On Sat, Mar 9, 2024 at 2:09 AM Muhammad Faiz <mfc...@gmail.com> wrote:
> On Fri, Mar 8, 2024 at 10:30 PM Paul B Mahol <one...@gmail.com> wrote: > > > On Fri, Mar 8, 2024 at 4:18 PM Muhammad Faiz <mfc...@gmail.com> wrote: > > > > > On Fri, Mar 8, 2024 at 5:40 PM Andreas Rheinhardt < > > > andreas.rheinha...@outlook.com> wrote: > > > > > > > Muhammad Faiz: > > > > > On Tue, Feb 6, 2024 at 3:58 PM Andreas Rheinhardt < > > > > > andreas.rheinha...@outlook.com> wrote: > > > > > > > > > >> Andreas Rheinhardt: > > > > >>> Obsolete since 4ca1fb9d2a91757c8c4c34dd456abf340e3f765f. > > > > >>> > > > > >>> Signed-off-by: Andreas Rheinhardt < > andreas.rheinha...@outlook.com> > > > > >>> --- > > > > >>> doc/filters.texi | 9 --- > > > > >>> libavfilter/Makefile | 1 - > > > > >>> libavfilter/allfilters.c | 2 - > > > > >>> libavfilter/fifo.c | 165 > > > --------------------------------------- > > > > >>> 4 files changed, 177 deletions(-) > > > > >>> delete mode 100644 libavfilter/fifo.c > > > > >>> > > > > >> > > > > >> Will apply in a few days unless there are objections. > > > > >> > > > > >> - Andreas > > > > >> > > > > >> > > > > > This breaks backward compatibility. > > > > > > > > > > Please revert. > > > > > > > > > > Thank's. > > > > > > > > What breaks that can't simply be fixed by removing the (a)fifo filter > > > > from the filterchain? > > > > > > > > - Andreas > > > > > > > > > > > I use afifo to optimize memory usage. > > > > > > > That statement sync with reality is questionable, fifo filters were > mainly > > used before .activate was added, > > for cases filters used >1 inputs. Now they should be irrelevant, unless > > there are bugs in code than this filters just hide more bugs. > > > > Yes, I've used it before .activate was added, and it worked, and still > works. > Although maybe the bugs have been fixed by .activate, what's wrong if > (a)fifo still exist? > As already wrote, lets repeat it again, hiding real bugs. Hiding is not correct word, it just adds yet another layer in filtergraph to accumulate memory/frames, with extra allocations per each frame for linking between prev and next frame. Which is pointless, as libavfilter internal code already have own queue mechanism. Check for which filters used after afifo, afifo count of queued frames reach > 1 than such filters needs fixing. Personally, my mpv visualizer modified script is not using afifo filters, and it works fine. > > Thank's > _______________________________________________ > ffmpeg-devel mailing list > ffmpeg-devel@ffmpeg.org > https://ffmpeg.org/mailman/listinfo/ffmpeg-devel > > To unsubscribe, visit link above, or email > ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe". > _______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-devel To unsubscribe, visit link above, or email ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".
