+ sub wq, mmsize - 1
+ jz .end
+ add srcq , 2*mmsize - 2
+ add dstUq, mmsize - 1
+ add dstVq, mmsize - 1
+.loop2:
+ movd m0, [srcq+wq*2] ; (byte) { Bx, Gx, Rx, xx }[0-3]
+ DEINTB 1, 0, 5, 4, 7 ; (word) { Gx, xx (m0/m4) or Bx, Rx (m1/m5) }[0-3]/[4-7]
+ pmaddwd m3, m1, coeffV1 ; (dword) { Bx*BV + Rx*RV }[0-3]
+ pmaddwd m2, m0, coeffV2 ; (dword) { Gx*GV }[0-3]
+ pmaddwd m1, coeffU1 ; (dword) { Bx*BU + Rx*RU }[0-3]
+ pmaddwd m0, coeffU2 ; (dword) { Gx*GU }[0-3]
+ paddd m3, m6 ; += rgb_UVrnd
+ paddd m1, m6 ; += rgb_UVrnd
+ paddd m2, m3 ; (dword) { V[0-3] }
+ paddd m0, m1 ; (dword) { U[0-3] }
+ psrad m0, 9
+ psrad m2, 9
+ packssdw m0, m0 ; (word) { U[0-7] }
+ packssdw m2, m2 ; (word) { V[0-7] }
+ movd [dstUq+wq], m0
+ movd [dstVq+wq], m2
+ add wq, 2
+ jl .loop2
+.end: