get division remainder of two 64-bit real numbers

; this code gets division remainder of two 64-bit real numbers (DP-FP) using SSE2 instruction set
; it's very slow I know
movsd xmm0, vl      ; copy left operand to xmm0
movsd xmm1, vr      ; copy right operand to xmm1
movsd xmm2, 0x7FFFFFFFFFFFFFFF
movsd xmm3, 0x8000000000000000
andpd xmm3, xmm0    ; remember the sign of left operand
andpd xmm0, xmm2    ; get abs value of left operand
andpd xmm1, xmm2    ; get abs value of right operand
start:
comisd xmm0, xmm1
jc end
subsd xmm0, xmm1
jmp start
orpd xmm0, xmm3     ; restore the sign of left operand
end:
mov [rdx], xmm0
ret