[FFmpeg-devel] [PATCH v2] fftools/ffmpeg_mux: fix reporting muxer EOF as error

["zhilizhao(赵志立)"]

> On Apr 23, 2023, at 17:42, Marton Balint <cus at passwd.hu> wrote:
> 
> On Sun, 23 Apr 2023, Anton Khirnov wrote:
> 
>> Quoting Marton Balint (2023-04-23 11:12:38)
>>> 
>>> 
>>> On Sat, 22 Apr 2023, Anton Khirnov wrote:
>>> 
>>>> Quoting Zhao Zhili (2023-04-22 14:56:34)
>>>>> From: Zhao Zhili <zhilizhao at tencent.com>
>>>>> 
>>>>> Fix #10327.
>>>>> 
>>>>> Signed-off-by: Zhao Zhili <zhilizhao at tencent.com>
>>>>> ---
>>>>> fftools/ffmpeg_mux.c | 12 +++++++++---
>>>>> 1 file changed, 9 insertions(+), 3 deletions(-)
>>>>> 
>>>>> diff --git a/fftools/ffmpeg_mux.c b/fftools/ffmpeg_mux.c
>>>>> index a2e8873ad2..0e1a5d7dc5 100644
>>>>> --- a/fftools/ffmpeg_mux.c
>>>>> +++ b/fftools/ffmpeg_mux.c
>>>>> @@ -214,9 +214,15 @@ static void *muxer_thread(void *arg)
>>>>>         ost = of->streams[stream_idx];
>>>>>         ret = sync_queue_process(mux, ost, ret < 0 ? NULL : pkt, &stream_eof);
>>>>>         av_packet_unref(pkt);
>>>>> -        if (ret == AVERROR_EOF && stream_eof)
>>>>> -            tq_receive_finish(mux->tq, stream_idx);
>>>>> -        else if (ret < 0) {
>>>>> +        if (ret == AVERROR_EOF) {
>>>>> +            if (stream_eof) {
>>>>> +                tq_receive_finish(mux->tq, stream_idx);
>>>>> +            } else {
>>>>> +                av_log(mux, AV_LOG_VERBOSE, "Muxer %s\n", av_err2str(ret));
>>>> 
>>>> That seems unnecesarily convoluted, given that we _know_ the error to be
>>>> EOF here. Also, please make it "Muxer returned EOF" to make it clear
>>>> what exactly is the source of EOF.
>>>> 
>>>> Otherwise ok, feel free to push with this change.
>>> 
>>> This seems like yet another clash of AVERROR_EOF error codes coming from
>>> different places with different semantics. For
>>> av_interleaved_write_frame(), AVERROR_EOF is an error condition, so
>>> file encoding should fail,
>> 
>> Why should it fail? I'd think a muxer returning EOF is the way to signal
>> non-error muxer-side termination.
> 
> That would be an API change. AVERROR_EOF is not special in any way from other error codes for av_interleaved_write_frame. A muxer cannot signal non-error muxer side termination with existing API.

Sorry I have pushed before seeing your comments.

It can happen in theory, but I’m not sure in which case muxer can return
EOF and should be treated as a real error condition.

A simple grep shows:

$ grep 'AVERROR_EOF'  libavformat/*enc.c

libavformat/hlsenc.c:        if (ret >= 0 || ret == AVERROR_EOF) <--- It changed ret to 0 afterwards.
libavformat/webmdashenc.c:    return AVERROR_EOF;                <—-- More like a bug on the caller side than a real error.

> 
> Regards,
> Marton
> _______________________________________________
> ffmpeg-devel mailing list
> ffmpeg-devel at ffmpeg.org
> https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
> 
> To unsubscribe, visit link above, or email
> ffmpeg-devel-request at ffmpeg.org with subject "unsubscribe".