Compute relative PTS on the client-side

The PTS received by MediaCodec are expressed relative to an arbitrary
clock origin. We consider the PTS of the first frame to be 0, and the
PTS of every other frame is relative to this first PTS (note that the
PTS is only used for recording, it is ignored for mirroring).

For simplicity, this relative PTS was computed on the server-side.

To prepare support for multiple stream (video and audio), send the
packet with its original PTS, and handle the PTS offset on the
client-side (by the recorder).

Since we can't know in advance which stream will produce the first
packet with the lowest PTS (a packet received later on one stream may
have a PTS lower than a packet received earlier on another stream),
computing the PTS on the server-side would require unnecessary waiting.

app/src/recorder.c

@@ -11,6 +11,8 @@
 /** Downcast packet_sink to recorder */
 #define DOWNCAST(SINK) container_of(SINK, struct sc_recorder, packet_sink)
 
+#define SC_PTS_ORIGIN_NONE UINT64_C(-1)
+
 static const AVRational SCRCPY_TIME_BASE = {1, 1000000}; // timestamps in us
 
 static const AVOutputFormat *
@@ -128,6 +130,8 @@ sc_recorder_write(struct sc_recorder *recorder, AVPacket *packet) {
         return true;
     }
 
+    LOGI("==== %" PRIu64, packet->pts);
+
     sc_recorder_rescale_packet(recorder, packet);
     return av_write_frame(recorder->ctx, packet) >= 0;
 }
@@ -169,6 +173,18 @@ run_recorder(void *data) {
 
         sc_mutex_unlock(&recorder->mutex);
 
+        if (recorder->pts_origin == SC_PTS_ORIGIN_NONE
+                && rec->packet->pts != AV_NOPTS_VALUE) {
+            // First PTS received
+            recorder->pts_origin = rec->packet->pts;
+        }
+
+        if (rec->packet->pts != AV_NOPTS_VALUE) {
+            // Set PTS relatve to the origin
+            rec->packet->pts -= recorder->pts_origin;
+            rec->packet->dts = rec->packet->pts;
+        }
+
         // recorder->previous is only written from this thread, no need to lock
         struct sc_record_packet *previous = recorder->previous;
         recorder->previous = rec;
@@ -243,6 +259,7 @@ sc_recorder_open(struct sc_recorder *recorder, const AVCodec *input_codec) {
     recorder->failed = false;
     recorder->header_written = false;
     recorder->previous = NULL;
+    recorder->pts_origin = SC_PTS_ORIGIN_NONE;
 
     const char *format_name = sc_recorder_get_format_name(recorder->format);
     assert(format_name);

app/src/recorder.h

@@ -28,6 +28,8 @@ struct sc_recorder {
     struct sc_size declared_frame_size;
     bool header_written;
 
+    uint64_t pts_origin;
+
     sc_thread thread;
     sc_mutex mutex;
     sc_cond queue_cond;

server/src/main/java/com/genymobile/scrcpy/ScreenEncoder.java

@@ -42,7 +42,6 @@ public class ScreenEncoder implements Device.RotationListener {
     private final int maxFps;
     private final boolean sendFrameMeta;
     private final boolean downsizeOnError;
-    private long ptsOrigin;
 
     private boolean firstFrameSent;
     private int consecutiveErrors;
@@ -218,10 +217,7 @@ public class ScreenEncoder implements Device.RotationListener {
         if ((bufferInfo.flags & MediaCodec.BUFFER_FLAG_CODEC_CONFIG) != 0) {
             pts = PACKET_FLAG_CONFIG; // non-media data packet
         } else {
-            if (ptsOrigin == 0) {
+            pts = bufferInfo.presentationTimeUs;
-                ptsOrigin = bufferInfo.presentationTimeUs;
-            }
-            pts = bufferInfo.presentationTimeUs - ptsOrigin;
             if ((bufferInfo.flags & MediaCodec.BUFFER_FLAG_KEY_FRAME) != 0) {
                 pts |= PACKET_FLAG_KEY_FRAME;
             }