RepoMirrors
/
ffmpeg
mirror of https://git.ffmpeg.org/ffmpeg.git
1
0
Fork 0
Code Issues Projects Releases Wiki Activity

swscale/x86/yuv2yuvX: use movq to load 8 bytes in all non-AVX2 functions 

mova expands to movq on non-XMM functions

Signed-off-by: James Almer <jamrial@gmail.com>
This commit is contained in:
James Almer 2021-02-18 12:09:11 -03:00
parent d512ebbaed
commit ebb48d85a0
1 changed files with 1 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
libswscale/x86/yuv2yuvX.asm
View File

@ -47,10 +47,8 @@ cglobal yuv2yuvX, 7, 7, 8, filter, filterSize, src, dest, dstW, dither, offset
%endif ; x86-64 %endif ; x86-64
%if cpuflag(avx2) %if cpuflag(avx2)
vpbroadcastq m3, [ditherq] vpbroadcastq m3, [ditherq]
%elif cpuflag(sse3)
movq xmm3, [ditherq]
%else %else
mova m3, [ditherq] movq xm3, [ditherq]
%endif ; avx2 %endif ; avx2
cmp offsetd, 0 cmp offsetd, 0
jz .offset jz .offset