From: Anton Khirnov <anton@khirnov.net>
To: FFmpeg development discussions and patches <ffmpeg-devel@ffmpeg.org>
Subject: Re: [FFmpeg-devel] [PATCH 1/3] avcodec/decode: move processing discard samples to its own function
Date: Sun, 09 Jul 2023 11:48:11 +0200
Message-ID: <168889609177.542.15252833672314979069@lain.khirnov.net> (raw)
In-Reply-To: <20230708190038.24324-1-jamrial@gmail.com>
Quoting James Almer (2023-07-08 21:00:36)
> -/*
> - * The core of the receive_frame_wrapper for the decoders implementing
> - * the simple API. Certain decoders might consume partial packets without
> - * returning any output, so this function needs to be called in a loop until it
> - * returns EAGAIN.
> - **/
> -static inline int decode_simple_internal(AVCodecContext *avctx, AVFrame *frame, int64_t *discarded_samples)
> +static int discard_samples(AVCodecContext *avctx, AVFrame *frame, int64_t *discarded_samples)
> {
> - AVCodecInternal *avci = avctx->internal;
> - AVPacket *const pkt = avci->in_pkt;
> - const FFCodec *const codec = ffcodec(avctx->codec);
> - int got_frame, actual_got_frame;
> - int ret;
> -
> - if (!pkt->data && !avci->draining) {
> - av_packet_unref(pkt);
> - ret = ff_decode_get_packet(avctx, pkt);
> - if (ret < 0 && ret != AVERROR_EOF)
> - return ret;
> - }
> -
> - // Some codecs (at least wma lossless) will crash when feeding drain packets
> - // after EOF was signaled.
> - if (avci->draining_done)
> - return AVERROR_EOF;
> -
> - if (!pkt->data &&
> - !(avctx->codec->capabilities & AV_CODEC_CAP_DELAY ||
> - avctx->active_thread_type & FF_THREAD_FRAME))
> - return AVERROR_EOF;
> -
> - got_frame = 0;
> -
> - if (HAVE_THREADS && avctx->active_thread_type & FF_THREAD_FRAME) {
> - ret = ff_thread_decode_frame(avctx, frame, &got_frame, pkt);
> - } else {
> - ret = codec->cb.decode(avctx, frame, &got_frame, pkt);
> -
> - if (!(codec->caps_internal & FF_CODEC_CAP_SETS_PKT_DTS))
> - frame->pkt_dts = pkt->dts;
> - if (avctx->codec->type == AVMEDIA_TYPE_VIDEO) {
> -#if FF_API_FRAME_PKT
> -FF_DISABLE_DEPRECATION_WARNINGS
> - if(!avctx->has_b_frames)
> - frame->pkt_pos = pkt->pos;
> -FF_ENABLE_DEPRECATION_WARNINGS
> -#endif
> - //FIXME these should be under if(!avctx->has_b_frames)
> - /* get_buffer is supposed to set frame parameters */
> - if (!(avctx->codec->capabilities & AV_CODEC_CAP_DR1)) {
> - if (!frame->sample_aspect_ratio.num) frame->sample_aspect_ratio = avctx->sample_aspect_ratio;
> - if (!frame->width) frame->width = avctx->width;
> - if (!frame->height) frame->height = avctx->height;
> - if (frame->format == AV_PIX_FMT_NONE) frame->format = avctx->pix_fmt;
> - }
> - }
> - }
> - emms_c();
> - actual_got_frame = got_frame;
> + AVCodecInternal *avci = avctx->internal;
> + int ret = 0;
>
> if (avctx->codec->type == AVMEDIA_TYPE_VIDEO) {
> if (frame->flags & AV_FRAME_FLAG_DISCARD)
> - got_frame = 0;
> + ret = AVERROR(EAGAIN);
It's quite strange an unexpected for a function called discard_samples()
to do anything with video. By leaving video processing in the caller you
also save a level of indentation.
> } else if (avctx->codec->type == AVMEDIA_TYPE_AUDIO) {
> uint8_t *side;
> size_t side_size;
> @@ -359,16 +304,10 @@ FF_ENABLE_DEPRECATION_WARNINGS
> uint8_t skip_reason = 0;
> uint8_t discard_reason = 0;
>
> - if (ret >= 0 && got_frame) {
> if (frame->format == AV_SAMPLE_FMT_NONE)
> frame->format = avctx->sample_fmt;
> - if (!frame->ch_layout.nb_channels) {
> - int ret2 = av_channel_layout_copy(&frame->ch_layout, &avctx->ch_layout);
> - if (ret2 < 0) {
> - ret = ret2;
> - got_frame = 0;
> - }
> - }
> + if (!frame->ch_layout.nb_channels)
> + ret = av_channel_layout_copy(&frame->ch_layout, &avctx->ch_layout);
> #if FF_API_OLD_CHANNEL_LAYOUT
> FF_DISABLE_DEPRECATION_WARNINGS
> if (!frame->channel_layout)
> @@ -380,7 +319,6 @@ FF_ENABLE_DEPRECATION_WARNINGS
> #endif
> if (!frame->sample_rate)
> frame->sample_rate = avctx->sample_rate;
> - }
So the rest of the code below was previously executed even when there is
no frame? The change seems correct, but not I wouldn't expect it in a
'move' commit, so perhaps mention it in the commit message.
>
> side= av_packet_get_side_data(avci->last_pkt_props, AV_PKT_DATA_SKIP_SAMPLES, &side_size);
> if(side && side_size>=10) {
> @@ -393,21 +331,21 @@ FF_ENABLE_DEPRECATION_WARNINGS
> discard_reason = AV_RL8(side + 9);
> }
>
> - if ((frame->flags & AV_FRAME_FLAG_DISCARD) && got_frame &&
> + if ((frame->flags & AV_FRAME_FLAG_DISCARD) && !ret &&
> !(avctx->flags2 & AV_CODEC_FLAG2_SKIP_MANUAL)) {
> avci->skip_samples = FFMAX(0, avci->skip_samples - frame->nb_samples);
> - got_frame = 0;
> *discarded_samples += frame->nb_samples;
> + ret = AVERROR(EAGAIN);
If I'm reading correctly, all the blocks below are conditioned on !ret,
so you might as well return here and drop those checks.
--
Anton Khirnov
_______________________________________________
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:[~2023-07-09 9:48 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-07-08 19:00 James Almer
2023-07-08 19:00 ` [FFmpeg-devel] [PATCH 2/3] avcodec/decode: check the output frame for discard samples with all decoders James Almer
2023-07-08 19:00 ` [FFmpeg-devel] [PATCH 3/3] avcodec/decode: fill missing frame fields for " James Almer
2023-07-09 9:49 ` Anton Khirnov
2023-07-09 12:05 ` James Almer
2023-07-09 12:58 ` Anton Khirnov
2023-07-08 19:38 ` [FFmpeg-devel] [PATCH 1/3] avcodec/decode: move processing discard samples to its own function Andreas Rheinhardt
2023-07-08 19:41 ` James Almer
2023-07-09 9:48 ` Anton Khirnov [this message]
2023-07-09 12:11 ` James Almer
2023-07-09 12:15 ` Anton Khirnov
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=168889609177.542.15252833672314979069@lain.khirnov.net \
--to=anton@khirnov.net \
--cc=ffmpeg-devel@ffmpeg.org \
/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