Message ID | 20230403100954.11563-1-anton@khirnov.net |
---|---|
State | Accepted |
Commit | 416e2661ea9c2b55d5b9e25d687f0d2a839d6cd6 |
Headers | show |
Series | [FFmpeg-devel] fftools/ffmpeg: make sure non-lavfi streams are closed on input EOF | expand |
Context | Check | Description |
---|---|---|
andriy/make_x86 | success | Make finished |
andriy/make_fate_x86 | success | Make fate finished |
On Mon, Apr 03, 2023 at 12:09:54PM +0200, Anton Khirnov wrote: > --- > Should be fixed by adding this patch to the set, at any point. > Also updated my ffmpeg_frame_size branch. > --- > fftools/ffmpeg.c | 1 + > 1 file changed, 1 insertion(+) Ive tested ffmpeg_frame_size and there are a few cases that change their output, but i failed to find one that by watching the output looked different I saw no crashes, infinite loops or other anomalies Thx [...]
Quoting Michael Niedermayer (2023-04-06 00:33:07) > On Mon, Apr 03, 2023 at 12:09:54PM +0200, Anton Khirnov wrote: > > --- > > Should be fixed by adding this patch to the set, at any point. > > Also updated my ffmpeg_frame_size branch. > > --- > > fftools/ffmpeg.c | 1 + > > 1 file changed, 1 insertion(+) > > Ive tested ffmpeg_frame_size and there are a few cases that change their > output, but i failed to find one that by watching the output looked different > I saw no crashes, infinite loops or other anomalies Thanks you for testing, I'd be interested in looking at some of those changed cases, if the samples are available. While changes in some corner cases are possible, it'd be nice to make sure that nothing actually breaks. Otherwise I'm going to push the set soonish.
diff --git a/fftools/ffmpeg.c b/fftools/ffmpeg.c index 438bee8fef..2f1f830507 100644 --- a/fftools/ffmpeg.c +++ b/fftools/ffmpeg.c @@ -3804,6 +3804,7 @@ static int process_input(int file_index) if (ost->ist == ist && (!ost->enc_ctx || ost->enc_ctx->codec_type == AVMEDIA_TYPE_SUBTITLE)) { OutputFile *of = output_files[ost->file_index]; + close_output_stream(ost); of_output_packet(of, ost->pkt, ost, 1); } }