[FFmpeg-devel] [PATCH] avfilter/vf_convolution: add x86 SIMD for filter_3x3()

Song, Ruiling ruiling.song at intel.com
Wed Jul 31 10:59:39 EEST 2019


> -----Original Message-----
> From: ffmpeg-devel [mailto:ffmpeg-devel-bounces at ffmpeg.org] On Behalf
> Of Paul B Mahol
> Sent: Wednesday, July 17, 2019 8:42 PM
> To: FFmpeg development discussions and patches <ffmpeg-
> devel at ffmpeg.org>
> Subject: Re: [FFmpeg-devel] [PATCH] avfilter/vf_convolution: add x86 SIMD
> for filter_3x3()
> 
> On 7/15/19, Song, Ruiling <ruiling.song at intel.com> wrote:
> >> -----Original Message-----
> >> From: Song, Ruiling
> >> Sent: Tuesday, July 9, 2019 9:15 AM
> >> To: ffmpeg-devel at ffmpeg.org
> >> Cc: Song, Ruiling <ruiling.song at intel.com>
> >> Subject: [PATCH] avfilter/vf_convolution: add x86 SIMD for filter_3x3()
> >>
> >> Tested using a simple command (apply edge enhance):
> >> ./ffmpeg_g -i ~/Downloads/bbb_sunflower_1080p_30fps_normal.mp4 \
> >>  -vf convolution="0 0 0 -1 1 0 0 0 0:0 0 0 -1 1 0 0 0 0:0 0 0 -1 1 0 0 0
> >> 0:0 0 0 -1 1 0 0
> >> 0 0:5:1:1:1:0:128:128:128" \
> >>  -an -vframes 1000 -f null /dev/null
> >>
> >> The fps increase from 151 to 270 on my local machine.
> >>
> >> Signed-off-by: Ruiling Song <ruiling.song at intel.com>
> > Ping?
> 
> Should be fine IFF output is exact with C version (under different
> parameters).
Thanks Paul, after fixing a bug in scalar code path, the v2 produces exact result as C version.
Have tested against many different parameters. Will apply in a few days.

Thanks!
Ruiling

> _______________________________________________
> 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".


More information about the ffmpeg-devel mailing list