lea r0, [r3*3]
DIFF_PIXELS_8 r1, r2, 0, r3, r0, rsp+gprsize
HADAMARD8
lea r0, [r3*3]
DIFF_PIXELS_8 r1, r2, 0, r3, r0, rsp+gprsize
HADAMARD8
TRANSPOSE8x8W 0, 1, 2, 3, 4, 5, 6, 7, 8
%else
TRANSPOSE8x8W 0, 1, 2, 3, 4, 5, 6, 7, [rsp+gprsize], [rsp+mmsize+gprsize]
TRANSPOSE8x8W 0, 1, 2, 3, 4, 5, 6, 7, 8
%else
TRANSPOSE8x8W 0, 1, 2, 3, 4, 5, 6, 7, [rsp+gprsize], [rsp+mmsize+gprsize]