$n <<= 1 while $n & 0xfff;

##</code><code>##

    for( c = 1; ( ~n & 1 ) && ( c < 4096 ); c <<= 1 ) n >>= 1;
    lcm = n * 4096;

##</code><code>##

    while( n & 0xfff ) n <<= 1;
    lcm = n;

##</code><code>##

C:\test\C>gcm
gcm : 2132901888
gcm2: 2132901888
gcm3: 2132901888
anonyM: gcm  for s=2147483648 & r=1 to 1073741824 took:33.850023994460
anonyM: gcm2 for s=2147483648 & r=1 to 1073741824 took:46.293298113614
anonyM: gcm3 for s=2147483648 & r=1 to 1073741824 took:64.208097030422

##</code><code>##

U64 gcm2( U64 max, U64 n ) {
    U32 b;
    U64 c, lcm;

    _BitScanForward64( &b, n );
    n >>= min( b, 12 );
    lcm = n * 4096;
    return ( max / lcm ) * lcm;
}

##</code><code>##

PUBLIC	gcm2
; Function compile flags: /Ogtpy
_TEXT	SEGMENT
max$ = 8
n$ = 16
gcm2	PROC

; 24   :     U32 b;
; 25   :     U64 c, lcm;
; 26   : 
; 27   :     _BitScanForward64( &b, n );

	bsf	rax, rdx
	mov	r8, rcx
	mov	r9, rdx

; 28   :     n >>= min( b, 12 );

	mov	ecx, 12
	cmp	eax, ecx
	cmovb	ecx, eax

; 29   :     lcm = n * 4096;
; 30   :     return ( max / lcm ) * lcm;

	xor	edx, edx
	mov	rax, r8
	shr	r9, cl
	shl	r9, 12
	div	r9
	imul	rax, r9

; 31   : }

	ret	0
gcm2	ENDP

##</code><code>##

PUBLIC	gcm
; Function compile flags: /Ogtpy
_TEXT	SEGMENT
max$ = 8
n$ = 16
gcm	PROC

; 16   :     U64 c, lcm;
; 17   : 
; 18   :     for( c = 1; ( ~n & 1 ) && ( c < 4096 ); c <<= 1 ) n >>= 1;

	movzx	eax, dl
	mov	r8d, 1
	mov	r9, rdx
	not	al
	test	al, r8b
	je	SHORT $LN1@gcm
$LL3@gcm:
	cmp	r8, 4096				; 00001000H
	jae	SHORT $LN1@gcm
	shr	r9, 1
	add	r8, r8
	movzx	eax, r9b
	not	al
	test	al, 1
	jne	SHORT $LL3@gcm
$LN1@gcm:

; 19   :     lcm = n * 4096;

	shl	r9, 12

; 20   :     return ( max / lcm ) * lcm;

	xor	edx, edx
	mov	rax, rcx
	div	r9
	imul	rax, r9

; 21   : }

	ret	0
gcm	ENDP

##</code><code>##

C:\test\C>gcm
anonyM: gcm  for s=2147483648 & r=1 to 1073741824 took: 33.920637491715 
anonyM: gcm2 for s=2147483648 & r=1 to 1073741824 took: 46.151947659089 
oiskuu: gcm  for s=2147483648 & r=1 to 1073741824 took:330.492011773110