On 12/3/17, Paul B Mahol <one...@gmail.com> wrote: > On 12/3/17, Paul B Mahol <one...@gmail.com> wrote: >> Signed-off-by: Paul B Mahol <one...@gmail.com> >> --- >> libavfilter/hflip.h | 38 ++++++++++++ >> libavfilter/vf_hflip.c | 133 >> ++++++++++++++++++++++++++-------------- >> libavfilter/x86/Makefile | 2 + >> libavfilter/x86/vf_hflip.asm | 98 +++++++++++++++++++++++++++++ >> libavfilter/x86/vf_hflip_init.c | 41 +++++++++++++ >> 5 files changed, 265 insertions(+), 47 deletions(-) >> create mode 100644 libavfilter/hflip.h >> create mode 100644 libavfilter/x86/vf_hflip.asm >> create mode 100644 libavfilter/x86/vf_hflip_init.c >> > > This is overall ~50% faster than pure C that gcc 6.3.0 gives with > vanilla options. >
By overall I mean this simple bench test: ffmpeg -f lavfi -i smptehdbars=hd1080 -vf hflip=threads=1 -f null - _______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel