudioMixer::CopyBufferProvider::reset()
-{
-    if (mBuffer.frameCount != 0) {
-        mTrackBufferProvider->releaseBuffer(&mBuffer);
-    }
-    mConsumed = 0;
-}
-
-AudioMixer::DownmixerBufferProvider::DownmixerBufferProvider(
-        audio_channel_mask_t inputChannelMask,
-        audio_channel_mask_t outputChannelMask, audio_format_t format,
-        uint32_t sampleRate, int32_t sessionId, size_t bufferFrameCount) :
-        CopyBufferProvider(
-            audio_bytes_per_sample(format) * audio_channel_count_from_out_mask(inputChannelMask),
-            audio_bytes_per_sample(format) * audio_channel_count_from_out_mask(outputChannelMask),
-            bufferFrameCount)  // set bufferFrameCount to 0 to do in-place
-{
-    ALOGV("DownmixerBufferProvider(%p)(%#x, %#x, %#x %u %d)",
-            this, inputChannelMask, outputChannelMask, format,
-            sampleRate, sessionId);
-    if (!sIsMultichannelCapable
-            || EffectCreate(&sDwnmFxDesc.uuid,
-                    sessionId,
-                    SESSION_ID_INVALID_AND_IGNORED,
-                    &mDownmixHandle) != 0) {
-         ALOGE("DownmixerBufferProvider() error creating downmixer effect");
-         mDownmixHandle = NULL;
-         return;
-     }
-     // channel input configuration will be overridden per-track
-     mDownmixConfig.inputCfg.channels = inputChannelMask;   // FIXME: Should be bits
-     mDownmixConfig.outputCfg.channels = outputChannelMask; // FIXME: should be bits
-     mDownmixConfig.inputCfg.format = format;
-     mDownmixConfig.outputCfg.format = format;
-     mDownmixConfig.inputCfg.samplingRate = sampleRate;
-     mDownmixConfig.outputCfg.samplingRate = sampleRate;
-     mDownmixConfig.inputCfg.accessMode = EFFECT_BUFFER_ACCESS_READ;
-     mDownmixConfig.outputCfg.accessMode = EFFECT_BUFFER_ACCESS_WRITE;
-     // input and output buffer provider, and frame count will not be used as the downmix effect
-     // process() function is called directly (see DownmixerBufferProvider::getNextBuffer())
-     mDownmixConfig.inputCfg.mask = EFFECT_CONFIG_SMP_RATE | EFFECT_CONFIG_CHANNELS |
-             EFFECT_CONFIG_FORMAT | EFFECT_CONFIG_ACC_MODE;
-     mDownmixConfig.outputCfg.mask = mDownmixConfig.inputCfg.mask;
-
-     int cmdStatus;
-     uint32_t replySize = sizeof(int);
-
-     // Configure downmixer
-     status_t status = (*mDownmixHandle)->command(mDownmixHandle,
-             EFFECT_CMD_SET_CONFIG /*cmdCode*/, sizeof(effect_config_t) /*cmdSize*/,
-             &mDownmixConfig /*pCmdData*/,
-             &replySize, &cmdStatus /*pReplyData*/);
-     if (status != 0 || cmdStatus != 0) {
-         ALOGE("DownmixerBufferProvider() error %d cmdStatus %d while configuring downmixer",
-                 status, cmdStatus);
-         EffectRelease(mDownmixHandle);
-         mDownmixHandle = NULL;
-         return;
-     }
-
-     // Enable downmixer
-     replySize = sizeof(int);
-     status = (*mDownmixHandle)->command(mDownmixHandle,
-             EFFECT_CMD_ENABLE /*cmdCode*/, 0 /*cmdSize*/, NULL /*pCmdData*/,
-             &replySize, &cmdStatus /*pReplyData*/);
-     if (status != 0 || cmdStatus != 0) {
-         ALOGE("DownmixerBufferProvider() error %d cmdStatus %d while enabling downmixer",
-                 status, cmdStatus);
-         EffectRelease(mDownmixHandle);
-         mDownmixHandle = NULL;
-         return;
-     }
-
-     // Set downmix type
-     // parameter size rounded for padding on 32bit boundary
-     const int psizePadded = ((sizeof(downmix_params_t) - 1)/sizeof(int) + 1) * sizeof(int);
-     const int downmixParamSize =
-             sizeof(effect_param_t) + psizePadded + sizeof(downmix_type_t);
-     effect_param_t * const param = (effect_param_t *) malloc(downmixParamSize);
-     param->psize = sizeof(downmix_params_t);
-     const downmix_params_t downmixParam = DOWNMIX_PARAM_TYPE;
-     memcpy(param->data, &downmixParam, param->psize);
-     const downmix_type_t downmixType = DOWNMIX_TYPE_FOLD;
-     param->vsize = sizeof(downmix_type_t);
-     memcpy(param->data + psizePadded, &downmixType, param->vsize);
-     replySize = sizeof(int);
-     status = (*mDownmixHandle)->command(mDownmixHandle,
-             EFFECT_CMD_SET_PARAM /* cmdCode */, downmixParamSize /* cmdSize */,
-             param /*pCmdData*/, &replySize, &cmdStatus /*pReplyData*/);
-     free(param);
-     if (status != 0 || cmdStatus != 0) {
-         ALOGE("DownmixerBufferProvider() error %d cmdStatus %d while setting downmix type",
-                 status, cmdStatus);
-         EffectRelease(mDownmixHandle);
-         mDownmixHandle = NULL;
-         return;
-     }
-     ALOGV("DownmixerBufferProvider() downmix type set to %d", (int) downmixType);
-}
-
-AudioMixer::DownmixerBufferProvider::~DownmixerBufferProvider()
-{
-    ALOGV("~DownmixerBufferProvider (%p)", this);
-    EffectRelease(mDownmixHandle);
-    mDownmixHandle = NULL;
-}
-
-void AudioMixer::DownmixerBufferProvider::copyFrames(void *dst, const void *src, size_t frames)
-{
-    mDownmixConfig.inputCfg.buffer.frameCount = frames;
-    mDownmixConfig.inputCfg.buffer.raw = const_cast<void *>(src);
-    mDownmixConfig.outputCfg.buffer.frameCount = frames;
-    mDownmixConfig.outputCfg.buffer.raw = dst;
-    // may be in-place if src == dst.
-    status_t res = (*mDownmixHandle)->process(mDownmixHandle,
-            &mDownmixConfig.inputCfg.buffer, &mDownmixConfig.outputCfg.buffer);
-    ALOGE_IF(res != OK, "DownmixBufferProvider error %d", res);
-}
-
-/* call once in a pthread_once handler. */
-/*static*/ status_t AudioMixer::DownmixerBufferProvider::init()
-{
-    // find multichannel downmix effect if we have to play multichannel content
-    uint32_t numEffects = 0;
-    int ret = EffectQueryNumberEffects(&numEffects);
-    if (ret != 0) {
-        ALOGE("AudioMixer() error %d querying number of effects", ret);
-        return NO_INIT;
-    }
-    ALOGV("EffectQueryNumberEffects() numEffects=%d", numEffects);
-
-    for (uint32_t i = 0 ; i < numEffects ; i++) {
-        if (EffectQueryEffect(i, &sDwnmFxDesc) == 0) {
-            ALOGV("effect %d is called %s", i, sDwnmFxDesc.name);
-            if (memcmp(&sDwnmFxDesc.type, EFFECT_UIID_DOWNMIX, sizeof(effect_uuid_t)) == 0) {
-                ALOGI("found effect \"%s\" from %s",
-                        sDwnmFxDesc.name, sDwnmFxDesc.implementor);
-                sIsMultichannelCapable = true;
-                break;
-            }
-        }
-    }
-    ALOGW_IF(!sIsMultichannelCapable, "unable to find downmix effect");
-    return NO_INIT;
-}
-
-/*static*/ bool AudioMixer::DownmixerBufferProvider::sIsMultichannelCapable = false;
-/*static*/ effect_descriptor_t AudioMixer::DownmixerBufferProvider::sDwnmFxDesc;
-
-AudioMixer::RemixBufferProvider::RemixBufferProvider(audio_channel_mask_t inputChannelMask,
-        audio_channel_mask_t outputChannelMask, audio_format_t format,
-        size_t bufferFrameCount) :
-        CopyBufferProvider(
-                audio_bytes_per_sample(format)
-                    * audio_channel_count_from_out_mask(inputChannelMask),
-                audio_bytes_per_sample(format)
-                    * audio_channel_count_from_out_mask(outputChannelMask),
-                bufferFrameCount),
-        mFormat(format),
-        mSampleSize(audio_bytes_per_sample(format)),
-        mInputChannels(audio_channel_count_from_out_mask(inputChannelMask)),
-        mOutputChannels(audio_channel_count_from_out_mask(outputChannelMask))
-{
-    ALOGV("RemixBufferProvider(%p)(%#x, %#x, %#x) %zu %zu",
-            this, format, inputChannelMask, outputChannelMask,
-            mInputChannels, mOutputChannels);
-    // TODO: consider channel representation in index array formulation
-    // We ignore channel representation, and just use the bits.
-    memcpy_by_index_array_initialization(mIdxAry, ARRAY_SIZE(mIdxAry),
-            audio_channel_mask_get_bits(outputChannelMask),
-            audio_channel_mask_get_bits(inputChannelMask));
-}
-
-void AudioMixer::RemixBufferProvider::copyFrames(void *dst, const void *src, size_t frames)
-{
-    memcpy_by_index_array(dst, mOutputChannels,
-            src, mInputChannels, mIdxAry, mSampleSize, frames);
-}
-
-AudioMixer::ReformatBufferProvider::ReformatBufferProvider(int32_t channels,
-        audio_format_t inputFormat, audio_format_t outputFormat,
-        size_t bufferFrameCount) :
-        CopyBufferProvider(
-            channels * audio_bytes_per_sample(inputFormat),
-            channels * audio_bytes_per_sample(outputFormat),
-            bufferFrameCount),
-        mChannels(channels),
-        mInputFormat(inputFormat),
-        mOutputFormat(outputFormat)
-{
-    ALOGV("ReformatBufferProvider(%p)(%d, %#x, %#x)", this, channels, inputFormat, outputFormat);
-}
-
-void AudioMixer::ReformatBufferProvider::copyFrames(void *dst, const void *src, size_t frames)
-{
-    memcpy_by_audio_format(dst, mOutputFormat, src, mInputFormat, frames * mChannels);
-}
-
 // ----------------------------------------------------------------------------
 
 // Ensure mConfiguredNames bitmask is initialized properly on all architectures.
@@ -407,6 +129,7 @@ AudioMixer::AudioMixer(size_t frameCount, uint32_t sampleRate, uint32_t maxNumTr
         t->resampler = NULL;
         t->downmixerBufferProvider = NULL;
         t->mReformatBufferProvider = NULL;
+        t->mTimestretchBufferProvider = NULL;
         t++;
     }
 
@@ -419,6 +142,7 @@ AudioMixer::~AudioMixer()
         delete t->resampler;
         delete t->downmixerBufferProvider;
         delete t->mReformatBufferProvider;
+        delete t->mTimestretchBufferProvider;
         t++;
     }
     delete [] mState.outputTemp;
@@ -430,6 +154,10 @@ void AudioMixer::setLog(NBLog::Writer *log)
     mState.mLog = log;
 }
 
+static inline audio_format_t selectMixerInFormat(audio_format_t inputFormat __unused) {
+    return kUseFloat && kUseNewMixer ? AUDIO_FORMAT_PCM_FLOAT : AUDIO_FORMAT_PCM_16_BIT;
+}
+
 int AudioMixer::getTrackName(audio_channel_mask_t channelMask,
         audio_format_t format, int sessionId)
 {
@@ -492,24 +220,25 @@ int AudioMixer::getTrackName(audio_channel_mask_t channelMask,
         t->mInputBufferProvider = NULL;
         t->mReformatBufferProvider = NULL;
         t->downmixerBufferProvider = NULL;
+        t->mPostDownmixReformatBufferProvider = NULL;
+        t->mTimestretchBufferProvider = NULL;
         t->mMixerFormat = AUDIO_FORMAT_PCM_16_BIT;
         t->mFormat = format;
-        t->mMixerInFormat = kUseFloat && kUseNewMixer
+        t->mMixerInFormat = selectMixerInFormat(format);
-                ? AUDIO_FORMAT_PCM_FLOAT : AUDIO_FORMAT_PCM_16_BIT;
+        t->mDownmixRequiresFormat = AUDIO_FORMAT_INVALID; // no format required
         t->mMixerChannelMask = audio_channel_mask_from_representation_and_bits(
                 AUDIO_CHANNEL_REPRESENTATION_POSITION, AUDIO_CHANNEL_OUT_STEREO);
         t->mMixerChannelCount = audio_channel_count_from_out_mask(t->mMixerChannelMask);
+        t->mPlaybackRate = AUDIO_PLAYBACK_RATE_DEFAULT;
         // Check the downmixing (or upmixing) requirements.
-        status_t status = initTrackDownmix(t, n);
+        status_t status = t->prepareForDownmix();
         if (status != OK) {
             ALOGE("AudioMixer::getTrackName invalid channelMask (%#x)", channelMask);
             return -1;
         }
-        // initTrackDownmix() may change the input format requirement.
+        // prepareForDownmix() may change mDownmixRequiresFormat
-        // If you desire floating point input to the mixer, it may change
-        // to integer because the downmixer requires integer to process.
         ALOGVV("mMixerFormat:%#x  mMixerInFormat:%#x\n", t->mMixerFormat, t->mMixerInFormat);
-        prepareTrackForReformat(t, n);
+        t->prepareForReformat();
         mTrackNames |= 1 << n;
         return TRACK0 + n;
     }
@@ -526,7 +255,7 @@ void AudioMixer::invalidateState(uint32_t mask)
  }
 
 // Called when channel masks have changed for a track name
-// TODO: Fix Downmixbufferprofider not to (possibly) change mixer input format,
+// TODO: Fix DownmixerBufferProvider not to (possibly) change mixer input format,
 // which will simplify this logic.
 bool AudioMixer::setChannelMasks(int name,
         audio_channel_mask_t trackChannelMask, audio_channel_mask_t mixerChannelMask) {
@@ -551,21 +280,18 @@ bool AudioMixer::setChannelMasks(int name,
 
     // channel masks have changed, does this track need a downmixer?
     // update to try using our desired format (if we aren't already using it)
-    const audio_format_t prevMixerInFormat = track.mMixerInFormat;
+    const audio_format_t prevDownmixerFormat = track.mDownmixRequiresFormat;
-    track.mMixerInFormat = kUseFloat && kUseNewMixer
+    const status_t status = mState.tracks[name].prepareForDownmix();
-            ? AUDIO_FORMAT_PCM_FLOAT : AUDIO_FORMAT_PCM_16_BIT;
-    const status_t status = initTrackDownmix(&mState.tracks[name], name);
     ALOGE_IF(status != OK,
-            "initTrackDownmix error %d, track channel mask %#x, mixer channel mask %#x",
+            "prepareForDownmix error %d, track channel mask %#x, mixer channel mask %#x",
             status, track.channelMask, track.mMixerChannelMask);
 
-    const bool mixerInFormatChanged = prevMixerInFormat != track.mMixerInFormat;
+    if (prevDownmixerFormat != track.mDownmixRequiresFormat) {
-    if (mixerInFormatChanged) {
+        track.prepareForReformat(); // because of downmixer, track format may change!
-        prepareTrackForReformat(&track, name); // because of downmixer, track format may change!
     }
 
-    if (track.resampler && (mixerInFormatChanged || mixerChannelCountChanged)) {
+    if (track.resampler && mixerChannelCountChanged) {
-        // resampler input format or channels may have changed.
+        // resampler channels may have changed.
         const uint32_t resetToSampleRate = track.sampleRate;
         delete track.resampler;
         track.resampler = NULL;
@@ -576,99 +302,129 @@ bool AudioMixer::setChannelMasks(int name,
     return true;
 }
 
-status_t AudioMixer::initTrackDownmix(track_t* pTrack, int trackName)
+void AudioMixer::track_t::unprepareForDownmix() {
-{
+    ALOGV("AudioMixer::unprepareForDownmix(%p)", this);
-    // Only remix (upmix or downmix) if the track and mixer/device channel masks
-    // are not the same and not handled internally, as mono -> stereo currently is.
-    if (pTrack->channelMask != pTrack->mMixerChannelMask
-            && !(pTrack->channelMask == AUDIO_CHANNEL_OUT_MONO
-                    && pTrack->mMixerChannelMask == AUDIO_CHANNEL_OUT_STEREO)) {
-        return prepareTrackForDownmix(pTrack, trackName);
-    }
-    // no remix necessary
-    unprepareTrackForDownmix(pTrack, trackName);
-    return NO_ERROR;
-}
 
-void AudioMixer::unprepareTrackForDownmix(track_t* pTrack, int trackName __unused) {
+    mDownmixRequiresFormat = AUDIO_FORMAT_INVALID;
-    ALOGV("AudioMixer::unprepareTrackForDownmix(%d)", trackName);
+    if (downmixerBufferProvider != NULL) {
-
-    if (pTrack->downmixerBufferProvider != NULL) {
         // this track had previously been configured with a downmixer, delete it
         ALOGV(" deleting old downmixer");
-        delete pTrack->downmixerBufferProvider;
+        delete downmixerBufferProvider;
-        pTrack->downmixerBufferProvider = NULL;
+        downmixerBufferProvider = NULL;
-        reconfigureBufferProviders(pTrack);
+        reconfigureBufferProviders();
     } else {
         ALOGV(" nothing to do, no downmixer to delete");
     }
 }
 
-status_t AudioMixer::prepareTrackForDownmix(track_t* pTrack, int trackName)
+status_t AudioMixer::track_t::prepareForDownmix()
 {
-    ALOGV("AudioMixer::prepareTrackForDownmix(%d) with mask 0x%x", trackName, pTrack->channelMask);
+    ALOGV("AudioMixer::prepareForDownmix(%p) with mask 0x%x",
+            this, channelMask);
 
     // discard the previous downmixer if there was one
-    unprepareTrackForDownmix(pTrack, trackName);
+    unprepareForDownmix();
-    if (DownmixerBufferProvider::isMultichannelCapable()) {
+    // MONO_HACK Only remix (upmix or downmix) if the track and mixer/device channel masks
-        DownmixerBufferProvider* pDbp = new DownmixerBufferProvider(pTrack->channelMask,
+    // are not the same and not handled internally, as mono -> stereo currently is.
-                pTrack->mMixerChannelMask,
+    if (channelMask == mMixerChannelMask
-                AUDIO_FORMAT_PCM_16_BIT /* TODO: use pTrack->mMixerInFormat, now only PCM 16 */,
+            || (channelMask == AUDIO_CHANNEL_OUT_MONO
-                pTrack->sampleRate, pTrack->sessionId, kCopyBufferFrameCount);
+                    && mMixerChannelMask == AUDIO_CHANNEL_OUT_STEREO)) {
+        return NO_ERROR;
+    }
+    // DownmixerBufferProvider is only used for position masks.
+    if (audio_channel_mask_get_representation(channelMask)
+                == AUDIO_CHANNEL_REPRESENTATION_POSITION
+            && DownmixerBufferProvider::isMultichannelCapable()) {
+        DownmixerBufferProvider* pDbp = new DownmixerBufferProvider(channelMask,
+                mMixerChannelMask,
+                AUDIO_FORMAT_PCM_16_BIT /* TODO: use mMixerInFormat, now only PCM 16 */,
+                sampleRate, sessionId, kCopyBufferFrameCount);
 
         if (pDbp->isValid()) { // if constructor completed properly
-            pTrack->mMixerInFormat = AUDIO_FORMAT_PCM_16_BIT; // PCM 16 bit required for downmix
+            mDownmixRequiresFormat = AUDIO_FORMAT_PCM_16_BIT; // PCM 16 bit required for downmix
-            pTrack->downmixerBufferProvider = pDbp;
+            downmixerBufferProvider = pDbp;
-            reconfigureBufferProviders(pTrack);
+            reconfigureBufferProviders();
             return NO_ERROR;
         }
         delete pDbp;
     }
 
     // Effect downmixer does not accept the channel conversion.  Let's use our remixer.
-    RemixBufferProvider* pRbp = new RemixBufferProvider(pTrack->channelMask,
+    RemixBufferProvider* pRbp = new RemixBufferProvider(channelMask,
-            pTrack->mMixerChannelMask, pTrack->mMixerInFormat, kCopyBufferFrameCount);
+            mMixerChannelMask, mMixerInFormat, kCopyBufferFrameCount);
     // Remix always finds a conversion whereas Downmixer effect above may fail.
-    pTrack->downmixerBufferProvider = pRbp;
+    downmixerBufferProvider = pRbp;
-    reconfigureBufferProviders(pTrack);
+    reconfigureBufferProviders();
     return NO_ERROR;
 }
 
-void AudioMixer::unprepareTrackForReformat(track_t* pTrack, int trackName __unused) {
+void AudioMixer::track_t::unprepareForReformat() {
-    ALOGV("AudioMixer::unprepareTrackForReformat(%d)", trackName);
+    ALOGV("AudioMixer::unprepareForReformat(%p)", this);
-    if (pTrack->mReformatBufferProvider != NULL) {
+    bool requiresReconfigure = false;
-        delete pTrack->mReformatBufferProvider;
+    if (mReformatBufferProvider != NULL) {
-        pTrack->mReformatBufferProvider = NULL;
+        delete mReformatBufferProvider;
-        reconfigureBufferProviders(pTrack);
+        mReformatBufferProvider = NULL;
+        requiresReconfigure = true;
+    }
+    if (mPostDownmixReformatBufferProvider != NULL) {
+        delete mPostDownmixReformatBufferProvider;
+        mPostDownmixReformatBufferProvider = NULL;
+        requiresReconfigure = true;
+    }
+    if (requiresReconfigure) {
+        reconfigureBufferProviders();
     }
 }
 
-status_t AudioMixer::prepareTrackForReformat(track_t* pTrack, int trackName)
+status_t AudioMixer::track_t::prepareForReformat()
 {
-    ALOGV("AudioMixer::prepareTrackForReformat(%d) with format %#x", trackName, pTrack->mFormat);
+    ALOGV("AudioMixer::prepareForReformat(%p) with format %#x", this, mFormat);
-    // discard the previous reformatter if there was one
+    // discard previous reformatters
-    unprepareTrackForReformat(pTrack, trackName);
+    unprepareForReformat();
-    // only configure reformatter if needed
+    // only configure reformatters as needed
-    if (pTrack->mFormat != pTrack->mMixerInFormat) {
+    const audio_format_t targetFormat = mDownmixRequiresFormat != AUDIO_FORMAT_INVALID
-        pTrack->mReformatBufferProvider = new ReformatBufferProvider(
+            ? mDownmixRequiresFormat : mMixerInFormat;
-                audio_channel_count_from_out_mask(pTrack->channelMask),
+    bool requiresReconfigure = false;
-                pTrack->mFormat, pTrack->mMixerInFormat,
+    if (mFormat != targetFormat) {
+        mReformatBufferProvider = new ReformatBufferProvider(
+                audio_channel_count_from_out_mask(channelMask),
+                mFormat,
+                targetFormat,
                 kCopyBufferFrameCount);
-        reconfigureBufferProviders(pTrack);
+        requiresReconfigure = true;
+    }
+    if (targetFormat != mMixerInFormat) {
+        mPostDownmixReformatBufferProvider = new ReformatBufferProvider(
+                audio_channel_count_from_out_mask(mMixerChannelMask),
+                targetFormat,
+                mMixerInFormat,
+                kCopyBufferFrameCount);
+        requiresReconfigure = true;
+    }
+    if (requiresReconfigure) {
+        reconfigureBufferProviders();
     }
     return NO_ERROR;
 }
 
-void AudioMixer::reconfigureBufferProviders(track_t* pTrack)
+void AudioMixer::track_t::reconfigureBufferProviders()
 {
-    pTrack->bufferProvider = pTrack->mInputBufferProvider;
+    bufferProvider = mInputBufferProvider;
-    if (pTrack->mReformatBufferProvider) {
+    if (mReformatBufferProvider) {
-        pTrack->mReformatBufferProvider->setBufferProvider(pTrack->bufferProvider);
+        mReformatBufferProvider->setBufferProvider(bufferProvider);
-        pTrack->bufferProvider = pTrack->mReformatBufferProvider;
+        bufferProvider = mReformatBufferProvider;
     }
-    if (pTrack->downmixerBufferProvider) {
+    if (downmixerBufferProvider) {
-        pTrack->downmixerBufferProvider->setBufferProvider(pTrack->bufferProvider);
+        downmixerBufferProvider->setBufferProvider(bufferProvider);
-        pTrack->bufferProvider = pTrack->downmixerBufferProvider;
+        bufferProvider = downmixerBufferProvider;
+    }
+    if (mPostDownmixReformatBufferProvider) {
+        mPostDownmixReformatBufferProvider->setBufferProvider(bufferProvider);
+        bufferProvider = mPostDownmixReformatBufferProvider;
+    }
+    if (mTimestretchBufferProvider) {
+        mTimestretchBufferProvider->setBufferProvider(bufferProvider);
+        bufferProvider = mTimestretchBufferProvider;
     }
 }
 
@@ -687,10 +443,12 @@ void AudioMixer::deleteTrackName(int name)
     delete track.resampler;
     track.resampler = NULL;
     // delete the downmixer
-    unprepareTrackForDownmix(&mState.tracks[name], name);
+    mState.tracks[name].unprepareForDownmix();
     // delete the reformatter
-    unprepareTrackForReformat(&mState.tracks[name], name);
+    mState.tracks[name].unprepareForReformat();
-
+    // delete the timestretch provider
+    delete track.mTimestretchBufferProvider;
+    track.mTimestretchBufferProvider = NULL;
     mTrackNames &= ~(1<<name);
 }
 
@@ -748,41 +506,99 @@ void AudioMixer::disable(int name)
 static inline bool setVolumeRampVariables(float newVolume, int32_t ramp,
         int16_t *pIntSetVolume, int32_t *pIntPrevVolume, int32_t *pIntVolumeInc,
         float *pSetVolume, float *pPrevVolume, float *pVolumeInc) {
+    // check floating point volume to see if it is identical to the previously
+    // set volume.
+    // We do not use a tolerance here (and reject changes too small)
+    // as it may be confusing to use a different value than the one set.
+    // If the resulting volume is too small to ramp, it is a direct set of the volume.
     if (newVolume == *pSetVolume) {
         return false;
     }
-    /* set the floating point volume variables */
+    if (newVolume < 0) {
-    if (ramp != 0) {
+        newVolume = 0; // should not have negative volumes
-        *pVolumeInc = (newVolume - *pSetVolume) / ramp;
-        *pPrevVolume = *pSetVolume;
     } else {
-        *pVolumeInc = 0;
+        switch (fpclassify(newVolume)) {
-        *pPrevVolume = newVolume;
+        case FP_SUBNORMAL:
+        case FP_NAN:
+            newVolume = 0;
+            break;
+        case FP_ZERO:
+            break; // zero volume is fine
+        case FP_INFINITE:
+            // Infinite volume could be handled consistently since
+            // floating point math saturates at infinities,
+            // but we limit volume to unity gain float.
+            // ramp = 0; break;
+            //
+            newVolume = AudioMixer::UNITY_GAIN_FLOAT;
+            break;
+        case FP_NORMAL:
+        default:
+            // Floating point does not have problems with overflow wrap
+            // that integer has.  However, we limit the volume to
+            // unity gain here.
+            // TODO: Revisit the volume limitation and perhaps parameterize.
+            if (newVolume > AudioMixer::UNITY_GAIN_FLOAT) {
+                newVolume = AudioMixer::UNITY_GAIN_FLOAT;
+            }
+            break;
+        }
     }
-    *pSetVolume = newVolume;
 
-    /* set the legacy integer volume variables */
+    // set floating point volume ramp
-    int32_t intVolume = newVolume * AudioMixer::UNITY_GAIN_INT;
+    if (ramp != 0) {
-    if (intVolume > AudioMixer::UNITY_GAIN_INT) {
+        // when the ramp completes, *pPrevVolume is set to *pSetVolume, so there
-        intVolume = AudioMixer::UNITY_GAIN_INT;
+        // is no computational mismatch; hence equality is checked here.
-    } else if (intVolume < 0) {
+        ALOGD_IF(*pPrevVolume != *pSetVolume, "previous float ramp hasn't finished,"
-        ALOGE("negative volume %.7g", newVolume);
+                " prev:%f  set_to:%f", *pPrevVolume, *pSetVolume);
-        intVolume = 0; // should never happen, but for safety check.
+        const float inc = (newVolume - *pPrevVolume) / ramp; // could be inf, nan, subnormal
+        const float maxv = max(newVolume, *pPrevVolume); // could be inf, cannot be nan, subnormal
+
+        if (isnormal(inc) // inc must be a normal number (no subnormals, infinite, nan)
+                && maxv + inc != maxv) { // inc must make forward progress
+            *pVolumeInc = inc;
+            // ramp is set now.
+            // Note: if newVolume is 0, then near the end of the ramp,
+            // it may be possible that the ramped volume may be subnormal or
+            // temporarily negative by a small amount or subnormal due to floating
+            // point inaccuracies.
+        } else {
+            ramp = 0; // ramp not allowed
         }
-    if (intVolume == *pIntSetVolume) {
+    }
-        *pIntVolumeInc = 0;
+
-        /* TODO: integer/float workaround: ignore floating volume ramp */
+    // compute and check integer volume, no need to check negative values
+    // The integer volume is limited to "unity_gain" to avoid wrapping and other
+    // audio artifacts, so it never reaches the range limit of U4.28.
+    // We safely use signed 16 and 32 bit integers here.
+    const float scaledVolume = newVolume * AudioMixer::UNITY_GAIN_INT; // not neg, subnormal, nan
+    const int32_t intVolume = (scaledVolume >= (float)AudioMixer::UNITY_GAIN_INT) ?
+            AudioMixer::UNITY_GAIN_INT : (int32_t)scaledVolume;
+
+    // set integer volume ramp
+    if (ramp != 0) {
+        // integer volume is U4.12 (to use 16 bit multiplies), but ramping uses U4.28.
+        // when the ramp completes, *pIntPrevVolume is set to *pIntSetVolume << 16, so there
+        // is no computational mismatch; hence equality is checked here.
+        ALOGD_IF(*pIntPrevVolume != *pIntSetVolume << 16, "previous int ramp hasn't finished,"
+                " prev:%d  set_to:%d", *pIntPrevVolume, *pIntSetVolume << 16);
+        const int32_t inc = ((intVolume << 16) - *pIntPrevVolume) / ramp;
+
+        if (inc != 0) { // inc must make forward progress
+            *pIntVolumeInc = inc;
+        } else {
+            ramp = 0; // ramp not allowed
+        }
+    }
+
+    // if no ramp, or ramp not allowed, then clear float and integer increments
+    if (ramp == 0) {
         *pVolumeInc = 0;
         *pPrevVolume = newVolume;
-        return true;
-    }
-    if (ramp != 0) {
-        *pIntVolumeInc = ((intVolume - *pIntSetVolume) << 16) / ramp;
-        *pIntPrevVolume = (*pIntVolumeInc == 0 ? intVolume : *pIntSetVolume) << 16;
-    } else {
         *pIntVolumeInc = 0;
         *pIntPrevVolume = intVolume << 16;
     }
+    *pSetVolume = newVolume;
     *pIntSetVolume = intVolume;
     return true;
 }
@@ -828,7 +644,7 @@ void AudioMixer::setParameter(int name, int target, int param, void *value)
                 ALOG_ASSERT(audio_is_linear_pcm(format), "Invalid format %#x", format);
                 track.mFormat = format;
                 ALOGV("setParameter(TRACK, FORMAT, %#x)", format);
-                prepareTrackForReformat(&track, name);
+                track.prepareForReformat();
                 invalidateState(1 << name);
             }
             } break;
@@ -912,6 +728,28 @@ void AudioMixer::setParameter(int name, int target, int param, void *value)
             }
         }
         break;
+        case TIMESTRETCH:
+            switch (param) {
+            case PLAYBACK_RATE: {
+                const AudioPlaybackRate *playbackRate =
+                        reinterpret_cast<AudioPlaybackRate*>(value);
+                ALOGW_IF(!isAudioPlaybackRateValid(*playbackRate),
+                        "bad parameters speed %f, pitch %f",playbackRate->mSpeed,
+                        playbackRate->mPitch);
+                if (track.setPlaybackRate(*playbackRate)) {
+                    ALOGV("setParameter(TIMESTRETCH, PLAYBACK_RATE, STRETCH_MODE, FALLBACK_MODE "
+                            "%f %f %d %d",
+                            playbackRate->mSpeed,
+                            playbackRate->mPitch,
+                            playbackRate->mStretchMode,
+                            playbackRate->mFallbackMode);
+                    // invalidateState(1 << name);
+                }
+            } break;
+            default:
+                LOG_ALWAYS_FATAL("setParameter timestretch: bad param %d", param);
+            }
+            break;
 
     default:
         LOG_ALWAYS_FATAL("setParameter: bad target %d", target);
@@ -931,11 +769,10 @@ bool AudioMixer::track_t::setResampler(uint32_t trackSampleRate, uint32_t devSam
                 // FIXME this is flawed for dynamic sample rates, as we choose the resampler
                 // quality level based on the initial ratio, but that could change later.
                 // Should have a way to distinguish tracks with static ratios vs. dynamic ratios.
-                if (!((trackSampleRate == 44100 && devSampleRate == 48000) ||
+                if (isMusicRate(trackSampleRate)) {
-                      (trackSampleRate == 48000 && devSampleRate == 44100))) {
-                    quality = AudioResampler::DYN_LOW_QUALITY;
-                } else {
                     quality = AudioResampler::DEFAULT_QUALITY;
+                } else {
+                    quality = AudioResampler::DYN_LOW_QUALITY;
                 }
 
                 // TODO: Remove MONO_HACK. Resampler sees #channels after the downmixer
@@ -957,6 +794,30 @@ bool AudioMixer::track_t::setResampler(uint32_t trackSampleRate, uint32_t devSam
     return false;
 }
 
+bool AudioMixer::track_t::setPlaybackRate(const AudioPlaybackRate &playbackRate)
+{
+    if ((mTimestretchBufferProvider == NULL &&
+            fabs(playbackRate.mSpeed - mPlaybackRate.mSpeed) < AUDIO_TIMESTRETCH_SPEED_MIN_DELTA &&
+            fabs(playbackRate.mPitch - mPlaybackRate.mPitch) < AUDIO_TIMESTRETCH_PITCH_MIN_DELTA) ||
+            isAudioPlaybackRateEqual(playbackRate, mPlaybackRate)) {
+        return false;
+    }
+    mPlaybackRate = playbackRate;
+    if (mTimestretchBufferProvider == NULL) {
+        // TODO: Remove MONO_HACK. Resampler sees #channels after the downmixer
+        // but if none exists, it is the channel count (1 for mono).
+        const int timestretchChannelCount = downmixerBufferProvider != NULL
+                ? mMixerChannelCount : channelCount;
+        mTimestretchBufferProvider = new TimestretchBufferProvider(timestretchChannelCount,
+                mMixerInFormat, sampleRate, playbackRate);
+        reconfigureBufferProviders();
+    } else {
+        reinterpret_cast<TimestretchBufferProvider*>(mTimestretchBufferProvider)
+                ->setPlaybackRate(playbackRate);
+    }
+    return true;
+}
+
 /* Checks to see if the volume ramp has completed and clears the increment
  * variables appropriately.
  *
@@ -974,7 +835,8 @@ inline void AudioMixer::track_t::adjustVolumeRamp(bool aux, bool useFloat)
 {
     if (useFloat) {
         for (uint32_t i = 0; i < MAX_NUM_VOLUMES; i++) {
-            if (mVolumeInc[i] != 0 && fabs(mVolume[i] - mPrevVolume[i]) <= fabs(mVolumeInc[i])) {
+            if ((mVolumeInc[i] > 0 && mPrevVolume[i] + mVolumeInc[i] >= mVolume[i]) ||
+                     (mVolumeInc[i] < 0 && mPrevVolume[i] + mVolumeInc[i] <= mVolume[i])) {
                 volumeInc[i] = 0;
                 prevVolume[i] = volume[i] << 16;
                 mVolumeInc[i] = 0.;
@@ -1032,10 +894,15 @@ void AudioMixer::setBufferProvider(int name, AudioBufferProvider* bufferProvider
     if (mState.tracks[name].mReformatBufferProvider != NULL) {
         mState.tracks[name].mReformatBufferProvider->reset();
     } else if (mState.tracks[name].downmixerBufferProvider != NULL) {
+        mState.tracks[name].downmixerBufferProvider->reset();
+    } else if (mState.tracks[name].mPostDownmixReformatBufferProvider != NULL) {
+        mState.tracks[name].mPostDownmixReformatBufferProvider->reset();
+    } else if (mState.tracks[name].mTimestretchBufferProvider != NULL) {
+        mState.tracks[name].mTimestretchBufferProvider->reset();
     }
 
     mState.tracks[name].mInputBufferProvider = bufferProvider;
-    reconfigureBufferProviders(&mState.tracks[name]);
+    mState.tracks[name].reconfigureBufferProviders();
 }
 
 
@@ -1114,7 +981,8 @@ void AudioMixer::process__validate(state_t* state, int64_t pts)
             } else {
                 if ((n & NEEDS_CHANNEL_COUNT__MASK) == NEEDS_CHANNEL_1){
                     t.hook = getTrackHook(
-                            t.mMixerChannelCount == 2 // TODO: MONO_HACK.
+                            (t.mMixerChannelMask == AUDIO_CHANNEL_OUT_STEREO  // TODO: MONO_HACK
+                                    && t.channelMask == AUDIO_CHANNEL_OUT_MONO)
                                 ? TRACKTYPE_NORESAMPLEMONO : TRACKTYPE_NORESAMPLE,
                             t.mMixerChannelCount,
                             t.mMixerInFormat, t.mMixerFormat);
@@ -2236,4 +2104,4 @@ AudioMixer::process_hook_t AudioMixer::getProcessHook(int processType, uint32_t
 }
 
 // ----------------------------------------------------------------------------
-}; // namespace android
+} // namespace android

02_day/test-mixer/AudioMixer.h

@@ -21,14 +21,16 @@
 #include <stdint.h>
 #include <sys/types.h>
 
+#include <hardware/audio_effect.h>
+#include <media/AudioBufferProvider.h>
+#include <media/AudioResamplerPublic.h>
+#include <media/nbaio/NBLog.h>
+#include <system/audio.h>
+#include <utils/Compat.h>
 #include <utils/threads.h>
 
-#include <media/AudioBufferProvider.h>
 #include "AudioResampler.h"
-
+#include "BufferProviders.h"
-#include <hardware/audio_effect.h>
-#include <system/audio.h>
-#include <media/nbaio/NBLog.h>
 
 // FIXME This is actually unity gain, which might not be max in future, expressed in U.12
 #define MAX_GAIN_INT AudioMixer::UNITY_GAIN_INT
@@ -58,7 +60,7 @@ public:
     static const uint32_t MAX_NUM_CHANNELS_TO_DOWNMIX = AUDIO_CHANNEL_COUNT_MAX;
 
     static const uint16_t UNITY_GAIN_INT = 0x1000;
-    static const float    UNITY_GAIN_FLOAT = 1.0f;
+    static const CONSTEXPR float UNITY_GAIN_FLOAT = 1.0f;
 
     enum { // names
 
@@ -72,6 +74,7 @@ public:
         RESAMPLE        = 0x3001,
         RAMP_VOLUME     = 0x3002, // ramp to new volume
         VOLUME          = 0x3003, // don't ramp
+        TIMESTRETCH     = 0x3004,
 
         // set Parameter names
         // for target TRACK
@@ -99,6 +102,9 @@ public:
         VOLUME0         = 0x4200,
         VOLUME1         = 0x4201,
         AUXLEVEL        = 0x4210,
+        // for target TIMESTRETCH
+        PLAYBACK_RATE   = 0x4300, // Configure timestretch on this track name;
+                                  // parameter 'value' is a pointer to the new playback rate.
     };
 
 
@@ -127,10 +133,16 @@ public:
     size_t      getUnreleasedFrames(int name) const;
 
     static inline bool isValidPcmTrackFormat(audio_format_t format) {
-        return format == AUDIO_FORMAT_PCM_16_BIT ||
+        switch (format) {
-                format == AUDIO_FORMAT_PCM_24_BIT_PACKED ||
+        case AUDIO_FORMAT_PCM_8_BIT:
-                format == AUDIO_FORMAT_PCM_32_BIT ||
+        case AUDIO_FORMAT_PCM_16_BIT:
-                format == AUDIO_FORMAT_PCM_FLOAT;
+        case AUDIO_FORMAT_PCM_24_BIT_PACKED:
+        case AUDIO_FORMAT_PCM_32_BIT:
+        case AUDIO_FORMAT_PCM_FLOAT:
+            return true;
+        default:
+            return false;
+        }
     }
 
 private:
@@ -153,7 +165,6 @@ private:
 
     struct state_t;
     struct track_t;
-    class CopyBufferProvider;
 
     typedef void (*hook_t)(track_t* t, int32_t* output, size_t numOutFrames, int32_t* temp,
                            int32_t* aux);
@@ -205,17 +216,38 @@ private:
         int32_t*           auxBuffer;
 
         // 16-byte boundary
+
+        /* Buffer providers are constructed to translate the track input data as needed.
+         *
+         * TODO: perhaps make a single PlaybackConverterProvider class to move
+         * all pre-mixer track buffer conversions outside the AudioMixer class.
+         *
+         * 1) mInputBufferProvider: The AudioTrack buffer provider.
+         * 2) mReformatBufferProvider: If not NULL, performs the audio reformat to
+         *    match either mMixerInFormat or mDownmixRequiresFormat, if the downmixer
+         *    requires reformat. For example, it may convert floating point input to
+         *    PCM_16_bit if that's required by the downmixer.
+         * 3) downmixerBufferProvider: If not NULL, performs the channel remixing to match
+         *    the number of channels required by the mixer sink.
+         * 4) mPostDownmixReformatBufferProvider: If not NULL, performs reformatting from
+         *    the downmixer requirements to the mixer engine input requirements.
+         * 5) mTimestretchBufferProvider: Adds timestretching for playback rate
+         */
         AudioBufferProvider*     mInputBufferProvider;    // externally provided buffer provider.
-        CopyBufferProvider*      mReformatBufferProvider; // provider wrapper for reformatting.
+        PassthruBufferProvider*  mReformatBufferProvider; // provider wrapper for reformatting.
-        CopyBufferProvider*      downmixerBufferProvider; // wrapper for channel conversion.
+        PassthruBufferProvider*  downmixerBufferProvider; // wrapper for channel conversion.
+        PassthruBufferProvider*  mPostDownmixReformatBufferProvider;
+        PassthruBufferProvider*  mTimestretchBufferProvider;
 
         int32_t     sessionId;
 
-        // 16-byte boundary
         audio_format_t mMixerFormat;     // output mix format: AUDIO_FORMAT_PCM_(FLOAT|16_BIT)
         audio_format_t mFormat;          // input track format
         audio_format_t mMixerInFormat;   // mix internal format AUDIO_FORMAT_PCM_(FLOAT|16_BIT)
                                          // each track must be converted to this format.
+        audio_format_t mDownmixRequiresFormat;  // required downmixer format
+                                                // AUDIO_FORMAT_PCM_16_BIT if 16 bit necessary
+                                                // AUDIO_FORMAT_INVALID if no required format
 
         float          mVolume[MAX_NUM_VOLUMES];     // floating point set volume
         float          mPrevVolume[MAX_NUM_VOLUMES]; // floating point previous volume
@@ -225,10 +257,11 @@ private:
         float          mPrevAuxLevel;                 // floating point prev aux level
         float          mAuxInc;                       // floating point aux increment
 
-        // 16-byte boundary
         audio_channel_mask_t mMixerChannelMask;
         uint32_t             mMixerChannelCount;
 
+        AudioPlaybackRate    mPlaybackRate;
+
         bool        needsRamp() { return (volumeInc[0] | volumeInc[1] | auxInc) != 0; }
         bool        setResampler(uint32_t trackSampleRate, uint32_t devSampleRate);
         bool        doesResample() const { return resampler != NULL; }
@@ -236,6 +269,13 @@ private:
         void        adjustVolumeRamp(bool aux, bool useFloat = false);
         size_t      getUnreleasedFrames() const { return resampler != NULL ?
                                                     resampler->getUnreleasedFrames() : 0; };
+
+        status_t    prepareForDownmix();
+        void        unprepareForDownmix();
+        status_t    prepareForReformat();
+        void        unprepareForReformat();
+        bool        setPlaybackRate(const AudioPlaybackRate &playbackRate);
+        void        reconfigureBufferProviders();
     };
 
     typedef void (*process_hook_t)(state_t* state, int64_t pts);
@@ -254,112 +294,6 @@ private:
         track_t         tracks[MAX_NUM_TRACKS] __attribute__((aligned(32)));
     };
 
-    // Base AudioBufferProvider class used for DownMixerBufferProvider, RemixBufferProvider,
-    // and ReformatBufferProvider.
-    // It handles a private buffer for use in converting format or channel masks from the
-    // input data to a form acceptable by the mixer.
-    // TODO: Make a ResamplerBufferProvider when integers are entirely removed from the
-    // processing pipeline.
-    class CopyBufferProvider : public AudioBufferProvider {
-    public:
-        // Use a private buffer of bufferFrameCount frames (each frame is outputFrameSize bytes).
-        // If bufferFrameCount is 0, no private buffer is created and in-place modification of
-        // the upstream buffer provider's buffers is performed by copyFrames().
-        CopyBufferProvider(size_t inputFrameSize, size_t outputFrameSize,
-                size_t bufferFrameCount);
-        virtual ~CopyBufferProvider();
-
-        // Overrides AudioBufferProvider methods
-        virtual status_t getNextBuffer(Buffer* buffer, int64_t pts);
-        virtual void releaseBuffer(Buffer* buffer);
-
-        // Other public methods
-
-        // call this to release the buffer to the upstream provider.
-        // treat it as an audio discontinuity for future samples.
-        virtual void reset();
-
-        // this function should be supplied by the derived class.  It converts
-        // #frames in the *src pointer to the *dst pointer.  It is public because
-        // some providers will allow this to work on arbitrary buffers outside
-        // of the internal buffers.
-        virtual void copyFrames(void *dst, const void *src, size_t frames) = 0;
-
-        // set the upstream buffer provider. Consider calling "reset" before this function.
-        void setBufferProvider(AudioBufferProvider *p) {
-            mTrackBufferProvider = p;
-        }
-
-    protected:
-        AudioBufferProvider* mTrackBufferProvider;
-        const size_t         mInputFrameSize;
-        const size_t         mOutputFrameSize;
-    private:
-        AudioBufferProvider::Buffer mBuffer;
-        const size_t         mLocalBufferFrameCount;
-        void*                mLocalBufferData;
-        size_t               mConsumed;
-    };
-
-    // DownmixerBufferProvider wraps a track AudioBufferProvider to provide
-    // position dependent downmixing by an Audio Effect.
-    class DownmixerBufferProvider : public CopyBufferProvider {
-    public:
-        DownmixerBufferProvider(audio_channel_mask_t inputChannelMask,
-                audio_channel_mask_t outputChannelMask, audio_format_t format,
-                uint32_t sampleRate, int32_t sessionId, size_t bufferFrameCount);
-        virtual ~DownmixerBufferProvider();
-        virtual void copyFrames(void *dst, const void *src, size_t frames);
-        bool isValid() const { return mDownmixHandle != NULL; }
-
-        static status_t init();
-        static bool isMultichannelCapable() { return sIsMultichannelCapable; }
-
-    protected:
-        effect_handle_t    mDownmixHandle;
-        effect_config_t    mDownmixConfig;
-
-        // effect descriptor for the downmixer used by the mixer
-        static effect_descriptor_t sDwnmFxDesc;
-        // indicates whether a downmix effect has been found and is usable by this mixer
-        static bool                sIsMultichannelCapable;
-        // FIXME: should we allow effects outside of the framework?
-        // We need to here. A special ioId that must be <= -2 so it does not map to a session.
-        static const int32_t SESSION_ID_INVALID_AND_IGNORED = -2;
-    };
-
-    // RemixBufferProvider wraps a track AudioBufferProvider to perform an
-    // upmix or downmix to the proper channel count and mask.
-    class RemixBufferProvider : public CopyBufferProvider {
-    public:
-        RemixBufferProvider(audio_channel_mask_t inputChannelMask,
-                audio_channel_mask_t outputChannelMask, audio_format_t format,
-                size_t bufferFrameCount);
-        virtual void copyFrames(void *dst, const void *src, size_t frames);
-
-    protected:
-        const audio_format_t mFormat;
-        const size_t         mSampleSize;
-        const size_t         mInputChannels;
-        const size_t         mOutputChannels;
-        int8_t               mIdxAry[sizeof(uint32_t)*8]; // 32 bits => channel indices
-    };
-
-    // ReformatBufferProvider wraps a track AudioBufferProvider to convert the input data
-    // to an acceptable mixer input format type.
-    class ReformatBufferProvider : public CopyBufferProvider {
-    public:
-        ReformatBufferProvider(int32_t channels,
-                audio_format_t inputFormat, audio_format_t outputFormat,
-                size_t bufferFrameCount);
-        virtual void copyFrames(void *dst, const void *src, size_t frames);
-
-    protected:
-        const int32_t        mChannels;
-        const audio_format_t mInputFormat;
-        const audio_format_t mOutputFormat;
-    };
-
     // bitmask of allocated track names, where bit 0 corresponds to TRACK0 etc.
     uint32_t        mTrackNames;
 
@@ -382,14 +316,6 @@ private:
     bool setChannelMasks(int name,
             audio_channel_mask_t trackChannelMask, audio_channel_mask_t mixerChannelMask);
 
-    // TODO: remove unused trackName/trackNum from functions below.
-    static status_t initTrackDownmix(track_t* pTrack, int trackName);
-    static status_t prepareTrackForDownmix(track_t* pTrack, int trackNum);
-    static void unprepareTrackForDownmix(track_t* pTrack, int trackName);
-    static status_t prepareTrackForReformat(track_t* pTrack, int trackNum);
-    static void unprepareTrackForReformat(track_t* pTrack, int trackName);
-    static void reconfigureBufferProviders(track_t* pTrack);
-
     static void track__genericResample(track_t* t, int32_t* out, size_t numFrames, int32_t* temp,
             int32_t* aux);
     static void track__nop(track_t* t, int32_t* out, size_t numFrames, int32_t* temp, int32_t* aux);
@@ -465,6 +391,6 @@ private:
 };
 
 // ----------------------------------------------------------------------------
-}; // namespace android
+} // namespace android
 
 #endif // ANDROID_AUDIO_MIXER_H

02_day/test-mixer/BufferProviders.cpp

@@ -0,0 +1,542 @@
+/*
+ * Copyright (C) 2015 The Android Open Source Project
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ *      http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#define LOG_TAG "BufferProvider"
+//#define LOG_NDEBUG 0
+
+#include <audio_effects/effect_downmix.h>
+#include <audio_utils/primitives.h>
+#include <audio_utils/format.h>
+#include <media/AudioResamplerPublic.h>
+#include <media/EffectsFactoryApi.h>
+
+#include <utils/Log.h>
+
+#include "Configuration.h"
+#include "BufferProviders.h"
+
+#ifndef ARRAY_SIZE
+#define ARRAY_SIZE(x) (sizeof(x)/sizeof((x)[0]))
+#endif
+
+namespace android {
+
+// ----------------------------------------------------------------------------
+
+template <typename T>
+static inline T min(const T& a, const T& b)
+{
+    return a < b ? a : b;
+}
+
+CopyBufferProvider::CopyBufferProvider(size_t inputFrameSize,
+        size_t outputFrameSize, size_t bufferFrameCount) :
+        mInputFrameSize(inputFrameSize),
+        mOutputFrameSize(outputFrameSize),
+        mLocalBufferFrameCount(bufferFrameCount),
+        mLocalBufferData(NULL),
+        mConsumed(0)
+{
+    ALOGV("CopyBufferProvider(%p)(%zu, %zu, %zu)", this,
+            inputFrameSize, outputFrameSize, bufferFrameCount);
+    LOG_ALWAYS_FATAL_IF(inputFrameSize < outputFrameSize && bufferFrameCount == 0,
+            "Requires local buffer if inputFrameSize(%zu) < outputFrameSize(%zu)",
+            inputFrameSize, outputFrameSize);
+    if (mLocalBufferFrameCount) {
+        (void)posix_memalign(&mLocalBufferData, 32, mLocalBufferFrameCount * mOutputFrameSize);
+    }
+    mBuffer.frameCount = 0;
+}
+
+CopyBufferProvider::~CopyBufferProvider()
+{
+    ALOGV("~CopyBufferProvider(%p)", this);
+    if (mBuffer.frameCount != 0) {
+        mTrackBufferProvider->releaseBuffer(&mBuffer);
+    }
+    free(mLocalBufferData);
+}
+
+status_t CopyBufferProvider::getNextBuffer(AudioBufferProvider::Buffer *pBuffer,
+        int64_t pts)
+{
+    //ALOGV("CopyBufferProvider(%p)::getNextBuffer(%p (%zu), %lld)",
+    //        this, pBuffer, pBuffer->frameCount, pts);
+    if (mLocalBufferFrameCount == 0) {
+        status_t res = mTrackBufferProvider->getNextBuffer(pBuffer, pts);
+        if (res == OK) {
+            copyFrames(pBuffer->raw, pBuffer->raw, pBuffer->frameCount);
+        }
+        return res;
+    }
+    if (mBuffer.frameCount == 0) {
+        mBuffer.frameCount = pBuffer->frameCount;
+        status_t res = mTrackBufferProvider->getNextBuffer(&mBuffer, pts);
+        // At one time an upstream buffer provider had
+        // res == OK and mBuffer.frameCount == 0, doesn't seem to happen now 7/18/2014.
+        //
+        // By API spec, if res != OK, then mBuffer.frameCount == 0.
+        // but there may be improper implementations.
+        ALOG_ASSERT(res == OK || mBuffer.frameCount == 0);
+        if (res != OK || mBuffer.frameCount == 0) { // not needed by API spec, but to be safe.
+            pBuffer->raw = NULL;
+            pBuffer->frameCount = 0;
+            return res;
+        }
+        mConsumed = 0;
+    }
+    ALOG_ASSERT(mConsumed < mBuffer.frameCount);
+    size_t count = min(mLocalBufferFrameCount, mBuffer.frameCount - mConsumed);
+    count = min(count, pBuffer->frameCount);
+    pBuffer->raw = mLocalBufferData;
+    pBuffer->frameCount = count;
+    copyFrames(pBuffer->raw, (uint8_t*)mBuffer.raw + mConsumed * mInputFrameSize,
+            pBuffer->frameCount);
+    return OK;
+}
+
+void CopyBufferProvider::releaseBuffer(AudioBufferProvider::Buffer *pBuffer)
+{
+    //ALOGV("CopyBufferProvider(%p)::releaseBuffer(%p(%zu))",
+    //        this, pBuffer, pBuffer->frameCount);
+    if (mLocalBufferFrameCount == 0) {
+        mTrackBufferProvider->releaseBuffer(pBuffer);
+        return;
+    }
+    // LOG_ALWAYS_FATAL_IF(pBuffer->frameCount == 0, "Invalid framecount");
+    mConsumed += pBuffer->frameCount; // TODO: update for efficiency to reuse existing content
+    if (mConsumed != 0 && mConsumed >= mBuffer.frameCount) {
+        mTrackBufferProvider->releaseBuffer(&mBuffer);
+        ALOG_ASSERT(mBuffer.frameCount == 0);
+    }
+    pBuffer->raw = NULL;
+    pBuffer->frameCount = 0;
+}
+
+void CopyBufferProvider::reset()
+{
+    if (mBuffer.frameCount != 0) {
+        mTrackBufferProvider->releaseBuffer(&mBuffer);
+    }
+    mConsumed = 0;
+}
+
+DownmixerBufferProvider::DownmixerBufferProvider(
+        audio_channel_mask_t inputChannelMask,
+        audio_channel_mask_t outputChannelMask, audio_format_t format,
+        uint32_t sampleRate, int32_t sessionId, size_t bufferFrameCount) :
+        CopyBufferProvider(
+            audio_bytes_per_sample(format) * audio_channel_count_from_out_mask(inputChannelMask),
+            audio_bytes_per_sample(format) * audio_channel_count_from_out_mask(outputChannelMask),
+            bufferFrameCount)  // set bufferFrameCount to 0 to do in-place
+{
+    ALOGV("DownmixerBufferProvider(%p)(%#x, %#x, %#x %u %d)",
+            this, inputChannelMask, outputChannelMask, format,
+            sampleRate, sessionId);
+    if (!sIsMultichannelCapable
+            || EffectCreate(&sDwnmFxDesc.uuid,
+                    sessionId,
+                    SESSION_ID_INVALID_AND_IGNORED,
+                    &mDownmixHandle) != 0) {
+         ALOGE("DownmixerBufferProvider() error creating downmixer effect");
+         mDownmixHandle = NULL;
+         return;
+     }
+     // channel input configuration will be overridden per-track
+     mDownmixConfig.inputCfg.channels = inputChannelMask;   // FIXME: Should be bits
+     mDownmixConfig.outputCfg.channels = outputChannelMask; // FIXME: should be bits
+     mDownmixConfig.inputCfg.format = format;
+     mDownmixConfig.outputCfg.format = format;
+     mDownmixConfig.inputCfg.samplingRate = sampleRate;
+     mDownmixConfig.outputCfg.samplingRate = sampleRate;
+     mDownmixConfig.inputCfg.accessMode = EFFECT_BUFFER_ACCESS_READ;
+     mDownmixConfig.outputCfg.accessMode = EFFECT_BUFFER_ACCESS_WRITE;
+     // input and output buffer provider, and frame count will not be used as the downmix effect
+     // process() function is called directly (see DownmixerBufferProvider::getNextBuffer())
+     mDownmixConfig.inputCfg.mask = EFFECT_CONFIG_SMP_RATE | EFFECT_CONFIG_CHANNELS |
+             EFFECT_CONFIG_FORMAT | EFFECT_CONFIG_ACC_MODE;
+     mDownmixConfig.outputCfg.mask = mDownmixConfig.inputCfg.mask;
+
+     int cmdStatus;
+     uint32_t replySize = sizeof(int);
+
+     // Configure downmixer
+     status_t status = (*mDownmixHandle)->command(mDownmixHandle,
+             EFFECT_CMD_SET_CONFIG /*cmdCode*/, sizeof(effect_config_t) /*cmdSize*/,
+             &mDownmixConfig /*pCmdData*/,
+             &replySize, &cmdStatus /*pReplyData*/);
+     if (status != 0 || cmdStatus != 0) {
+         ALOGE("DownmixerBufferProvider() error %d cmdStatus %d while configuring downmixer",
+                 status, cmdStatus);
+         EffectRelease(mDownmixHandle);
+         mDownmixHandle = NULL;
+         return;
+     }
+
+     // Enable downmixer
+     replySize = sizeof(int);
+     status = (*mDownmixHandle)->command(mDownmixHandle,
+             EFFECT_CMD_ENABLE /*cmdCode*/, 0 /*cmdSize*/, NULL /*pCmdData*/,
+             &replySize, &cmdStatus /*pReplyData*/);
+     if (status != 0 || cmdStatus != 0) {
+         ALOGE("DownmixerBufferProvider() error %d cmdStatus %d while enabling downmixer",
+                 status, cmdStatus);
+         EffectRelease(mDownmixHandle);
+         mDownmixHandle = NULL;
+         return;
+     }
+
+     // Set downmix type
+     // parameter size rounded for padding on 32bit boundary
+     const int psizePadded = ((sizeof(downmix_params_t) - 1)/sizeof(int) + 1) * sizeof(int);
+     const int downmixParamSize =
+             sizeof(effect_param_t) + psizePadded + sizeof(downmix_type_t);
+     effect_param_t * const param = (effect_param_t *) malloc(downmixParamSize);
+     param->psize = sizeof(downmix_params_t);
+     const downmix_params_t downmixParam = DOWNMIX_PARAM_TYPE;
+     memcpy(param->data, &downmixParam, param->psize);
+     const downmix_type_t downmixType = DOWNMIX_TYPE_FOLD;
+     param->vsize = sizeof(downmix_type_t);
+     memcpy(param->data + psizePadded, &downmixType, param->vsize);
+     replySize = sizeof(int);
+     status = (*mDownmixHandle)->command(mDownmixHandle,
+             EFFECT_CMD_SET_PARAM /* cmdCode */, downmixParamSize /* cmdSize */,
+             param /*pCmdData*/, &replySize, &cmdStatus /*pReplyData*/);
+     free(param);
+     if (status != 0 || cmdStatus != 0) {
+         ALOGE("DownmixerBufferProvider() error %d cmdStatus %d while setting downmix type",
+                 status, cmdStatus);
+         EffectRelease(mDownmixHandle);
+         mDownmixHandle = NULL;
+         return;
+     }
+     ALOGV("DownmixerBufferProvider() downmix type set to %d", (int) downmixType);
+}
+
+DownmixerBufferProvider::~DownmixerBufferProvider()
+{
+    ALOGV("~DownmixerBufferProvider (%p)", this);
+    EffectRelease(mDownmixHandle);
+    mDownmixHandle = NULL;
+}
+
+void DownmixerBufferProvider::copyFrames(void *dst, const void *src, size_t frames)
+{
+    mDownmixConfig.inputCfg.buffer.frameCount = frames;
+    mDownmixConfig.inputCfg.buffer.raw = const_cast<void *>(src);
+    mDownmixConfig.outputCfg.buffer.frameCount = frames;
+    mDownmixConfig.outputCfg.buffer.raw = dst;
+    // may be in-place if src == dst.
+    status_t res = (*mDownmixHandle)->process(mDownmixHandle,
+            &mDownmixConfig.inputCfg.buffer, &mDownmixConfig.outputCfg.buffer);
+    ALOGE_IF(res != OK, "DownmixBufferProvider error %d", res);
+}
+
+/* call once in a pthread_once handler. */
+/*static*/ status_t DownmixerBufferProvider::init()
+{
+    // find multichannel downmix effect if we have to play multichannel content
+    uint32_t numEffects = 0;
+    int ret = EffectQueryNumberEffects(&numEffects);
+    if (ret != 0) {
+        ALOGE("AudioMixer() error %d querying number of effects", ret);
+        return NO_INIT;
+    }
+    ALOGV("EffectQueryNumberEffects() numEffects=%d", numEffects);
+
+    for (uint32_t i = 0 ; i < numEffects ; i++) {
+        if (EffectQueryEffect(i, &sDwnmFxDesc) == 0) {
+            ALOGV("effect %d is called %s", i, sDwnmFxDesc.name);
+            if (memcmp(&sDwnmFxDesc.type, EFFECT_UIID_DOWNMIX, sizeof(effect_uuid_t)) == 0) {
+                ALOGI("found effect \"%s\" from %s",
+                        sDwnmFxDesc.name, sDwnmFxDesc.implementor);
+                sIsMultichannelCapable = true;
+                break;
+            }
+        }
+    }
+    ALOGW_IF(!sIsMultichannelCapable, "unable to find downmix effect");
+    return NO_INIT;
+}
+
+/*static*/ bool DownmixerBufferProvider::sIsMultichannelCapable = false;
+/*static*/ effect_descriptor_t DownmixerBufferProvider::sDwnmFxDesc;
+
+RemixBufferProvider::RemixBufferProvider(audio_channel_mask_t inputChannelMask,
+        audio_channel_mask_t outputChannelMask, audio_format_t format,
+        size_t bufferFrameCount) :
+        CopyBufferProvider(
+                audio_bytes_per_sample(format)
+                    * audio_channel_count_from_out_mask(inputChannelMask),
+                audio_bytes_per_sample(format)
+                    * audio_channel_count_from_out_mask(outputChannelMask),
+                bufferFrameCount),
+        mFormat(format),
+        mSampleSize(audio_bytes_per_sample(format)),
+        mInputChannels(audio_channel_count_from_out_mask(inputChannelMask)),
+        mOutputChannels(audio_channel_count_from_out_mask(outputChannelMask))
+{
+    ALOGV("RemixBufferProvider(%p)(%#x, %#x, %#x) %zu %zu",
+            this, format, inputChannelMask, outputChannelMask,
+            mInputChannels, mOutputChannels);
+    (void) memcpy_by_index_array_initialization_from_channel_mask(
+            mIdxAry, ARRAY_SIZE(mIdxAry), outputChannelMask, inputChannelMask);
+}
+
+void RemixBufferProvider::copyFrames(void *dst, const void *src, size_t frames)
+{
+    memcpy_by_index_array(dst, mOutputChannels,
+            src, mInputChannels, mIdxAry, mSampleSize, frames);
+}
+
+ReformatBufferProvider::ReformatBufferProvider(int32_t channelCount,
+        audio_format_t inputFormat, audio_format_t outputFormat,
+        size_t bufferFrameCount) :
+        CopyBufferProvider(
+                channelCount * audio_bytes_per_sample(inputFormat),
+                channelCount * audio_bytes_per_sample(outputFormat),
+                bufferFrameCount),
+        mChannelCount(channelCount),
+        mInputFormat(inputFormat),
+        mOutputFormat(outputFormat)
+{
+    ALOGV("ReformatBufferProvider(%p)(%u, %#x, %#x)",
+            this, channelCount, inputFormat, outputFormat);
+}
+
+void ReformatBufferProvider::copyFrames(void *dst, const void *src, size_t frames)
+{
+    memcpy_by_audio_format(dst, mOutputFormat, src, mInputFormat, frames * mChannelCount);
+}
+
+TimestretchBufferProvider::TimestretchBufferProvider(int32_t channelCount,
+        audio_format_t format, uint32_t sampleRate, const AudioPlaybackRate &playbackRate) :
+        mChannelCount(channelCount),
+        mFormat(format),
+        mSampleRate(sampleRate),
+        mFrameSize(channelCount * audio_bytes_per_sample(format)),
+        mLocalBufferFrameCount(0),
+        mLocalBufferData(NULL),
+        mRemaining(0),
+        mSonicStream(sonicCreateStream(sampleRate, mChannelCount)),
+        mFallbackFailErrorShown(false),
+        mAudioPlaybackRateValid(false)
+{
+    LOG_ALWAYS_FATAL_IF(mSonicStream == NULL,
+            "TimestretchBufferProvider can't allocate Sonic stream");
+
+    setPlaybackRate(playbackRate);
+    ALOGV("TimestretchBufferProvider(%p)(%u, %#x, %u %f %f %d %d)",
+            this, channelCount, format, sampleRate, playbackRate.mSpeed,
+            playbackRate.mPitch, playbackRate.mStretchMode, playbackRate.mFallbackMode);
+    mBuffer.frameCount = 0;
+}
+
+TimestretchBufferProvider::~TimestretchBufferProvider()
+{
+    ALOGV("~TimestretchBufferProvider(%p)", this);
+    sonicDestroyStream(mSonicStream);
+    if (mBuffer.frameCount != 0) {
+        mTrackBufferProvider->releaseBuffer(&mBuffer);
+    }
+    free(mLocalBufferData);
+}
+
+status_t TimestretchBufferProvider::getNextBuffer(
+        AudioBufferProvider::Buffer *pBuffer, int64_t pts)
+{
+    ALOGV("TimestretchBufferProvider(%p)::getNextBuffer(%p (%zu), %lld)",
+            this, pBuffer, pBuffer->frameCount, pts);
+
+    // BYPASS
+    //return mTrackBufferProvider->getNextBuffer(pBuffer, pts);
+
+    // check if previously processed data is sufficient.
+    if (pBuffer->frameCount <= mRemaining) {
+        ALOGV("previous sufficient");
+        pBuffer->raw = mLocalBufferData;
+        return OK;
+    }
+
+    // do we need to resize our buffer?
+    if (pBuffer->frameCount > mLocalBufferFrameCount) {
+        void *newmem;
+        if (posix_memalign(&newmem, 32, pBuffer->frameCount * mFrameSize) == OK) {
+            if (mRemaining != 0) {
+                memcpy(newmem, mLocalBufferData, mRemaining * mFrameSize);
+            }
+            free(mLocalBufferData);
+            mLocalBufferData = newmem;
+            mLocalBufferFrameCount = pBuffer->frameCount;
+        }
+    }
+
+    // need to fetch more data
+    const size_t outputDesired = pBuffer->frameCount - mRemaining;
+    size_t dstAvailable;
+    do {
+        mBuffer.frameCount = mPlaybackRate.mSpeed == AUDIO_TIMESTRETCH_SPEED_NORMAL
+                ? outputDesired : outputDesired * mPlaybackRate.mSpeed + 1;
+
+        status_t res = mTrackBufferProvider->getNextBuffer(&mBuffer, pts);
+
+        ALOG_ASSERT(res == OK || mBuffer.frameCount == 0);
+        if (res != OK || mBuffer.frameCount == 0) { // not needed by API spec, but to be safe.
+            ALOGV("upstream provider cannot provide data");
+            if (mRemaining == 0) {
+                pBuffer->raw = NULL;
+                pBuffer->frameCount = 0;
+                return res;
+            } else { // return partial count
+                pBuffer->raw = mLocalBufferData;
+                pBuffer->frameCount = mRemaining;
+                return OK;
+            }
+        }
+
+        // time-stretch the data
+        dstAvailable = min(mLocalBufferFrameCount - mRemaining, outputDesired);
+        size_t srcAvailable = mBuffer.frameCount;
+        processFrames((uint8_t*)mLocalBufferData + mRemaining * mFrameSize, &dstAvailable,
+                mBuffer.raw, &srcAvailable);
+
+        // release all data consumed
+        mBuffer.frameCount = srcAvailable;
+        mTrackBufferProvider->releaseBuffer(&mBuffer);
+    } while (dstAvailable == 0); // try until we get output data or upstream provider fails.
+
+    // update buffer vars with the actual data processed and return with buffer
+    mRemaining += dstAvailable;
+
+    pBuffer->raw = mLocalBufferData;
+    pBuffer->frameCount = mRemaining;
+
+    return OK;
+}
+
+void TimestretchBufferProvider::releaseBuffer(AudioBufferProvider::Buffer *pBuffer)
+{
+    ALOGV("TimestretchBufferProvider(%p)::releaseBuffer(%p (%zu))",
+       this, pBuffer, pBuffer->frameCount);
+
+    // BYPASS
+    //return mTrackBufferProvider->releaseBuffer(pBuffer);
+
+    // LOG_ALWAYS_FATAL_IF(pBuffer->frameCount == 0, "Invalid framecount");
+    if (pBuffer->frameCount < mRemaining) {
+        memcpy(mLocalBufferData,
+                (uint8_t*)mLocalBufferData + pBuffer->frameCount * mFrameSize,
+                (mRemaining - pBuffer->frameCount) * mFrameSize);
+        mRemaining -= pBuffer->frameCount;
+    } else if (pBuffer->frameCount == mRemaining) {
+        mRemaining = 0;
+    } else {
+        LOG_ALWAYS_FATAL("Releasing more frames(%zu) than available(%zu)",
+                pBuffer->frameCount, mRemaining);
+    }
+
+    pBuffer->raw = NULL;
+    pBuffer->frameCount = 0;
+}
+
+void TimestretchBufferProvider::reset()
+{
+    mRemaining = 0;
+}
+
+status_t TimestretchBufferProvider::setPlaybackRate(const AudioPlaybackRate &playbackRate)
+{
+    mPlaybackRate = playbackRate;
+    mFallbackFailErrorShown = false;
+    sonicSetSpeed(mSonicStream, mPlaybackRate.mSpeed);
+    //TODO: pitch is ignored for now
+    //TODO: optimize: if parameters are the same, don't do any extra computation.
+
+    mAudioPlaybackRateValid = isAudioPlaybackRateValid(mPlaybackRate);
+    return OK;
+}
+
+void TimestretchBufferProvider::processFrames(void *dstBuffer, size_t *dstFrames,
+        const void *srcBuffer, size_t *srcFrames)
+{
+    ALOGV("processFrames(%zu %zu)  remaining(%zu)", *dstFrames, *srcFrames, mRemaining);
+    // Note dstFrames is the required number of frames.
+
+    // Ensure consumption from src is as expected.
+    //TODO: add logic to track "very accurate" consumption related to speed, original sampling
+    //rate, actual frames processed.
+    const size_t targetSrc = *dstFrames * mPlaybackRate.mSpeed;
+    if (*srcFrames < targetSrc) { // limit dst frames to that possible
+        *dstFrames = *srcFrames / mPlaybackRate.mSpeed;
+    } else if (*srcFrames > targetSrc + 1) {
+        *srcFrames = targetSrc + 1;
+    }
+
+    if (!mAudioPlaybackRateValid) {
+        //fallback mode
+        if (*dstFrames > 0) {
+            switch(mPlaybackRate.mFallbackMode) {
+            case AUDIO_TIMESTRETCH_FALLBACK_CUT_REPEAT:
+                if (*dstFrames <= *srcFrames) {
+                      size_t copySize = mFrameSize * *dstFrames;
+                      memcpy(dstBuffer, srcBuffer, copySize);
+                  } else {
+                      // cyclically repeat the source.
+                      for (size_t count = 0; count < *dstFrames; count += *srcFrames) {
+                          size_t remaining = min(*srcFrames, *dstFrames - count);
+                          memcpy((uint8_t*)dstBuffer + mFrameSize * count,
+                                  srcBuffer, mFrameSize * remaining);
+                      }
+                  }
+                break;
+            case AUDIO_TIMESTRETCH_FALLBACK_DEFAULT:
+            case AUDIO_TIMESTRETCH_FALLBACK_MUTE:
+                memset(dstBuffer,0, mFrameSize * *dstFrames);
+                break;
+            case AUDIO_TIMESTRETCH_FALLBACK_FAIL:
+            default:
+                if(!mFallbackFailErrorShown) {
+                    ALOGE("invalid parameters in TimestretchBufferProvider fallbackMode:%d",
+                            mPlaybackRate.mFallbackMode);
+                    mFallbackFailErrorShown = true;
+                }
+                break;
+            }
+        }
+    } else {
+        switch (mFormat) {
+        case AUDIO_FORMAT_PCM_FLOAT:
+            if (sonicWriteFloatToStream(mSonicStream, (float*)srcBuffer, *srcFrames) != 1) {
+                ALOGE("sonicWriteFloatToStream cannot realloc");
+                *srcFrames = 0; // cannot consume all of srcBuffer
+            }
+            *dstFrames = sonicReadFloatFromStream(mSonicStream, (float*)dstBuffer, *dstFrames);
+            break;
+        case AUDIO_FORMAT_PCM_16_BIT:
+            if (sonicWriteShortToStream(mSonicStream, (short*)srcBuffer, *srcFrames) != 1) {
+                ALOGE("sonicWriteShortToStream cannot realloc");
+                *srcFrames = 0; // cannot consume all of srcBuffer
+            }
+            *dstFrames = sonicReadShortFromStream(mSonicStream, (short*)dstBuffer, *dstFrames);
+            break;
+        default:
+            // could also be caught on construction
+            LOG_ALWAYS_FATAL("invalid format %#x for TimestretchBufferProvider", mFormat);
+        }
+    }
+}
+// ----------------------------------------------------------------------------
+} // namespace android

02_day/test-mixer/test-mixer.cpp

@@ -1,19 +1,3 @@
-/*
- * Copyright (C) 2014 The Android Open Source Project
- *
- * Licensed under the Apache License, Version 2.0 (the "License");
- * you may not use this file except in compliance with the License.
- * You may obtain a copy of the License at
- *
- *      http://www.apache.org/licenses/LICENSE-2.0
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
 #include <stdio.h>
 #include <inttypes.h>
 #include <math.h>
@@ -39,7 +23,7 @@ static void usage(const char* name) {
     fprintf(stderr, "Usage: %s [-f] [-m] [-c channels]"
                     " [-s sample-rate] [-o <output-file>] [-a <aux-buffer-file>] [-P csv]"
                     " (<input-file> | <command>)+\n", name);
-    fprintf(stderr, "    -f    enable floating point input track\n");
+    fprintf(stderr, "    -f    enable floating point input track by default\n");
     fprintf(stderr, "    -m    enable floating point mixer output\n");
     fprintf(stderr, "    -c    number of mixer output channels\n");
     fprintf(stderr, "    -s    mixer sample-rate\n");
@@ -47,8 +31,8 @@ static void usage(const char* name) {
     fprintf(stderr, "    -a    <aux-buffer-file>\n");
     fprintf(stderr, "    -P    # frames provided per call to resample() in CSV format\n");
     fprintf(stderr, "    <input-file> is a WAV file\n");
-    fprintf(stderr, "    <command> can be 'sine:<channels>,<frequency>,<samplerate>'\n");
+    fprintf(stderr, "    <command> can be 'sine:[(i|f),]<channels>,<frequency>,<samplerate>'\n");
-    fprintf(stderr, "                     'chirp:<channels>,<samplerate>'\n");
+    fprintf(stderr, "                     'chirp:[(i|f),]<channels>,<samplerate>'\n");
 }
 
 static int writeFile(const char *filename, const void *buffer,
@@ -78,6 +62,18 @@ static int writeFile(const char *filename, const void *buffer,
     return EXIT_SUCCESS;
 }
 
+const char *parseFormat(const char *s, bool *useFloat) {
+    if (!strncmp(s, "f,", 2)) {
+        *useFloat = true;
+        return s + 2;
+    }
+    if (!strncmp(s, "i,", 2)) {
+        *useFloat = false;
+        return s + 2;
+    }
+    return s;
+}
+
 int main(int argc, char* argv[]) {
     const char* const progname = argv[0];
     bool useInputFloat = false;
@@ -88,8 +84,9 @@ int main(int argc, char* argv[]) {
     std::vector<int> Pvalues;
     const char* outputFilename = NULL;
     const char* auxFilename = NULL;
-    std::vector<int32_t> Names;
+    std::vector<int32_t> names;
-    std::vector<SignalProvider> Providers;
+    std::vector<SignalProvider> providers;
+    std::vector<audio_format_t> formats;
 
     for (int ch; (ch = getopt(argc, argv, "fmc:s:o:a:P:")) != -1;) {
         switch (ch) {
@@ -138,54 +135,65 @@ int main(int argc, char* argv[]) {
     size_t outputFrames = 0;
 
     // create providers for each track
-    Providers.resize(argc);
+    names.resize(argc);
+    providers.resize(argc);
+    formats.resize(argc);
     for (int i = 0; i < argc; ++i) {
         static const char chirp[] = "chirp:";
         static const char sine[] = "sine:";
-        static const double kSeconds = 10;
+        static const double kSeconds = 1;
+        bool useFloat = useInputFloat;
 
         if (!strncmp(argv[i], chirp, strlen(chirp))) {
             std::vector<int> v;
+            const char *s = parseFormat(argv[i] + strlen(chirp), &useFloat);
 
-            parseCSV(argv[i] + strlen(chirp), v);
+            parseCSV(s, v);
             if (v.size() == 2) {
                 printf("creating chirp(%d %d)\n", v[0], v[1]);
-                if (useInputFloat) {
+                if (useFloat) {
-                    Providers[i].setChirp<float>(v[0], 0, v[1]/2, v[1], kSeconds);
+                    providers[i].setChirp<float>(v[0], 0, v[1]/2, v[1], kSeconds);
+                    formats[i] = AUDIO_FORMAT_PCM_FLOAT;
                 } else {
-                    Providers[i].setChirp<int16_t>(v[0], 0, v[1]/2, v[1], kSeconds);
+                    providers[i].setChirp<int16_t>(v[0], 0, v[1]/2, v[1], kSeconds);
+                    formats[i] = AUDIO_FORMAT_PCM_16_BIT;
                 }
-                Providers[i].setIncr(Pvalues);
+                providers[i].setIncr(Pvalues);
             } else {
                 fprintf(stderr, "malformed input '%s'\n", argv[i]);
             }
         } else if (!strncmp(argv[i], sine, strlen(sine))) {
             std::vector<int> v;
+            const char *s = parseFormat(argv[i] + strlen(sine), &useFloat);
 
-            parseCSV(argv[i] + strlen(sine), v);
+            parseCSV(s, v);
             if (v.size() == 3) {
                 printf("creating sine(%d %d %d)\n", v[0], v[1], v[2]);
-                if (useInputFloat) {
+                if (useFloat) {
-                    Providers[i].setSine<float>(v[0], v[1], v[2], kSeconds);
+                    providers[i].setSine<float>(v[0], v[1], v[2], kSeconds);
+                    formats[i] = AUDIO_FORMAT_PCM_FLOAT;
                 } else {
-                    Providers[i].setSine<int16_t>(v[0], v[1], v[2], kSeconds);
+                    providers[i].setSine<int16_t>(v[0], v[1], v[2], kSeconds);
+                    formats[i] = AUDIO_FORMAT_PCM_16_BIT;
                 }
-                Providers[i].setIncr(Pvalues);
+                providers[i].setIncr(Pvalues);
             } else {
                 fprintf(stderr, "malformed input '%s'\n", argv[i]);
             }
         } else {
             printf("creating filename(%s)\n", argv[i]);
             if (useInputFloat) {
-                Providers[i].setFile<float>(argv[i]);
+                providers[i].setFile<float>(argv[i]);
+                formats[i] = AUDIO_FORMAT_PCM_FLOAT;
             } else {
-                Providers[i].setFile<short>(argv[i]);
+                providers[i].setFile<short>(argv[i]);
+                formats[i] = AUDIO_FORMAT_PCM_16_BIT;
             }
-            Providers[i].setIncr(Pvalues);
+            providers[i].setIncr(Pvalues);
         }
         // calculate the number of output frames
-        size_t nframes = (int64_t) Providers[i].getNumFrames() * outputSampleRate
+        size_t nframes = (int64_t) providers[i].getNumFrames() * outputSampleRate
-                / Providers[i].getSampleRate();
+                / providers[i].getSampleRate();
         if (i == 0 || outputFrames > nframes) { // choose minimum for outputFrames
             outputFrames = nframes;
         }
@@ -213,22 +221,20 @@ int main(int argc, char* argv[]) {
     // create the mixer.
     const size_t mixerFrameCount = 320; // typical numbers may range from 240 or 960
     AudioMixer *mixer = new AudioMixer(mixerFrameCount, outputSampleRate);
-    audio_format_t inputFormat = useInputFloat
-            ? AUDIO_FORMAT_PCM_FLOAT : AUDIO_FORMAT_PCM_16_BIT;
     audio_format_t mixerFormat = useMixerFloat
             ? AUDIO_FORMAT_PCM_FLOAT : AUDIO_FORMAT_PCM_16_BIT;
-    float f = AudioMixer::UNITY_GAIN_FLOAT / Providers.size(); // normalize volume by # tracks
+    float f = AudioMixer::UNITY_GAIN_FLOAT / providers.size(); // normalize volume by # tracks
     static float f0; // zero
 
     // set up the tracks.
-    for (size_t i = 0; i < Providers.size(); ++i) {
+    for (size_t i = 0; i < providers.size(); ++i) {
-        //printf("track %d out of %d\n", i, Providers.size());
+        //printf("track %d out of %d\n", i, providers.size());
-        uint32_t channelMask = audio_channel_out_mask_from_count(Providers[i].getNumChannels());
+        uint32_t channelMask = audio_channel_out_mask_from_count(providers[i].getNumChannels());
         int32_t name = mixer->getTrackName(channelMask,
-                inputFormat, AUDIO_SESSION_OUTPUT_MIX);
+                formats[i], AUDIO_SESSION_OUTPUT_MIX);
         ALOG_ASSERT(name >= 0);
-        Names.push_back(name);
+        names[i] = name;
-        mixer->setBufferProvider(name, &Providers[i]);
+        mixer->setBufferProvider(name, &providers[i]);
         mixer->setParameter(name, AudioMixer::TRACK, AudioMixer::MAIN_BUFFER,
                 (void *)outputAddr);
         mixer->setParameter(
@@ -240,7 +246,7 @@ int main(int argc, char* argv[]) {
                 name,
                 AudioMixer::TRACK,
                 AudioMixer::FORMAT,
-                (void *)(uintptr_t)inputFormat);
+                (void *)(uintptr_t)formats[i]);
         mixer->setParameter(
                 name,
                 AudioMixer::TRACK,
@@ -255,7 +261,7 @@ int main(int argc, char* argv[]) {
                 name,
                 AudioMixer::RESAMPLE,
                 AudioMixer::SAMPLE_RATE,
-                (void *)(uintptr_t)Providers[i].getSampleRate());
+                (void *)(uintptr_t)providers[i].getSampleRate());
         if (useRamp) {
             mixer->setParameter(name, AudioMixer::VOLUME, AudioMixer::VOLUME0, &f0);
             mixer->setParameter(name, AudioMixer::VOLUME, AudioMixer::VOLUME1, &f0);
@@ -277,11 +283,11 @@ int main(int argc, char* argv[]) {
     // pump the mixer to process data.
     size_t i;
     for (i = 0; i < outputFrames - mixerFrameCount; i += mixerFrameCount) {
-        for (size_t j = 0; j < Names.size(); ++j) {
+        for (size_t j = 0; j < names.size(); ++j) {
-            mixer->setParameter(Names[j], AudioMixer::TRACK, AudioMixer::MAIN_BUFFER,
+            mixer->setParameter(names[j], AudioMixer::TRACK, AudioMixer::MAIN_BUFFER,
                     (char *) outputAddr + i * outputFrameSize);
             if (auxFilename) {
-                mixer->setParameter(Names[j], AudioMixer::TRACK, AudioMixer::AUX_BUFFER,
+                mixer->setParameter(names[j], AudioMixer::TRACK, AudioMixer::AUX_BUFFER,
                         (char *) auxAddr + i * auxFrameSize);
             }
         }