From 41b0afab2902115022c4327d3c59b3c2ecfb4b45 Mon Sep 17 00:00:00 2001 From: Roland McGrath Date: Thu, 26 Sep 2002 22:32:54 +0000 Subject: 2002-09-26 Roland McGrath * stdlib/longlong.h: Replaced with current version from GCC mainline, last modified 2002-09-22 Kazu Hirata . --- stdlib/longlong.h | 988 ++++++++++++++++++++---------------------------------- 1 file changed, 370 insertions(+), 618 deletions(-) diff --git a/stdlib/longlong.h b/stdlib/longlong.h index ab2e9bd17d..476f36474f 100644 --- a/stdlib/longlong.h +++ b/stdlib/longlong.h @@ -1,5 +1,7 @@ /* longlong.h -- definitions for mixed size 32/64 bit arithmetic. - Copyright (C) 1991,92,94,95,96,97,98,99,2000,2001 Free Software Foundation, Inc. + Copyright (C) 1991, 1992, 1994, 1995, 1996, 1997, 1998, 1999, 2000 + Free Software Foundation, Inc. + This file is part of the GNU C Library. The GNU C Library is free software; you can redistribute it and/or @@ -66,7 +68,7 @@ is rounded towards 0. 5) count_leading_zeros(count, x) counts the number of zero-bits from the - msb to the first non-zero bit in the UWtype X. This is the number of + msb to the first nonzero bit in the UWtype X. This is the number of steps X needs to be shifted left to set the msb. Undefined for X == 0, unless the symbol COUNT_LEADING_ZEROS_0 is defined to some value. @@ -107,51 +109,6 @@ #define __AND_CLOBBER_CC , "cc" #endif /* __GNUC__ < 2 */ -#if (defined (__a29k__) || defined (_AM29K)) && W_TYPE_SIZE == 32 -#define add_ssaaaa(sh, sl, ah, al, bh, bl) \ - __asm__ ("add %1,%4,%5\n" \ - "addc %0,%2,%3" \ - : "=r" ((USItype) (sh)), \ - "=&r" ((USItype) (sl)) \ - : "%r" ((USItype) (ah)), \ - "rI" ((USItype) (bh)), \ - "%r" ((USItype) (al)), \ - "rI" ((USItype) (bl))) -#define sub_ddmmss(sh, sl, ah, al, bh, bl) \ - __asm__ ("sub %1,%4,%5\n" \ - "subc %0,%2,%3" \ - : "=r" ((USItype) (sh)), \ - "=&r" ((USItype) (sl)) \ - : "r" ((USItype) (ah)), \ - "rI" ((USItype) (bh)), \ - "r" ((USItype) (al)), \ - "rI" ((USItype) (bl))) -#define umul_ppmm(xh, xl, m0, m1) \ - do { \ - USItype __m0 = (m0), __m1 = (m1); \ - __asm__ ("multiplu %0,%1,%2" \ - : "=r" ((USItype) (xl)) \ - : "r" (__m0), \ - "r" (__m1)); \ - __asm__ ("multmu %0,%1,%2" \ - : "=r" ((USItype) (xh)) \ - : "r" (__m0), \ - "r" (__m1)); \ - } while (0) -#define udiv_qrnnd(q, r, n1, n0, d) \ - __asm__ ("dividu %0,%3,%4" \ - : "=r" ((USItype) (q)), \ - "=q" ((USItype) (r)) \ - : "1" ((USItype) (n1)), \ - "r" ((USItype) (n0)), \ - "r" ((USItype) (d))) -#define count_leading_zeros(count, x) \ - __asm__ ("clz %0,%1" \ - : "=r" ((USItype) (count)) \ - : "r" ((USItype) (x))) -#define COUNT_LEADING_ZEROS_0 32 -#endif /* __a29k__ */ - #if defined (__alpha) && W_TYPE_SIZE == 64 #define umul_ppmm(ph, pl, m0, m1) \ do { \ @@ -169,15 +126,47 @@ (q) = __udiv_qrnnd (&__r, (n1), (n0), (d)); \ (r) = __r; \ } while (0) -extern UDItype __udiv_qrnnd (UDItype *, UDItype, UDItype, UDItype); +extern UDItype __udiv_qrnnd PARAMS ((UDItype *, UDItype, UDItype, UDItype)); #define UDIV_TIME 220 #endif /* LONGLONG_STANDALONE */ +#ifdef __alpha_cix__ +#define count_leading_zeros(COUNT,X) \ + __asm__("ctlz %1,%0" : "=r"(COUNT) : "r"(X)) +#define count_trailing_zeros(COUNT,X) \ + __asm__("cttz %1,%0" : "=r"(COUNT) : "r"(X)) +#define COUNT_LEADING_ZEROS_0 64 +#else +extern const UQItype __clz_tab[]; +#define count_leading_zeros(COUNT,X) \ + do { \ + UDItype __xr = (X), __t, __a; \ + __asm__("cmpbge $31,%1,%0" : "=r"(__t) : "r"(__xr)); \ + __a = __clz_tab[__t ^ 0xff] - 1; \ + __asm__("extbl %1,%2,%0" : "=r"(__t) : "r"(__xr), "r"(__a)); \ + (COUNT) = 64 - (__clz_tab[__t] + __a*8); \ + } while (0) +#define count_trailing_zeros(COUNT,X) \ + do { \ + UDItype __xr = (X), __t, __a; \ + __asm__("cmpbge $31,%1,%0" : "=r"(__t) : "r"(__xr)); \ + __t = ~__t & -~__t; \ + __a = ((__t & 0xCC) != 0) * 2; \ + __a += ((__t & 0xF0) != 0) * 4; \ + __a += ((__t & 0xAA) != 0); \ + __asm__("extbl %1,%2,%0" : "=r"(__t) : "r"(__xr), "r"(__a)); \ + __a <<= 3; \ + __t &= -__t; \ + __a += ((__t & 0xCC) != 0) * 2; \ + __a += ((__t & 0xF0) != 0) * 4; \ + __a += ((__t & 0xAA) != 0); \ + (COUNT) = __a; \ + } while (0) +#endif /* __alpha_cix__ */ #endif /* __alpha */ #if defined (__arc__) && W_TYPE_SIZE == 32 #define add_ssaaaa(sh, sl, ah, al, bh, bl) \ - __asm__ ("add.f %1, %4, %5\n" \ - "adc %0, %2, %3" \ + __asm__ ("add.f %1, %4, %5\n\tadc %0, %2, %3" \ : "=r" ((USItype) (sh)), \ "=&r" ((USItype) (sl)) \ : "%r" ((USItype) (ah)), \ @@ -185,18 +174,17 @@ extern UDItype __udiv_qrnnd (UDItype *, UDItype, UDItype, UDItype); "%r" ((USItype) (al)), \ "rIJ" ((USItype) (bl))) #define sub_ddmmss(sh, sl, ah, al, bh, bl) \ - __asm__ ("sub.f %1, %4, %5\n" \ - "sbc %0, %2, %3" \ + __asm__ ("sub.f %1, %4, %5\n\tsbc %0, %2, %3" \ : "=r" ((USItype) (sh)), \ "=&r" ((USItype) (sl)) \ : "r" ((USItype) (ah)), \ "rIJ" ((USItype) (bh)), \ "r" ((USItype) (al)), \ "rIJ" ((USItype) (bl))) -/* Call libgcc1 routine. */ +/* Call libgcc routine. */ #define umul_ppmm(w1, w0, u, v) \ do { \ - DIunion __w; \ + DWunion __w; \ __w.ll = __umulsidi3 (u, v); \ w1 = __w.s.high; \ w0 = __w.s.low; \ @@ -207,8 +195,7 @@ UDItype __umulsidi3 (USItype, USItype); #if defined (__arm__) && W_TYPE_SIZE == 32 #define add_ssaaaa(sh, sl, ah, al, bh, bl) \ - __asm__ ("adds %1, %4, %5\n" \ - "adc %0, %2, %3" \ + __asm__ ("adds %1, %4, %5\n\tadc %0, %2, %3" \ : "=r" ((USItype) (sh)), \ "=&r" ((USItype) (sl)) \ : "%r" ((USItype) (ah)), \ @@ -216,8 +203,7 @@ UDItype __umulsidi3 (USItype, USItype); "%r" ((USItype) (al)), \ "rI" ((USItype) (bl))) #define sub_ddmmss(sh, sl, ah, al, bh, bl) \ - __asm__ ("subs %1, %4, %5\n" \ - "sbc %0, %2, %3" \ + __asm__ ("subs %1, %4, %5\n\tsbc %0, %2, %3" \ : "=r" ((USItype) (sh)), \ "=&r" ((USItype) (sl)) \ : "r" ((USItype) (ah)), \ @@ -227,18 +213,18 @@ UDItype __umulsidi3 (USItype, USItype); #define umul_ppmm(xh, xl, a, b) \ {register USItype __t0, __t1, __t2; \ __asm__ ("%@ Inlined umul_ppmm\n" \ - "mov %2, %5, lsr #16\n" \ - "mov %0, %6, lsr #16\n" \ - "bic %3, %5, %2, lsl #16\n" \ - "bic %4, %6, %0, lsl #16\n" \ - "mul %1, %3, %4\n" \ - "mul %4, %2, %4\n" \ - "mul %3, %0, %3\n" \ - "mul %0, %2, %0\n" \ - "adds %3, %4, %3\n" \ - "addcs %0, %0, #65536\n" \ - "adds %1, %1, %3, lsl #16\n" \ - "adc %0, %0, %3, lsr #16" \ + " mov %2, %5, lsr #16\n" \ + " mov %0, %6, lsr #16\n" \ + " bic %3, %5, %2, lsl #16\n" \ + " bic %4, %6, %0, lsl #16\n" \ + " mul %1, %3, %4\n" \ + " mul %4, %2, %4\n" \ + " mul %3, %0, %3\n" \ + " mul %0, %2, %0\n" \ + " adds %3, %4, %3\n" \ + " addcs %0, %0, #65536\n" \ + " adds %1, %1, %3, lsl #16\n" \ + " adc %0, %0, %3, lsr #16" \ : "=&r" ((USItype) (xh)), \ "=r" ((USItype) (xl)), \ "=&r" (__t0), "=&r" (__t1), "=r" (__t2) \ @@ -248,77 +234,9 @@ UDItype __umulsidi3 (USItype, USItype); #define UDIV_TIME 100 #endif /* __arm__ */ -#if defined (__clipper__) && W_TYPE_SIZE == 32 -#define umul_ppmm(w1, w0, u, v) \ - ({union {UDItype __ll; \ - struct {USItype __l, __h;} __i; \ - } __xx; \ - __asm__ ("mulwux %2,%0" \ - : "=r" (__xx.__ll) \ - : "%0" ((USItype) (u)), \ - "r" ((USItype) (v))); \ - (w1) = __xx.__i.__h; (w0) = __xx.__i.__l;}) -#define smul_ppmm(w1, w0, u, v) \ - ({union {DItype __ll; \ - struct {SItype __l, __h;} __i; \ - } __xx; \ - __asm__ ("mulwx %2,%0" \ - : "=r" (__xx.__ll) \ - : "%0" ((SItype) (u)), \ - "r" ((SItype) (v))); \ - (w1) = __xx.__i.__h; (w0) = __xx.__i.__l;}) -#define __umulsidi3(u, v) \ - ({UDItype __w; \ - __asm__ ("mulwux %2,%0" \ - : "=r" (__w) \ - : "%0" ((USItype) (u)), \ - "r" ((USItype) (v))); \ - __w; }) -#endif /* __clipper__ */ - -#if defined (__gmicro__) && W_TYPE_SIZE == 32 -#define add_ssaaaa(sh, sl, ah, al, bh, bl) \ - __asm__ ("add.w %5,%1\n" \ - "addx %3,%0" \ - : "=g" ((USItype) (sh)), \ - "=&g" ((USItype) (sl)) \ - : "%0" ((USItype) (ah)), \ - "g" ((USItype) (bh)), \ - "%1" ((USItype) (al)), \ - "g" ((USItype) (bl))) -#define sub_ddmmss(sh, sl, ah, al, bh, bl) \ - __asm__ ("sub.w %5,%1\n" \ - "subx %3,%0" \ - : "=g" ((USItype) (sh)), \ - "=&g" ((USItype) (sl)) \ - : "0" ((USItype) (ah)), \ - "g" ((USItype) (bh)), \ - "1" ((USItype) (al)), \ - "g" ((USItype) (bl))) -#define umul_ppmm(ph, pl, m0, m1) \ - __asm__ ("mulx %3,%0,%1" \ - : "=g" ((USItype) (ph)), \ - "=r" ((USItype) (pl)) \ - : "%0" ((USItype) (m0)), \ - "g" ((USItype) (m1))) -#define udiv_qrnnd(q, r, nh, nl, d) \ - __asm__ ("divx %4,%0,%1" \ - : "=g" ((USItype) (q)), \ - "=r" ((USItype) (r)) \ - : "1" ((USItype) (nh)), \ - "0" ((USItype) (nl)), \ - "g" ((USItype) (d))) -#define count_leading_zeros(count, x) \ - __asm__ ("bsch/1 %1,%0" \ - : "=g" (count) \ - : "g" ((USItype) (x)), \ - "0" ((USItype) 0)) -#endif - #if defined (__hppa) && W_TYPE_SIZE == 32 #define add_ssaaaa(sh, sl, ah, al, bh, bl) \ - __asm__ ("add %4,%5,%1\n" \ - "addc %2,%3,%0" \ + __asm__ ("add %4,%5,%1\n\taddc %2,%3,%0" \ : "=r" ((USItype) (sh)), \ "=&r" ((USItype) (sl)) \ : "%rM" ((USItype) (ah)), \ @@ -326,8 +244,7 @@ UDItype __umulsidi3 (USItype, USItype); "%rM" ((USItype) (al)), \ "rM" ((USItype) (bl))) #define sub_ddmmss(sh, sl, ah, al, bh, bl) \ - __asm__ ("sub %4,%5,%1\n" \ - "subb %2,%3,%0" \ + __asm__ ("sub %4,%5,%1\n\tsubb %2,%3,%0" \ : "=r" ((USItype) (sh)), \ "=&r" ((USItype) (sl)) \ : "rM" ((USItype) (ah)), \ @@ -355,25 +272,25 @@ UDItype __umulsidi3 (USItype, USItype); #endif #define UDIV_TIME 40 #define count_leading_zeros(count, x) \ - do { \ - USItype __tmp; \ - __asm__ ( \ - "ldi 1,%0\n" \ - "extru,= %1,15,16,%%r0 ; Bits 31..16 zero?\n" \ - "extru,tr %1,15,16,%1 ; No. Shift down, skip add.\n" \ - "ldo 16(%0),%0 ; Yes. Perform add.\n" \ - "extru,= %1,23,8,%%r0 ; Bits 15..8 zero?\n" \ - "extru,tr %1,23,8,%1 ; No. Shift down, skip add.\n" \ - "ldo 8(%0),%0 ; Yes. Perform add.\n" \ - "extru,= %1,27,4,%%r0 ; Bits 7..4 zero?\n" \ - "extru,tr %1,27,4,%1 ; No. Shift down, skip add.\n" \ - "ldo 4(%0),%0 ; Yes. Perform add.\n" \ - "extru,= %1,29,2,%%r0 ; Bits 3..2 zero?\n" \ - "extru,tr %1,29,2,%1 ; No. Shift down, skip add.\n" \ - "ldo 2(%0),%0 ; Yes. Perform add.\n" \ - "extru %1,30,1,%1 ; Extract bit 1.\n" \ - "sub %0,%1,%0 ; Subtract it." \ - : "=r" (count), "=r" (__tmp) : "1" (x)); \ + do { \ + USItype __tmp; \ + __asm__ ( \ + "ldi 1,%0\n" \ +" extru,= %1,15,16,%%r0 ; Bits 31..16 zero?\n" \ +" extru,tr %1,15,16,%1 ; No. Shift down, skip add.\n"\ +" ldo 16(%0),%0 ; Yes. Perform add.\n" \ +" extru,= %1,23,8,%%r0 ; Bits 15..8 zero?\n" \ +" extru,tr %1,23,8,%1 ; No. Shift down, skip add.\n"\ +" ldo 8(%0),%0 ; Yes. Perform add.\n" \ +" extru,= %1,27,4,%%r0 ; Bits 7..4 zero?\n" \ +" extru,tr %1,27,4,%1 ; No. Shift down, skip add.\n"\ +" ldo 4(%0),%0 ; Yes. Perform add.\n" \ +" extru,= %1,29,2,%%r0 ; Bits 3..2 zero?\n" \ +" extru,tr %1,29,2,%1 ; No. Shift down, skip add.\n"\ +" ldo 2(%0),%0 ; Yes. Perform add.\n" \ +" extru %1,30,1,%1 ; Extract bit 1.\n" \ +" sub %0,%1,%0 ; Subtract it.\n" \ + : "=r" (count), "=r" (__tmp) : "1" (x)); \ } while (0) #endif @@ -420,8 +337,7 @@ UDItype __umulsidi3 (USItype, USItype); #if (defined (__i386__) || defined (__i486__)) && W_TYPE_SIZE == 32 #define add_ssaaaa(sh, sl, ah, al, bh, bl) \ - __asm__ ("addl %5,%1\n" \ - "adcl %3,%0" \ + __asm__ ("addl %5,%1\n\tadcl %3,%0" \ : "=r" ((USItype) (sh)), \ "=&r" ((USItype) (sl)) \ : "%0" ((USItype) (ah)), \ @@ -429,8 +345,7 @@ UDItype __umulsidi3 (USItype, USItype); "%1" ((USItype) (al)), \ "g" ((USItype) (bl))) #define sub_ddmmss(sh, sl, ah, al, bh, bl) \ - __asm__ ("subl %5,%1\n" \ - "sbbl %3,%0" \ + __asm__ ("subl %5,%1\n\tsbbl %3,%0" \ : "=r" ((USItype) (sh)), \ "=&r" ((USItype) (sl)) \ : "0" ((USItype) (ah)), \ @@ -443,13 +358,13 @@ UDItype __umulsidi3 (USItype, USItype); "=d" ((USItype) (w1)) \ : "%0" ((USItype) (u)), \ "rm" ((USItype) (v))) -#define udiv_qrnnd(q, r, n1, n0, d) \ +#define udiv_qrnnd(q, r, n1, n0, dv) \ __asm__ ("divl %4" \ : "=a" ((USItype) (q)), \ "=d" ((USItype) (r)) \ : "0" ((USItype) (n0)), \ "1" ((USItype) (n1)), \ - "rm" ((USItype) (d))) + "rm" ((USItype) (dv))) #define count_leading_zeros(count, x) \ do { \ USItype __cbtmp; \ @@ -463,47 +378,6 @@ UDItype __umulsidi3 (USItype, USItype); #define UDIV_TIME 40 #endif /* 80x86 */ -#if defined (__i860__) && W_TYPE_SIZE == 32 -#if 0 -/* Make sure these patterns really improve the code before - switching them on. */ -#define add_ssaaaa(sh, sl, ah, al, bh, bl) \ - do { \ - union \ - { \ - DItype __ll; \ - struct {USItype __l, __h;} __i; \ - } __a, __b, __s; \ - __a.__i.__l = (al); \ - __a.__i.__h = (ah); \ - __b.__i.__l = (bl); \ - __b.__i.__h = (bh); \ - __asm__ ("fiadd.dd %1,%2,%0" \ - : "=f" (__s.__ll) \ - : "%f" (__a.__ll), "f" (__b.__ll)); \ - (sh) = __s.__i.__h; \ - (sl) = __s.__i.__l; \ - } while (0) -#define sub_ddmmss(sh, sl, ah, al, bh, bl) \ - do { \ - union \ - { \ - DItype __ll; \ - struct {USItype __l, __h;} __i; \ - } __a, __b, __s; \ - __a.__i.__l = (al); \ - __a.__i.__h = (ah); \ - __b.__i.__l = (bl); \ - __b.__i.__h = (bh); \ - __asm__ ("fisub.dd %1,%2,%0" \ - : "=f" (__s.__ll) \ - : "%f" (__a.__ll), "f" (__b.__ll)); \ - (sh) = __s.__i.__h; \ - (sl) = __s.__i.__l; \ - } while (0) -#endif -#endif /* __i860__ */ - #if defined (__i960__) && W_TYPE_SIZE == 32 #define umul_ppmm(w1, w0, u, v) \ ({union {UDItype __ll; \ @@ -526,9 +400,7 @@ UDItype __umulsidi3 (USItype, USItype); #if defined (__M32R__) && W_TYPE_SIZE == 32 #define add_ssaaaa(sh, sl, ah, al, bh, bl) \ /* The cmp clears the condition bit. */ \ - __asm__ ("cmp %0,%0\n" \ - "addx %%5,%1\n" \ - "addx %%3,%0" \ + __asm__ ("cmp %0,%0\n\taddx %%5,%1\n\taddx %%3,%0" \ : "=r" ((USItype) (sh)), \ "=&r" ((USItype) (sl)) \ : "%0" ((USItype) (ah)), \ @@ -538,9 +410,7 @@ UDItype __umulsidi3 (USItype, USItype); : "cbit") #define sub_ddmmss(sh, sl, ah, al, bh, bl) \ /* The cmp clears the condition bit. */ \ - __asm__ ("cmp %0,%0\n" \ - "subx %5,%1\n" \ - "subx %3,%0" \ + __asm__ ("cmp %0,%0\n\tsubx %5,%1\n\tsubx %3,%0" \ : "=r" ((USItype) (sh)), \ "=&r" ((USItype) (sl)) \ : "0" ((USItype) (ah)), \ @@ -552,8 +422,7 @@ UDItype __umulsidi3 (USItype, USItype); #if defined (__mc68000__) && W_TYPE_SIZE == 32 #define add_ssaaaa(sh, sl, ah, al, bh, bl) \ - __asm__ ("add%.l %5,%1\n" \ - "addx%.l %3,%0" \ + __asm__ ("add%.l %5,%1\n\taddx%.l %3,%0" \ : "=d" ((USItype) (sh)), \ "=&d" ((USItype) (sl)) \ : "%0" ((USItype) (ah)), \ @@ -561,8 +430,7 @@ UDItype __umulsidi3 (USItype, USItype); "%1" ((USItype) (al)), \ "g" ((USItype) (bl))) #define sub_ddmmss(sh, sl, ah, al, bh, bl) \ - __asm__ ("sub%.l %5,%1\n" \ - "subx%.l %3,%0" \ + __asm__ ("sub%.l %5,%1\n\tsubx%.l %3,%0" \ : "=d" ((USItype) (sh)), \ "=&d" ((USItype) (sl)) \ : "0" ((USItype) (ah)), \ @@ -570,12 +438,11 @@ UDItype __umulsidi3 (USItype, USItype); "1" ((USItype) (al)), \ "g" ((USItype) (bl))) -/* The '020, '030, '040 and CPU32 have 32x32->64 and 64/32->32q-32r. */ +/* The '020, '030, '040 and CPU32 have 32x32->64 and 64/32->32q-32r. */ #if defined (__mc68020__) || defined(mc68020) \ || defined(__mc68030__) || defined(mc68030) \ || defined(__mc68040__) || defined(mc68040) \ - || defined(__mcpu32__) || defined(mcpu32) \ - || defined(__NeXT__) + || defined(__mcpu32__) || defined(mcpu32) #define umul_ppmm(w1, w0, u, v) \ __asm__ ("mulu%.l %3,%1:%0" \ : "=d" ((USItype) (w0)), \ @@ -604,31 +471,31 @@ UDItype __umulsidi3 (USItype, USItype); /* %/ inserts REGISTER_PREFIX, %# inserts IMMEDIATE_PREFIX. */ #define umul_ppmm(xh, xl, a, b) \ __asm__ ("| Inlined umul_ppmm\n" \ -" move%.l %2,%/d0\n" \ -" move%.l %3,%/d1\n" \ -" move%.l %/d0,%/d2\n" \ -" swap %/d0\n" \ -" move%.l %/d1,%/d3\n" \ -" swap %/d1\n" \ -" move%.w %/d2,%/d4\n" \ -" mulu %/d3,%/d4\n" \ -" mulu %/d1,%/d2\n" \ -" mulu %/d0,%/d3\n" \ -" mulu %/d0,%/d1\n" \ -" move%.l %/d4,%/d0\n" \ -" eor%.w %/d0,%/d0\n" \ -" swap %/d0\n" \ -" add%.l %/d0,%/d2\n" \ -" add%.l %/d3,%/d2\n" \ -" jcc 1f\n" \ -" add%.l %#65536,%/d1\n" \ -"1: swap %/d2\n" \ -" moveq %#0,%/d0\n" \ -" move%.w %/d2,%/d0\n" \ -" move%.w %/d4,%/d2\n" \ -" move%.l %/d2,%1\n" \ -" add%.l %/d1,%/d0\n" \ -" move%.l %/d0,%0" \ + " move%.l %2,%/d0\n" \ + " move%.l %3,%/d1\n" \ + " move%.l %/d0,%/d2\n" \ + " swap %/d0\n" \ + " move%.l %/d1,%/d3\n" \ + " swap %/d1\n" \ + " move%.w %/d2,%/d4\n" \ + " mulu %/d3,%/d4\n" \ + " mulu %/d1,%/d2\n" \ + " mulu %/d0,%/d3\n" \ + " mulu %/d0,%/d1\n" \ + " move%.l %/d4,%/d0\n" \ + " eor%.w %/d0,%/d0\n" \ + " swap %/d0\n" \ + " add%.l %/d0,%/d2\n" \ + " add%.l %/d3,%/d2\n" \ + " jcc 1f\n" \ + " add%.l %#65536,%/d1\n" \ + "1: swap %/d2\n" \ + " moveq %#0,%/d0\n" \ + " move%.w %/d2,%/d0\n" \ + " move%.w %/d4,%/d2\n" \ + " move%.l %/d2,%1\n" \ + " add%.l %/d1,%/d0\n" \ + " move%.l %/d0,%0" \ : "=g" ((USItype) (xh)), \ "=g" ((USItype) (xl)) \ : "g" ((USItype) (a)), \ @@ -639,12 +506,11 @@ UDItype __umulsidi3 (USItype, USItype); #endif /* not mcf5200 */ #endif /* not mc68020 */ -/* The '020, '030, '040 and '060 have bitfield insns. */ +/* The '020, '030, '040 and '060 have bitfield insns. */ #if defined (__mc68020__) || defined(mc68020) \ || defined(__mc68030__) || defined(mc68030) \ || defined(__mc68040__) || defined(mc68040) \ - || defined(__mc68060__) || defined(mc68060) \ - || defined(__NeXT__) + || defined(__mc68060__) || defined(mc68060) #define count_leading_zeros(count, x) \ __asm__ ("bfffo %1{%b2:%b2},%0" \ : "=d" ((USItype) (count)) \ @@ -654,8 +520,7 @@ UDItype __umulsidi3 (USItype, USItype); #if defined (__m88000__) && W_TYPE_SIZE == 32 #define add_ssaaaa(sh, sl, ah, al, bh, bl) \ - __asm__ ("addu.co %1,%r4,%r5\n" \ - "addu.ci %0,%r2,%r3" \ + __asm__ ("addu.co %1,%r4,%r5\n\taddu.ci %0,%r2,%r3" \ : "=r" ((USItype) (sh)), \ "=&r" ((USItype) (sl)) \ : "%rJ" ((USItype) (ah)), \ @@ -663,8 +528,7 @@ UDItype __umulsidi3 (USItype, USItype); "%rJ" ((USItype) (al)), \ "rJ" ((USItype) (bl))) #define sub_ddmmss(sh, sl, ah, al, bh, bl) \ - __asm__ ("subu.co %1,%r4,%r5\n" \ - "subu.ci %0,%r2,%r3" \ + __asm__ ("subu.co %1,%r4,%r5\n\tsubu.ci %0,%r2,%r3" \ : "=r" ((USItype) (sh)), \ "=&r" ((USItype) (sl)) \ : "rJ" ((USItype) (ah)), \ @@ -751,267 +615,147 @@ UDItype __umulsidi3 (USItype, USItype); "g" ((USItype) (d))); \ (r) = __xx.__i.__l; (q) = __xx.__i.__h; }) #define count_trailing_zeros(count,x) \ - do { - __asm__ ("ffsd %2,%0" \ - : "=r" ((USItype) (count)) \ - : "0" ((USItype) 0), \ - "r" ((USItype) (x))); \ + do { \ + __asm__ ("ffsd %2,%0" \ + : "=r" ((USItype) (count)) \ + : "0" ((USItype) 0), \ + "r" ((USItype) (x))); \ } while (0) #endif /* __ns32000__ */ -#if (defined (_ARCH_PPC) || defined (_IBMR2)) -#if W_TYPE_SIZE == 32 +/* FIXME: We should test _IBMR2 here when we add assembly support for the + system vendor compilers. + FIXME: What's needed for gcc PowerPC VxWorks? __vxworks__ is not good + enough, since that hits ARM and m68k too. */ +#if (defined (_ARCH_PPC) /* AIX */ \ + || defined (_ARCH_PWR) /* AIX */ \ + || defined (_ARCH_COM) /* AIX */ \ + || defined (__powerpc__) /* gcc */ \ + || defined (__POWERPC__) /* BEOS */ \ + || defined (__ppc__) /* Darwin */ \ + || defined (PPC) /* GNU/Linux, SysV */ \ + ) && W_TYPE_SIZE == 32 #define add_ssaaaa(sh, sl, ah, al, bh, bl) \ do { \ if (__builtin_constant_p (bh) && (bh) == 0) \ __asm__ ("{a%I4|add%I4c} %1,%3,%4\n\t{aze|addze} %0,%2" \ - : "=r" ((USItype) (sh)), \ - "=&r" ((USItype) (sl)) \ - : "%r" ((USItype) (ah)), \ - "%r" ((USItype) (al)), \ - "rI" ((USItype) (bl))); \ - else if (__builtin_constant_p (bh) && (bh) ==~(USItype) 0) \ + : "=r" (sh), "=&r" (sl) : "r" (ah), "%r" (al), "rI" (bl));\ + else if (__builtin_constant_p (bh) && (bh) == ~(USItype) 0) \ __asm__ ("{a%I4|add%I4c} %1,%3,%4\n\t{ame|addme} %0,%2" \ - : "=r" ((USItype) (sh)), \ - "=&r" ((USItype) (sl)) \ - : "%r" ((USItype) (ah)), \ - "%r" ((USItype) (al)), \ - "rI" ((USItype) (bl))); \ + : "=r" (sh), "=&r" (sl) : "r" (ah), "%r" (al), "rI" (bl));\ else \ __asm__ ("{a%I5|add%I5c} %1,%4,%5\n\t{ae|adde} %0,%2,%3" \ - : "=r" ((USItype) (sh)), \ - "=&r" ((USItype) (sl)) \ - : "%r" ((USItype) (ah)), \ - "r" ((USItype) (bh)), \ - "%r" ((USItype) (al)), \ - "rI" ((USItype) (bl))); \ + : "=r" (sh), "=&r" (sl) \ + : "%r" (ah), "r" (bh), "%r" (al), "rI" (bl)); \ } while (0) #define sub_ddmmss(sh, sl, ah, al, bh, bl) \ do { \ if (__builtin_constant_p (ah) && (ah) == 0) \ __asm__ ("{sf%I3|subf%I3c} %1,%4,%3\n\t{sfze|subfze} %0,%2" \ - : "=r" ((USItype) (sh)), \ - "=&r" ((USItype) (sl)) \ - : "r" ((USItype) (bh)), \ - "rI" ((USItype) (al)), \ - "r" ((USItype) (bl))); \ - else if (__builtin_constant_p (ah) && (ah) ==~(USItype) 0) \ + : "=r" (sh), "=&r" (sl) : "r" (bh), "rI" (al), "r" (bl));\ + else if (__builtin_constant_p (ah) && (ah) == ~(USItype) 0) \ __asm__ ("{sf%I3|subf%I3c} %1,%4,%3\n\t{sfme|subfme} %0,%2" \ - : "=r" ((USItype) (sh)), \ - "=&r" ((USItype) (sl)) \ - : "r" ((USItype) (bh)), \ - "rI" ((USItype) (al)), \ - "r" ((USItype) (bl))); \ + : "=r" (sh), "=&r" (sl) : "r" (bh), "rI" (al), "r" (bl));\ else if (__builtin_constant_p (bh) && (bh) == 0) \ __asm__ ("{sf%I3|subf%I3c} %1,%4,%3\n\t{ame|addme} %0,%2" \ - : "=r" ((USItype) (sh)), \ - "=&r" ((USItype) (sl)) \ - : "r" ((USItype) (ah)), \ - "rI" ((USItype) (al)), \ - "r" ((USItype) (bl))); \ - else if (__builtin_constant_p (bh) && (bh) ==~(USItype) 0) \ + : "=r" (sh), "=&r" (sl) : "r" (ah), "rI" (al), "r" (bl));\ + else if (__builtin_constant_p (bh) && (bh) == ~(USItype) 0) \ __asm__ ("{sf%I3|subf%I3c} %1,%4,%3\n\t{aze|addze} %0,%2" \ - : "=r" ((USItype) (sh)), \ - "=&r" ((USItype) (sl)) \ - : "r" ((USItype) (ah)), \ - "rI" ((USItype) (al)), \ - "r" ((USItype) (bl))); \ + : "=r" (sh), "=&r" (sl) : "r" (ah), "rI" (al), "r" (bl));\ else \ __asm__ ("{sf%I4|subf%I4c} %1,%5,%4\n\t{sfe|subfe} %0,%3,%2" \ - : "=r" ((USItype) (sh)), \ - "=&r" ((USItype) (sl)) \ - : "r" ((USItype) (ah)), \ - "r" ((USItype) (bh)), \ - "rI" ((USItype) (al)), \ - "r" ((USItype) (bl))); \ + : "=r" (sh), "=&r" (sl) \ + : "r" (ah), "r" (bh), "rI" (al), "r" (bl)); \ } while (0) #define count_leading_zeros(count, x) \ - __asm__ ("{cntlz|cntlzw} %0,%1" \ - : "=r" ((USItype) (count)) \ - : "r" ((USItype) (x))) + __asm__ ("{cntlz|cntlzw} %0,%1" : "=r" (count) : "r" (x)) #define COUNT_LEADING_ZEROS_0 32 -#if defined (_ARCH_PPC) +#if defined (_ARCH_PPC) || defined (__powerpc__) || defined (__POWERPC__) \ + || defined (__ppc__) || defined (PPC) || defined (__vxworks__) #define umul_ppmm(ph, pl, m0, m1) \ do { \ USItype __m0 = (m0), __m1 = (m1); \ - __asm__ ("mulhwu %0,%1,%2" \ - : "=r" ((USItype) ph) \ - : "%r" (__m0), \ - "r" (__m1)); \ + __asm__ ("mulhwu %0,%1,%2" : "=r" (ph) : "%r" (m0), "r" (m1)); \ (pl) = __m0 * __m1; \ } while (0) #define UMUL_TIME 15 #define smul_ppmm(ph, pl, m0, m1) \ do { \ SItype __m0 = (m0), __m1 = (m1); \ - __asm__ ("mulhw %0,%1,%2" \ - : "=r" ((SItype) ph) \ - : "%r" (__m0), \ - "r" (__m1)); \ + __asm__ ("mulhw %0,%1,%2" : "=r" (ph) : "%r" (m0), "r" (m1)); \ (pl) = __m0 * __m1; \ } while (0) #define SMUL_TIME 14 #define UDIV_TIME 120 -#else -#define umul_ppmm(xh, xl, m0, m1) \ - do { \ - USItype __m0 = (m0), __m1 = (m1); \ - __asm__ ("mul %0,%2,%3" \ - : "=r" ((USItype) (xh)), \ - "=q" ((USItype) (xl)) \ - : "r" (__m0), \ - "r" (__m1)); \ - (xh) += ((((SItype) __m0 >> 31) & __m1) \ - + (((SItype) __m1 >> 31) & __m0)); \ - } while (0) +#elif defined (_ARCH_PWR) #define UMUL_TIME 8 #define smul_ppmm(xh, xl, m0, m1) \ - __asm__ ("mul %0,%2,%3" \ - : "=r" ((SItype) (xh)), \ - "=q" ((SItype) (xl)) \ - : "r" (m0), \ - "r" (m1)) + __asm__ ("mul %0,%2,%3" : "=r" (xh), "=q" (xl) : "r" (m0), "r" (m1)) #define SMUL_TIME 4 #define sdiv_qrnnd(q, r, nh, nl, d) \ - __asm__ ("div %0,%2,%4" \ - : "=r" ((SItype) (q)), "=q" ((SItype) (r)) \ - : "r" ((SItype) (nh)), "1" ((SItype) (nl)), "r" ((SItype) (d))) + __asm__ ("div %0,%2,%4" : "=r" (q), "=q" (r) : "r" (nh), "1" (nl), "r" (d)) #define UDIV_TIME 100 #endif -#else /* W_TYPE_SIZE != 32. */ -/* Must be powerpc64. */ +#endif /* 32-bit POWER architecture variants. */ + +/* We should test _IBMR2 here when we add assembly support for the system + vendor compilers. */ +#if (defined (_ARCH_PPC64) || defined (__powerpc64__)) && W_TYPE_SIZE == 64 #define add_ssaaaa(sh, sl, ah, al, bh, bl) \ do { \ if (__builtin_constant_p (bh) && (bh) == 0) \ __asm__ ("{a%I4|add%I4c} %1,%3,%4\n\t{aze|addze} %0,%2" \ - : "=r" ((UDItype) (sh)), \ - "=&r" ((UDItype) (sl)) \ - : "%r" ((UDItype) (ah)), \ - "%r" ((UDItype) (al)), \ - "rI" ((UDItype) (bl))); \ - else if (__builtin_constant_p (bh) && (bh) ==~(UDItype) 0) \ + : "=r" (sh), "=&r" (sl) : "r" (ah), "%r" (al), "rI" (bl));\ + else if (__builtin_constant_p (bh) && (bh) == ~(UDItype) 0) \ __asm__ ("{a%I4|add%I4c} %1,%3,%4\n\t{ame|addme} %0,%2" \ - : "=r" ((UDItype) (sh)), \ - "=&r" ((UDItype) (sl)) \ - : "%r" ((UDItype) (ah)), \ - "%r" ((UDItype) (al)), \ - "rI" ((UDItype) (bl))); \ + : "=r" (sh), "=&r" (sl) : "r" (ah), "%r" (al), "rI" (bl));\ else \ __asm__ ("{a%I5|add%I5c} %1,%4,%5\n\t{ae|adde} %0,%2,%3" \ - : "=r" ((UDItype) (sh)), \ - "=&r" ((UDItype) (sl)) \ - : "%r" ((UDItype) (ah)), \ - "r" ((UDItype) (bh)), \ - "%r" ((UDItype) (al)), \ - "rI" ((UDItype) (bl))); \ + : "=r" (sh), "=&r" (sl) \ + : "%r" (ah), "r" (bh), "%r" (al), "rI" (bl)); \ } while (0) #define sub_ddmmss(sh, sl, ah, al, bh, bl) \ do { \ if (__builtin_constant_p (ah) && (ah) == 0) \ __asm__ ("{sf%I3|subf%I3c} %1,%4,%3\n\t{sfze|subfze} %0,%2" \ - : "=r" ((UDItype) (sh)), \ - "=&r" ((UDItype) (sl)) \ - : "r" ((UDItype) (bh)), \ - "rI" ((UDItype) (al)), \ - "r" ((UDItype) (bl))); \ - else if (__builtin_constant_p (ah) && (ah) ==~(UDItype) 0) \ + : "=r" (sh), "=&r" (sl) : "r" (bh), "rI" (al), "r" (bl));\ + else if (__builtin_constant_p (ah) && (ah) == ~(UDItype) 0) \ __asm__ ("{sf%I3|subf%I3c} %1,%4,%3\n\t{sfme|subfme} %0,%2" \ - : "=r" ((UDItype) (sh)), \ - "=&r" ((UDItype) (sl)) \ - : "r" ((UDItype) (bh)), \ - "rI" ((UDItype) (al)), \ - "r" ((UDItype) (bl))); \ + : "=r" (sh), "=&r" (sl) : "r" (bh), "rI" (al), "r" (bl));\ else if (__builtin_constant_p (bh) && (bh) == 0) \ __asm__ ("{sf%I3|subf%I3c} %1,%4,%3\n\t{ame|addme} %0,%2" \ - : "=r" ((UDItype) (sh)), \ - "=&r" ((UDItype) (sl)) \ - : "r" ((UDItype) (ah)), \ - "rI" ((UDItype) (al)), \ - "r" ((UDItype) (bl))); \ - else if (__builtin_constant_p (bh) && (bh) ==~(UDItype) 0) \ + : "=r" (sh), "=&r" (sl) : "r" (ah), "rI" (al), "r" (bl));\ + else if (__builtin_constant_p (bh) && (bh) == ~(UDItype) 0) \ __asm__ ("{sf%I3|subf%I3c} %1,%4,%3\n\t{aze|addze} %0,%2" \ - : "=r" ((UDItype) (sh)), \ - "=&r" ((UDItype) (sl)) \ - : "r" ((UDItype) (ah)), \ - "rI" ((UDItype) (al)), \ - "r" ((UDItype) (bl))); \ + : "=r" (sh), "=&r" (sl) : "r" (ah), "rI" (al), "r" (bl));\ else \ __asm__ ("{sf%I4|subf%I4c} %1,%5,%4\n\t{sfe|subfe} %0,%3,%2" \ - : "=r" ((UDItype) (sh)), \ - "=&r" ((UDItype) (sl)) \ - : "r" ((UDItype) (ah)), \ - "r" ((UDItype) (bh)), \ - "rI" ((UDItype) (al)), \ - "r" ((UDItype) (bl))); \ + : "=r" (sh), "=&r" (sl) \ + : "r" (ah), "r" (bh), "rI" (al), "r" (bl)); \ } while (0) - #define count_leading_zeros(count, x) \ - __asm__ ("{cntlz|cntlzd} %0,%1" \ - : "=r" (count) \ - : "r" ((UDItype) (x))) + __asm__ ("cntlzd %0,%1" : "=r" (count) : "r" (x)) #define COUNT_LEADING_ZEROS_0 64 - #define umul_ppmm(ph, pl, m0, m1) \ do { \ UDItype __m0 = (m0), __m1 = (m1); \ - __asm__ ("mulhdu %0,%1,%2" \ - : "=r" ((UDItype) ph) \ - : "%r" (__m0), \ - "r" (__m1)); \ + __asm__ ("mulhdu %0,%1,%2" : "=r" (ph) : "%r" (m0), "r" (m1)); \ (pl) = __m0 * __m1; \ } while (0) -#define UMUL_TIME 16 - +#define UMUL_TIME 15 #define smul_ppmm(ph, pl, m0, m1) \ do { \ DItype __m0 = (m0), __m1 = (m1); \ - __asm__ ("mulhd %0,%1,%2" \ - : "=r" ((DItype) ph) \ - : "%r" (__m0), \ - "r" (__m1)); \ + __asm__ ("mulhd %0,%1,%2" : "=r" (ph) : "%r" (m0), "r" (m1)); \ (pl) = __m0 * __m1; \ } while (0) -#define SMUL_TIME 16 -#define UDIV_TIME 72 -#endif /* W_TYPE_SIZE == 32 */ -#endif /* Power architecture variants. */ - -#if defined (__pyr__) && W_TYPE_SIZE == 32 -#define add_ssaaaa(sh, sl, ah, al, bh, bl) \ - __asm__ ("addw %5,%1\n" \ - "addwc %3,%0" \ - : "=r" ((USItype) (sh)), \ - "=&r" ((USItype) (sl)) \ - : "%0" ((USItype) (ah)), \ - "g" ((USItype) (bh)), \ - "%1" ((USItype) (al)), \ - "g" ((USItype) (bl))) -#define sub_ddmmss(sh, sl, ah, al, bh, bl) \ - __asm__ ("subw %5,%1\n" \ - "subwb %3,%0" \ - : "=r" ((USItype) (sh)), \ - "=&r" ((USItype) (sl)) \ - : "0" ((USItype) (ah)), \ - "g" ((USItype) (bh)), \ - "1" ((USItype) (al)), \ - "g" ((USItype) (bl))) -/* This insn works on Pyramids with AP, XP, or MI CPUs, but not with SP. */ -#define umul_ppmm(w1, w0, u, v) \ - ({union {UDItype __ll; \ - struct {USItype __h, __l;} __i; \ - } __xx; \ - __asm__ ("movw %1,%R0\n" \ - "uemul %2,%0" \ - : "=&r" (__xx.__ll) \ - : "g" ((USItype) (u)), \ - "g" ((USItype) (v))); \ - (w1) = __xx.__i.__h; (w0) = __xx.__i.__l;}) -#endif /* __pyr__ */ +#define SMUL_TIME 14 /* ??? */ +#define UDIV_TIME 120 /* ??? */ +#endif /* 64-bit PowerPC. */ #if defined (__ibm032__) /* RT/ROMP */ && W_TYPE_SIZE == 32 #define add_ssaaaa(sh, sl, ah, al, bh, bl) \ - __asm__ ("a %1,%5\n" \ - "ae %0,%3" \ + __asm__ ("a %1,%5\n\tae %0,%3" \ : "=r" ((USItype) (sh)), \ "=&r" ((USItype) (sl)) \ : "%0" ((USItype) (ah)), \ @@ -1019,8 +763,7 @@ UDItype __umulsidi3 (USItype, USItype); "%1" ((USItype) (al)), \ "r" ((USItype) (bl))) #define sub_ddmmss(sh, sl, ah, al, bh, bl) \ - __asm__ ("s %1,%5\n" \ - "se %0,%3" \ + __asm__ ("s %1,%5\n\tse %0,%3" \ : "=r" ((USItype) (sh)), \ "=&r" ((USItype) (sl)) \ : "0" ((USItype) (ah)), \ @@ -1031,26 +774,26 @@ UDItype __umulsidi3 (USItype, USItype); do { \ USItype __m0 = (m0), __m1 = (m1); \ __asm__ ( \ - "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" \ + "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" ((USItype) (ph)), \ "=r" ((USItype) (pl)) \ : "%r" (__m0), \ @@ -1080,9 +823,7 @@ UDItype __umulsidi3 (USItype, USItype); #if defined (__sh2__) && W_TYPE_SIZE == 32 #define umul_ppmm(w1, w0, u, v) \ __asm__ ( \ - "dmulu.l %2,%3\n" \ - "sts macl,%1\n" \ - "sts mach,%0" \ + "dmulu.l %2,%3\n\tsts macl,%1\n\tsts mach,%0" \ : "=r" ((USItype)(w1)), \ "=r" ((USItype)(w0)) \ : "r" ((USItype)(u)), \ @@ -1091,11 +832,25 @@ UDItype __umulsidi3 (USItype, USItype); #define UMUL_TIME 5 #endif -#if defined (__sparc__) && !defined(__arch64__) \ - && !defined(__sparcv9) && W_TYPE_SIZE == 32 +#if defined (__SH5__) && __SHMEDIA__ && W_TYPE_SIZE == 32 +#define __umulsidi3(u,v) ((UDItype)(USItype)u*(USItype)v) +#define count_leading_zeros(count, x) \ + do \ + { \ + UDItype x_ = (USItype)(x); \ + SItype c_; \ + \ + __asm__ ("nsb %1, %0" : "=r" (c_) : "r" (x_)); \ + (count) = c_ - 31; \ + } \ + while (0) +#define COUNT_LEADING_ZEROS_0 32 +#endif + +#if defined (__sparc__) && !defined (__arch64__) && !defined (__sparcv9) \ + && W_TYPE_SIZE == 32 #define add_ssaaaa(sh, sl, ah, al, bh, bl) \ - __asm__ ("addcc %r4,%5,%1\n" \ - "addx %r2,%3,%0" \ + __asm__ ("addcc %r4,%5,%1\n\taddx %r2,%3,%0" \ : "=r" ((USItype) (sh)), \ "=&r" ((USItype) (sl)) \ : "%rJ" ((USItype) (ah)), \ @@ -1104,8 +859,7 @@ UDItype __umulsidi3 (USItype, USItype); "rI" ((USItype) (bl)) \ __CLOBBER_CC) #define sub_ddmmss(sh, sl, ah, al, bh, bl) \ - __asm__ ("subcc %r4,%5,%1\n" \ - "subx %r2,%3,%0" \ + __asm__ ("subcc %r4,%5,%1\n\tsubx %r2,%3,%0" \ : "=r" ((USItype) (sh)), \ "=&r" ((USItype) (sl)) \ : "rJ" ((USItype) (ah)), \ @@ -1120,13 +874,13 @@ UDItype __umulsidi3 (USItype, USItype); "=r" ((USItype) (w0)) \ : "r" ((USItype) (u)), \ "r" ((USItype) (v))) -#define udiv_qrnnd(q, r, n1, n0, d) \ +#define udiv_qrnnd(__q, __r, __n1, __n0, __d) \ __asm__ ("mov %2,%%y;nop;nop;nop;udiv %3,%4,%0;umul %0,%4,%1;sub %3,%1,%1"\ - : "=&r" ((USItype) (q)), \ - "=&r" ((USItype) (r)) \ - : "r" ((USItype) (n1)), \ - "r" ((USItype) (n0)), \ - "r" ((USItype) (d))) + : "=&r" ((USItype) (__q)), \ + "=&r" ((USItype) (__r)) \ + : "r" ((USItype) (__n1)), \ + "r" ((USItype) (__n0)), \ + "r" ((USItype) (__d))) #else #if defined (__sparclite__) /* This has hardware multiply but not divide. It also has two additional @@ -1139,43 +893,43 @@ UDItype __umulsidi3 (USItype, USItype); "r" ((USItype) (v))) #define udiv_qrnnd(q, r, n1, n0, d) \ __asm__ ("! Inlined udiv_qrnnd\n" \ -" wr %%g0,%2,%%y ! Not a delayed write for sparclite\n" \ -" tst %%g0\n" \ -" divscc %3,%4,%%g1\n" \ -" divscc %%g1,%4,%%g1\n" \ -" divscc %%g1,%4,%%g1\n" \ -" divscc %%g1,%4,%%g1\n" \ -" divscc %%g1,%4,%%g1\n" \ -" divscc %%g1,%4,%%g1\n" \ -" divscc %%g1,%4,%%g1\n" \ -" divscc %%g1,%4,%%g1\n" \ -" divscc %%g1,%4,%%g1\n" \ -" divscc %%g1,%4,%%g1\n" \ -" divscc %%g1,%4,%%g1\n" \ -" divscc %%g1,%4,%%g1\n" \ -" divscc %%g1,%4,%%g1\n" \ -" divscc %%g1,%4,%%g1\n" \ -" divscc %%g1,%4,%%g1\n" \ -" divscc %%g1,%4,%%g1\n" \ -" divscc %%g1,%4,%%g1\n" \ -" divscc %%g1,%4,%%g1\n" \ -" divscc %%g1,%4,%%g1\n" \ -" divscc %%g1,%4,%%g1\n" \ -" divscc %%g1,%4,%%g1\n" \ -" divscc %%g1,%4,%%g1\n" \ -" divscc %%g1,%4,%%g1\n" \ -" divscc %%g1,%4,%%g1\n" \ -" divscc %%g1,%4,%%g1\n" \ -" divscc %%g1,%4,%%g1\n" \ -" divscc %%g1,%4,%%g1\n" \ -" divscc %%g1,%4,%%g1\n" \ -" divscc %%g1,%4,%%g1\n" \ -" divscc %%g1,%4,%%g1\n" \ -" divscc %%g1,%4,%%g1\n" \ -" divscc %%g1,%4,%0\n" \ -" rd %%y,%1\n" \ -" bl,a 1f\n" \ -" add %1,%4,%1\n" \ +" wr %%g0,%2,%%y ! Not a delayed write for sparclite\n" \ +" tst %%g0\n" \ +" divscc %3,%4,%%g1\n" \ +" divscc %%g1,%4,%%g1\n" \ +" divscc %%g1,%4,%%g1\n" \ +" divscc %%g1,%4,%%g1\n" \ +" divscc %%g1,%4,%%g1\n" \ +" divscc %%g1,%4,%%g1\n" \ +" divscc %%g1,%4,%%g1\n" \ +" divscc %%g1,%4,%%g1\n" \ +" divscc %%g1,%4,%%g1\n" \ +" divscc %%g1,%4,%%g1\n" \ +" divscc %%g1,%4,%%g1\n" \ +" divscc %%g1,%4,%%g1\n" \ +" divscc %%g1,%4,%%g1\n" \ +" divscc %%g1,%4,%%g1\n" \ +" divscc %%g1,%4,%%g1\n" \ +" divscc %%g1,%4,%%g1\n" \ +" divscc %%g1,%4,%%g1\n" \ +" divscc %%g1,%4,%%g1\n" \ +" divscc %%g1,%4,%%g1\n" \ +" divscc %%g1,%4,%%g1\n" \ +" divscc %%g1,%4,%%g1\n" \ +" divscc %%g1,%4,%%g1\n" \ +" divscc %%g1,%4,%%g1\n" \ +" divscc %%g1,%4,%%g1\n" \ +" divscc %%g1,%4,%%g1\n" \ +" divscc %%g1,%4,%%g1\n" \ +" divscc %%g1,%4,%%g1\n" \ +" divscc %%g1,%4,%%g1\n" \ +" divscc %%g1,%4,%%g1\n" \ +" divscc %%g1,%4,%%g1\n" \ +" divscc %%g1,%4,%%g1\n" \ +" divscc %%g1,%4,%0\n" \ +" rd %%y,%1\n" \ +" bl,a 1f\n" \ +" add %1,%4,%1\n" \ "1: ! End of inline udiv_qrnnd" \ : "=r" ((USItype) (q)), \ "=r" ((USItype) (r)) \ @@ -1185,10 +939,10 @@ UDItype __umulsidi3 (USItype, USItype); : "g1" __AND_CLOBBER_CC) #define UDIV_TIME 37 #define count_leading_zeros(count, x) \ - do { \ - __asm__ ("scan %1,1,%0" \ - : "=r" ((USItype) (count)) \ - : "r" ((USItype) (x))); \ + do { \ + __asm__ ("scan %1,1,%0" \ + : "=r" ((USItype) (count)) \ + : "r" ((USItype) (x))); \ } while (0) /* Early sparclites return 63 for an argument of 0, but they warn that future implementations might change this. Therefore, leave COUNT_LEADING_ZEROS_0 @@ -1198,45 +952,45 @@ UDItype __umulsidi3 (USItype, USItype); (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\n" \ -" wr %%g0,%2,%%y ! SPARC has 0-3 delay insn after a wr\n"\ -" sra %3,31,%%o5 ! Don't move this insn\n" \ -" and %2,%%o5,%%o5 ! 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,%%o5,%0\n" \ -" rd %%y,%1" \ +" wr %%g0,%2,%%y ! SPARC has 0-3 delay insn after a wr\n"\ +" sra %3,31,%%o5 ! Don't move this insn\n" \ +" and %2,%%o5,%%o5 ! 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,%%o5,%0\n" \ +" rd %%y,%1" \ : "=r" ((USItype) (w1)), \ "=r" ((USItype) (w0)) \ : "%rI" ((USItype) (u)), \ @@ -1244,96 +998,96 @@ UDItype __umulsidi3 (USItype, USItype); : "g1", "o5" __AND_CLOBBER_CC) #define UMUL_TIME 39 /* 39 instructions */ /* 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) \ + The default udiv_qrnnd (in C) is more than 10 times slower! */ +#define udiv_qrnnd(__q, __r, __n1, __n0, __d) \ __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" \ +" 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" \ +" subcc %1,%2,%%g0\n" \ " bcs 3f\n" \ -" addxcc %0,%0,%0 ! shift n1n0 and a q-bit in lsb\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" \ +" 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" \ +" 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" \ +" 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" ((USItype) (q)), \ - "=&r" ((USItype) (r)) \ - : "r" ((USItype) (d)), \ - "1" ((USItype) (n1)), \ - "0" ((USItype) (n0)) : "g1" __AND_CLOBBER_CC) -#define UDIV_TIME (3+7*32) /* 7 instructions/iteration. 32 iterations. */ +" ! End of inline udiv_qrnnd" \ + : "=&r" ((USItype) (__q)), \ + "=&r" ((USItype) (__r)) \ + : "r" ((USItype) (__d)), \ + "1" ((USItype) (__n1)), \ + "0" ((USItype) (__n0)) : "g1" __AND_CLOBBER_CC) +#define UDIV_TIME (3+7*32) /* 7 instructions/iteration. 32 iterations. */ #endif /* __sparclite__ */ #endif /* __sparc_v8__ */ -#endif /* __sparc__ */ +#endif /* sparc32 */ -#if ((defined (__sparc__) && defined (__arch64__)) \ - || defined (__sparcv9)) && W_TYPE_SIZE == 64 +#if ((defined (__sparc__) && defined (__arch64__)) || defined (__sparcv9)) \ + && W_TYPE_SIZE == 64 #define add_ssaaaa(sh, sl, ah, al, bh, bl) \ - __asm__ ("addcc %r4,%5,%1\n" \ - "add %r2,%3,%0\n" \ - "bcs,a,pn %%xcc, 1f\n" \ - "add %0, 1, %0\n" \ + __asm__ ("addcc %r4,%5,%1\n\t" \ + "add %r2,%3,%0\n\t" \ + "bcs,a,pn %%xcc, 1f\n\t" \ + "add %0, 1, %0\n" \ "1:" \ - : "=r" ((UDItype)(sh)), \ - "=&r" ((UDItype)(sl)) \ - : "%rJ" ((UDItype)(ah)), \ - "rI" ((UDItype)(bh)), \ - "%rJ" ((UDItype)(al)), \ - "rI" ((UDItype)(bl)) \ + : "=r" ((UDItype)(sh)), \ + "=&r" ((UDItype)(sl)) \ + : "%rJ" ((UDItype)(ah)), \ + "rI" ((UDItype)(bh)), \ + "%rJ" ((UDItype)(al)), \ + "rI" ((UDItype)(bl)) \ __CLOBBER_CC) -#define sub_ddmmss(sh, sl, ah, al, bh, bl) \ - __asm__ ("subcc %r4,%5,%1\n" \ - "sub %r2,%3,%0\n" \ - "bcs,a,pn %%xcc, 1f\n" \ - "sub %0, 1, %0\n" \ +#define sub_ddmmss(sh, sl, ah, al, bh, bl) \ + __asm__ ("subcc %r4,%5,%1\n\t" \ + "sub %r2,%3,%0\n\t" \ + "bcs,a,pn %%xcc, 1f\n\t" \ + "sub %0, 1, %0\n\t" \ "1:" \ - : "=r" ((UDItype)(sh)), \ - "=&r" ((UDItype)(sl)) \ - : "rJ" ((UDItype)(ah)), \ - "rI" ((UDItype)(bh)), \ - "rJ" ((UDItype)(al)), \ - "rI" ((UDItype)(bl)) \ + : "=r" ((UDItype)(sh)), \ + "=&r" ((UDItype)(sl)) \ + : "rJ" ((UDItype)(ah)), \ + "rI" ((UDItype)(bh)), \ + "rJ" ((UDItype)(al)), \ + "rI" ((UDItype)(bl)) \ __CLOBBER_CC) #define umul_ppmm(wh, wl, u, v) \ do { \ UDItype tmp1, tmp2, tmp3, tmp4; \ __asm__ __volatile__ ( \ - "srl %7,0,%3\n" \ - "mulx %3,%6,%1\n" \ - "srlx %6,32,%2\n" \ - "mulx %2,%3,%4\n" \ - "sllx %4,32,%5\n" \ - "srl %6,0,%3\n" \ - "sub %1,%5,%5\n" \ - "srlx %5,32,%5\n" \ - "addcc %4,%5,%4\n" \ - "srlx %7,32,%5\n" \ - "mulx %3,%5,%3\n" \ - "mulx %2,%5,%5\n" \ - "sethi %%hi(0x80000000),%2\n" \ - "addcc %4,%3,%4\n" \ - "srlx %4,32,%4\n" \ - "add %2,%2,%2\n" \ - "movcc %%xcc,%%g0,%2\n" \ - "addcc %5,%4,%5\n" \ - "sllx %3,32,%3\n" \ - "add %1,%3,%1\n" \ + "srl %7,0,%3\n\t" \ + "mulx %3,%6,%1\n\t" \ + "srlx %6,32,%2\n\t" \ + "mulx %2,%3,%4\n\t" \ + "sllx %4,32,%5\n\t" \ + "srl %6,0,%3\n\t" \ + "sub %1,%5,%5\n\t" \ + "srlx %5,32,%5\n\t" \ + "addcc %4,%5,%4\n\t" \ + "srlx %7,32,%5\n\t" \ + "mulx %3,%5,%3\n\t" \ + "mulx %2,%5,%5\n\t" \ + "sethi %%hi(0x80000000),%2\n\t" \ + "addcc %4,%3,%4\n\t" \ + "srlx %4,32,%4\n\t" \ + "add %2,%2,%2\n\t" \ + "movcc %%xcc,%%g0,%2\n\t" \ + "addcc %5,%4,%5\n\t" \ + "sllx %3,32,%3\n\t" \ + "add %1,%3,%1\n\t" \ "add %5,%2,%0" \ : "=r" ((UDItype)(wh)), \ "=&r" ((UDItype)(wl)), \ @@ -1348,8 +1102,7 @@ UDItype __umulsidi3 (USItype, USItype); #if defined (__vax__) && W_TYPE_SIZE == 32 #define add_ssaaaa(sh, sl, ah, al, bh, bl) \ - __asm__ ("addl2 %5,%1\n" \ - "adwc %3,%0" \ + __asm__ ("addl2 %5,%1\n\tadwc %3,%0" \ : "=g" ((USItype) (sh)), \ "=&g" ((USItype) (sl)) \ : "%0" ((USItype) (ah)), \ @@ -1357,8 +1110,7 @@ UDItype __umulsidi3 (USItype, USItype); "%1" ((USItype) (al)), \ "g" ((USItype) (bl))) #define sub_ddmmss(sh, sl, ah, al, bh, bl) \ - __asm__ ("subl2 %5,%1\n" \ - "sbwc %3,%0" \ + __asm__ ("subl2 %5,%1\n\tsbwc %3,%0" \ : "=g" ((USItype) (sh)), \ "=&g" ((USItype) (sl)) \ : "0" ((USItype) (ah)), \ @@ -1470,7 +1222,7 @@ UDItype __umulsidi3 (USItype, USItype); __x1 += __ll_highpart (__x0);/* this can't give carry */ \ __x1 += __x2; /* but this indeed can */ \ if (__x1 < __x2) /* did we get it? */ \ - __x3 += __ll_B; /* yes, add it in the proper pos. */ \ + __x3 += __ll_B; /* yes, add it in the proper pos. */ \ \ (w1) = __x3 + __ll_highpart (__x1); \ (w0) = __ll_lowpart (__x1) * __ll_B + __ll_lowpart (__x0); \ @@ -1479,7 +1231,7 @@ UDItype __umulsidi3 (USItype, USItype); #if !defined (__umulsidi3) #define __umulsidi3(u, v) \ - ({DIunion __w; \ + ({DWunion __w; \ umul_ppmm (__w.s.high, __w.s.low, u, v); \ __w.ll; }) #endif -- cgit 1.4.1