VideoCaptureInput enforce VideoFrame::render_time to be generated by webrtc clock.

render_time time field (means capture time for sender side) is used by rtcp SenderReport to calculate offset since last frame and to estimate rtp timestamp for the time SenderReport should be send at.
mapping between rtp timestamp and ntp time in SenderReport is used for stream synchronization.

calculation of rtp_timestamp (using ntp_time of incoming video frame) for rtp packets is unchanged.

BUG=webrtc:5433, webrtc:5504, webrtc:5505

Review URL: https://codereview.webrtc.org/1693443002

Cr-Commit-Position: refs/heads/master@{#11820}

webrtc/call/call_perf_tests.cc

@@ -383,9 +383,7 @@ void CallPerfTest::TestAudioVideoSync(FecMode fec,
   VoiceEngine::Delete(voice_engine);
 }
 
-// TODO(danilchap): Reenable after adding support for frame capture clock
-// that is not in sync with local TickTime clock.
-TEST_F(CallPerfTest, DISABLED_PlaysOutAudioAndVideoInSyncWithVideoNtpDrift) {
+TEST_F(CallPerfTest, PlaysOutAudioAndVideoInSyncWithVideoNtpDrift) {
   TestAudioVideoSync(FecMode::kOff, CreateOrder::kAudioFirst,
                      DriftingClock::PercentsFaster(10.0f),
                      DriftingClock::kNoDrift, DriftingClock::kNoDrift);

webrtc/video/video_capture_input.cc

@@ -17,8 +17,6 @@
 #include "webrtc/modules/video_capture/video_capture_factory.h"
 #include "webrtc/modules/video_processing/include/video_processing.h"
 #include "webrtc/modules/video_render/video_render_defines.h"
-#include "webrtc/system_wrappers/include/clock.h"
-#include "webrtc/system_wrappers/include/tick_util.h"
 #include "webrtc/video/overuse_frame_detector.h"
 #include "webrtc/video/send_statistics_proxy.h"
 #include "webrtc/video/vie_encoder.h"
@@ -36,10 +34,13 @@ VideoCaptureInput::VideoCaptureInput(VideoCaptureCallback* frame_callback,
       encoder_thread_(EncoderThreadFunction, this, "EncoderThread"),
       capture_event_(false, false),
       stop_(0),
+      // TODO(danilchap): Pass clock from outside to ensure it is same clock
+      // rtcp module use to calculate offset since last frame captured
+      // to estimate rtp timestamp for SenderReport.
+      clock_(Clock::GetRealTimeClock()),
       last_captured_timestamp_(0),
-      delta_ntp_internal_ms_(
-          Clock::GetRealTimeClock()->CurrentNtpInMilliseconds() -
-          TickTime::MillisecondTimestamp()),
+      delta_ntp_internal_ms_(clock_->CurrentNtpInMilliseconds() -
+                             clock_->TimeInMilliseconds()),
       overuse_detector_(overuse_detector) {
   encoder_thread_.Start();
   encoder_thread_.SetPriority(rtc::kHighPriority);
@@ -62,18 +63,20 @@ void VideoCaptureInput::IncomingCapturedFrame(const VideoFrame& video_frame) {
 
   VideoFrame incoming_frame = video_frame;
 
-  if (incoming_frame.ntp_time_ms() != 0) {
-    // If a NTP time stamp is set, this is the time stamp we will use.
-    incoming_frame.set_render_time_ms(incoming_frame.ntp_time_ms() -
-                                      delta_ntp_internal_ms_);
-  } else {  // NTP time stamp not set.
-    int64_t render_time = incoming_frame.render_time_ms() != 0
-                              ? incoming_frame.render_time_ms()
-                              : TickTime::MillisecondTimestamp();
+  // Local time in webrtc time base.
+  int64_t current_time = clock_->TimeInMilliseconds();
+  incoming_frame.set_render_time_ms(current_time);
 
-    incoming_frame.set_render_time_ms(render_time);
-    incoming_frame.set_ntp_time_ms(render_time + delta_ntp_internal_ms_);
+  // Capture time may come from clock with an offset and drift from clock_.
+  int64_t capture_ntp_time_ms;
+  if (video_frame.ntp_time_ms() != 0) {
+    capture_ntp_time_ms = video_frame.ntp_time_ms();
+  } else if (video_frame.render_time_ms() != 0) {
+    capture_ntp_time_ms = video_frame.render_time_ms() + delta_ntp_internal_ms_;
+  } else {
+    capture_ntp_time_ms = current_time + delta_ntp_internal_ms_;
   }
+  incoming_frame.set_ntp_time_ms(capture_ntp_time_ms);
 
   // Convert NTP time, in ms, to RTP timestamp.
   const int kMsToRtpTimestamp = 90;

webrtc/video/video_capture_input.h

@@ -23,6 +23,7 @@
 #include "webrtc/modules/video_coding/include/video_codec_interface.h"
 #include "webrtc/modules/video_coding/include/video_coding.h"
 #include "webrtc/modules/video_processing/include/video_processing.h"
+#include "webrtc/system_wrappers/include/clock.h"
 #include "webrtc/typedefs.h"
 #include "webrtc/video_send_stream.h"
 
@@ -68,6 +69,7 @@ class VideoCaptureInput : public webrtc::VideoCaptureInput {
   volatile int stop_;
 
   VideoFrame captured_frame_ GUARDED_BY(crit_);
+  Clock* const clock_;
   // Used to make sure incoming time stamp is increasing for every frame.
   int64_t last_captured_timestamp_;
   // Delta used for translating between NTP and internal timestamps.

webrtc/video/video_capture_input_unittest.cc

@@ -249,8 +249,7 @@ bool EqualFrames(const VideoFrame& frame1, const VideoFrame& frame2) {
 bool EqualTextureFrames(const VideoFrame& frame1, const VideoFrame& frame2) {
   return ((frame1.native_handle() == frame2.native_handle()) &&
           (frame1.width() == frame2.width()) &&
-          (frame1.height() == frame2.height()) &&
-          (frame1.render_time_ms() == frame2.render_time_ms()));
+          (frame1.height() == frame2.height()));
 }
 
 bool EqualBufferFrames(const VideoFrame& frame1, const VideoFrame& frame2) {
@@ -259,7 +258,6 @@ bool EqualBufferFrames(const VideoFrame& frame1, const VideoFrame& frame2) {
           (frame1.stride(kYPlane) == frame2.stride(kYPlane)) &&
           (frame1.stride(kUPlane) == frame2.stride(kUPlane)) &&
           (frame1.stride(kVPlane) == frame2.stride(kVPlane)) &&
-          (frame1.render_time_ms() == frame2.render_time_ms()) &&
           (frame1.allocated_size(kYPlane) == frame2.allocated_size(kYPlane)) &&
           (frame1.allocated_size(kUPlane) == frame2.allocated_size(kUPlane)) &&
           (frame1.allocated_size(kVPlane) == frame2.allocated_size(kVPlane)) &&

webrtc/video/video_send_stream_tests.cc

@@ -1257,7 +1257,6 @@ void ExpectEqualTextureFrames(const VideoFrame& frame1,
   EXPECT_EQ(frame1.native_handle(), frame2.native_handle());
   EXPECT_EQ(frame1.width(), frame2.width());
   EXPECT_EQ(frame1.height(), frame2.height());
-  EXPECT_EQ(frame1.render_time_ms(), frame2.render_time_ms());
 }
 
 void ExpectEqualBufferFrames(const VideoFrame& frame1,
@@ -1267,7 +1266,6 @@ void ExpectEqualBufferFrames(const VideoFrame& frame1,
   EXPECT_EQ(frame1.stride(kYPlane), frame2.stride(kYPlane));
   EXPECT_EQ(frame1.stride(kUPlane), frame2.stride(kUPlane));
   EXPECT_EQ(frame1.stride(kVPlane), frame2.stride(kVPlane));
-  EXPECT_EQ(frame1.render_time_ms(), frame2.render_time_ms());
   ASSERT_EQ(frame1.allocated_size(kYPlane), frame2.allocated_size(kYPlane));
   EXPECT_EQ(0,
             memcmp(frame1.buffer(kYPlane),