Modified Windows buffer size to avoid audio glitches.

remoting/host/audio_capturer_win.cc

Index: remoting/host/audio_capturer_win.cc
diff --git a/remoting/host/audio_capturer_win.cc b/remoting/host/audio_capturer_win.cc
index c648a0ab0727c5c776e51a9bf66db3c4100bc8e4..43fdfd8182ab702b919b35d24fcb344e641764d8 100644
--- a/remoting/host/audio_capturer_win.cc
+++ b/remoting/host/audio_capturer_win.cc
@@ -9,6 +9,7 @@
 #include <mmreg.h>
 #include <mmsystem.h>
 
+#include <algorithm>
 #include <stdlib.h>
 
 #include "base/basictypes.h"
@@ -27,13 +28,20 @@ const int kChannels = 2;
 const int kBitsPerSample = 16;
 const int kBitsPerByte = 8;
 // Conversion factor from 100ns to 1ms.
-const int kHnsToMs = 10000;
+const int k100nsInMillisecond = 10000;

[Wez, 2012/08/16 20:20:20]

nit: k100nsPerMillisecond, to avoid ambiguity over how "in" is being used here,
and to match the constants above?

[kxing, 2012/08/16 20:38:46]

Done.

 
 // Tolerance for catching packets of silence. If all samples have absolute
 // value less than this threshold, the packet will be counted as a packet of
 // silence. A value of 2 was chosen, because Windows can give samples of 1 and
 // -1, even when no audio is playing.
 const int kSilenceThreshold = 2;
+
+// Lower bound for timer intervals, in milliseconds.
+const int kMinTimerInterval = 30;
+
+// Upper bound for the timer precision error, in milliseconds.
+// Timers are supposed to be accurate to 20ms, so we use 30ms to be safe.
+const int kMaxExpectedTimerLag = 30;
 }  // namespace
 
 namespace remoting {
@@ -128,8 +136,12 @@ bool AudioCapturerWin::Start(const PacketCapturedCallback& callback) {
     LOG(ERROR) << "IAudioClient::GetDevicePeriod failed. Error " << hr;
     return false;
   }
+  // We round up, if |device_period| / |k100nsInMillisecond|
+  // is not a whole number.
+  int device_period_in_milliseconds =
+      1 + (device_period - 1) / k100nsInMillisecond;

[Wez, 2012/08/16 20:20:20]

nit: Bracket the division to make it clear to the reader that we're subtracting
1*100ns, and adding 1ms.

[kxing, 2012/08/16 20:38:46]

Done.

   audio_device_period_ = base::TimeDelta::FromMilliseconds(
-      device_period / kChannels / kHnsToMs);
+      std::max(device_period_in_milliseconds, kMinTimerInterval));
 
   // Get the wave format.
   hr = audio_client_->GetMixFormat(&wave_format_ex_);
@@ -193,12 +205,14 @@ bool AudioCapturerWin::Start(const PacketCapturedCallback& callback) {
   }
 
   // Initialize the IAudioClient.
-  hr = audio_client_->Initialize(AUDCLNT_SHAREMODE_SHARED,
-                                 AUDCLNT_STREAMFLAGS_LOOPBACK,
-                                 0,
-                                 0,
-                                 wave_format_ex_,
-                                 NULL);
+  hr = audio_client_->Initialize(
+      AUDCLNT_SHAREMODE_SHARED,
+      AUDCLNT_STREAMFLAGS_LOOPBACK,
+      (kMaxExpectedTimerLag + audio_device_period_.InMilliseconds()) *
+      k100nsInMillisecond,
+      0,
+      wave_format_ex_,
+      NULL);
   if (FAILED(hr)) {
     LOG(ERROR) << "Failed to initialize IAudioClient. Error " << hr;
     return false;
@@ -275,8 +289,8 @@ void AudioCapturerWin::DoCapture() {
     }
 
     if (!IsPacketOfSilence(
-          reinterpret_cast<const int16*>(data),
-          frames * kChannels)) {
+            reinterpret_cast<const int16*>(data),
+            frames * kChannels)) {
       scoped_ptr<AudioPacket> packet =
           scoped_ptr<AudioPacket>(new AudioPacket());
       packet->add_data(data, frames * wave_format_ex_->nBlockAlign);