$NetBSD: patch-ac,v 1.2 2004/01/16 00:59:18 kristerw Exp $

--- src/runtime/Integer/asmlong.h.orig	2003-02-27 10:19:30.000000000 +0000
+++ src/runtime/Integer/asmlong.h	2003-12-16 16:33:42.000000000 +0000
@@ -1,13 +1,13 @@
 #if defined (__a29k__) || defined (___AM29K__)
 #define add_ssaaaa(sh, sl, ah, al, bh, bl) \
-  __asm__ ("add %1,%4,%5
-	addc %0,%2,%3"							\
+  __asm__ ("add %1,%4,%5\n"	\
+"	addc %0,%2,%3"		\
 	 : "=r" ((unsigned long int)(sh)), "=&r" ((unsigned long int)(sl)) \
 	 : "%r" ((unsigned long int)(ah)), "rI" ((unsigned long int)(bh)), \
 	   "%r" ((unsigned long int)(al)), "rI" ((unsigned long int)(bl)))
 #define sub_ddmmss(sh, sl, ah, al, bh, bl) \
-  __asm__ ("sub %1,%4,%5
-	subc %0,%2,%3"							\
+  __asm__ ("sub %1,%4,%5\n"	\
+"	subc %0,%2,%3"		\
 	 : "=r" ((unsigned long int)(sh)), "=&r" ((unsigned long int)(sl)) \
 	 : "r" ((unsigned long int)(ah)), "rI" ((unsigned long int)(bh)), \
 	   "r" ((unsigned long int)(al)), "rI" ((unsigned long int)(bl)))
@@ -31,14 +31,14 @@
 
 #if defined (__arm__)
 #define add_ssaaaa(sh, sl, ah, al, bh, bl) \
-  __asm__ ("adds %1,%4,%5
-	adc %0,%2,%3"							\
+  __asm__ ("adds %1,%4,%5\n"	\
+"	adc %0,%2,%3"		\
 	: "=r" ((unsigned long int)(sh)), "=&r" ((unsigned long int)(sl)) \
 	: "%r" ((unsigned long int)(ah)), "rI" ((unsigned long int)(bh)), \
 	  "%r" ((unsigned long int)(al)), "rI" ((unsigned long int)(bl)))
 #define sub_ddmmss(sh, sl, ah, al, bh, bl) \
-  __asm__ ("subs %1,%4,%5
-	sbc %0,%2,%3"							\
+  __asm__ ("subs %1,%4,%5\n"	\
+"	sbc %0,%2,%3"		\
 	: "=r" ((unsigned long int)(sh)), "=&r" ((unsigned long int)(sl)) \
 	: "r" ((unsigned long int)(ah)), "rI" ((unsigned long int)(bh)), \
 	  "r" ((unsigned long int)(al)), "rI" ((unsigned long int)(bl)))
@@ -46,14 +46,14 @@
 
 #if defined (__gmicro__)
 #define add_ssaaaa(sh, sl, ah, al, bh, bl) \
-  __asm__ ("add.w %5,%1
-	addx %3,%0"							\
+  __asm__ ("add.w %5,%1\n"	\
+"	addx %3,%0"		\
        : "=g" ((unsigned long int)(sh)), "=&g" ((unsigned long int)(sl))\
        : "%0" ((unsigned long int)(ah)), "g" ((unsigned long int)(bh)),	\
 	 "%1" ((unsigned long int)(al)), "g" ((unsigned long int)(bl)))
 #define sub_ddmmss(sh, sl, ah, al, bh, bl) \
-  __asm__ ("sub.w %5,%1
-	subx %3,%0"							\
+  __asm__ ("sub.w %5,%1\n"	\
+"	subx %3,%0"		\
        : "=g" ((unsigned long int)(sh)), "=&g" ((unsigned long int)(sl))\
        : "0" ((unsigned long int)(ah)), "g" ((unsigned long int)(bh)),	\
 	 "1" ((unsigned long int)(al)), "g" ((unsigned long int)(bl)))
@@ -74,14 +74,14 @@
 
 #if defined (__hppa)
 #define add_ssaaaa(sh, sl, ah, al, bh, bl) \
-  __asm__ ("add %4,%5,%1
-        addc %2,%3,%0"				\
+  __asm__ ("add %4,%5,%1\n"	\
+"        addc %2,%3,%0"		\
 	: "=r" ((unsigned long int)(sh)), "=&r" ((unsigned long int)(sl))\
 	: "%r" ((unsigned long int)(ah)), "r" ((unsigned long int)(bh)),\
 	  "%r" ((unsigned long int)(al)), "r" ((unsigned long int)(bl)))
 #define sub_ddmmss(sh, sl, ah, al, bh, bl) \
-  __asm__ ("sub %5,%4,%1
-        subb %3,%2,%0"				\
+  __asm__ ("sub %5,%4,%1\n"	\
+"        subb %3,%2,%0"		\
 	: "=r" ((unsigned long int)(sh)), "=&r" ((unsigned long int)(sl))\
 	: "r" ((unsigned long int)(ah)), "r" ((unsigned long int)(bh)),	\
 	  "r" ((unsigned long int)(al)), "r" ((unsigned long int)(bl)))
@@ -89,14 +89,14 @@
 
 #if defined (__i386__) || defined (__i486__)
 #define add_ssaaaa(sh, sl, ah, al, bh, bl) \
-  __asm__ ("addl %5,%1
-	adcl %3,%0"							\
+  __asm__ ("addl %5,%1\n"	\
+"	adcl %3,%0"		\
        : "=r" ((unsigned long int)(sh)), "=&r" ((unsigned long int)(sl))\
        : "%0" ((unsigned long int)(ah)), "g" ((unsigned long int)(bh)),	\
 	 "%1" ((unsigned long int)(al)), "g" ((unsigned long int)(bl)))
 #define sub_ddmmss(sh, sl, ah, al, bh, bl) \
-  __asm__ ("subl %5,%1
-	sbbl %3,%0"							\
+  __asm__ ("subl %5,%1\n"	\
+"	sbbl %3,%0"		\
        : "=r" ((unsigned long int)(sh)), "=&r" ((unsigned long int)(sl))\
        : "0" ((unsigned long int)(ah)), "g" ((unsigned long int)(bh)),	\
 	 "1" ((unsigned long int)(al)), "g" ((unsigned long int)(bl)))
@@ -157,14 +157,14 @@
 
 #if defined (___IBMR2__) /* IBM RS6000 */
 #define add_ssaaaa(sh, sl, ah, al, bh, bl) \
-  __asm__ ("a %1,%4,%5
-	ae %0,%2,%3"							\
+  __asm__ ("a %1,%4,%5\n"	\
+"	ae %0,%2,%3"		\
 	: "=r" ((unsigned long int)(sh)), "=&r" ((unsigned long int)(sl))\
 	: "%r" ((unsigned long int)(ah)), "r" ((unsigned long int)(bh)),\
 	  "%r" ((unsigned long int)(al)), "r" ((unsigned long int)(bl)))
 #define sub_ddmmss(sh, sl, ah, al, bh, bl) \
-  __asm__ ("sf %1,%5,%4
-	sfe %0,%3,%2"							\
+  __asm__ ("sf %1,%5,%4\n"	\
+"	sfe %0,%3,%2"		\
 	: "=r" ((unsigned long int)(sh)), "=&r" ((unsigned long int)(sl))\
 	: "r" ((unsigned long int)(ah)), "r" ((unsigned long int)(bh)),	\
 	  "r" ((unsigned long int)(al)), "r" ((unsigned long int)(bl)))
@@ -264,32 +264,32 @@
 	: "od" ((unsigned long int)(x)), "n" (0))
 #else /* not mc68020 */
 #define umul_ppmm(xh, xl, a, b) \
-  __asm__ ("| Inlined umul_ppmm
-	movel	%2,d0
-	movel	%3,d1
-	movel	d0,d2
-	swap	d0
-	movel	d1,d3
-	swap	d1
-	movew	d2,d4
-	mulu	d3,d4
-	mulu	d1,d2
-	mulu	d0,d3
-	mulu	d0,d1
-	movel	d4,d0
-	eorw	d0,d0
-	swap	d0
-	addl	d0,d2
-	addl	d3,d2
-	jcc	1f
-	addl	#65536,d1
-1:	swap	d2
-	moveq	#0,d0
-	movew	d2,d0
-	movew	d4,d2
-	movel	d2,%1
-	addl	d1,d0
-	movel	d0,%0"							\
+  __asm__ ("| Inlined umul_ppmm\n"	\
+"	movel	%2,d0\n"	\
+"	movel	%3,d1\n"	\
+"	movel	d0,d2\n"	\
+"	swap	d0\n"		\
+"	movel	d1,d3\n"	\
+"	swap	d1\n"		\
+"	movew	d2,d4\n"	\
+"	mulu	d3,d4\n"	\
+"	mulu	d1,d2\n"	\
+"	mulu	d0,d3\n"	\
+"	mulu	d0,d1\n"	\
+"	movel	d4,d0\n"	\
+"	eorw	d0,d0\n"	\
+"	swap	d0\n"		\
+"	addl	d0,d2\n"	\
+"	addl	d3,d2\n"	\
+"	jcc	1f\n"		\
+"	addl	#65536,d1\n"	\
+"1:	swap	d2\n"		\
+"	moveq	#0,d0\n"	\
+"	movew	d2,d0\n"	\
+"	movew	d4,d2\n"	\
+"	movel	d2,%1\n"	\
+"	addl	d1,d0\n"	\
+"	movel	d0,%0"		\
        : "=g" ((unsigned long int)(xh)), "=g" ((unsigned long int)(xl))	\
        :"g" ((unsigned long int)(a)), "g" ((unsigned long int)(b))	\
        : "d0", "d1", "d2", "d3", "d4")
@@ -298,14 +298,12 @@
 
 #if defined (__m88000__)
 #define add_ssaaaa(sh, sl, ah, al, bh, bl) \
-  __asm__ ("addu.co %1,%r4,%r5
-	addu.ci %0,%r2,%r3"						\
+  __asm__ ("addu.co %1,%r4,%r5\n\taddu.ci %0,%r2,%r3"			\
 	: "=r" ((unsigned long int)(sh)), "=&r" ((unsigned long int)(sl))\
 	: "%rJ" ((unsigned long int)(ah)), "rJ" ((unsigned long int)(bh)),\
 	  "%rJ" ((unsigned long int)(al)), "rJ" ((unsigned long int)(bl)))
 #define sub_ddmmss(sh, sl, ah, al, bh, bl) \
-  __asm__ ("subu.co %1,%r4,%r5
-	subu.ci %0,%r2,%r3"						\
+  __asm__ ("subu.co %1,%r4,%r5\n\t subu.ci %0,%r2,%r3"			\
 	: "=r" ((unsigned long int)(sh)), "=&r" ((unsigned long int)(sl))\
 	: "rJ" ((unsigned long int)(ah)), "rJ" ((unsigned long int)(bh)),\
 	  "rJ" ((unsigned long int)(al)), "rJ" ((unsigned long int)(bl)))
@@ -322,9 +320,9 @@
 
 #if defined (__mips__)
 #define umul_ppmm(w1, w0, u, v) \
-  __asm__ ("multu %2,%3
-	mflo %0
-	mfhi %1"							\
+  __asm__ ("multu %2,%3\n"	\
+"	mflo %0\n"		\
+"	mfhi %1"		\
 	: "=r" ((unsigned long int)(w0)), "=r" ((unsigned long int)(w1))\
 	: "r" ((unsigned long int)(u)), "r" ((unsigned long int)(v)))
 #define UMUL_TIME 5
@@ -338,11 +336,11 @@
 	: "%0" ((unsigned long int)(u)), "g" ((unsigned long int)(v)));	\
       __w; })
 #define div_qrnnd(q, r, n1, n0, d) \
-  __asm__ ("movd %2,r0
-	movd %3,r1
-	deid %4,r0
-	movd r1,%0
-	movd r0,%1"							\
+  __asm__ ("movd %2,r0\n"	\
+"	movd %3,r1\n"	\
+"	deid %4,r0\n"	\
+"	movd r1,%0\n"	\
+"	movd r0,%1"	\
 	: "=g" ((unsigned long int)(q)), "=g" ((unsigned long int)(r))	\
 	: "g" ((unsigned long int)(n0)), "g" ((unsigned long int)(n1)),	\
 	  "g" ((unsigned long int)(d)) : "r0", "r1")
@@ -350,23 +348,23 @@
 
 #if defined (__pyr__)
 #define add_ssaaaa(sh, sl, ah, al, bh, bl) \
-  __asm__ ("addw	%5,%1
-	addwc	%3,%0"							\
+  __asm__ ("addw	%5,%1\n"	\
+"	addwc	%3,%0"							\
 	: "=r" ((unsigned long int)(sh)), "=&r" ((unsigned long int)(sl))\
 	: "%0" ((unsigned long int)(ah)), "g" ((unsigned long int)(bh)),\
 	  "%1" ((unsigned long int)(al)), "g" ((unsigned long int)(bl)))
 #define sub_ddmmss(sh, sl, ah, al, bh, bl) \
-  __asm__ ("subw	%5,%1
-	subwb	%3,%0"							\
+  __asm__ ("subw	%5,%1\n"	\
+"	subwb	%3,%0"							\
 	: "=r" ((unsigned long int)(sh)), "=&r" ((unsigned long int)(sl))\
 	: "0" ((unsigned long int)(ah)), "g" ((unsigned long int)(bh)),	\
 	  "1" ((unsigned long int)(al)), "g" ((unsigned long int)(bl)))
 /* This insn doesn't work on ancient pyramids.  */
 #define umul_ppmm(w1, w0, u, v) \
-  __asm__ ("movw %2,tr11
-	uemul %3,tr10
-	movw tr10,%0
-	movw tr11,%1"							\
+  __asm__ ("movw %2,tr11\n"	\
+"	uemul %3,tr10\n"	\
+"	movw tr10,%0\n"		\
+"	movw tr11,%1"							\
 	: "=r" ((unsigned long int)(w1)), "=r" ((unsigned long int)(w0))\
 	: "r" ((unsigned long int)(u)), "r" ((unsigned long int)(v))	\
 	: "tr10", "tr11")
@@ -374,14 +372,14 @@
 
 #if defined (__ibm032__) /* RT/ROMP */
 #define add_ssaaaa(sh, sl, ah, al, bh, bl) \
-  __asm__ ("a %1,%5
-	ae %0,%3"							\
+  __asm__ ("a %1,%5\n"	\
+"	ae %0,%3"							\
 	: "=r" ((unsigned long int)(sh)), "=&r" ((unsigned long int)(sl))\
 	: "%0" ((unsigned long int)(ah)), "r" ((unsigned long int)(bh)),\
 	  "%1" ((unsigned long int)(al)), "r" ((unsigned long int)(bl)))
 #define sub_ddmmss(sh, sl, ah, al, bh, bl) \
-  __asm__ ("s %1,%5
-	se %0,%3"							\
+  __asm__ ("s %1,%5\n"	\
+"	se %0,%3"							\
 	: "=r" ((unsigned long int)(sh)), "=&r" ((unsigned long int)(sl))\
 	: "0" ((unsigned long int)(ah)), "r" ((unsigned long int)(bh)),	\
 	  "1" ((unsigned long int)(al)), "r" ((unsigned long int)(bl)))
@@ -389,26 +387,26 @@
   do {									\
     unsigned long int __m0 = (m0), __m1 = (m1);				\
     __asm__ (								\
-       "s	r2,r2
-	mts	r10,%2
-	m	r2,%3
-	m	r2,%3
-	m	r2,%3
-	m	r2,%3
-	m	r2,%3
-	m	r2,%3
-	m	r2,%3
-	m	r2,%3
-	m	r2,%3
-	m	r2,%3
-	m	r2,%3
-	m	r2,%3
-	m	r2,%3
-	m	r2,%3
-	m	r2,%3
-	m	r2,%3
-	cas	%0,r2,r0
-	mfs	r10,%1"							\
+       "s	r2,r2\n"	\
+"	mts	r10,%2\n"	\
+"	m	r2,%3\n"	\
+"	m	r2,%3\n"	\
+"	m	r2,%3\n"	\
+"	m	r2,%3\n"	\
+"	m	r2,%3\n"	\
+"	m	r2,%3\n"	\
+"	m	r2,%3\n"	\
+"	m	r2,%3\n"	\
+"	m	r2,%3\n"	\
+"	m	r2,%3\n"	\
+"	m	r2,%3\n"	\
+"	m	r2,%3\n"	\
+"	m	r2,%3\n"	\
+"	m	r2,%3\n"	\
+"	m	r2,%3\n"	\
+"	m	r2,%3\n"	\
+"	cas	%0,r2,r0\n"	\
+"	mfs	r10,%1"							\
        : "=r" ((unsigned long int)(ph)), "=r" ((unsigned long int)(pl))	\
        : "%r" (__m0), "r" (__m1)					\
        : "r2");								\
@@ -433,14 +431,14 @@
 
 #if defined (__sparc__)
 #define add_ssaaaa(sh, sl, ah, al, bh, bl) \
-  __asm__ ("addcc %4,%5,%1
-	addx %2,%3,%0"							\
+  __asm__ ("addcc %4,%5,%1\n"	\
+"	addx %2,%3,%0"		\
 	: "=r" ((unsigned long int)(sh)), "=&r" ((unsigned long int)(sl))\
 	: "%r" ((unsigned long int)(ah)), "rI" ((unsigned long int)(bh)),\
 	  "%r" ((unsigned long int)(al)), "rI" ((unsigned long int)(bl)))
 #define sub_ddmmss(sh, sl, ah, al, bh, bl) \
-  __asm__ ("subcc %4,%5,%1
-	subx %2,%3,%0"							\
+  __asm__ ("subcc %4,%5,%1\n"	\
+"	subx %2,%3,%0"		\
 	: "=r" ((unsigned long int)(sh)), "=&r" ((unsigned long int)(sl))\
 	: "r" ((unsigned long int)(ah)), "rI" ((unsigned long int)(bh)),\
 	  "r" ((unsigned long int)(al)), "rI" ((unsigned long int)(bl)))
@@ -458,46 +456,46 @@
 /* SPARC without integer multiplication and divide instructions.
    (i.e. at least Sun4/20,40,60,65,75,110,260,280,330,360,380,470,490) */
 #define umul_ppmm(w1, w0, u, v) \
-  __asm__ ("! Inlined umul_ppmm
-	wr	%%g0,%2,%%y	! SPARC has 0-3 delay insn after a wr
-	sra	%3,31,%%g2	! Don't move this insn
-	and	%2,%%g2,%%g2	! Don't move this insn
-	andcc	%%g0,0,%%g1	! Don't move this insn
-	mulscc	%%g1,%3,%%g1
-	mulscc	%%g1,%3,%%g1
-	mulscc	%%g1,%3,%%g1
-	mulscc	%%g1,%3,%%g1
-	mulscc	%%g1,%3,%%g1
-	mulscc	%%g1,%3,%%g1
-	mulscc	%%g1,%3,%%g1
-	mulscc	%%g1,%3,%%g1
-	mulscc	%%g1,%3,%%g1
-	mulscc	%%g1,%3,%%g1
-	mulscc	%%g1,%3,%%g1
-	mulscc	%%g1,%3,%%g1
-	mulscc	%%g1,%3,%%g1
-	mulscc	%%g1,%3,%%g1
-	mulscc	%%g1,%3,%%g1
-	mulscc	%%g1,%3,%%g1
-	mulscc	%%g1,%3,%%g1
-	mulscc	%%g1,%3,%%g1
-	mulscc	%%g1,%3,%%g1
-	mulscc	%%g1,%3,%%g1
-	mulscc	%%g1,%3,%%g1
-	mulscc	%%g1,%3,%%g1
-	mulscc	%%g1,%3,%%g1
-	mulscc	%%g1,%3,%%g1
-	mulscc	%%g1,%3,%%g1
-	mulscc	%%g1,%3,%%g1
-	mulscc	%%g1,%3,%%g1
-	mulscc	%%g1,%3,%%g1
-	mulscc	%%g1,%3,%%g1
-	mulscc	%%g1,%3,%%g1
-	mulscc	%%g1,%3,%%g1
-	mulscc	%%g1,%3,%%g1
-	mulscc	%%g1,0,%%g1
-	add	%%g1,%%g2,%0
-	rd	%%y,%1"							\
+  __asm__ ("! Inlined umul_ppmm\n"	\
+"	wr	%%g0,%2,%%y	! SPARC has 0-3 delay insn after a wr\n" \
+"	sra	%3,31,%%g2	! Don't move this insn\n"	\
+"	and	%2,%%g2,%%g2	! Don't move this insn\n"	\
+"	andcc	%%g0,0,%%g1	! Don't move this insn\n"	\
+"	mulscc	%%g1,%3,%%g1\n"	\
+"	mulscc	%%g1,%3,%%g1\n"	\
+"	mulscc	%%g1,%3,%%g1\n"	\
+"	mulscc	%%g1,%3,%%g1\n"	\
+"	mulscc	%%g1,%3,%%g1\n"	\
+"	mulscc	%%g1,%3,%%g1\n"	\
+"	mulscc	%%g1,%3,%%g1\n"	\
+"	mulscc	%%g1,%3,%%g1\n"	\
+"	mulscc	%%g1,%3,%%g1\n"	\
+"	mulscc	%%g1,%3,%%g1\n"	\
+"	mulscc	%%g1,%3,%%g1\n"	\
+"	mulscc	%%g1,%3,%%g1\n"	\
+"	mulscc	%%g1,%3,%%g1\n"	\
+"	mulscc	%%g1,%3,%%g1\n"	\
+"	mulscc	%%g1,%3,%%g1\n"	\
+"	mulscc	%%g1,%3,%%g1\n"	\
+"	mulscc	%%g1,%3,%%g1\n"	\
+"	mulscc	%%g1,%3,%%g1\n"	\
+"	mulscc	%%g1,%3,%%g1\n"	\
+"	mulscc	%%g1,%3,%%g1\n"	\
+"	mulscc	%%g1,%3,%%g1\n"	\
+"	mulscc	%%g1,%3,%%g1\n"	\
+"	mulscc	%%g1,%3,%%g1\n"	\
+"	mulscc	%%g1,%3,%%g1\n"	\
+"	mulscc	%%g1,%3,%%g1\n"	\
+"	mulscc	%%g1,%3,%%g1\n"	\
+"	mulscc	%%g1,%3,%%g1\n"	\
+"	mulscc	%%g1,%3,%%g1\n"	\
+"	mulscc	%%g1,%3,%%g1\n"	\
+"	mulscc	%%g1,%3,%%g1\n"	\
+"	mulscc	%%g1,%3,%%g1\n"	\
+"	mulscc	%%g1,%3,%%g1\n"	\
+"	mulscc	%%g1,0,%%g1\n"	\
+"	add	%%g1,%%g2,%0\n"	\
+"	rd	%%y,%1"							\
 	: "=r" ((unsigned long int)(w1)), "=r" ((unsigned long int)(w0))\
 	: "%rI" ((unsigned long int)(u)), "r" ((unsigned long int)(v))	\
        : "%g1", "%g2")
@@ -505,30 +503,30 @@
 /* It's quite necessary to add this much assembler for the sparc.
    The default udiv_qrnnd (in C) is more than 10 times slower!  */
 #define udiv_qrnnd(q, r, n1, n0, d) \
-  __asm__ ("! Inlined udiv_qrnnd
-	mov	32,%%g1
-	subcc	%1,%2,%%g0
-1:	bcs	5f
-	 addxcc %0,%0,%0	! shift n1n0 and a q-bit in lsb
-	sub	%1,%2,%1	! this kills msb of n
-	addx	%1,%1,%1	! so this can't give carry
-	subcc	%%g1,1,%%g1
-2:	bne	1b
-	 subcc	%1,%2,%%g0
-	bcs	3f
-	 addxcc %0,%0,%0	! shift n1n0 and a q-bit in lsb
-	b	3f
-	 sub	%1,%2,%1	! this kills msb of n
-4:	sub	%1,%2,%1
-5:	addxcc	%1,%1,%1
-	bcc	2b
-	 subcc	%%g1,1,%%g1
-! Got carry from n.  Subtract next step to cancel this carry.
-	bne	4b
-	 addcc	%0,%0,%0	! shift n1n0 and a 0-bit in lsb
-	sub	%1,%2,%1
-3:	xnor	%0,0,%0
-	! End of inline udiv_qrnnd"					\
+  __asm__ ("! Inlined udiv_qrnnd\n"					\
+"	mov	32,%%g1\n"						\
+"	subcc	%1,%2,%%g0\n"						\
+"1:	bcs	5f\n"							\
+"	 addxcc %0,%0,%0	! shift n1n0 and a q-bit in lsb\n"	\
+"	sub	%1,%2,%1	! this kills msb of n\n"		\
+"	addx	%1,%1,%1	! so this can't give carry\n"		\
+"	subcc	%%g1,1,%%g1\n"						\
+"2:	bne	1b\n"							\
+"	 subcc	%1,%2,%%g0\n"						\
+"	bcs	3f\n"							\
+"	 addxcc %0,%0,%0	! shift n1n0 and a q-bit in lsb\n"	\
+"	b	3f\n"							\
+"	 sub	%1,%2,%1	! this kills msb of n\n"		\
+"4:	sub	%1,%2,%1\n"						\
+"5:	addxcc	%1,%1,%1\n"						\
+"	bcc	2b\n"							\
+"	 subcc	%%g1,1,%%g1\n"						\
+"! Got carry from n.  Subtract next step to cancel this carry.\n"	\
+"	bne	4b\n"							\
+"	 addcc	%0,%0,%0	! shift n1n0 and a 0-bit in lsb\n"	\
+"	sub	%1,%2,%1\n"						\
+"3:	xnor	%0,0,%0\n"						\
+"	! End of inline udiv_qrnnd"					\
 	: "=r&" ((unsigned long int)(q)), "=r&" ((unsigned long int)(r))\
 	: "r" ((unsigned long int)(d)), "1" ((unsigned long int)(n1)),	\
 	  "0" ((unsigned long int)(n0)) : "%g1")
@@ -538,14 +536,14 @@
 
 #if defined (__vax__)
 #define add_ssaaaa(sh, sl, ah, al, bh, bl) \
-  __asm__ ("addl2 %5,%1
-	adwc %3,%0"							\
+  __asm__ ("addl2 %5,%1\n"	\
+"	adwc %3,%0"							\
 	: "=g" ((unsigned long int)(sh)), "=&g" ((unsigned long int)(sl))\
 	: "%0" ((unsigned long int)(ah)), "g" ((unsigned long int)(bh)),\
 	  "%1" ((unsigned long int)(al)), "g" ((unsigned long int)(bl)))
 #define sub_ddmmss(sh, sl, ah, al, bh, bl) \
-  __asm__ ("subl2 %5,%1
-	sbwc %3,%0"							\
+  __asm__ ("subl2 %5,%1\n"	\
+"	sbwc %3,%0"							\
 	: "=g" ((unsigned long int)(sh)), "=&g" ((unsigned long int)(sl))\
 	: "0" ((unsigned long int)(ah)), "g" ((unsigned long int)(bh)),	\
 	  "1" ((unsigned long int)(al)), "g" ((unsigned long int)(bl)))