From: Zhao Zhili <quinkblack@foxmail.com> To: ffmpeg-devel@ffmpeg.org Cc: Zhao Zhili <zhilizhao@tencent.com> Subject: [FFmpeg-devel] [PATCH 5/7] avcodec/mediacodecenc: remove the strategy to create DTS Date: Mon, 5 Dec 2022 01:12:26 +0800 Message-ID: <tencent_E2E9936B95CC490CEEB505497300C8539305@qq.com> (raw) In-Reply-To: <20221204171228.50160-1-quinkblack@foxmail.com> From: Zhao Zhili <zhilizhao@tencent.com> Use input PTS as DTS has multiple problems: 1. If there is no reordering, it's better to just use the output PTS as DTS, since encoder may change the timestamp value (do it on purpose or rounding error). 2. If there is reordering, input PTS should be shift a few frames as DTS to satisfy the requirement of PTS >= DTS. I can't find a reliable way to determine how many frames to be shift. For example, we don't known if the encoder use hierarchical B frames. The max_num_reorder_frames can be get from VUI, but VUI is optional. 3. Encoder dropping frames makes the case worse. Android has an BITRATE_MODE_CBR_FD option to allow it explicitly. --- libavcodec/mediacodecenc.c | 28 +--------------------------- 1 file changed, 1 insertion(+), 27 deletions(-) diff --git a/libavcodec/mediacodecenc.c b/libavcodec/mediacodecenc.c index dce4a96d1f..4b0397d1ea 100644 --- a/libavcodec/mediacodecenc.c +++ b/libavcodec/mediacodecenc.c @@ -64,18 +64,6 @@ typedef struct MediaCodecEncContext { uint8_t *extradata; int extradata_size; - - // Since MediaCodec doesn't output DTS, use a timestamp queue to save pts - // of AVFrame and generate DTS for AVPacket. - // - // This doesn't work when use Surface as input, in that case frames can be - // sent to encoder without our notice. One exception is frames come from - // our MediaCodec decoder wrapper, since we can control it's render by - // av_mediacodec_release_buffer. - int64_t timestamps[32]; - int ts_head; - int ts_tail; - int eof_sent; AVFrame *frame; @@ -368,11 +356,6 @@ static int mediacodec_receive(AVCodecContext *avctx, } memcpy(pkt->data + extradata_size, out_buf + out_info.offset, out_info.size); pkt->pts = av_rescale_q(out_info.presentationTimeUs, AV_TIME_BASE_Q, avctx->time_base); - if (s->ts_tail != s->ts_head) { - pkt->dts = s->timestamps[s->ts_tail]; - s->ts_tail = (s->ts_tail + 1) % FF_ARRAY_ELEMS(s->timestamps); - } - if (out_info.flags & ff_AMediaCodec_getBufferFlagKeyFrame(codec)) pkt->flags |= AV_PKT_FLAG_KEY; ret = 0; @@ -437,14 +420,8 @@ static int mediacodec_send(AVCodecContext *avctx, return ff_AMediaCodec_signalEndOfInputStream(codec); } - - if (frame->data[3]) { - pts = av_rescale_q(frame->pts, avctx->time_base, AV_TIME_BASE_Q); - s->timestamps[s->ts_head] = frame->pts; - s->ts_head = (s->ts_head + 1) % FF_ARRAY_ELEMS(s->timestamps); - + if (frame->data[3]) av_mediacodec_release_buffer((AVMediaCodecBuffer *)frame->data[3], 1); - } return 0; } @@ -463,9 +440,6 @@ static int mediacodec_send(AVCodecContext *avctx, copy_frame_to_buffer(avctx, frame, input_buf, input_size); pts = av_rescale_q(frame->pts, avctx->time_base, AV_TIME_BASE_Q); - - s->timestamps[s->ts_head] = frame->pts; - s->ts_head = (s->ts_head + 1) % FF_ARRAY_ELEMS(s->timestamps); } else { flags |= ff_AMediaCodec_getBufferFlagEndOfStream(codec); s->eof_sent = 1; -- 2.25.1 _______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-devel To unsubscribe, visit link above, or email ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
next prev parent reply other threads:[~2022-12-04 17:13 UTC|newest] Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top [not found] <20221204171228.50160-1-quinkblack@foxmail.com> 2022-12-04 17:12 ` [FFmpeg-devel] [PATCH 2/7] avcodec/mediacodecenc: make each encoder has its own option Zhao Zhili 2022-12-05 4:23 ` Steven Liu 2022-12-05 5:29 ` [FFmpeg-devel] [Internet]Re: " "zhilizhao(赵志立)" 2022-12-04 17:12 ` [FFmpeg-devel] [PATCH 3/7] avcodec/mediacodecenc: add level option Zhao Zhili 2022-12-04 17:12 ` [FFmpeg-devel] [PATCH 4/7] avcodec/mediacodecenc: use bsf to handle crop Zhao Zhili 2022-12-04 17:12 ` Zhao Zhili [this message] 2022-12-04 17:12 ` [FFmpeg-devel] [PATCH 6/7] avcodec/mediacodecenc: add max-bframes support Zhao Zhili 2022-12-04 17:12 ` [FFmpeg-devel] [PATCH 7/7] avcodec/mediacodecenc: add pts_as_dts option Zhao Zhili
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=tencent_E2E9936B95CC490CEEB505497300C8539305@qq.com \ --to=quinkblack@foxmail.com \ --cc=ffmpeg-devel@ffmpeg.org \ --cc=zhilizhao@tencent.com \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: link
Git Inbox Mirror of the ffmpeg-devel mailing list - see https://ffmpeg.org/mailman/listinfo/ffmpeg-devel This inbox may be cloned and mirrored by anyone: git clone --mirror https://master.gitmailbox.com/ffmpegdev/0 ffmpegdev/git/0.git # If you have public-inbox 1.1+ installed, you may # initialize and index your mirror using the following commands: public-inbox-init -V2 ffmpegdev ffmpegdev/ https://master.gitmailbox.com/ffmpegdev \ ffmpegdev@gitmailbox.com public-inbox-index ffmpegdev Example config snippet for mirrors. AGPL code for this site: git clone https://public-inbox.org/public-inbox.git