modules/audio_device/audio_device_unittest.cc

/*
 *  Copyright (c) 2017 The WebRTC project authors. All Rights Reserved.
 *
 *  Use of this source code is governed by a BSD-style license
 *  that can be found in the LICENSE file in the root of the source
 *  tree. An additional intellectual property rights grant can be found
 *  in the file PATENTS.  All contributing project authors may
 *  be found in the AUTHORS file in the root of the source tree.
 */

#include <algorithm>
#include <cstring>
#include <memory>
#include <numeric>

#include "absl/types/optional.h"
#include "api/array_view.h"
#include "modules/audio_device/audio_device_impl.h"
#include "modules/audio_device/include/audio_device.h"
#include "modules/audio_device/include/mock_audio_transport.h"
#include "rtc_base/buffer.h"
#include "rtc_base/criticalsection.h"
#include "rtc_base/event.h"
#include "rtc_base/logging.h"
#include "rtc_base/numerics/safe_conversions.h"
#include "rtc_base/race_checker.h"
#include "rtc_base/scoped_ref_ptr.h"
#include "rtc_base/thread_annotations.h"
#include "rtc_base/thread_checker.h"
#include "rtc_base/timeutils.h"
#include "system_wrappers/include/sleep.h"
#include "test/gmock.h"
#include "test/gtest.h"
#ifdef WEBRTC_WIN
#include "modules/audio_device/include/audio_device_factory.h"
#include "modules/audio_device/win/core_audio_utility_win.h"
#endif

using ::testing::_;
using ::testing::AtLeast;
using ::testing::Ge;
using ::testing::Invoke;
using ::testing::NiceMock;
using ::testing::NotNull;
using ::testing::Mock;

namespace webrtc {
namespace {

// Using a #define for AUDIO_DEVICE since we will call *different* versions of
// the ADM functions, depending on the ID type.
#if defined(WEBRTC_WIN)
#define AUDIO_DEVICE_ID (AudioDeviceModule::WindowsDeviceType::kDefaultDevice)
#else
#define AUDIO_DEVICE_ID (0u)
#endif  // defined(WEBRTC_WIN)

// #define ENABLE_DEBUG_PRINTF
#ifdef ENABLE_DEBUG_PRINTF
#define PRINTD(...) fprintf(stderr, __VA_ARGS__);
#else
#define PRINTD(...) ((void)0)
#endif
#define PRINT(...) fprintf(stderr, __VA_ARGS__);

// Don't run these tests in combination with sanitizers.
// TODO(webrtc:9778): Re-enable on THREAD_SANITIZER?
#if !defined(ADDRESS_SANITIZER) && !defined(MEMORY_SANITIZER) && \
    !defined(THREAD_SANITIZER)
#define SKIP_TEST_IF_NOT(requirements_satisfied) \
  do {                                           \
    if (!requirements_satisfied) {               \
      return;                                    \
    }                                            \
  } while (false)
#else
// Or if other audio-related requirements are not met.
#define SKIP_TEST_IF_NOT(requirements_satisfied) \
  do {                                           \
    return;                                      \
  } while (false)
#endif

// Number of callbacks (input or output) the tests waits for before we set
// an event indicating that the test was OK.
static constexpr size_t kNumCallbacks = 10;
// Max amount of time we wait for an event to be set while counting callbacks.
static constexpr size_t kTestTimeOutInMilliseconds = 10 * 1000;
// Average number of audio callbacks per second assuming 10ms packet size.
static constexpr size_t kNumCallbacksPerSecond = 100;
// Run the full-duplex test during this time (unit is in seconds).
static constexpr size_t kFullDuplexTimeInSec = 5;
// Length of round-trip latency measurements. Number of deteced impulses
// shall be kImpulseFrequencyInHz * kMeasureLatencyTimeInSec - 1 since the
// last transmitted pulse is not used.
static constexpr size_t kMeasureLatencyTimeInSec = 10;
// Sets the number of impulses per second in the latency test.
static constexpr size_t kImpulseFrequencyInHz = 1;
// Utilized in round-trip latency measurements to avoid capturing noise samples.
static constexpr int kImpulseThreshold = 1000;

enum class TransportType {
  kInvalid,
  kPlay,
  kRecord,
  kPlayAndRecord,
};

// Interface for processing the audio stream. Real implementations can e.g.
// run audio in loopback, read audio from a file or perform latency
// measurements.
class AudioStream {
 public:
  virtual void Write(rtc::ArrayView<const int16_t> source) = 0;
  virtual void Read(rtc::ArrayView<int16_t> destination) = 0;

  virtual ~AudioStream() = default;
};

// Converts index corresponding to position within a 10ms buffer into a
// delay value in milliseconds.
// Example: index=240, frames_per_10ms_buffer=480 => 5ms as output.
int IndexToMilliseconds(size_t index, size_t frames_per_10ms_buffer) {
  return rtc::checked_cast<int>(
      10.0 * (static_cast<double>(index) / frames_per_10ms_buffer) + 0.5);
}

}  // namespace

// Simple first in first out (FIFO) class that wraps a list of 16-bit audio
// buffers of fixed size and allows Write and Read operations. The idea is to
// store recorded audio buffers (using Write) and then read (using Read) these
// stored buffers with as short delay as possible when the audio layer needs
// data to play out. The number of buffers in the FIFO will stabilize under
// normal conditions since there will be a balance between Write and Read calls.
// The container is a std::list container and access is protected with a lock
// since both sides (playout and recording) are driven by its own thread.
// Note that, we know by design that the size of the audio buffer will not
// change over time and that both sides will in most cases use the same size.
class FifoAudioStream : public AudioStream {
 public:
  void Write(rtc::ArrayView<const int16_t> source) override {
    RTC_DCHECK_RUNS_SERIALIZED(&race_checker_);
    const size_t size = [&] {
      rtc::CritScope lock(&lock_);
      fifo_.push_back(Buffer16(source.data(), source.size()));
      return fifo_.size();
    }();
    if (size > max_size_) {
      max_size_ = size;
    }
    // Add marker once per second to signal that audio is active.
    if (write_count_++ % 100 == 0) {
      PRINT(".");
    }
    written_elements_ += size;
  }

  void Read(rtc::ArrayView<int16_t> destination) override {
    rtc::CritScope lock(&lock_);
    if (fifo_.empty()) {
      std::fill(destination.begin(), destination.end(), 0);
    } else {
      const Buffer16& buffer = fifo_.front();
      if (buffer.size() == destination.size()) {
        // Default case where input and output uses same sample rate and
        // channel configuration. No conversion is needed.
        std::copy(buffer.begin(), buffer.end(), destination.begin());
      } else if (destination.size() == 2 * buffer.size()) {
        // Recorded input signal in |buffer| is in mono. Do channel upmix to
        // match stereo output (1 -> 2).
        for (size_t i = 0; i < buffer.size(); ++i) {
          destination[2 * i] = buffer[i];
          destination[2 * i + 1] = buffer[i];
        }
      } else if (buffer.size() == 2 * destination.size()) {
        // Recorded input signal in |buffer| is in stereo. Do channel downmix
        // to match mono output (2 -> 1).
        for (size_t i = 0; i < destination.size(); ++i) {
          destination[i] =
              (static_cast<int32_t>(buffer[2 * i]) + buffer[2 * i + 1]) / 2;
        }
      } else {
        RTC_NOTREACHED() << "Required conversion is not support";
      }
      fifo_.pop_front();
    }
  }

  size_t size() const {
    rtc::CritScope lock(&lock_);
    return fifo_.size();
  }

  size_t max_size() const {
    RTC_DCHECK_RUNS_SERIALIZED(&race_checker_);
    return max_size_;
  }

  size_t average_size() const {
    RTC_DCHECK_RUNS_SERIALIZED(&race_checker_);
    return 0.5 + static_cast<float>(written_elements_ / write_count_);
  }

  using Buffer16 = rtc::BufferT<int16_t>;

  rtc::CriticalSection lock_;
  rtc::RaceChecker race_checker_;

  std::list<Buffer16> fifo_ RTC_GUARDED_BY(lock_);
  size_t write_count_ RTC_GUARDED_BY(race_checker_) = 0;
  size_t max_size_ RTC_GUARDED_BY(race_checker_) = 0;
  size_t written_elements_ RTC_GUARDED_BY(race_checker_) = 0;
};

// Inserts periodic impulses and measures the latency between the time of
// transmission and time of receiving the same impulse.
class LatencyAudioStream : public AudioStream {
 public:
  LatencyAudioStream() {
    // Delay thread checkers from being initialized until first callback from
    // respective thread.
    read_thread_checker_.DetachFromThread();
    write_thread_checker_.DetachFromThread();
  }

  // Insert periodic impulses in first two samples of |destination|.
  void Read(rtc::ArrayView<int16_t> destination) override {
    RTC_DCHECK_RUN_ON(&read_thread_checker_);
    if (read_count_ == 0) {
      PRINT("[");
    }
    read_count_++;
    std::fill(destination.begin(), destination.end(), 0);
    if (read_count_ % (kNumCallbacksPerSecond / kImpulseFrequencyInHz) == 0) {
      PRINT(".");
      {
        rtc::CritScope lock(&lock_);
        if (!pulse_time_) {
          pulse_time_ = rtc::TimeMillis();
        }
      }
      constexpr int16_t impulse = std::numeric_limits<int16_t>::max();
      std::fill_n(destination.begin(), 2, impulse);
    }
  }

  // Detect received impulses in |source|, derive time between transmission and
  // detection and add the calculated delay to list of latencies.
  void Write(rtc::ArrayView<const int16_t> source) override {
    RTC_DCHECK_RUN_ON(&write_thread_checker_);
    RTC_DCHECK_RUNS_SERIALIZED(&race_checker_);
    rtc::CritScope lock(&lock_);
    write_count_++;
    if (!pulse_time_) {
      // Avoid detection of new impulse response until a new impulse has
      // been transmitted (sets |pulse_time_| to value larger than zero).
      return;
    }
    // Find index (element position in vector) of the max element.
    const size_t index_of_max =
        std::max_element(source.begin(), source.end()) - source.begin();
    // Derive time between transmitted pulse and received pulse if the level
    // is high enough (removes noise).
    const size_t max = source[index_of_max];
    if (max > kImpulseThreshold) {
      PRINTD("(%zu, %zu)", max, index_of_max);
      int64_t now_time = rtc::TimeMillis();
      int extra_delay = IndexToMilliseconds(index_of_max, source.size());
      PRINTD("[%d]", rtc::checked_cast<int>(now_time - pulse_time_));
      PRINTD("[%d]", extra_delay);
      // Total latency is the difference between transmit time and detection
      // tome plus the extra delay within the buffer in which we detected the
      // received impulse. It is transmitted at sample 0 but can be received
      // at sample N where N > 0. The term |extra_delay| accounts for N and it
      // is a value between 0 and 10ms.
      latencies_.push_back(now_time - *pulse_time_ + extra_delay);
      pulse_time_.reset();
    } else {
      PRINTD("-");
    }
  }

  size_t num_latency_values() const {
    RTC_DCHECK_RUNS_SERIALIZED(&race_checker_);
    return latencies_.size();
  }

  int min_latency() const {
    RTC_DCHECK_RUNS_SERIALIZED(&race_checker_);
    if (latencies_.empty())
      return 0;
    return *std::min_element(latencies_.begin(), latencies_.end());
  }

  int max_latency() const {
    RTC_DCHECK_RUNS_SERIALIZED(&race_checker_);
    if (latencies_.empty())
      return 0;
    return *std::max_element(latencies_.begin(), latencies_.end());
  }

  int average_latency() const {
    RTC_DCHECK_RUNS_SERIALIZED(&race_checker_);
    if (latencies_.empty())
      return 0;
    return 0.5 + static_cast<double>(
                     std::accumulate(latencies_.begin(), latencies_.end(), 0)) /
                     latencies_.size();
  }

  void PrintResults() const {
    RTC_DCHECK_RUNS_SERIALIZED(&race_checker_);
    PRINT("] ");
    for (auto it = latencies_.begin(); it != latencies_.end(); ++it) {
      PRINTD("%d ", *it);
    }
    PRINT("\n");
    PRINT("[..........] [min, max, avg]=[%d, %d, %d] ms\n", min_latency(),
          max_latency(), average_latency());
  }

  rtc::CriticalSection lock_;
  rtc::RaceChecker race_checker_;
  rtc::ThreadChecker read_thread_checker_;
  rtc::ThreadChecker write_thread_checker_;

  absl::optional<int64_t> pulse_time_ RTC_GUARDED_BY(lock_);
  std::vector<int> latencies_ RTC_GUARDED_BY(race_checker_);
  size_t read_count_ RTC_GUARDED_BY(read_thread_checker_) = 0;
  size_t write_count_ RTC_GUARDED_BY(write_thread_checker_) = 0;
};

// Mocks the AudioTransport object and proxies actions for the two callbacks
// (RecordedDataIsAvailable and NeedMorePlayData) to different implementations
// of AudioStreamInterface.
class MockAudioTransport : public test::MockAudioTransport {
 public:
  explicit MockAudioTransport(TransportType type) : type_(type) {}
  ~MockAudioTransport() {}

  // Set default actions of the mock object. We are delegating to fake
  // implementation where the number of callbacks is counted and an event
  // is set after a certain number of callbacks. Audio parameters are also
  // checked.
  void HandleCallbacks(rtc::Event* event,
                       AudioStream* audio_stream,
                       int num_callbacks) {
    event_ = event;
    audio_stream_ = audio_stream;
    num_callbacks_ = num_callbacks;
    if (play_mode()) {
      ON_CALL(*this, NeedMorePlayData(_, _, _, _, _, _, _, _))
          .WillByDefault(
              Invoke(this, &MockAudioTransport::RealNeedMorePlayData));
    }
    if (rec_mode()) {
      ON_CALL(*this, RecordedDataIsAvailable(_, _, _, _, _, _, _, _, _, _))
          .WillByDefault(
              Invoke(this, &MockAudioTransport::RealRecordedDataIsAvailable));
    }
  }

  // Special constructor used in manual tests where the user wants to run audio
  // until e.g. a keyboard key is pressed. The event flag is set to nullptr by
  // default since it is up to the user to stop the test. See e.g.
  // DISABLED_RunPlayoutAndRecordingInFullDuplexAndWaitForEnterKey().
  void HandleCallbacks(AudioStream* audio_stream) {
    HandleCallbacks(nullptr, audio_stream, 0);
  }

  int32_t RealRecordedDataIsAvailable(const void* audio_buffer,
                                      const size_t samples_per_channel,
                                      const size_t bytes_per_frame,
                                      const size_t channels,
                                      const uint32_t sample_rate,
                                      const uint32_t total_delay_ms,
                                      const int32_t clock_drift,
                                      const uint32_t current_mic_level,
                                      const bool typing_status,
                                      uint32_t& new_mic_level) {
    EXPECT_TRUE(rec_mode()) << "No test is expecting these callbacks.";
    // Store audio parameters once in the first callback. For all other
    // callbacks, verify that the provided audio parameters are maintained and
    // that each callback corresponds to 10ms for any given sample rate.
    if (!record_parameters_.is_complete()) {
      record_parameters_.reset(sample_rate, channels, samples_per_channel);
    } else {
      EXPECT_EQ(samples_per_channel, record_parameters_.frames_per_buffer());
      EXPECT_EQ(bytes_per_frame, record_parameters_.GetBytesPerFrame());
      EXPECT_EQ(channels, record_parameters_.channels());
      EXPECT_EQ(static_cast<int>(sample_rate),
                record_parameters_.sample_rate());
      EXPECT_EQ(samples_per_channel,
                record_parameters_.frames_per_10ms_buffer());
    }
    {
      rtc::CritScope lock(&lock_);
      rec_count_++;
    }
    // Write audio data to audio stream object if one has been injected.
    if (audio_stream_) {
      audio_stream_->Write(
          rtc::MakeArrayView(static_cast<const int16_t*>(audio_buffer),
                             samples_per_channel * channels));
    }
    // Signal the event after given amount of callbacks.
    if (event_ && ReceivedEnoughCallbacks()) {
      event_->Set();
    }
    return 0;
  }

  int32_t RealNeedMorePlayData(const size_t samples_per_channel,
                               const size_t bytes_per_frame,
                               const size_t channels,
                               const uint32_t sample_rate,
                               void* audio_buffer,
                               size_t& samples_out,
                               int64_t* elapsed_time_ms,
                               int64_t* ntp_time_ms) {
    EXPECT_TRUE(play_mode()) << "No test is expecting these callbacks.";
    // Store audio parameters once in the first callback. For all other
    // callbacks, verify that the provided audio parameters are maintained and
    // that each callback corresponds to 10ms for any given sample rate.
    if (!playout_parameters_.is_complete()) {
      playout_parameters_.reset(sample_rate, channels, samples_per_channel);
    } else {
      EXPECT_EQ(samples_per_channel, playout_parameters_.frames_per_buffer());
      EXPECT_EQ(bytes_per_frame, playout_parameters_.GetBytesPerFrame());
      EXPECT_EQ(channels, playout_parameters_.channels());
      EXPECT_EQ(static_cast<int>(sample_rate),
                playout_parameters_.sample_rate());
      EXPECT_EQ(samples_per_channel,
                playout_parameters_.frames_per_10ms_buffer());
    }
    {
      rtc::CritScope lock(&lock_);
      play_count_++;
    }
    samples_out = samples_per_channel * channels;
    // Read audio data from audio stream object if one has been injected.
    if (audio_stream_) {
      audio_stream_->Read(rtc::MakeArrayView(
          static_cast<int16_t*>(audio_buffer), samples_per_channel * channels));
    } else {
      // Fill the audio buffer with zeros to avoid disturbing audio.
      const size_t num_bytes = samples_per_channel * bytes_per_frame;
      std::memset(audio_buffer, 0, num_bytes);
    }
    // Signal the event after given amount of callbacks.
    if (event_ && ReceivedEnoughCallbacks()) {
      event_->Set();
    }
    return 0;
  }

  bool ReceivedEnoughCallbacks() {
    bool recording_done = false;
    if (rec_mode()) {
      rtc::CritScope lock(&lock_);
      recording_done = rec_count_ >= num_callbacks_;
    } else {
      recording_done = true;
    }
    bool playout_done = false;
    if (play_mode()) {
      rtc::CritScope lock(&lock_);
      playout_done = play_count_ >= num_callbacks_;
    } else {
      playout_done = true;
    }
    return recording_done && playout_done;
  }

  bool play_mode() const {
    return type_ == TransportType::kPlay ||
           type_ == TransportType::kPlayAndRecord;
  }

  bool rec_mode() const {
    return type_ == TransportType::kRecord ||
           type_ == TransportType::kPlayAndRecord;
  }

  void ResetCallbackCounters() {
    rtc::CritScope lock(&lock_);
    if (play_mode()) {
      play_count_ = 0;
    }
    if (rec_mode()) {
      rec_count_ = 0;
    }
  }

 private:
  rtc::CriticalSection lock_;
  TransportType type_ = TransportType::kInvalid;
  rtc::Event* event_ = nullptr;
  AudioStream* audio_stream_ = nullptr;
  size_t num_callbacks_ = 0;
  size_t play_count_ RTC_GUARDED_BY(lock_) = 0;
  size_t rec_count_ RTC_GUARDED_BY(lock_) = 0;
  AudioParameters playout_parameters_;
  AudioParameters record_parameters_;
};

// AudioDeviceTest test fixture.
class AudioDeviceTest
    : public ::testing::TestWithParam<webrtc::AudioDeviceModule::AudioLayer> {
 protected:
  AudioDeviceTest() : audio_layer_(GetParam()), event_(false, false) {
// TODO(webrtc:9778): Re-enable on THREAD_SANITIZER?
#if !defined(ADDRESS_SANITIZER) && !defined(MEMORY_SANITIZER) && \
    !defined(WEBRTC_DUMMY_AUDIO_BUILD) && !defined(THREAD_SANITIZER)
    rtc::LogMessage::LogToDebug(rtc::LS_INFO);
    // Add extra logging fields here if needed for debugging.
    rtc::LogMessage::LogTimestamps();
    rtc::LogMessage::LogThreads();
    audio_device_ = CreateAudioDevice();
    EXPECT_NE(audio_device_.get(), nullptr);
    AudioDeviceModule::AudioLayer audio_layer;
    int got_platform_audio_layer =
        audio_device_->ActiveAudioLayer(&audio_layer);
    // First, ensure that a valid audio layer can be activated.
    if (got_platform_audio_layer != 0) {
      requirements_satisfied_ = false;
    }
    // Next, verify that the ADM can be initialized.
    if (requirements_satisfied_) {
      requirements_satisfied_ = (audio_device_->Init() == 0);
    }
    // Finally, ensure that at least one valid device exists in each direction.
    if (requirements_satisfied_) {
      const int16_t num_playout_devices = audio_device_->PlayoutDevices();
      const int16_t num_record_devices = audio_device_->RecordingDevices();
      requirements_satisfied_ =
          num_playout_devices > 0 && num_record_devices > 0;
    }
#else
    requirements_satisfied_ = false;
#endif
    if (requirements_satisfied_) {
      EXPECT_EQ(0, audio_device_->SetPlayoutDevice(AUDIO_DEVICE_ID));
      EXPECT_EQ(0, audio_device_->InitSpeaker());
      EXPECT_EQ(0, audio_device_->StereoPlayoutIsAvailable(&stereo_playout_));
      EXPECT_EQ(0, audio_device_->SetStereoPlayout(stereo_playout_));
      EXPECT_EQ(0, audio_device_->SetRecordingDevice(AUDIO_DEVICE_ID));
      EXPECT_EQ(0, audio_device_->InitMicrophone());
      // Avoid asking for input stereo support and always record in mono
      // since asking can cause issues in combination with remote desktop.
      // See https://bugs.chromium.org/p/webrtc/issues/detail?id=7397 for
      // details.
      EXPECT_EQ(0, audio_device_->SetStereoRecording(false));
    }
  }

  virtual ~AudioDeviceTest() {
    if (audio_device_) {
      EXPECT_EQ(0, audio_device_->Terminate());
    }
  }

  bool requirements_satisfied() const { return requirements_satisfied_; }
  rtc::Event* event() { return &event_; }
  AudioDeviceModule::AudioLayer audio_layer() const { return audio_layer_; }

  // AudioDeviceModuleForTest extends the default ADM interface with some extra
  // test methods. Intended for usage in tests only and requires a unique
  // factory method. See CreateAudioDevice() for details.
  const rtc::scoped_refptr<AudioDeviceModuleForTest>& audio_device() const {
    return audio_device_;
  }

  rtc::scoped_refptr<AudioDeviceModuleForTest> CreateAudioDevice() {
    // Use the default factory for kPlatformDefaultAudio and a special factory
    // CreateWindowsCoreAudioAudioDeviceModuleForTest() for kWindowsCoreAudio2.
    // The value of |audio_layer_| is set at construction by GetParam() and two
    // different layers are tested on Windows only.
    if (audio_layer_ == AudioDeviceModule::kPlatformDefaultAudio) {
      return AudioDeviceModule::CreateForTest(audio_layer_);
    } else if (audio_layer_ == AudioDeviceModule::kWindowsCoreAudio2) {
#ifdef WEBRTC_WIN
      // We must initialize the COM library on a thread before we calling any of
      // the library functions. All COM functions in the ADM will return
      // CO_E_NOTINITIALIZED otherwise.
      com_initializer_ = absl::make_unique<webrtc_win::ScopedCOMInitializer>(
          webrtc_win::ScopedCOMInitializer::kMTA);
      EXPECT_TRUE(com_initializer_->Succeeded());
      EXPECT_TRUE(webrtc_win::core_audio_utility::IsSupported());
      EXPECT_TRUE(webrtc_win::core_audio_utility::IsMMCSSSupported());
      return CreateWindowsCoreAudioAudioDeviceModuleForTest();
#else
      return nullptr;
#endif
    } else {
      return nullptr;
    }
  }

  void StartPlayout() {
    EXPECT_FALSE(audio_device()->Playing());
    EXPECT_EQ(0, audio_device()->InitPlayout());
    EXPECT_TRUE(audio_device()->PlayoutIsInitialized());
    EXPECT_EQ(0, audio_device()->StartPlayout());
    EXPECT_TRUE(audio_device()->Playing());
  }

  void StopPlayout() {
    EXPECT_EQ(0, audio_device()->StopPlayout());
    EXPECT_FALSE(audio_device()->Playing());
    EXPECT_FALSE(audio_device()->PlayoutIsInitialized());
  }

  void StartRecording() {
    EXPECT_FALSE(audio_device()->Recording());
    EXPECT_EQ(0, audio_device()->InitRecording());
    EXPECT_TRUE(audio_device()->RecordingIsInitialized());
    EXPECT_EQ(0, audio_device()->StartRecording());
    EXPECT_TRUE(audio_device()->Recording());
  }

  void StopRecording() {
    EXPECT_EQ(0, audio_device()->StopRecording());
    EXPECT_FALSE(audio_device()->Recording());
    EXPECT_FALSE(audio_device()->RecordingIsInitialized());
  }

  bool NewWindowsAudioDeviceModuleIsUsed() {
#ifdef WEBRTC_WIN
    AudioDeviceModule::AudioLayer audio_layer;
    EXPECT_EQ(0, audio_device()->ActiveAudioLayer(&audio_layer));
    if (audio_layer == AudioDeviceModule::kWindowsCoreAudio2) {
      // Default device is always added as first element in the list and the
      // default communication device as the second element. Hence, the list
      // contains two extra elements in this case.
      return true;
    }
#endif
    return false;
  }

 private:
#ifdef WEBRTC_WIN
  // Windows Core Audio based ADM needs to run on a COM initialized thread.
  std::unique_ptr<webrtc_win::ScopedCOMInitializer> com_initializer_;
#endif
  AudioDeviceModule::AudioLayer audio_layer_;
  bool requirements_satisfied_ = true;
  rtc::Event event_;
  rtc::scoped_refptr<AudioDeviceModuleForTest> audio_device_;
  bool stereo_playout_ = false;
};

// Instead of using the test fixture, verify that the different factory methods
// work as intended.
TEST(AudioDeviceTestWin, ConstructDestructWithFactory) {
  rtc::scoped_refptr<AudioDeviceModule> audio_device;
  // The default factory should work for all platforms when a default ADM is
  // requested.
  audio_device =
      AudioDeviceModule::Create(AudioDeviceModule::kPlatformDefaultAudio);
  EXPECT_TRUE(audio_device);
  audio_device = nullptr;
#ifdef WEBRTC_WIN
  // For Windows, the old factory method creates an ADM where the platform-
  // specific parts are implemented by an AudioDeviceGeneric object. Verify
  // that the old factory can't be used in combination with the latest audio
  // layer AudioDeviceModule::kWindowsCoreAudio2.
  audio_device =
      AudioDeviceModule::Create(AudioDeviceModule::kWindowsCoreAudio2);
  EXPECT_FALSE(audio_device);
  audio_device = nullptr;
  // Instead, ensure that the new dedicated factory method called
  // CreateWindowsCoreAudioAudioDeviceModule() can be used on Windows and that
  // it sets the audio layer to kWindowsCoreAudio2 implicitly. Note that, the
  // new ADM for Windows must be created on a COM thread.
  webrtc_win::ScopedCOMInitializer com_initializer(
      webrtc_win::ScopedCOMInitializer::kMTA);
  EXPECT_TRUE(com_initializer.Succeeded());
  audio_device = CreateWindowsCoreAudioAudioDeviceModule();
  EXPECT_TRUE(audio_device);
  AudioDeviceModule::AudioLayer audio_layer;
  EXPECT_EQ(0, audio_device->ActiveAudioLayer(&audio_layer));
  EXPECT_EQ(audio_layer, AudioDeviceModule::kWindowsCoreAudio2);
#endif
}

// Uses the test fixture to create, initialize and destruct the ADM.
TEST_P(AudioDeviceTest, ConstructDestructDefault) {}

TEST_P(AudioDeviceTest, InitTerminate) {
  SKIP_TEST_IF_NOT(requirements_satisfied());
  // Initialization is part of the test fixture.
  EXPECT_TRUE(audio_device()->Initialized());
  EXPECT_EQ(0, audio_device()->Terminate());
  EXPECT_FALSE(audio_device()->Initialized());
}

// Enumerate all available and active output devices.
TEST_P(AudioDeviceTest, PlayoutDeviceNames) {
  SKIP_TEST_IF_NOT(requirements_satisfied());
  char device_name[kAdmMaxDeviceNameSize];
  char unique_id[kAdmMaxGuidSize];
  int num_devices = audio_device()->PlayoutDevices();
  if (NewWindowsAudioDeviceModuleIsUsed()) {
    num_devices += 2;
  }
  EXPECT_GT(num_devices, 0);
  for (int i = 0; i < num_devices; ++i) {
    EXPECT_EQ(0, audio_device()->PlayoutDeviceName(i, device_name, unique_id));
  }
  EXPECT_EQ(-1, audio_device()->PlayoutDeviceName(num_devices, device_name,
                                                  unique_id));
}

// Enumerate all available and active input devices.
TEST_P(AudioDeviceTest, RecordingDeviceNames) {
  SKIP_TEST_IF_NOT(requirements_satisfied());
  char device_name[kAdmMaxDeviceNameSize];
  char unique_id[kAdmMaxGuidSize];
  int num_devices = audio_device()->RecordingDevices();
  if (NewWindowsAudioDeviceModuleIsUsed()) {
    num_devices += 2;
  }
  EXPECT_GT(num_devices, 0);
  for (int i = 0; i < num_devices; ++i) {
    EXPECT_EQ(0,
              audio_device()->RecordingDeviceName(i, device_name, unique_id));
  }
  EXPECT_EQ(-1, audio_device()->RecordingDeviceName(num_devices, device_name,
                                                    unique_id));
}

// Counts number of active output devices and ensure that all can be selected.
TEST_P(AudioDeviceTest, SetPlayoutDevice) {
  SKIP_TEST_IF_NOT(requirements_satisfied());
  int num_devices = audio_device()->PlayoutDevices();
  if (NewWindowsAudioDeviceModuleIsUsed()) {
    num_devices += 2;
  }
  EXPECT_GT(num_devices, 0);
  // Verify that all available playout devices can be set (not enabled yet).
  for (int i = 0; i < num_devices; ++i) {
    EXPECT_EQ(0, audio_device()->SetPlayoutDevice(i));
  }
  EXPECT_EQ(-1, audio_device()->SetPlayoutDevice(num_devices));
#ifdef WEBRTC_WIN
  // On Windows, verify the alternative method where the user can select device
  // by role.
  EXPECT_EQ(
      0, audio_device()->SetPlayoutDevice(AudioDeviceModule::kDefaultDevice));
  EXPECT_EQ(0, audio_device()->SetPlayoutDevice(
                   AudioDeviceModule::kDefaultCommunicationDevice));
#endif
}

// Counts number of active input devices and ensure that all can be selected.
TEST_P(AudioDeviceTest, SetRecordingDevice) {
  SKIP_TEST_IF_NOT(requirements_satisfied());
  int num_devices = audio_device()->RecordingDevices();
  if (NewWindowsAudioDeviceModuleIsUsed()) {
    num_devices += 2;
  }
  EXPECT_GT(num_devices, 0);
  // Verify that all available recording devices can be set (not enabled yet).
  for (int i = 0; i < num_devices; ++i) {
    EXPECT_EQ(0, audio_device()->SetRecordingDevice(i));
  }
  EXPECT_EQ(-1, audio_device()->SetRecordingDevice(num_devices));
#ifdef WEBRTC_WIN
  // On Windows, verify the alternative method where the user can select device
  // by role.
  EXPECT_EQ(
      0, audio_device()->SetRecordingDevice(AudioDeviceModule::kDefaultDevice));
  EXPECT_EQ(0, audio_device()->SetRecordingDevice(
                   AudioDeviceModule::kDefaultCommunicationDevice));
#endif
}

// Tests Start/Stop playout without any registered audio callback.
TEST_P(AudioDeviceTest, StartStopPlayout) {
  SKIP_TEST_IF_NOT(requirements_satisfied());
  StartPlayout();
  StopPlayout();
}

// Tests Start/Stop recording without any registered audio callback.
TEST_P(AudioDeviceTest, StartStopRecording) {
  SKIP_TEST_IF_NOT(requirements_satisfied());
  StartRecording();
  StopRecording();
}

// Tests Init/Stop/Init recording without any registered audio callback.
// See https://bugs.chromium.org/p/webrtc/issues/detail?id=8041 for details
// on why this test is useful.
TEST_P(AudioDeviceTest, InitStopInitRecording) {
  SKIP_TEST_IF_NOT(requirements_satisfied());
  EXPECT_EQ(0, audio_device()->InitRecording());
  EXPECT_TRUE(audio_device()->RecordingIsInitialized());
  StopRecording();
  EXPECT_EQ(0, audio_device()->InitRecording());
  StopRecording();
}

// Tests Init/Stop/Init recording while playout is active.
TEST_P(AudioDeviceTest, InitStopInitRecordingWhilePlaying) {
  SKIP_TEST_IF_NOT(requirements_satisfied());
  StartPlayout();
  EXPECT_EQ(0, audio_device()->InitRecording());
  EXPECT_TRUE(audio_device()->RecordingIsInitialized());
  StopRecording();
  EXPECT_EQ(0, audio_device()->InitRecording());
  StopRecording();
  StopPlayout();
}

// Tests Init/Stop/Init playout without any registered audio callback.
TEST_P(AudioDeviceTest, InitStopInitPlayout) {
  SKIP_TEST_IF_NOT(requirements_satisfied());
  EXPECT_EQ(0, audio_device()->InitPlayout());
  EXPECT_TRUE(audio_device()->PlayoutIsInitialized());
  StopPlayout();
  EXPECT_EQ(0, audio_device()->InitPlayout());
  StopPlayout();
}

// Tests Init/Stop/Init playout while recording is active.
TEST_P(AudioDeviceTest, InitStopInitPlayoutWhileRecording) {
  SKIP_TEST_IF_NOT(requirements_satisfied());
  StartRecording();
  EXPECT_EQ(0, audio_device()->InitPlayout());
  EXPECT_TRUE(audio_device()->PlayoutIsInitialized());
  StopPlayout();
  EXPECT_EQ(0, audio_device()->InitPlayout());
  StopPlayout();
  StopRecording();
}

// TODO(henrika): restart without intermediate destruction is currently only
// supported on Windows.
#ifdef WEBRTC_WIN
// Tests Start/Stop playout followed by a second session (emulates a restart
// triggered by a user using public APIs).
TEST_P(AudioDeviceTest, StartStopPlayoutWithExternalRestart) {
  SKIP_TEST_IF_NOT(requirements_satisfied());
  StartPlayout();
  StopPlayout();
  // Restart playout without destroying the ADM in between. Ensures that we
  // support: Init(), Start(), Stop(), Init(), Start(), Stop().
  StartPlayout();
  StopPlayout();
}

// Tests Start/Stop recording followed by a second session (emulates a restart
// triggered by a user using public APIs).
TEST_P(AudioDeviceTest, StartStopRecordingWithExternalRestart) {
  SKIP_TEST_IF_NOT(requirements_satisfied());
  StartRecording();
  StopRecording();
  // Restart recording without destroying the ADM in between.  Ensures that we
  // support: Init(), Start(), Stop(), Init(), Start(), Stop().
  StartRecording();
  StopRecording();
}

// Tests Start/Stop playout followed by a second session (emulates a restart
// triggered by an internal callback e.g. corresponding to a device switch).
// Note that, internal restart is only supported in combination with the latest
// Windows ADM.
TEST_P(AudioDeviceTest, StartStopPlayoutWithInternalRestart) {
  SKIP_TEST_IF_NOT(requirements_satisfied());
  if (audio_layer() != AudioDeviceModule::kWindowsCoreAudio2) {
    return;
  }
  MockAudioTransport mock(TransportType::kPlay);
  mock.HandleCallbacks(event(), nullptr, kNumCallbacks);
  EXPECT_CALL(mock, NeedMorePlayData(_, _, _, _, NotNull(), _, _, _))
      .Times(AtLeast(kNumCallbacks));
  EXPECT_EQ(0, audio_device()->RegisterAudioCallback(&mock));
  StartPlayout();
  event()->Wait(kTestTimeOutInMilliseconds);
  EXPECT_TRUE(audio_device()->Playing());
  // Restart playout but without stopping the internal audio thread.
  // This procedure uses a non-public test API and it emulates what happens
  // inside the ADM when e.g. a device is removed.
  EXPECT_EQ(0, audio_device()->RestartPlayoutInternally());

  // Run basic tests of public APIs while a restart attempt is active.
  // These calls should now be very thin and not trigger any new actions.
  EXPECT_EQ(-1, audio_device()->StopPlayout());
  EXPECT_TRUE(audio_device()->Playing());
  EXPECT_TRUE(audio_device()->PlayoutIsInitialized());
  EXPECT_EQ(0, audio_device()->InitPlayout());
  EXPECT_EQ(0, audio_device()->StartPlayout());

  // Wait until audio has restarted and a new sequence of audio callbacks
  // becomes active.
  // TODO(henrika): is it possible to verify that the internal state transition
  // is Stop->Init->Start?
  ASSERT_TRUE(Mock::VerifyAndClearExpectations(&mock));
  mock.ResetCallbackCounters();
  EXPECT_CALL(mock, NeedMorePlayData(_, _, _, _, NotNull(), _, _, _))
      .Times(AtLeast(kNumCallbacks));
  event()->Wait(kTestTimeOutInMilliseconds);
  EXPECT_TRUE(audio_device()->Playing());
  // Stop playout and the audio thread after successful internal restart.
  StopPlayout();
}

// Tests Start/Stop recording followed by a second session (emulates a restart
// triggered by an internal callback e.g. corresponding to a device switch).
// Note that, internal restart is only supported in combination with the latest
// Windows ADM.
TEST_P(AudioDeviceTest, StartStopRecordingWithInternalRestart) {
  SKIP_TEST_IF_NOT(requirements_satisfied());
  if (audio_layer() != AudioDeviceModule::kWindowsCoreAudio2) {
    return;
  }
  MockAudioTransport mock(TransportType::kRecord);
  mock.HandleCallbacks(event(), nullptr, kNumCallbacks);
  EXPECT_CALL(mock, RecordedDataIsAvailable(NotNull(), _, _, _, _, Ge(0u), 0, _,
                                            false, _))
      .Times(AtLeast(kNumCallbacks));
  EXPECT_EQ(0, audio_device()->RegisterAudioCallback(&mock));
  StartRecording();
  event()->Wait(kTestTimeOutInMilliseconds);
  EXPECT_TRUE(audio_device()->Recording());
  // Restart recording but without stopping the internal audio thread.
  // This procedure uses a non-public test API and it emulates what happens
  // inside the ADM when e.g. a device is removed.
  EXPECT_EQ(0, audio_device()->RestartRecordingInternally());

  // Run basic tests of public APIs while a restart attempt is active.
  // These calls should now be very thin and not trigger any new actions.
  EXPECT_EQ(-1, audio_device()->StopRecording());
  EXPECT_TRUE(audio_device()->Recording());
  EXPECT_TRUE(audio_device()->RecordingIsInitialized());
  EXPECT_EQ(0, audio_device()->InitRecording());
  EXPECT_EQ(0, audio_device()->StartRecording());

  // Wait until audio has restarted and a new sequence of audio callbacks
  // becomes active.
  // TODO(henrika): is it possible to verify that the internal state transition
  // is Stop->Init->Start?
  ASSERT_TRUE(Mock::VerifyAndClearExpectations(&mock));
  mock.ResetCallbackCounters();
  EXPECT_CALL(mock, RecordedDataIsAvailable(NotNull(), _, _, _, _, Ge(0u), 0, _,
                                            false, _))
      .Times(AtLeast(kNumCallbacks));
  event()->Wait(kTestTimeOutInMilliseconds);
  EXPECT_TRUE(audio_device()->Recording());
  // Stop recording and the audio thread after successful internal restart.
  StopRecording();
}
#endif  // #ifdef WEBRTC_WIN

// Start playout and verify that the native audio layer starts asking for real
// audio samples to play out using the NeedMorePlayData() callback.
// Note that we can't add expectations on audio parameters in EXPECT_CALL
// since parameter are not provided in the each callback. We therefore test and
// verify the parameters in the fake audio transport implementation instead.
TEST_P(AudioDeviceTest, StartPlayoutVerifyCallbacks) {
  SKIP_TEST_IF_NOT(requirements_satisfied());
  MockAudioTransport mock(TransportType::kPlay);
  mock.HandleCallbacks(event(), nullptr, kNumCallbacks);
  EXPECT_CALL(mock, NeedMorePlayData(_, _, _, _, NotNull(), _, _, _))
      .Times(AtLeast(kNumCallbacks));
  EXPECT_EQ(0, audio_device()->RegisterAudioCallback(&mock));
  StartPlayout();
  event()->Wait(kTestTimeOutInMilliseconds);
  StopPlayout();
}

// Start recording and verify that the native audio layer starts providing real
// audio samples using the RecordedDataIsAvailable() callback.
TEST_P(AudioDeviceTest, StartRecordingVerifyCallbacks) {
  SKIP_TEST_IF_NOT(requirements_satisfied());
  MockAudioTransport mock(TransportType::kRecord);
  mock.HandleCallbacks(event(), nullptr, kNumCallbacks);
  EXPECT_CALL(mock, RecordedDataIsAvailable(NotNull(), _, _, _, _, Ge(0u), 0, _,
                                            false, _))
      .Times(AtLeast(kNumCallbacks));
  EXPECT_EQ(0, audio_device()->RegisterAudioCallback(&mock));
  StartRecording();
  event()->Wait(kTestTimeOutInMilliseconds);
  StopRecording();
}

// Start playout and recording (full-duplex audio) and verify that audio is
// active in both directions.
TEST_P(AudioDeviceTest, StartPlayoutAndRecordingVerifyCallbacks) {
  SKIP_TEST_IF_NOT(requirements_satisfied());
  MockAudioTransport mock(TransportType::kPlayAndRecord);
  mock.HandleCallbacks(event(), nullptr, kNumCallbacks);
  EXPECT_CALL(mock, NeedMorePlayData(_, _, _, _, NotNull(), _, _, _))
      .Times(AtLeast(kNumCallbacks));
  EXPECT_CALL(mock, RecordedDataIsAvailable(NotNull(), _, _, _, _, Ge(0u), 0, _,
                                            false, _))
      .Times(AtLeast(kNumCallbacks));
  EXPECT_EQ(0, audio_device()->RegisterAudioCallback(&mock));
  StartPlayout();
  StartRecording();
  event()->Wait(kTestTimeOutInMilliseconds);
  StopRecording();
  StopPlayout();
}

// Start playout and recording and store recorded data in an intermediate FIFO
// buffer from which the playout side then reads its samples in the same order
// as they were stored. Under ideal circumstances, a callback sequence would
// look like: ...+-+-+-+-+-+-+-..., where '+' means 'packet recorded' and '-'
// means 'packet played'. Under such conditions, the FIFO would contain max 1,
// with an average somewhere in (0,1) depending on how long the packets are
// buffered. However, under more realistic conditions, the size
// of the FIFO will vary more due to an unbalance between the two sides.
// This test tries to verify that the device maintains a balanced callback-
// sequence by running in loopback for a few seconds while measuring the size
// (max and average) of the FIFO. The size of the FIFO is increased by the
// recording side and decreased by the playout side.
TEST_P(AudioDeviceTest, RunPlayoutAndRecordingInFullDuplex) {
  SKIP_TEST_IF_NOT(requirements_satisfied());
  NiceMock<MockAudioTransport> mock(TransportType::kPlayAndRecord);
  FifoAudioStream audio_stream;
  mock.HandleCallbacks(event(), &audio_stream,
                       kFullDuplexTimeInSec * kNumCallbacksPerSecond);
  EXPECT_EQ(0, audio_device()->RegisterAudioCallback(&mock));
  // Run both sides using the same channel configuration to avoid conversions
  // between mono/stereo while running in full duplex mode. Also, some devices
  // (mainly on Windows) do not support mono.
  EXPECT_EQ(0, audio_device()->SetStereoPlayout(true));
  EXPECT_EQ(0, audio_device()->SetStereoRecording(true));
  StartPlayout();
  StartRecording();
  event()->Wait(static_cast<int>(
      std::max(kTestTimeOutInMilliseconds, 1000 * kFullDuplexTimeInSec)));
  StopRecording();
  StopPlayout();
  // This thresholds is set rather high to accommodate differences in hardware
  // in several devices. The main idea is to capture cases where a very large
  // latency is built up. See http://bugs.webrtc.org/7744 for examples on
  // bots where relatively large average latencies can happen.
  EXPECT_LE(audio_stream.average_size(), 25u);
  PRINT("\n");
}

// Runs audio in full duplex until user hits Enter. Intended as a manual test
// to ensure that the audio quality is good and that real device switches works
// as intended.
TEST_P(AudioDeviceTest,
       DISABLED_RunPlayoutAndRecordingInFullDuplexAndWaitForEnterKey) {
  SKIP_TEST_IF_NOT(requirements_satisfied());
  if (audio_layer() != AudioDeviceModule::kWindowsCoreAudio2) {
    return;
  }
  NiceMock<MockAudioTransport> mock(TransportType::kPlayAndRecord);
  FifoAudioStream audio_stream;
  mock.HandleCallbacks(&audio_stream);
  EXPECT_EQ(0, audio_device()->RegisterAudioCallback(&mock));
  EXPECT_EQ(0, audio_device()->SetStereoPlayout(true));
  EXPECT_EQ(0, audio_device()->SetStereoRecording(true));
  // Ensure that the sample rate for both directions are identical so that we
  // always can listen to our own voice. Will lead to rate conversion (and
  // higher latency) if the native sample rate is not 48kHz.
  EXPECT_EQ(0, audio_device()->SetPlayoutSampleRate(48000));
  EXPECT_EQ(0, audio_device()->SetRecordingSampleRate(48000));
  StartPlayout();
  StartRecording();
  do {
    PRINT("Loopback audio is active at 48kHz. Press Enter to stop.\n");
  } while (getchar() != '\n');
  StopRecording();
  StopPlayout();
}

// Measures loopback latency and reports the min, max and average values for
// a full duplex audio session.
// The latency is measured like so:
// - Insert impulses periodically on the output side.
// - Detect the impulses on the input side.
// - Measure the time difference between the transmit time and receive time.
// - Store time differences in a vector and calculate min, max and average.
// This test needs the '--gtest_also_run_disabled_tests' flag to run and also
// some sort of audio feedback loop. E.g. a headset where the mic is placed
// close to the speaker to ensure highest possible echo. It is also recommended
// to run the test at highest possible output volume.
TEST_P(AudioDeviceTest, DISABLED_MeasureLoopbackLatency) {
  SKIP_TEST_IF_NOT(requirements_satisfied());
  NiceMock<MockAudioTransport> mock(TransportType::kPlayAndRecord);
  LatencyAudioStream audio_stream;
  mock.HandleCallbacks(event(), &audio_stream,
                       kMeasureLatencyTimeInSec * kNumCallbacksPerSecond);
  EXPECT_EQ(0, audio_device()->RegisterAudioCallback(&mock));
  EXPECT_EQ(0, audio_device()->SetStereoPlayout(true));
  EXPECT_EQ(0, audio_device()->SetStereoRecording(true));
  StartPlayout();
  StartRecording();
  event()->Wait(static_cast<int>(
      std::max(kTestTimeOutInMilliseconds, 1000 * kMeasureLatencyTimeInSec)));
  StopRecording();
  StopPlayout();
  // Verify that a sufficient number of transmitted impulses are detected.
  EXPECT_GE(audio_stream.num_latency_values(),
            static_cast<size_t>(
                kImpulseFrequencyInHz * kMeasureLatencyTimeInSec - 2));
  // Print out min, max and average delay values for debugging purposes.
  audio_stream.PrintResults();
}

#ifdef WEBRTC_WIN
// Test two different audio layers (or rather two different Core Audio
// implementations) for Windows.
INSTANTIATE_TEST_CASE_P(
    AudioLayerWin,
    AudioDeviceTest,
    ::testing::Values(AudioDeviceModule::kPlatformDefaultAudio,
                      AudioDeviceModule::kWindowsCoreAudio2));
#else
// For all platforms but Windows, only test the default audio layer.
INSTANTIATE_TEST_CASE_P(
    AudioLayer,
    AudioDeviceTest,
    ::testing::Values(AudioDeviceModule::kPlatformDefaultAudio));
#endif

}  // namespace webrtc