[FFmpeg-devel] [PATCH] vf_blend: Add SSE2 optimization for multiply

Timothy Gu timothygu99 at gmail.com
Mon Feb 8 22:36:08 CET 2016


On Mon, Feb 08, 2016 at 09:51:55PM +0100, Paul B Mahol wrote:
> On 2/8/16, Timothy Gu <timothygu99 at gmail.com> wrote:
> > 5 times faster than C, 3 times overall.
> > ---
> >
> > Removes constant.c changes in accordance with James's comments
> >
> > ---
> >  libavfilter/x86/vf_blend.asm    | 29 +++++++++++++++++++++++++++++
> >  libavfilter/x86/vf_blend_init.c |  2 ++
> >  2 files changed, 31 insertions(+)
> >
> 
> If output is always bitexact with C, should be OK for me.

Yes.

Pushed, thanks.

Timothy


More information about the ffmpeg-devel mailing list