On 2017-01-15 22:55:50 +0200, Martin Storsjö wrote: > This adds lots of extra .ifs, but speeds it up by a couple cycles, > by avoiding stalls. > --- > libavcodec/arm/vp9lpf_neon.S | 8 ++++++-- > 1 file changed, 6 insertions(+), 2 deletions(-) > > diff --git a/libavcodec/arm/vp9lpf_neon.S b/libavcodec/arm/vp9lpf_neon.S > index 9be4cef..e31c807 100644 > --- a/libavcodec/arm/vp9lpf_neon.S > +++ b/libavcodec/arm/vp9lpf_neon.S > @@ -181,16 +181,20 @@ > > vmovl.u8 q0, d22 @ p1 > vmovl.u8 q1, d25 @ q1 > +.if \wd >= 8 > + vmov r2, r3, d6 > +.endif > vaddw.s8 q0, q0, \tmp3 @ p1 + f > vsubw.s8 q1, q1, \tmp3 @ q1 - f > +.if \wd >= 8 > + orrs r2, r2, r3 > +.endif > vqmovun.s16 d0, q0 @ out p1 > vqmovun.s16 d2, q1 @ out q1 > vbit d22, d0, d5 @ if (!hev && fm && !flat8in) > vbit d25, d2, d5 > > .if \wd >= 8 > - vmov r2, r3, d6 > - orrs r2, r2, r3 > @ If no pixels need flat8in, jump to flat8out > @ (or to a writeout of the inner 4 pixels, for wd=8) > beq 6f
ok Janne _______________________________________________ libav-devel mailing list [email protected] https://lists.libav.org/mailman/listinfo/libav-devel
