[FFmpeg-cvslog] nvenc: generate dts properly

[Anton Khirnov]

ffmpeg | branch: master | Anton Khirnov <anton at khirnov.net> | Fri Jan  8 17:08:39 2016 +0100| [c59fec783d6540dd96540b079d753ee4a6ad2e58] | committer: Anton Khirnov

nvenc: generate dts properly

When there is a non-zero decoding delay due to reordering, the first dts
should be lower than the first pts (since the first packet fed to the
decoder does not produce any output).

Use the same scheme used in mpegvideo_enc (which comes from x264
originally) -- wait for first two timestamps and extrapolate linearly to
the past to produce the first dts value.

> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=c59fec783d6540dd96540b079d753ee4a6ad2e58
---

 libavcodec/nvenc.c |   49 +++++++++++++++++++++++++++++++++++++++++++++----
 libavcodec/nvenc.h |    5 ++++-
 2 files changed, 49 insertions(+), 5 deletions(-)

diff --git a/libavcodec/nvenc.c b/libavcodec/nvenc.c
index e23ec88..2dfea94 100644
--- a/libavcodec/nvenc.c
+++ b/libavcodec/nvenc.c
@@ -663,7 +663,6 @@ static int nvenc_setup_encoder(AVCodecContext *avctx)
 
     if (avctx->gop_size > 0) {
         if (avctx->max_b_frames > 0) {
-            ctx->last_dts = -2;
             /* 0 is intra-only,
              * 1 is I/P only,
              * 2 is one B Frame,
@@ -681,6 +680,9 @@ static int nvenc_setup_encoder(AVCodecContext *avctx)
     if (ctx->config.frameIntervalP > 1)
         avctx->max_b_frames = ctx->config.frameIntervalP - 1;
 
+    ctx->initial_pts[0] = AV_NOPTS_VALUE;
+    ctx->initial_pts[1] = AV_NOPTS_VALUE;
+
     nvenc_setup_rate_control(avctx);
 
     if (avctx->flags & AV_CODEC_FLAG_INTERLACED_DCT) {
@@ -1049,13 +1051,35 @@ static inline int nvenc_dequeue_surface(AVFifoBuffer *f,
     return av_fifo_generic_read(f, surf, sizeof(*surf), NULL);
 }
 
-static int nvenc_set_timestamp(NVENCContext *ctx,
+static int nvenc_set_timestamp(AVCodecContext *avctx,
                                NV_ENC_LOCK_BITSTREAM *params,
                                AVPacket *pkt)
 {
+    NVENCContext *ctx = avctx->priv_data;
+
     pkt->pts      = params->outputTimeStamp;
     pkt->duration = params->outputDuration;
 
+    /* generate the first dts by linearly extrapolating the
+     * first two pts values to the past */
+    if (avctx->max_b_frames > 0 && !ctx->first_packet_output &&
+        ctx->initial_pts[1] != AV_NOPTS_VALUE) {
+        int64_t ts0 = ctx->initial_pts[0], ts1 = ctx->initial_pts[1];
+        int64_t delta;
+
+        if ((ts0 < 0 && ts1 > INT64_MAX + ts0) ||
+            (ts0 > 0 && ts1 < INT64_MIN + ts0))
+            return AVERROR(ERANGE);
+        delta = ts1 - ts0;
+
+        if ((delta < 0 && ts0 > INT64_MAX + delta) ||
+            (delta > 0 && ts0 < INT64_MIN + delta))
+            return AVERROR(ERANGE);
+        pkt->dts = ts0 - delta;
+
+        ctx->first_packet_output = 1;
+        return 0;
+    }
     return nvenc_dequeue_timestamp(ctx->timestamps, &pkt->dts);
 }
 
@@ -1090,7 +1114,7 @@ static int nvenc_get_frame(AVCodecContext *avctx, AVPacket *pkt)
 
     out->busy = out->in->locked = 0;
 
-    ret = nvenc_set_timestamp(ctx, &params, pkt);
+    ret = nvenc_set_timestamp(avctx, &params, pkt);
     if (ret < 0)
         return ret;
 
@@ -1119,6 +1143,18 @@ FF_ENABLE_DEPRECATION_WARNINGS
     return 0;
 }
 
+static int output_ready(AVCodecContext *avctx, int flush)
+{
+    NVENCContext *ctx = avctx->priv_data;
+
+    /* when B-frames are enabled, we wait for two initial timestamps to
+     * calculate the first dts */
+    if (!flush && avctx->max_b_frames > 0 &&
+        (ctx->initial_pts[0] == AV_NOPTS_VALUE || ctx->initial_pts[1] == AV_NOPTS_VALUE))
+        return 0;
+    return av_fifo_size(ctx->ready) > 0;
+}
+
 int ff_nvenc_encode_frame(AVCodecContext *avctx, AVPacket *pkt,
                           const AVFrame *frame, int *got_packet)
 {
@@ -1162,6 +1198,11 @@ int ff_nvenc_encode_frame(AVCodecContext *avctx, AVPacket *pkt,
         ret = nvenc_enqueue_timestamp(ctx->timestamps, frame->pts);
         if (ret < 0)
             return ret;
+
+        if (ctx->initial_pts[0] == AV_NOPTS_VALUE)
+            ctx->initial_pts[0] = frame->pts;
+        else if (ctx->initial_pts[1] == AV_NOPTS_VALUE)
+            ctx->initial_pts[1] = frame->pts;
     } else {
         params.encodePicFlags = NV_ENC_PIC_FLAG_EOS;
     }
@@ -1185,7 +1226,7 @@ int ff_nvenc_encode_frame(AVCodecContext *avctx, AVPacket *pkt,
         }
     }
 
-    if (av_fifo_size(ctx->ready) > 0) {
+    if (output_ready(avctx, !frame)) {
         ret = nvenc_get_frame(avctx, pkt);
         if (ret < 0)
             return ret;
diff --git a/libavcodec/nvenc.h b/libavcodec/nvenc.h
index 8819b3c..32d3345 100644
--- a/libavcodec/nvenc.h
+++ b/libavcodec/nvenc.h
@@ -112,7 +112,10 @@ typedef struct NVENCContext {
     AVFifoBuffer *timestamps;
     AVFifoBuffer *pending, *ready;
 
-    int64_t last_dts;
+    /* timestamps of the first two frames, for computing the first dts
+     * when b-frames are present */
+    int64_t initial_pts[2];
+    int first_packet_output;
 
     void *nvenc_ctx;