Converting a 128 integer into 80 bit floating point

Hi asm wizards!

I am adding 128 bit integer sup****t to my compiler system, and I need to 
convert an int128 into a long double (80 bits floating point)

I developed this macro for doing this. Do you see something wrong here?

I have tested it and seems to be working but I am not 100% sure.

	push	$64
	fildl	(%%rsp)
	leaq	some_address,%%rcx
;;
;; load higher part of the 128 bit number into the FPU
	fildt	8(%%rcx)
;; multiply by 2^64
	fscale
;;
;; load lower part of the 128 bit number
	fildt	(%%rcx)
	faddp
;; cleanup the FPU stack
	fstpl	%%st(1)
;; cleanup the stack
	addq	$8,%%rsp


Thanks in advance


-- 
jacob navia
jacob at jacob point remcomp point fr
logiciels/informatique
http://www.cs.virginia.edu/~lcc-win32