Bug 717026 - Ensure libsydneyaudio windows block size is an even multiple of channels*frame_size. r=kinetik

content/media/nsBuiltinDecoderStateMachine.cpp

@@ -631,7 +631,6 @@ void nsBuiltinDecoderStateMachine::AudioLoop()
   LOG(PR_LOG_DEBUG, ("%p Begun audio thread/loop", mDecoder.get()));
   PRInt64 audioDuration = 0;
   PRInt64 audioStartTime = -1;
-  PRInt64 framesWritten = 0;
   PRUint32 channels, rate;
   double volume = -1;
   bool setVolume;
@@ -746,6 +745,7 @@ void nsBuiltinDecoderStateMachine::AudioLoop()
       break;
     }
 
+    PRInt64 framesWritten = 0;
     if (missingFrames > 0) {
       // The next audio chunk begins some time after the end of the last chunk
       // we pushed to the audio hardware. We must push silence into the audio
@@ -780,18 +780,22 @@ void nsBuiltinDecoderStateMachine::AudioLoop()
     bool seeking = false;
     {
       ReentrantMonitorAutoEnter mon(mDecoder->GetReentrantMonitor());
-      if (framesWritten < minWriteFrames) {
-        // We've not written minWriteFrames in the last write, the audio
-        // may not start playing. Write silence to ensure we've got enough
-        // written to start playback.
-        PRInt64 minToWrite = minWriteFrames - framesWritten;
-        if (minToWrite < PR_UINT32_MAX / channels) {
+      PRInt64 unplayedFrames = audioDuration % minWriteFrames;
+      if (minWriteFrames > 1 && unplayedFrames > 0) {
+        // Sound is written by libsydneyaudio to the hardware in blocks of
+        // frames of size minWriteFrames. So if the number of frames we've
+        // written isn't an exact multiple of minWriteFrames, we'll have
+        // left over audio data which hasn't yet been written to the hardware,
+        // and so that audio will not start playing. Write silence to ensure
+        // the last block gets pushed to hardware, so that playback starts.
+        PRInt64 framesToWrite = minWriteFrames - unplayedFrames;
+        if (framesToWrite < PR_UINT32_MAX / channels) {
           // Write silence manually rather than using PlaySilence(), so that
           // the AudioAPI doesn't get a copy of the audio frames.
-          PRUint32 numSamples = minToWrite * channels;
+          PRUint32 numSamples = framesToWrite * channels;
           nsAutoArrayPtr<AudioDataValue> buf(new AudioDataValue[numSamples]);
           memset(buf.get(), 0, numSamples * sizeof(AudioDataValue));
-          mAudioStream->Write(buf, minToWrite);
+          mAudioStream->Write(buf, framesToWrite);
         }
       }
 

media/libsydneyaudio/src/sydney_audio_waveapi.c

@@ -169,10 +169,12 @@ int sa_stream_create_pcm(sa_stream_t **s,
   _s->deviceName = DEFAULT_DEVICE_NAME;
   _s->device = DEFAULT_DEVICE;
   _s->playing = 0;
-  _s->blockSize = BYTES_PER_SAMPLE * ((rate * nchannels * BLOCK_DURATION_MS) / 1000);
+  _s->blockSize = BYTES_PER_SAMPLE * nchannels * ((rate * BLOCK_DURATION_MS) / 1000);
   /* Other parts of the code assumes that the block size is evenly
      divisible by 2. */
   assert((_s->blockSize & 1) != 1);
+  assert((_s->blockSize % BYTES_PER_SAMPLE) == 0);
+  assert(((_s->blockSize / BYTES_PER_SAMPLE) % nchannels) == 0);
   *s = _s; 
   return SA_SUCCESS;
 }