Re: [FFmpeg-devel] [PATCH] x86/takdsp: use arithmetic shift instructions
On 10/9/2015 11:48 PM, Ronald S. Bultje wrote: > Hi, > > On Fri, Oct 9, 2015 at 10:42 PM, James Almer wrote: > >> p1 and p2 are int32_t. >> >> Signed-off-by: James Almer >> --- >> libavcodec/x86/takdsp.asm | 8 >> 1 file changed, 4 insertions(+), 4 deletions(-) > > > ok. > > Ronald Pushed, thanks. ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] x86/takdsp: use arithmetic shift instructions
Hi, On Fri, Oct 9, 2015 at 10:42 PM, James Almer wrote: > p1 and p2 are int32_t. > > Signed-off-by: James Almer > --- > libavcodec/x86/takdsp.asm | 8 > 1 file changed, 4 insertions(+), 4 deletions(-) ok. Ronald ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
[FFmpeg-devel] [PATCH] x86/takdsp: use arithmetic shift instructions
p1 and p2 are int32_t. Signed-off-by: James Almer --- libavcodec/x86/takdsp.asm | 8 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/libavcodec/x86/takdsp.asm b/libavcodec/x86/takdsp.asm index d3c6a13..5f3ded3 100644 --- a/libavcodec/x86/takdsp.asm +++ b/libavcodec/x86/takdsp.asm @@ -75,8 +75,8 @@ cglobal tak_decorrelate_sm, 3, 3, 6, p1, p2, length mova m4, [p2q+lengthq+mmsize] mova m2, m1 mova m5, m4 -psrldm2, 1 -psrldm5, 1 +psradm2, 1 +psradm5, 1 psubdm0, m2 psubdm3, m5 padddm1, m0 @@ -104,10 +104,10 @@ cglobal tak_decorrelate_sf, 3, 3, 5, p1, p2, length, dshift, dfactor .loop: mova m0, [p1q+lengthq] mova m1, [p2q+lengthq] -psrldm1, m2 +psradm1, m2 pmulld m1, m3 padddm1, m4 -psrldm1, 8 +psradm1, 8 pslldm1, m2 psubdm1, m0 mova [p1q+lengthq], m1 -- 2.6.0 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
