| Message ID | 20230612033311.363757-1-fei.w.wang@intel.com |
|---|---|
| State | New |
| Headers | show |
| Series | [FFmpeg-devel,v1] fftools/ffmpeg_dec: Don't keep sending frame to filters in flushing | expand |
| Context | Check | Description |
|---|---|---|
| yinshiyou/make_loongarch64 | success | Make finished |
| yinshiyou/make_fate_loongarch64 | success | Make fate finished |
| andriy/make_x86 | success | Make finished |
| andriy/make_fate_x86 | success | Make fate finished |
On Ma, 2023-06-12 at 11:33 +0800, Fei Wang wrote: > Filter may has a limited frame pool size. Do not always send frame to > filters without reaping. > > Fix the regression of commit 5fa00b38e6. > > Example cmd: > $ ffmpeg -threads 20 -init_hw_device vaapi=hw:/dev/dri/renderD128 \ > -hwaccel_output_format vaapi -hwaccel vaapi -i avc_1080P.mp4 \ > -vf 'scale_vaapi=w=720:h=480' -f null - > > Signed-off-by: Fei Wang <fei.w.wang@intel.com> > --- > fftools/ffmpeg_dec.c | 6 ++++++ > 1 file changed, 6 insertions(+) > > diff --git a/fftools/ffmpeg_dec.c b/fftools/ffmpeg_dec.c > index 799be63215..d6b34402f8 100644 > --- a/fftools/ffmpeg_dec.c > +++ b/fftools/ffmpeg_dec.c > @@ -531,6 +531,12 @@ int dec_packet(InputStream *ist, const AVPacket *pkt, int > no_eof) > av_frame_unref(frame); > if (ret < 0) > exit_program(1); > + > + // During flushing, break out to reap filter once send a frame to > filters to > + // avoid drain out filter's output frame pool. Especially for HW > filters which > + // always have a limited frame pool size. > + if (!pkt) > + return 0; Looks good and it works well for me. I'm going to push it if no objection Thanks Haihao > } > } >
On So, 2023-06-25 at 06:20 +0000, Xiang, Haihao wrote: > On Ma, 2023-06-12 at 11:33 +0800, Fei Wang wrote: > > Filter may has a limited frame pool size. Do not always send frame to > > filters without reaping. > > > > Fix the regression of commit 5fa00b38e6. > > > > Example cmd: > > $ ffmpeg -threads 20 -init_hw_device vaapi=hw:/dev/dri/renderD128 \ > > -hwaccel_output_format vaapi -hwaccel vaapi -i avc_1080P.mp4 \ > > -vf 'scale_vaapi=w=720:h=480' -f null - > > > > Signed-off-by: Fei Wang <fei.w.wang@intel.com> > > --- > > fftools/ffmpeg_dec.c | 6 ++++++ > > 1 file changed, 6 insertions(+) > > > > diff --git a/fftools/ffmpeg_dec.c b/fftools/ffmpeg_dec.c > > index 799be63215..d6b34402f8 100644 > > --- a/fftools/ffmpeg_dec.c > > +++ b/fftools/ffmpeg_dec.c > > @@ -531,6 +531,12 @@ int dec_packet(InputStream *ist, const AVPacket *pkt, > > int > > no_eof) > > av_frame_unref(frame); > > if (ret < 0) > > exit_program(1); > > + > > + // During flushing, break out to reap filter once send a frame to > > filters to > > + // avoid drain out filter's output frame pool. Especially for HW > > filters which > > + // always have a limited frame pool size. > > + if (!pkt) > > + return 0; > > Looks good and it works well for me. I'm going to push it if no objection > There is a conflict when applying your patch to the latest FFmpeg, could you rebase the patch ? Thanks Haihao
On Mon, 2023-06-26 at 05:01 +0000, Xiang, Haihao wrote: > On So, 2023-06-25 at 06:20 +0000, Xiang, Haihao wrote: > > On Ma, 2023-06-12 at 11:33 +0800, Fei Wang wrote: > > > Filter may has a limited frame pool size. Do not always send > > > frame to > > > filters without reaping. > > > > > > Fix the regression of commit 5fa00b38e6. > > > > > > Example cmd: > > > $ ffmpeg -threads 20 -init_hw_device > > > vaapi=hw:/dev/dri/renderD128 \ > > > -hwaccel_output_format vaapi -hwaccel vaapi -i > > > avc_1080P.mp4 \ > > > -vf 'scale_vaapi=w=720:h=480' -f null - > > > > > > Signed-off-by: Fei Wang <fei.w.wang@intel.com> > > > --- > > > fftools/ffmpeg_dec.c | 6 ++++++ > > > 1 file changed, 6 insertions(+) > > > > > > diff --git a/fftools/ffmpeg_dec.c b/fftools/ffmpeg_dec.c > > > index 799be63215..d6b34402f8 100644 > > > --- a/fftools/ffmpeg_dec.c > > > +++ b/fftools/ffmpeg_dec.c > > > @@ -531,6 +531,12 @@ int dec_packet(InputStream *ist, const > > > AVPacket *pkt, > > > int > > > no_eof) > > > av_frame_unref(frame); > > > if (ret < 0) > > > exit_program(1); > > > + > > > + // During flushing, break out to reap filter once send a > > > frame to > > > filters to > > > + // avoid drain out filter's output frame pool. > > > Especially for HW > > > filters which > > > + // always have a limited frame pool size. > > > + if (!pkt) > > > + return 0; > > > > Looks good and it works well for me. I'm going to push it if no > > objection > > > > There is a conflict when applying your patch to the latest FFmpeg, > could you > rebase the patch ? Rebased with V2. Thanks Fei > > Thanks > Haihao > > _______________________________________________ > 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".
diff --git a/fftools/ffmpeg_dec.c b/fftools/ffmpeg_dec.c index 799be63215..d6b34402f8 100644 --- a/fftools/ffmpeg_dec.c +++ b/fftools/ffmpeg_dec.c @@ -531,6 +531,12 @@ int dec_packet(InputStream *ist, const AVPacket *pkt, int no_eof) av_frame_unref(frame); if (ret < 0) exit_program(1); + + // During flushing, break out to reap filter once send a frame to filters to + // avoid drain out filter's output frame pool. Especially for HW filters which + // always have a limited frame pool size. + if (!pkt) + return 0; } }
Filter may has a limited frame pool size. Do not always send frame to filters without reaping. Fix the regression of commit 5fa00b38e6. Example cmd: $ ffmpeg -threads 20 -init_hw_device vaapi=hw:/dev/dri/renderD128 \ -hwaccel_output_format vaapi -hwaccel vaapi -i avc_1080P.mp4 \ -vf 'scale_vaapi=w=720:h=480' -f null - Signed-off-by: Fei Wang <fei.w.wang@intel.com> --- fftools/ffmpeg_dec.c | 6 ++++++ 1 file changed, 6 insertions(+)