diff options
Diffstat (limited to 'libgcrypt-1.4.6/mpi')
-rw-r--r-- | libgcrypt-1.4.6/mpi/longlong.h | 3156 | ||||
-rw-r--r-- | libgcrypt-1.4.6/mpi/mpi-bit.c | 730 | ||||
-rw-r--r-- | libgcrypt-1.4.6/mpi/mpi-div.c | 716 | ||||
-rw-r--r-- | libgcrypt-1.4.6/mpi/mpi-scan.c | 264 | ||||
-rw-r--r-- | libgcrypt-1.4.6/mpi/mpicoder.c | 1503 | ||||
-rw-r--r-- | libgcrypt-1.4.6/mpi/mpiutil.c | 923 |
6 files changed, 3647 insertions, 3645 deletions
diff --git a/libgcrypt-1.4.6/mpi/longlong.h b/libgcrypt-1.4.6/mpi/longlong.h index 586c4cc..b736d49 100644 --- a/libgcrypt-1.4.6/mpi/longlong.h +++ b/libgcrypt-1.4.6/mpi/longlong.h @@ -1,1578 +1,1578 @@ -/* longlong.h -- definitions for mixed size 32/64 bit arithmetic.
- Note: I added some stuff for use with gnupg
-
-Copyright (C) 1991, 1992, 1993, 1994, 1996, 1998,
- 2000, 2001, 2002, 2003, 2004 Free Software Foundation, Inc.
-
-This file is free software; you can redistribute it and/or modify
-it under the terms of the GNU Lesser General Public License as published by
-the Free Software Foundation; either version 2.1 of the License, or (at your
-option) any later version.
-
-This file is distributed in the hope that it will be useful, but
-WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
-or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public
-License for more details.
-
-You should have received a copy of the GNU Library General Public License
-along with this file; see the file COPYING.LIB. If not, write to
-the Free Software Foundation, Inc., 59 Temple Place - Suite 330, Boston,
-MA 02111-1307, USA. */
-
-/* You have to define the following before including this file:
-
- UWtype -- An unsigned type, default type for operations (typically a "word")
- UHWtype -- An unsigned type, at least half the size of UWtype.
- UDWtype -- An unsigned type, at least twice as large a UWtype
- W_TYPE_SIZE -- size in bits of UWtype
-
- SItype, USItype -- Signed and unsigned 32 bit types.
- DItype, UDItype -- Signed and unsigned 64 bit types.
-
- On a 32 bit machine UWtype should typically be USItype;
- on a 64 bit machine, UWtype should typically be UDItype.
-*/
-
-#define __BITS4 (W_TYPE_SIZE / 4)
-#define __ll_B ((UWtype) 1 << (W_TYPE_SIZE / 2))
-#define __ll_lowpart(t) ((UWtype) (t) & (__ll_B - 1))
-#define __ll_highpart(t) ((UWtype) (t) >> (W_TYPE_SIZE / 2))
-
-/* This is used to make sure no undesirable sharing between different libraries
- that use this file takes place. */
-#ifndef __MPN
-#define __MPN(x) __##x
-#endif
-
-/* Define auxiliary asm macros.
-
- 1) umul_ppmm(high_prod, low_prod, multipler, multiplicand) multiplies two
- UWtype integers MULTIPLER and MULTIPLICAND, and generates a two UWtype
- word product in HIGH_PROD and LOW_PROD.
-
- 2) __umulsidi3(a,b) multiplies two UWtype integers A and B, and returns a
- UDWtype product. This is just a variant of umul_ppmm.
-
- 3) udiv_qrnnd(quotient, remainder, high_numerator, low_numerator,
- denominator) divides a UDWtype, composed by the UWtype integers
- HIGH_NUMERATOR and LOW_NUMERATOR, by DENOMINATOR and places the quotient
- in QUOTIENT and the remainder in REMAINDER. HIGH_NUMERATOR must be less
- than DENOMINATOR for correct operation. If, in addition, the most
- significant bit of DENOMINATOR must be 1, then the pre-processor symbol
- UDIV_NEEDS_NORMALIZATION is defined to 1.
-
- 4) sdiv_qrnnd(quotient, remainder, high_numerator, low_numerator,
- denominator). Like udiv_qrnnd but the numbers are signed. The quotient
- 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
- 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.
-
- 6) count_trailing_zeros(count, x) like count_leading_zeros, but counts
- from the least significant end.
-
- 7) add_ssaaaa(high_sum, low_sum, high_addend_1, low_addend_1,
- high_addend_2, low_addend_2) adds two UWtype integers, composed by
- HIGH_ADDEND_1 and LOW_ADDEND_1, and HIGH_ADDEND_2 and LOW_ADDEND_2
- respectively. The result is placed in HIGH_SUM and LOW_SUM. Overflow
- (i.e. carry out) is not stored anywhere, and is lost.
-
- 8) sub_ddmmss(high_difference, low_difference, high_minuend, low_minuend,
- high_subtrahend, low_subtrahend) subtracts two two-word UWtype integers,
- composed by HIGH_MINUEND_1 and LOW_MINUEND_1, and HIGH_SUBTRAHEND_2 and
- LOW_SUBTRAHEND_2 respectively. The result is placed in HIGH_DIFFERENCE
- and LOW_DIFFERENCE. Overflow (i.e. carry out) is not stored anywhere,
- and is lost.
-
- If any of these macros are left undefined for a particular CPU,
- C macros are used. */
-
-/* The CPUs come in alphabetical order below.
-
- Please add support for more CPUs here, or improve the current support
- for the CPUs below! */
-
-#ifdef __riscos__
-#pragma continue_after_hash_error
-#else /* !__riscos__ */
-#if defined (__GNUC__) && !defined (NO_ASM)
-
-/* We sometimes need to clobber "cc" with gcc2, but that would not be
- understood by gcc1. Use cpp to avoid major code duplication. */
-#if __GNUC__ < 2
-#define __CLOBBER_CC
-#define __AND_CLOBBER_CC
-#else /* __GNUC__ >= 2 */
-#define __CLOBBER_CC : "cc"
-#define __AND_CLOBBER_CC , "cc"
-#endif /* __GNUC__ < 2 */
-
-
-/***************************************
- ************** A29K *****************
- ***************************************/
-#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 { \
- UDItype __m0 = (m0), __m1 = (m1); \
- __asm__ ("umulh %r1,%2,%0" \
- : "=r" ((UDItype) ph) \
- : "%rJ" (__m0), \
- "rI" (__m1)); \
- (pl) = __m0 * __m1; \
- } while (0)
-#define UMUL_TIME 46
-#ifndef LONGLONG_STANDALONE
-#define udiv_qrnnd(q, r, n1, n0, d) \
- do { UDItype __r; \
- (q) = __udiv_qrnnd (&__r, (n1), (n0), (d)); \
- (r) = __r; \
- } while (0)
-extern UDItype __udiv_qrnnd ();
-#define UDIV_TIME 220
-#endif /* LONGLONG_STANDALONE */
-#endif /* __alpha */
-
-/***************************************
- ************** ARM ******************
- ***************************************/
-#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" \
- : "=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__ ("subs %1, %4, %5\n" \
- "sbc %0, %2, %3" \
- : "=r" ((USItype)(sh)), \
- "=&r" ((USItype)(sl)) \
- : "r" ((USItype)(ah)), \
- "rI" ((USItype)(bh)), \
- "r" ((USItype)(al)), \
- "rI" ((USItype)(bl)))
-#if defined __ARM_ARCH_2__ || defined __ARM_ARCH_3__
-#define umul_ppmm(xh, xl, a, b) \
- __asm__ ("%@ Inlined umul_ppmm\n" \
- "mov %|r0, %2, lsr #16 @ AAAA\n" \
- "mov %|r2, %3, lsr #16 @ BBBB\n" \
- "bic %|r1, %2, %|r0, lsl #16 @ aaaa\n" \
- "bic %0, %3, %|r2, lsl #16 @ bbbb\n" \
- "mul %1, %|r1, %|r2 @ aaaa * BBBB\n" \
- "mul %|r2, %|r0, %|r2 @ AAAA * BBBB\n" \
- "mul %|r1, %0, %|r1 @ aaaa * bbbb\n" \
- "mul %0, %|r0, %0 @ AAAA * bbbb\n" \
- "adds %|r0, %1, %0 @ central sum\n" \
- "addcs %|r2, %|r2, #65536\n" \
- "adds %1, %|r1, %|r0, lsl #16\n" \
- "adc %0, %|r2, %|r0, lsr #16" \
- : "=&r" ((USItype)(xh)), \
- "=r" ((USItype)(xl)) \
- : "r" ((USItype)(a)), \
- "r" ((USItype)(b)) \
- : "r0", "r1", "r2")
-#else
-#define umul_ppmm(xh, xl, a, b) \
- __asm__ ("%@ Inlined umul_ppmm\n" \
- "umull %r1, %r0, %r2, %r3" \
- : "=&r" ((USItype)(xh)), \
- "=r" ((USItype)(xl)) \
- : "r" ((USItype)(a)), \
- "r" ((USItype)(b)) \
- : "r0", "r1")
-#endif
-#define UMUL_TIME 20
-#define UDIV_TIME 100
-#endif /* __arm__ */
-
-/***************************************
- ************** CLIPPER **************
- ***************************************/
-#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__ */
-
-
-/***************************************
- ************** GMICRO ***************
- ***************************************/
-#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
-
-
-/***************************************
- ************** HPPA *****************
- ***************************************/
-#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" \
- : "=r" ((USItype)(sh)), \
- "=&r" ((USItype)(sl)) \
- : "%rM" ((USItype)(ah)), \
- "rM" ((USItype)(bh)), \
- "%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" \
- : "=r" ((USItype)(sh)), \
- "=&r" ((USItype)(sl)) \
- : "rM" ((USItype)(ah)), \
- "rM" ((USItype)(bh)), \
- "rM" ((USItype)(al)), \
- "rM" ((USItype)(bl)))
-#if defined (_PA_RISC1_1)
-#define umul_ppmm(wh, wl, u, v) \
- do { \
- union {UDItype __ll; \
- struct {USItype __h, __l;} __i; \
- } __xx; \
- __asm__ (" xmpyu %1,%2,%0" \
- : "=*f" (__xx.__ll) \
- : "*f" ((USItype)(u)), \
- "*f" ((USItype)(v))); \
- (wh) = __xx.__i.__h; \
- (wl) = __xx.__i.__l; \
- } while (0)
-#define UMUL_TIME 8
-#define UDIV_TIME 60
-#else
-#define UMUL_TIME 40
-#define UDIV_TIME 80
-#endif
-#ifndef LONGLONG_STANDALONE
-#define udiv_qrnnd(q, r, n1, n0, d) \
- do { USItype __r; \
- (q) = __udiv_qrnnd (&__r, (n1), (n0), (d)); \
- (r) = __r; \
- } while (0)
-extern USItype __udiv_qrnnd ();
-#endif /* LONGLONG_STANDALONE */
-#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)); \
- } while (0)
-#endif /* hppa */
-
-
-/***************************************
- ************** I370 *****************
- ***************************************/
-#if (defined (__i370__) || defined (__mvs__)) && W_TYPE_SIZE == 32
-#define umul_ppmm(xh, xl, m0, m1) \
- do { \
- union {UDItype __ll; \
- struct {USItype __h, __l;} __i; \
- } __xx; \
- USItype __m0 = (m0), __m1 = (m1); \
- __asm__ ("mr %0,%3" \
- : "=r" (__xx.__i.__h), \
- "=r" (__xx.__i.__l) \
- : "%1" (__m0), \
- "r" (__m1)); \
- (xh) = __xx.__i.__h; (xl) = __xx.__i.__l; \
- (xh) += ((((SItype) __m0 >> 31) & __m1) \
- + (((SItype) __m1 >> 31) & __m0)); \
- } while (0)
-#define smul_ppmm(xh, xl, m0, m1) \
- do { \
- union {DItype __ll; \
- struct {USItype __h, __l;} __i; \
- } __xx; \
- __asm__ ("mr %0,%3" \
- : "=r" (__xx.__i.__h), \
- "=r" (__xx.__i.__l) \
- : "%1" (m0), \
- "r" (m1)); \
- (xh) = __xx.__i.__h; (xl) = __xx.__i.__l; \
- } while (0)
-#define sdiv_qrnnd(q, r, n1, n0, d) \
- do { \
- union {DItype __ll; \
- struct {USItype __h, __l;} __i; \
- } __xx; \
- __xx.__i.__h = n1; __xx.__i.__l = n0; \
- __asm__ ("dr %0,%2" \
- : "=r" (__xx.__ll) \
- : "0" (__xx.__ll), "r" (d)); \
- (q) = __xx.__i.__l; (r) = __xx.__i.__h; \
- } while (0)
-#endif
-
-
-/***************************************
- ************** I386 *****************
- ***************************************/
-#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" \
- : "=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__ ("subl %5,%1\n" \
- "sbbl %3,%0" \
- : "=r" ((USItype)(sh)), \
- "=&r" ((USItype)(sl)) \
- : "0" ((USItype)(ah)), \
- "g" ((USItype)(bh)), \
- "1" ((USItype)(al)), \
- "g" ((USItype)(bl)))
-#define umul_ppmm(w1, w0, u, v) \
- __asm__ ("mull %3" \
- : "=a" ((USItype)(w0)), \
- "=d" ((USItype)(w1)) \
- : "%0" ((USItype)(u)), \
- "rm" ((USItype)(v)))
-#define udiv_qrnnd(q, r, n1, n0, d) \
- __asm__ ("divl %4" \
- : "=a" ((USItype)(q)), \
- "=d" ((USItype)(r)) \
- : "0" ((USItype)(n0)), \
- "1" ((USItype)(n1)), \
- "rm" ((USItype)(d)))
-#define count_leading_zeros(count, x) \
- do { \
- USItype __cbtmp; \
- __asm__ ("bsrl %1,%0" \
- : "=r" (__cbtmp) : "rm" ((USItype)(x))); \
- (count) = __cbtmp ^ 31; \
- } while (0)
-#define count_trailing_zeros(count, x) \
- __asm__ ("bsfl %1,%0" : "=r" (count) : "rm" ((USItype)(x)))
-#ifndef UMUL_TIME
-#define UMUL_TIME 40
-#endif
-#ifndef UDIV_TIME
-#define UDIV_TIME 40
-#endif
-#endif /* 80x86 */
-
-
-/***************************************
- ************** I860 *****************
- ***************************************/
-#if defined (__i860__) && W_TYPE_SIZE == 32
-#define rshift_rhlc(r,h,l,c) \
- __asm__ ("shr %3,r0,r0\n" \
- "shrd %1,%2,%0" \
- "=r" (r) : "r" (h), "r" (l), "rn" (c))
-#endif /* i860 */
-
-/***************************************
- ************** I960 *****************
- ***************************************/
-#if defined (__i960__) && W_TYPE_SIZE == 32
-#define add_ssaaaa(sh, sl, ah, al, bh, bl) \
- __asm__ ("cmpo 1,0\n" \
- "addc %5,%4,%1\n" \
- "addc %3,%2,%0" \
- : "=r" ((USItype)(sh)), \
- "=&r" ((USItype)(sl)) \
- : "%dI" ((USItype)(ah)), \
- "dI" ((USItype)(bh)), \
- "%dI" ((USItype)(al)), \
- "dI" ((USItype)(bl)))
-#define sub_ddmmss(sh, sl, ah, al, bh, bl) \
- __asm__ ("cmpo 0,0\n" \
- "subc %5,%4,%1\n" \
- "subc %3,%2,%0" \
- : "=r" ((USItype)(sh)), \
- "=&r" ((USItype)(sl)) \
- : "dI" ((USItype)(ah)), \
- "dI" ((USItype)(bh)), \
- "dI" ((USItype)(al)), \
- "dI" ((USItype)(bl)))
-#define umul_ppmm(w1, w0, u, v) \
- ({union {UDItype __ll; \
- struct {USItype __l, __h;} __i; \
- } __xx; \
- __asm__ ("emul %2,%1,%0" \
- : "=d" (__xx.__ll) \
- : "%dI" ((USItype)(u)), \
- "dI" ((USItype)(v))); \
- (w1) = __xx.__i.__h; (w0) = __xx.__i.__l;})
-#define __umulsidi3(u, v) \
- ({UDItype __w; \
- __asm__ ("emul %2,%1,%0" \
- : "=d" (__w) \
- : "%dI" ((USItype)(u)), \
- "dI" ((USItype)(v))); \
- __w; })
-#define udiv_qrnnd(q, r, nh, nl, d) \
- do { \
- union {UDItype __ll; \
- struct {USItype __l, __h;} __i; \
- } __nn; \
- __nn.__i.__h = (nh); __nn.__i.__l = (nl); \
- __asm__ ("ediv %d,%n,%0" \
- : "=d" (__rq.__ll) \
- : "dI" (__nn.__ll), \
- "dI" ((USItype)(d))); \
- (r) = __rq.__i.__l; (q) = __rq.__i.__h; \
- } while (0)
-#define count_leading_zeros(count, x) \
- do { \
- USItype __cbtmp; \
- __asm__ ("scanbit %1,%0" \
- : "=r" (__cbtmp) \
- : "r" ((USItype)(x))); \
- (count) = __cbtmp ^ 31; \
- } while (0)
-#define COUNT_LEADING_ZEROS_0 (-32) /* sic */
-#if defined (__i960mx) /* what is the proper symbol to test??? */
-#define rshift_rhlc(r,h,l,c) \
- do { \
- union {UDItype __ll; \
- struct {USItype __l, __h;} __i; \
- } __nn; \
- __nn.__i.__h = (h); __nn.__i.__l = (l); \
- __asm__ ("shre %2,%1,%0" \
- : "=d" (r) : "dI" (__nn.__ll), "dI" (c)); \
- }
-#endif /* i960mx */
-#endif /* i960 */
-
-
-/***************************************
- ************** 68000 ****************
- ***************************************/
-#if (defined (__mc68000__) || defined (__mc68020__) || defined (__NeXT__) || defined(mc68020)) && W_TYPE_SIZE == 32
-#define add_ssaaaa(sh, sl, ah, al, bh, bl) \
- __asm__ ("add%.l %5,%1\n" \
- "addx%.l %3,%0" \
- : "=d" ((USItype)(sh)), \
- "=&d" ((USItype)(sl)) \
- : "%0" ((USItype)(ah)), \
- "d" ((USItype)(bh)), \
- "%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" \
- : "=d" ((USItype)(sh)), \
- "=&d" ((USItype)(sl)) \
- : "0" ((USItype)(ah)), \
- "d" ((USItype)(bh)), \
- "1" ((USItype)(al)), \
- "g" ((USItype)(bl)))
-#if (defined (__mc68020__) || defined (__NeXT__) || defined(mc68020))
-#define umul_ppmm(w1, w0, u, v) \
- __asm__ ("mulu%.l %3,%1:%0" \
- : "=d" ((USItype)(w0)), \
- "=d" ((USItype)(w1)) \
- : "%0" ((USItype)(u)), \
- "dmi" ((USItype)(v)))
-#define UMUL_TIME 45
-#define udiv_qrnnd(q, r, n1, n0, d) \
- __asm__ ("divu%.l %4,%1:%0" \
- : "=d" ((USItype)(q)), \
- "=d" ((USItype)(r)) \
- : "0" ((USItype)(n0)), \
- "1" ((USItype)(n1)), \
- "dmi" ((USItype)(d)))
-#define UDIV_TIME 90
-#define sdiv_qrnnd(q, r, n1, n0, d) \
- __asm__ ("divs%.l %4,%1:%0" \
- : "=d" ((USItype)(q)), \
- "=d" ((USItype)(r)) \
- : "0" ((USItype)(n0)), \
- "1" ((USItype)(n1)), \
- "dmi" ((USItype)(d)))
-#define count_leading_zeros(count, x) \
- __asm__ ("bfffo %1{%b2:%b2},%0" \
- : "=d" ((USItype)(count)) \
- : "od" ((USItype)(x)), "n" (0))
-#define COUNT_LEADING_ZEROS_0 32
-#else /* not mc68020 */
-#define umul_ppmm(xh, xl, a, b) \
- do { USItype __umul_tmp1, __umul_tmp2; \
- __asm__ ("| Inlined umul_ppmm \n" \
- " move%.l %5,%3 \n" \
- " move%.l %2,%0 \n" \
- " move%.w %3,%1 \n" \
- " swap %3 \n" \
- " swap %0 \n" \
- " mulu %2,%1 \n" \
- " mulu %3,%0 \n" \
- " mulu %2,%3 \n" \
- " swap %2 \n" \
- " mulu %5,%2 \n" \
- " add%.l %3,%2 \n" \
- " jcc 1f \n" \
- " add%.l %#0x10000,%0 \n" \
- "1: move%.l %2,%3 \n" \
- " clr%.w %2 \n" \
- " swap %2 \n" \
- " swap %3 \n" \
- " clr%.w %3 \n" \
- " add%.l %3,%1 \n" \
- " addx%.l %2,%0 \n" \
- " | End inlined umul_ppmm" \
- : "=&d" ((USItype)(xh)), "=&d" ((USItype)(xl)), \
- "=d" (__umul_tmp1), "=&d" (__umul_tmp2) \
- : "%2" ((USItype)(a)), "d" ((USItype)(b))); \
- } while (0)
-#define UMUL_TIME 100
-#define UDIV_TIME 400
-#endif /* not mc68020 */
-#endif /* mc68000 */
-
-
-/***************************************
- ************** 88000 ****************
- ***************************************/
-#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" \
- : "=r" ((USItype)(sh)), \
- "=&r" ((USItype)(sl)) \
- : "%rJ" ((USItype)(ah)), \
- "rJ" ((USItype)(bh)), \
- "%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" \
- : "=r" ((USItype)(sh)), \
- "=&r" ((USItype)(sl)) \
- : "rJ" ((USItype)(ah)), \
- "rJ" ((USItype)(bh)), \
- "rJ" ((USItype)(al)), \
- "rJ" ((USItype)(bl)))
-#define count_leading_zeros(count, x) \
- do { \
- USItype __cbtmp; \
- __asm__ ("ff1 %0,%1" \
- : "=r" (__cbtmp) \
- : "r" ((USItype)(x))); \
- (count) = __cbtmp ^ 31; \
- } while (0)
-#define COUNT_LEADING_ZEROS_0 63 /* sic */
-#if defined (__m88110__)
-#define umul_ppmm(wh, wl, u, v) \
- do { \
- union {UDItype __ll; \
- struct {USItype __h, __l;} __i; \
- } __x; \
- __asm__ ("mulu.d %0,%1,%2" : "=r" (__x.__ll) : "r" (u), "r" (v)); \
- (wh) = __x.__i.__h; \
- (wl) = __x.__i.__l; \
- } while (0)
-#define udiv_qrnnd(q, r, n1, n0, d) \
- ({union {UDItype __ll; \
- struct {USItype __h, __l;} __i; \
- } __x, __q; \
- __x.__i.__h = (n1); __x.__i.__l = (n0); \
- __asm__ ("divu.d %0,%1,%2" \
- : "=r" (__q.__ll) : "r" (__x.__ll), "r" (d)); \
- (r) = (n0) - __q.__l * (d); (q) = __q.__l; })
-#define UMUL_TIME 5
-#define UDIV_TIME 25
-#else
-#define UMUL_TIME 17
-#define UDIV_TIME 150
-#endif /* __m88110__ */
-#endif /* __m88000__ */
-
-/***************************************
- ************** MIPS *****************
- ***************************************/
-#if defined (__mips__) && W_TYPE_SIZE == 32
-#if (__GNUC__ >= 5) || (__GNUC__ == 4 && __GNUC_MINOR__ >= 4)
-#define umul_ppmm(w1, w0, u, v) \
- do { \
- UDItype _r; \
- _r = (UDItype) u * v; \
- (w1) = _r >> 32; \
- (w0) = (USItype) _r; \
- } while (0)
-#elif __GNUC__ > 2 || __GNUC_MINOR__ >= 7
-#define umul_ppmm(w1, w0, u, v) \
- __asm__ ("multu %2,%3" \
- : "=l" ((USItype)(w0)), \
- "=h" ((USItype)(w1)) \
- : "d" ((USItype)(u)), \
- "d" ((USItype)(v)))
-#else
-#define umul_ppmm(w1, w0, u, v) \
- __asm__ ("multu %2,%3 \n" \
- "mflo %0 \n" \
- "mfhi %1" \
- : "=d" ((USItype)(w0)), \
- "=d" ((USItype)(w1)) \
- : "d" ((USItype)(u)), \
- "d" ((USItype)(v)))
-#endif
-#define UMUL_TIME 10
-#define UDIV_TIME 100
-#endif /* __mips__ */
-
-/***************************************
- ************** MIPS/64 **************
- ***************************************/
-#if (defined (__mips) && __mips >= 3) && W_TYPE_SIZE == 64
-#if (__GNUC__ >= 5) || (__GNUC__ == 4 && __GNUC_MINOR__ >= 4)
-typedef unsigned int UTItype __attribute__ ((mode (TI)));
-#define umul_ppmm(w1, w0, u, v) \
- do { \
- UTItype _r; \
- _r = (UTItype) u * v; \
- (w1) = _r >> 64; \
- (w0) = (UDItype) _r; \
- } while (0)
-#elif __GNUC__ > 2 || __GNUC_MINOR__ >= 7
-#define umul_ppmm(w1, w0, u, v) \
- __asm__ ("dmultu %2,%3" \
- : "=l" ((UDItype)(w0)), \
- "=h" ((UDItype)(w1)) \
- : "d" ((UDItype)(u)), \
- "d" ((UDItype)(v)))
-#else
-#define umul_ppmm(w1, w0, u, v) \
- __asm__ ("dmultu %2,%3 \n" \
- "mflo %0 \n" \
- "mfhi %1" \
- : "=d" ((UDItype)(w0)), \
- "=d" ((UDItype)(w1)) \
- : "d" ((UDItype)(u)), \
- "d" ((UDItype)(v)))
-#endif
-#define UMUL_TIME 20
-#define UDIV_TIME 140
-#endif /* __mips__ */
-
-
-/***************************************
- ************** 32000 ****************
- ***************************************/
-#if defined (__ns32000__) && W_TYPE_SIZE == 32
-#define umul_ppmm(w1, w0, u, v) \
- ({union {UDItype __ll; \
- struct {USItype __l, __h;} __i; \
- } __xx; \
- __asm__ ("meid %2,%0" \
- : "=g" (__xx.__ll) \
- : "%0" ((USItype)(u)), \
- "g" ((USItype)(v))); \
- (w1) = __xx.__i.__h; (w0) = __xx.__i.__l;})
-#define __umulsidi3(u, v) \
- ({UDItype __w; \
- __asm__ ("meid %2,%0" \
- : "=g" (__w) \
- : "%0" ((USItype)(u)), \
- "g" ((USItype)(v))); \
- __w; })
-#define udiv_qrnnd(q, r, n1, n0, d) \
- ({union {UDItype __ll; \
- struct {USItype __l, __h;} __i; \
- } __xx; \
- __xx.__i.__h = (n1); __xx.__i.__l = (n0); \
- __asm__ ("deid %2,%0" \
- : "=g" (__xx.__ll) \
- : "0" (__xx.__ll), \
- "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))); \
- } while (0)
-#endif /* __ns32000__ */
-
-
-/***************************************
- ************** PPC ******************
- ***************************************/
-#if (defined (_ARCH_PPC) || defined (_IBMR2)) && 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) \
- __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))); \
- 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))); \
- } 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) \
- __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))); \
- 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) \
- __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))); \
- 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))); \
- } while (0)
-#define count_leading_zeros(count, x) \
- __asm__ ("{cntlz|cntlzw} %0,%1" \
- : "=r" ((USItype)(count)) \
- : "r" ((USItype)(x)))
-#define COUNT_LEADING_ZEROS_0 32
-#if defined (_ARCH_PPC)
-#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)); \
- (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)); \
- (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)
-#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))
-#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)))
-#define UDIV_TIME 100
-#endif
-#endif /* Power architecture variants. */
-
-/* Powerpc 64 bit support taken from gmp-4.1.2. */
-/* We should test _IBMR2 here when we add assembly support for the system
- vendor compilers. */
-#if 0 /* Not yet enabled because we don't have hardware for a test. */
-#if (defined (_ARCH_PPC) || defined (__powerpc__)) && 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" (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" (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" (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" (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" (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" (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" (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" (sh), "=&r" (sl) \
- : "r" (ah), "r" (bh), "rI" (al), "r" (bl)); \
- } while (0)
-#define count_leading_zeros(count, 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" (ph) : "%r" (m0), "r" (m1)); \
- (pl) = __m0 * __m1; \
- } while (0)
-#define UMUL_TIME 15
-#define smul_ppmm(ph, pl, m0, m1) \
- do { \
- DItype __m0 = (m0), __m1 = (m1); \
- __asm__ ("mulhd %0,%1,%2" : "=r" (ph) : "%r" (m0), "r" (m1)); \
- (pl) = __m0 * __m1; \
- } while (0)
-#define SMUL_TIME 14 /* ??? */
-#define UDIV_TIME 120 /* ??? */
-#endif /* 64-bit PowerPC. */
-#endif /* if 0 */
-
-/***************************************
- ************** PYR ******************
- ***************************************/
-#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__ */
-
-
-/***************************************
- ************** RT/ROMP **************
- ***************************************/
-#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" \
- : "=r" ((USItype)(sh)), \
- "=&r" ((USItype)(sl)) \
- : "%0" ((USItype)(ah)), \
- "r" ((USItype)(bh)), \
- "%1" ((USItype)(al)), \
- "r" ((USItype)(bl)))
-#define sub_ddmmss(sh, sl, ah, al, bh, bl) \
- __asm__ ("s %1,%5\n" \
- "se %0,%3" \
- : "=r" ((USItype)(sh)), \
- "=&r" ((USItype)(sl)) \
- : "0" ((USItype)(ah)), \
- "r" ((USItype)(bh)), \
- "1" ((USItype)(al)), \
- "r" ((USItype)(bl)))
-#define umul_ppmm(ph, pl, m0, m1) \
- 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" \
- : "=r" ((USItype)(ph)), \
- "=r" ((USItype)(pl)) \
- : "%r" (__m0), \
- "r" (__m1) \
- : "r2"); \
- (ph) += ((((SItype) __m0 >> 31) & __m1) \
- + (((SItype) __m1 >> 31) & __m0)); \
- } while (0)
-#define UMUL_TIME 20
-#define UDIV_TIME 200
-#define count_leading_zeros(count, x) \
- do { \
- if ((x) >= 0x10000) \
- __asm__ ("clz %0,%1" \
- : "=r" ((USItype)(count)) \
- : "r" ((USItype)(x) >> 16)); \
- else \
- { \
- __asm__ ("clz %0,%1" \
- : "=r" ((USItype)(count)) \
- : "r" ((USItype)(x))); \
- (count) += 16; \
- } \
- } while (0)
-#endif /* RT/ROMP */
-
-
-/***************************************
- ************** SH2 ******************
- ***************************************/
-#if (defined (__sh2__) || defined(__sh3__) || defined(__SH4__) ) \
- && W_TYPE_SIZE == 32
-#define umul_ppmm(w1, w0, u, v) \
- __asm__ ( \
- "dmulu.l %2,%3\n" \
- "sts macl,%1\n" \
- "sts mach,%0" \
- : "=r" ((USItype)(w1)), \
- "=r" ((USItype)(w0)) \
- : "r" ((USItype)(u)), \
- "r" ((USItype)(v)) \
- : "macl", "mach")
-#define UMUL_TIME 5
-#endif
-
-/***************************************
- ************** SPARC ****************
- ***************************************/
-#if defined (__sparc__) && W_TYPE_SIZE == 32
-#define add_ssaaaa(sh, sl, ah, al, bh, bl) \
- __asm__ ("addcc %r4,%5,%1\n" \
- "addx %r2,%3,%0" \
- : "=r" ((USItype)(sh)), \
- "=&r" ((USItype)(sl)) \
- : "%rJ" ((USItype)(ah)), \
- "rI" ((USItype)(bh)), \
- "%rJ" ((USItype)(al)), \
- "rI" ((USItype)(bl)) \
- __CLOBBER_CC)
-#define sub_ddmmss(sh, sl, ah, al, bh, bl) \
- __asm__ ("subcc %r4,%5,%1\n" \
- "subx %r2,%3,%0" \
- : "=r" ((USItype)(sh)), \
- "=&r" ((USItype)(sl)) \
- : "rJ" ((USItype)(ah)), \
- "rI" ((USItype)(bh)), \
- "rJ" ((USItype)(al)), \
- "rI" ((USItype)(bl)) \
- __CLOBBER_CC)
-#if defined (__sparc_v8__)
-/* Don't match immediate range because, 1) it is not often useful,
- 2) the 'I' flag thinks of the range as a 13 bit signed interval,
- while we want to match a 13 bit interval, sign extended to 32 bits,
- but INTERPRETED AS UNSIGNED. */
-#define umul_ppmm(w1, w0, u, v) \
- __asm__ ("umul %2,%3,%1;rd %%y,%0" \
- : "=r" ((USItype)(w1)), \
- "=r" ((USItype)(w0)) \
- : "r" ((USItype)(u)), \
- "r" ((USItype)(v)))
-#define UMUL_TIME 5
-#ifndef SUPERSPARC /* SuperSPARC's udiv only handles 53 bit dividends */
-#define udiv_qrnnd(q, r, n1, n0, d) \
- do { \
- USItype __q; \
- __asm__ ("mov %1,%%y;nop;nop;nop;udiv %2,%3,%0" \
- : "=r" ((USItype)(__q)) \
- : "r" ((USItype)(n1)), \
- "r" ((USItype)(n0)), \
- "r" ((USItype)(d))); \
- (r) = (n0) - __q * (d); \
- (q) = __q; \
- } while (0)
-#define UDIV_TIME 25
-#endif /* SUPERSPARC */
-#else /* ! __sparc_v8__ */
-#if defined (__sparclite__)
-/* This has hardware multiply but not divide. It also has two additional
- instructions scan (ffs from high bit) and divscc. */
-#define umul_ppmm(w1, w0, u, v) \
- __asm__ ("umul %2,%3,%1;rd %%y,%0" \
- : "=r" ((USItype)(w1)), \
- "=r" ((USItype)(w0)) \
- : "r" ((USItype)(u)), \
- "r" ((USItype)(v)))
-#define UMUL_TIME 5
-#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" \
- "1: ! End of inline udiv_qrnnd" \
- : "=r" ((USItype)(q)), \
- "=r" ((USItype)(r)) \
- : "r" ((USItype)(n1)), \
- "r" ((USItype)(n0)), \
- "rI" ((USItype)(d)) \
- : "%g1" __AND_CLOBBER_CC)
-#define UDIV_TIME 37
-#define count_leading_zeros(count, x) \
- __asm__ ("scan %1,0,%0" \
- : "=r" ((USItype)(x)) \
- : "r" ((USItype)(count)))
-/* Early sparclites return 63 for an argument of 0, but they warn that future
- implementations might change this. Therefore, leave COUNT_LEADING_ZEROS_0
- undefined. */
-#endif /* __sparclite__ */
-#endif /* __sparc_v8__ */
-/* Default to sparc v7 versions of umul_ppmm and udiv_qrnnd. */
-#ifndef umul_ppmm
-#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,%%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" ((USItype)(w1)), \
- "=r" ((USItype)(w0)) \
- : "%rI" ((USItype)(u)), \
- "r" ((USItype)(v)) \
- : "%g1", "%g2" __AND_CLOBBER_CC)
-#define UMUL_TIME 39 /* 39 instructions */
-#endif
-#ifndef udiv_qrnnd
-#ifndef LONGLONG_STANDALONE
-#define udiv_qrnnd(q, r, n1, n0, d) \
- do { USItype __r; \
- (q) = __udiv_qrnnd (&__r, (n1), (n0), (d)); \
- (r) = __r; \
- } while (0)
-extern USItype __udiv_qrnnd ();
-#define UDIV_TIME 140
-#endif /* LONGLONG_STANDALONE */
-#endif /* udiv_qrnnd */
-#endif /* __sparc__ */
-
-
-/***************************************
- ************** VAX ******************
- ***************************************/
-#if defined (__vax__) && W_TYPE_SIZE == 32
-#define add_ssaaaa(sh, sl, ah, al, bh, bl) \
- __asm__ ("addl2 %5,%1\n" \
- "adwc %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__ ("subl2 %5,%1\n" \
- "sbwc %3,%0" \
- : "=g" ((USItype)(sh)), \
- "=&g" ((USItype)(sl)) \
- : "0" ((USItype)(ah)), \
- "g" ((USItype)(bh)), \
- "1" ((USItype)(al)), \
- "g" ((USItype)(bl)))
-#define umul_ppmm(xh, xl, m0, m1) \
- do { \
- union {UDItype __ll; \
- struct {USItype __l, __h;} __i; \
- } __xx; \
- USItype __m0 = (m0), __m1 = (m1); \
- __asm__ ("emul %1,%2,$0,%0" \
- : "=g" (__xx.__ll) \
- : "g" (__m0), \
- "g" (__m1)); \
- (xh) = __xx.__i.__h; (xl) = __xx.__i.__l; \
- (xh) += ((((SItype) __m0 >> 31) & __m1) \
- + (((SItype) __m1 >> 31) & __m0)); \
- } while (0)
-#define sdiv_qrnnd(q, r, n1, n0, d) \
- do { \
- union {DItype __ll; \
- struct {SItype __l, __h;} __i; \
- } __xx; \
- __xx.__i.__h = n1; __xx.__i.__l = n0; \
- __asm__ ("ediv %3,%2,%0,%1" \
- : "=g" (q), "=g" (r) \
- : "g" (__xx.__ll), "g" (d)); \
- } while (0)
-#endif /* __vax__ */
-
-
-/***************************************
- ************** Z8000 ****************
- ***************************************/
-#if defined (__z8000__) && W_TYPE_SIZE == 16
-#define add_ssaaaa(sh, sl, ah, al, bh, bl) \
- __asm__ ("add %H1,%H5\n\tadc %H0,%H3" \
- : "=r" ((unsigned int)(sh)), \
- "=&r" ((unsigned int)(sl)) \
- : "%0" ((unsigned int)(ah)), \
- "r" ((unsigned int)(bh)), \
- "%1" ((unsigned int)(al)), \
- "rQR" ((unsigned int)(bl)))
-#define sub_ddmmss(sh, sl, ah, al, bh, bl) \
- __asm__ ("sub %H1,%H5\n\tsbc %H0,%H3" \
- : "=r" ((unsigned int)(sh)), \
- "=&r" ((unsigned int)(sl)) \
- : "0" ((unsigned int)(ah)), \
- "r" ((unsigned int)(bh)), \
- "1" ((unsigned int)(al)), \
- "rQR" ((unsigned int)(bl)))
-#define umul_ppmm(xh, xl, m0, m1) \
- do { \
- union {long int __ll; \
- struct {unsigned int __h, __l;} __i; \
- } __xx; \
- unsigned int __m0 = (m0), __m1 = (m1); \
- __asm__ ("mult %S0,%H3" \
- : "=r" (__xx.__i.__h), \
- "=r" (__xx.__i.__l) \
- : "%1" (__m0), \
- "rQR" (__m1)); \
- (xh) = __xx.__i.__h; (xl) = __xx.__i.__l; \
- (xh) += ((((signed int) __m0 >> 15) & __m1) \
- + (((signed int) __m1 >> 15) & __m0)); \
- } while (0)
-#endif /* __z8000__ */
-
-#endif /* __GNUC__ */
-#endif /* !__riscos__ */
-
-
-/***************************************
- *********** Generic Versions ********
- ***************************************/
-#if !defined (umul_ppmm) && defined (__umulsidi3)
-#define umul_ppmm(ph, pl, m0, m1) \
- { \
- UDWtype __ll = __umulsidi3 (m0, m1); \
- ph = (UWtype) (__ll >> W_TYPE_SIZE); \
- pl = (UWtype) __ll; \
- }
-#endif
-
-#if !defined (__umulsidi3)
-#define __umulsidi3(u, v) \
- ({UWtype __hi, __lo; \
- umul_ppmm (__hi, __lo, u, v); \
- ((UDWtype) __hi << W_TYPE_SIZE) | __lo; })
-#endif
-
-/* If this machine has no inline assembler, use C macros. */
-
-#if !defined (add_ssaaaa)
-#define add_ssaaaa(sh, sl, ah, al, bh, bl) \
- do { \
- UWtype __x; \
- __x = (al) + (bl); \
- (sh) = (ah) + (bh) + (__x < (al)); \
- (sl) = __x; \
- } while (0)
-#endif
-
-#if !defined (sub_ddmmss)
-#define sub_ddmmss(sh, sl, ah, al, bh, bl) \
- do { \
- UWtype __x; \
- __x = (al) - (bl); \
- (sh) = (ah) - (bh) - (__x > (al)); \
- (sl) = __x; \
- } while (0)
-#endif
-
-#if !defined (umul_ppmm)
-#define umul_ppmm(w1, w0, u, v) \
- do { \
- UWtype __x0, __x1, __x2, __x3; \
- UHWtype __ul, __vl, __uh, __vh; \
- UWtype __u = (u), __v = (v); \
- \
- __ul = __ll_lowpart (__u); \
- __uh = __ll_highpart (__u); \
- __vl = __ll_lowpart (__v); \
- __vh = __ll_highpart (__v); \
- \
- __x0 = (UWtype) __ul * __vl; \
- __x1 = (UWtype) __ul * __vh; \
- __x2 = (UWtype) __uh * __vl; \
- __x3 = (UWtype) __uh * __vh; \
- \
- __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. */ \
- \
- (w1) = __x3 + __ll_highpart (__x1); \
- (w0) = (__ll_lowpart (__x1) << W_TYPE_SIZE/2) + __ll_lowpart (__x0);\
- } while (0)
-#endif
-
-#if !defined (umul_ppmm)
-#define smul_ppmm(w1, w0, u, v) \
- do { \
- UWtype __w1; \
- UWtype __m0 = (u), __m1 = (v); \
- umul_ppmm (__w1, w0, __m0, __m1); \
- (w1) = __w1 - (-(__m0 >> (W_TYPE_SIZE - 1)) & __m1) \
- - (-(__m1 >> (W_TYPE_SIZE - 1)) & __m0); \
- } while (0)
-#endif
-
-/* Define this unconditionally, so it can be used for debugging. */
-#define __udiv_qrnnd_c(q, r, n1, n0, d) \
- do { \
- UWtype __d1, __d0, __q1, __q0, __r1, __r0, __m; \
- __d1 = __ll_highpart (d); \
- __d0 = __ll_lowpart (d); \
- \
- __r1 = (n1) % __d1; \
- __q1 = (n1) / __d1; \
- __m = (UWtype) __q1 * __d0; \
- __r1 = __r1 * __ll_B | __ll_highpart (n0); \
- if (__r1 < __m) \
- { \
- __q1--, __r1 += (d); \
- if (__r1 >= (d)) /* i.e. we didn't get carry when adding to __r1 */\
- if (__r1 < __m) \
- __q1--, __r1 += (d); \
- } \
- __r1 -= __m; \
- \
- __r0 = __r1 % __d1; \
- __q0 = __r1 / __d1; \
- __m = (UWtype) __q0 * __d0; \
- __r0 = __r0 * __ll_B | __ll_lowpart (n0); \
- if (__r0 < __m) \
- { \
- __q0--, __r0 += (d); \
- if (__r0 >= (d)) \
- if (__r0 < __m) \
- __q0--, __r0 += (d); \
- } \
- __r0 -= __m; \
- \
- (q) = (UWtype) __q1 * __ll_B | __q0; \
- (r) = __r0; \
- } while (0)
-
-/* If the processor has no udiv_qrnnd but sdiv_qrnnd, go through
- __udiv_w_sdiv (defined in libgcc or elsewhere). */
-#if !defined (udiv_qrnnd) && defined (sdiv_qrnnd)
-#define udiv_qrnnd(q, r, nh, nl, d) \
- do { \
- UWtype __r; \
- (q) = __MPN(udiv_w_sdiv) (&__r, nh, nl, d); \
- (r) = __r; \
- } while (0)
-#endif
-
-/* If udiv_qrnnd was not defined for this processor, use __udiv_qrnnd_c. */
-#if !defined (udiv_qrnnd)
-#define UDIV_NEEDS_NORMALIZATION 1
-#define udiv_qrnnd __udiv_qrnnd_c
-#endif
-
-#if !defined (count_leading_zeros)
-extern
-#ifdef __STDC__
-const
-#endif
-unsigned char _gcry_clz_tab[];
-#define MPI_INTERNAL_NEED_CLZ_TAB 1
-#define count_leading_zeros(count, x) \
- do { \
- UWtype __xr = (x); \
- UWtype __a; \
- \
- if (W_TYPE_SIZE <= 32) \
- { \
- __a = __xr < ((UWtype) 1 << 2*__BITS4) \
- ? (__xr < ((UWtype) 1 << __BITS4) ? 0 : __BITS4) \
- : (__xr < ((UWtype) 1 << 3*__BITS4) ? 2*__BITS4 : 3*__BITS4);\
- } \
- else \
- { \
- for (__a = W_TYPE_SIZE - 8; __a > 0; __a -= 8) \
- if (((__xr >> __a) & 0xff) != 0) \
- break; \
- } \
- \
- (count) = W_TYPE_SIZE - (_gcry_clz_tab[__xr >> __a] + __a); \
- } while (0)
-/* This version gives a well-defined value for zero. */
-#define COUNT_LEADING_ZEROS_0 W_TYPE_SIZE
-#endif
-
-#if !defined (count_trailing_zeros)
-/* Define count_trailing_zeros using count_leading_zeros. The latter might be
- defined in asm, but if it is not, the C version above is good enough. */
-#define count_trailing_zeros(count, x) \
- do { \
- UWtype __ctz_x = (x); \
- UWtype __ctz_c; \
- count_leading_zeros (__ctz_c, __ctz_x & -__ctz_x); \
- (count) = W_TYPE_SIZE - 1 - __ctz_c; \
- } while (0)
-#endif
-
-#ifndef UDIV_NEEDS_NORMALIZATION
-#define UDIV_NEEDS_NORMALIZATION 0
-#endif
+/* longlong.h -- definitions for mixed size 32/64 bit arithmetic. + Note: I added some stuff for use with gnupg + +Copyright (C) 1991, 1992, 1993, 1994, 1996, 1998, + 2000, 2001, 2002, 2003, 2004 Free Software Foundation, Inc. + +This file is free software; you can redistribute it and/or modify +it under the terms of the GNU Lesser General Public License as published by +the Free Software Foundation; either version 2.1 of the License, or (at your +option) any later version. + +This file is distributed in the hope that it will be useful, but +WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY +or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public +License for more details. + +You should have received a copy of the GNU Library General Public License +along with this file; see the file COPYING.LIB. If not, write to +the Free Software Foundation, Inc., 59 Temple Place - Suite 330, Boston, +MA 02111-1307, USA. */ + +/* You have to define the following before including this file: + + UWtype -- An unsigned type, default type for operations (typically a "word") + UHWtype -- An unsigned type, at least half the size of UWtype. + UDWtype -- An unsigned type, at least twice as large a UWtype + W_TYPE_SIZE -- size in bits of UWtype + + SItype, USItype -- Signed and unsigned 32 bit types. + DItype, UDItype -- Signed and unsigned 64 bit types. + + On a 32 bit machine UWtype should typically be USItype; + on a 64 bit machine, UWtype should typically be UDItype. +*/ + +#define __BITS4 (W_TYPE_SIZE / 4) +#define __ll_B ((UWtype) 1 << (W_TYPE_SIZE / 2)) +#define __ll_lowpart(t) ((UWtype) (t) & (__ll_B - 1)) +#define __ll_highpart(t) ((UWtype) (t) >> (W_TYPE_SIZE / 2)) + +/* This is used to make sure no undesirable sharing between different libraries + that use this file takes place. */ +#ifndef __MPN +#define __MPN(x) __##x +#endif + +/* Define auxiliary asm macros. + + 1) umul_ppmm(high_prod, low_prod, multipler, multiplicand) multiplies two + UWtype integers MULTIPLER and MULTIPLICAND, and generates a two UWtype + word product in HIGH_PROD and LOW_PROD. + + 2) __umulsidi3(a,b) multiplies two UWtype integers A and B, and returns a + UDWtype product. This is just a variant of umul_ppmm. + + 3) udiv_qrnnd(quotient, remainder, high_numerator, low_numerator, + denominator) divides a UDWtype, composed by the UWtype integers + HIGH_NUMERATOR and LOW_NUMERATOR, by DENOMINATOR and places the quotient + in QUOTIENT and the remainder in REMAINDER. HIGH_NUMERATOR must be less + than DENOMINATOR for correct operation. If, in addition, the most + significant bit of DENOMINATOR must be 1, then the pre-processor symbol + UDIV_NEEDS_NORMALIZATION is defined to 1. + + 4) sdiv_qrnnd(quotient, remainder, high_numerator, low_numerator, + denominator). Like udiv_qrnnd but the numbers are signed. The quotient + 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 + 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. + + 6) count_trailing_zeros(count, x) like count_leading_zeros, but counts + from the least significant end. + + 7) add_ssaaaa(high_sum, low_sum, high_addend_1, low_addend_1, + high_addend_2, low_addend_2) adds two UWtype integers, composed by + HIGH_ADDEND_1 and LOW_ADDEND_1, and HIGH_ADDEND_2 and LOW_ADDEND_2 + respectively. The result is placed in HIGH_SUM and LOW_SUM. Overflow + (i.e. carry out) is not stored anywhere, and is lost. + + 8) sub_ddmmss(high_difference, low_difference, high_minuend, low_minuend, + high_subtrahend, low_subtrahend) subtracts two two-word UWtype integers, + composed by HIGH_MINUEND_1 and LOW_MINUEND_1, and HIGH_SUBTRAHEND_2 and + LOW_SUBTRAHEND_2 respectively. The result is placed in HIGH_DIFFERENCE + and LOW_DIFFERENCE. Overflow (i.e. carry out) is not stored anywhere, + and is lost. + + If any of these macros are left undefined for a particular CPU, + C macros are used. */ + +/* The CPUs come in alphabetical order below. + + Please add support for more CPUs here, or improve the current support + for the CPUs below! */ + +#ifdef __riscos__ +#pragma continue_after_hash_error +#else /* !__riscos__ */ +#if defined (__GNUC__) && !defined (NO_ASM) + +/* We sometimes need to clobber "cc" with gcc2, but that would not be + understood by gcc1. Use cpp to avoid major code duplication. */ +#if __GNUC__ < 2 +#define __CLOBBER_CC +#define __AND_CLOBBER_CC +#else /* __GNUC__ >= 2 */ +#define __CLOBBER_CC : "cc" +#define __AND_CLOBBER_CC , "cc" +#endif /* __GNUC__ < 2 */ + + +/*************************************** + ************** A29K ***************** + ***************************************/ +#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 { \ + UDItype __m0 = (m0), __m1 = (m1); \ + __asm__ ("umulh %r1,%2,%0" \ + : "=r" ((UDItype) ph) \ + : "%rJ" (__m0), \ + "rI" (__m1)); \ + (pl) = __m0 * __m1; \ + } while (0) +#define UMUL_TIME 46 +#ifndef LONGLONG_STANDALONE +#define udiv_qrnnd(q, r, n1, n0, d) \ + do { UDItype __r; \ + (q) = __udiv_qrnnd (&__r, (n1), (n0), (d)); \ + (r) = __r; \ + } while (0) +extern UDItype __udiv_qrnnd (); +#define UDIV_TIME 220 +#endif /* LONGLONG_STANDALONE */ +#endif /* __alpha */ + +/*************************************** + ************** ARM ****************** + ***************************************/ +#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" \ + : "=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__ ("subs %1, %4, %5\n" \ + "sbc %0, %2, %3" \ + : "=r" ((USItype)(sh)), \ + "=&r" ((USItype)(sl)) \ + : "r" ((USItype)(ah)), \ + "rI" ((USItype)(bh)), \ + "r" ((USItype)(al)), \ + "rI" ((USItype)(bl))) +#if defined __ARM_ARCH_2__ || defined __ARM_ARCH_3__ +#define umul_ppmm(xh, xl, a, b) \ + __asm__ ("%@ Inlined umul_ppmm\n" \ + "mov %|r0, %2, lsr #16 @ AAAA\n" \ + "mov %|r2, %3, lsr #16 @ BBBB\n" \ + "bic %|r1, %2, %|r0, lsl #16 @ aaaa\n" \ + "bic %0, %3, %|r2, lsl #16 @ bbbb\n" \ + "mul %1, %|r1, %|r2 @ aaaa * BBBB\n" \ + "mul %|r2, %|r0, %|r2 @ AAAA * BBBB\n" \ + "mul %|r1, %0, %|r1 @ aaaa * bbbb\n" \ + "mul %0, %|r0, %0 @ AAAA * bbbb\n" \ + "adds %|r0, %1, %0 @ central sum\n" \ + "addcs %|r2, %|r2, #65536\n" \ + "adds %1, %|r1, %|r0, lsl #16\n" \ + "adc %0, %|r2, %|r0, lsr #16" \ + : "=&r" ((USItype)(xh)), \ + "=r" ((USItype)(xl)) \ + : "r" ((USItype)(a)), \ + "r" ((USItype)(b)) \ + : "r0", "r1", "r2") +#else +#define umul_ppmm(xh, xl, a, b) \ + __asm__ ("%@ Inlined umul_ppmm\n" \ + "umull %r1, %r0, %r2, %r3" \ + : "=&r" ((USItype)(xh)), \ + "=r" ((USItype)(xl)) \ + : "r" ((USItype)(a)), \ + "r" ((USItype)(b)) \ + : "r0", "r1") +#endif +#define UMUL_TIME 20 +#define UDIV_TIME 100 +#endif /* __arm__ */ + +/*************************************** + ************** CLIPPER ************** + ***************************************/ +#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__ */ + + +/*************************************** + ************** GMICRO *************** + ***************************************/ +#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 + + +/*************************************** + ************** HPPA ***************** + ***************************************/ +#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" \ + : "=r" ((USItype)(sh)), \ + "=&r" ((USItype)(sl)) \ + : "%rM" ((USItype)(ah)), \ + "rM" ((USItype)(bh)), \ + "%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" \ + : "=r" ((USItype)(sh)), \ + "=&r" ((USItype)(sl)) \ + : "rM" ((USItype)(ah)), \ + "rM" ((USItype)(bh)), \ + "rM" ((USItype)(al)), \ + "rM" ((USItype)(bl))) +#if defined (_PA_RISC1_1) +#define umul_ppmm(wh, wl, u, v) \ + do { \ + union {UDItype __ll; \ + struct {USItype __h, __l;} __i; \ + } __xx; \ + __asm__ (" xmpyu %1,%2,%0" \ + : "=*f" (__xx.__ll) \ + : "*f" ((USItype)(u)), \ + "*f" ((USItype)(v))); \ + (wh) = __xx.__i.__h; \ + (wl) = __xx.__i.__l; \ + } while (0) +#define UMUL_TIME 8 +#define UDIV_TIME 60 +#else +#define UMUL_TIME 40 +#define UDIV_TIME 80 +#endif +#ifndef LONGLONG_STANDALONE +#define udiv_qrnnd(q, r, n1, n0, d) \ + do { USItype __r; \ + (q) = __udiv_qrnnd (&__r, (n1), (n0), (d)); \ + (r) = __r; \ + } while (0) +extern USItype __udiv_qrnnd (); +#endif /* LONGLONG_STANDALONE */ +#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)); \ + } while (0) +#endif /* hppa */ + + +/*************************************** + ************** I370 ***************** + ***************************************/ +#if (defined (__i370__) || defined (__mvs__)) && W_TYPE_SIZE == 32 +#define umul_ppmm(xh, xl, m0, m1) \ + do { \ + union {UDItype __ll; \ + struct {USItype __h, __l;} __i; \ + } __xx; \ + USItype __m0 = (m0), __m1 = (m1); \ + __asm__ ("mr %0,%3" \ + : "=r" (__xx.__i.__h), \ + "=r" (__xx.__i.__l) \ + : "%1" (__m0), \ + "r" (__m1)); \ + (xh) = __xx.__i.__h; (xl) = __xx.__i.__l; \ + (xh) += ((((SItype) __m0 >> 31) & __m1) \ + + (((SItype) __m1 >> 31) & __m0)); \ + } while (0) +#define smul_ppmm(xh, xl, m0, m1) \ + do { \ + union {DItype __ll; \ + struct {USItype __h, __l;} __i; \ + } __xx; \ + __asm__ ("mr %0,%3" \ + : "=r" (__xx.__i.__h), \ + "=r" (__xx.__i.__l) \ + : "%1" (m0), \ + "r" (m1)); \ + (xh) = __xx.__i.__h; (xl) = __xx.__i.__l; \ + } while (0) +#define sdiv_qrnnd(q, r, n1, n0, d) \ + do { \ + union {DItype __ll; \ + struct {USItype __h, __l;} __i; \ + } __xx; \ + __xx.__i.__h = n1; __xx.__i.__l = n0; \ + __asm__ ("dr %0,%2" \ + : "=r" (__xx.__ll) \ + : "0" (__xx.__ll), "r" (d)); \ + (q) = __xx.__i.__l; (r) = __xx.__i.__h; \ + } while (0) +#endif + + +/*************************************** + ************** I386 ***************** + ***************************************/ +#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" \ + : "=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__ ("subl %5,%1\n" \ + "sbbl %3,%0" \ + : "=r" ((USItype)(sh)), \ + "=&r" ((USItype)(sl)) \ + : "0" ((USItype)(ah)), \ + "g" ((USItype)(bh)), \ + "1" ((USItype)(al)), \ + "g" ((USItype)(bl))) +#define umul_ppmm(w1, w0, u, v) \ + __asm__ ("mull %3" \ + : "=a" ((USItype)(w0)), \ + "=d" ((USItype)(w1)) \ + : "%0" ((USItype)(u)), \ + "rm" ((USItype)(v))) +#define udiv_qrnnd(q, r, n1, n0, d) \ + __asm__ ("divl %4" \ + : "=a" ((USItype)(q)), \ + "=d" ((USItype)(r)) \ + : "0" ((USItype)(n0)), \ + "1" ((USItype)(n1)), \ + "rm" ((USItype)(d))) +#define count_leading_zeros(count, x) \ + do { \ + USItype __cbtmp; \ + __asm__ ("bsrl %1,%0" \ + : "=r" (__cbtmp) : "rm" ((USItype)(x))); \ + (count) = __cbtmp ^ 31; \ + } while (0) +#define count_trailing_zeros(count, x) \ + __asm__ ("bsfl %1,%0" : "=r" (count) : "rm" ((USItype)(x))) +#ifndef UMUL_TIME +#define UMUL_TIME 40 +#endif +#ifndef UDIV_TIME +#define UDIV_TIME 40 +#endif +#endif /* 80x86 */ + + +/*************************************** + ************** I860 ***************** + ***************************************/ +#if defined (__i860__) && W_TYPE_SIZE == 32 +#define rshift_rhlc(r,h,l,c) \ + __asm__ ("shr %3,r0,r0\n" \ + "shrd %1,%2,%0" \ + "=r" (r) : "r" (h), "r" (l), "rn" (c)) +#endif /* i860 */ + +/*************************************** + ************** I960 ***************** + ***************************************/ +#if defined (__i960__) && W_TYPE_SIZE == 32 +#define add_ssaaaa(sh, sl, ah, al, bh, bl) \ + __asm__ ("cmpo 1,0\n" \ + "addc %5,%4,%1\n" \ + "addc %3,%2,%0" \ + : "=r" ((USItype)(sh)), \ + "=&r" ((USItype)(sl)) \ + : "%dI" ((USItype)(ah)), \ + "dI" ((USItype)(bh)), \ + "%dI" ((USItype)(al)), \ + "dI" ((USItype)(bl))) +#define sub_ddmmss(sh, sl, ah, al, bh, bl) \ + __asm__ ("cmpo 0,0\n" \ + "subc %5,%4,%1\n" \ + "subc %3,%2,%0" \ + : "=r" ((USItype)(sh)), \ + "=&r" ((USItype)(sl)) \ + : "dI" ((USItype)(ah)), \ + "dI" ((USItype)(bh)), \ + "dI" ((USItype)(al)), \ + "dI" ((USItype)(bl))) +#define umul_ppmm(w1, w0, u, v) \ + ({union {UDItype __ll; \ + struct {USItype __l, __h;} __i; \ + } __xx; \ + __asm__ ("emul %2,%1,%0" \ + : "=d" (__xx.__ll) \ + : "%dI" ((USItype)(u)), \ + "dI" ((USItype)(v))); \ + (w1) = __xx.__i.__h; (w0) = __xx.__i.__l;}) +#define __umulsidi3(u, v) \ + ({UDItype __w; \ + __asm__ ("emul %2,%1,%0" \ + : "=d" (__w) \ + : "%dI" ((USItype)(u)), \ + "dI" ((USItype)(v))); \ + __w; }) +#define udiv_qrnnd(q, r, nh, nl, d) \ + do { \ + union {UDItype __ll; \ + struct {USItype __l, __h;} __i; \ + } __nn; \ + __nn.__i.__h = (nh); __nn.__i.__l = (nl); \ + __asm__ ("ediv %d,%n,%0" \ + : "=d" (__rq.__ll) \ + : "dI" (__nn.__ll), \ + "dI" ((USItype)(d))); \ + (r) = __rq.__i.__l; (q) = __rq.__i.__h; \ + } while (0) +#define count_leading_zeros(count, x) \ + do { \ + USItype __cbtmp; \ + __asm__ ("scanbit %1,%0" \ + : "=r" (__cbtmp) \ + : "r" ((USItype)(x))); \ + (count) = __cbtmp ^ 31; \ + } while (0) +#define COUNT_LEADING_ZEROS_0 (-32) /* sic */ +#if defined (__i960mx) /* what is the proper symbol to test??? */ +#define rshift_rhlc(r,h,l,c) \ + do { \ + union {UDItype __ll; \ + struct {USItype __l, __h;} __i; \ + } __nn; \ + __nn.__i.__h = (h); __nn.__i.__l = (l); \ + __asm__ ("shre %2,%1,%0" \ + : "=d" (r) : "dI" (__nn.__ll), "dI" (c)); \ + } +#endif /* i960mx */ +#endif /* i960 */ + + +/*************************************** + ************** 68000 **************** + ***************************************/ +#if (defined (__mc68000__) || defined (__mc68020__) || defined (__NeXT__) || defined(mc68020)) && W_TYPE_SIZE == 32 +#define add_ssaaaa(sh, sl, ah, al, bh, bl) \ + __asm__ ("add%.l %5,%1\n" \ + "addx%.l %3,%0" \ + : "=d" ((USItype)(sh)), \ + "=&d" ((USItype)(sl)) \ + : "%0" ((USItype)(ah)), \ + "d" ((USItype)(bh)), \ + "%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" \ + : "=d" ((USItype)(sh)), \ + "=&d" ((USItype)(sl)) \ + : "0" ((USItype)(ah)), \ + "d" ((USItype)(bh)), \ + "1" ((USItype)(al)), \ + "g" ((USItype)(bl))) +#if (defined (__mc68020__) || defined (__NeXT__) || defined(mc68020)) +#define umul_ppmm(w1, w0, u, v) \ + __asm__ ("mulu%.l %3,%1:%0" \ + : "=d" ((USItype)(w0)), \ + "=d" ((USItype)(w1)) \ + : "%0" ((USItype)(u)), \ + "dmi" ((USItype)(v))) +#define UMUL_TIME 45 +#define udiv_qrnnd(q, r, n1, n0, d) \ + __asm__ ("divu%.l %4,%1:%0" \ + : "=d" ((USItype)(q)), \ + "=d" ((USItype)(r)) \ + : "0" ((USItype)(n0)), \ + "1" ((USItype)(n1)), \ + "dmi" ((USItype)(d))) +#define UDIV_TIME 90 +#define sdiv_qrnnd(q, r, n1, n0, d) \ + __asm__ ("divs%.l %4,%1:%0" \ + : "=d" ((USItype)(q)), \ + "=d" ((USItype)(r)) \ + : "0" ((USItype)(n0)), \ + "1" ((USItype)(n1)), \ + "dmi" ((USItype)(d))) +#define count_leading_zeros(count, x) \ + __asm__ ("bfffo %1{%b2:%b2},%0" \ + : "=d" ((USItype)(count)) \ + : "od" ((USItype)(x)), "n" (0)) +#define COUNT_LEADING_ZEROS_0 32 +#else /* not mc68020 */ +#define umul_ppmm(xh, xl, a, b) \ + do { USItype __umul_tmp1, __umul_tmp2; \ + __asm__ ("| Inlined umul_ppmm \n" \ + " move%.l %5,%3 \n" \ + " move%.l %2,%0 \n" \ + " move%.w %3,%1 \n" \ + " swap %3 \n" \ + " swap %0 \n" \ + " mulu %2,%1 \n" \ + " mulu %3,%0 \n" \ + " mulu %2,%3 \n" \ + " swap %2 \n" \ + " mulu %5,%2 \n" \ + " add%.l %3,%2 \n" \ + " jcc 1f \n" \ + " add%.l %#0x10000,%0 \n" \ + "1: move%.l %2,%3 \n" \ + " clr%.w %2 \n" \ + " swap %2 \n" \ + " swap %3 \n" \ + " clr%.w %3 \n" \ + " add%.l %3,%1 \n" \ + " addx%.l %2,%0 \n" \ + " | End inlined umul_ppmm" \ + : "=&d" ((USItype)(xh)), "=&d" ((USItype)(xl)), \ + "=d" (__umul_tmp1), "=&d" (__umul_tmp2) \ + : "%2" ((USItype)(a)), "d" ((USItype)(b))); \ + } while (0) +#define UMUL_TIME 100 +#define UDIV_TIME 400 +#endif /* not mc68020 */ +#endif /* mc68000 */ + + +/*************************************** + ************** 88000 **************** + ***************************************/ +#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" \ + : "=r" ((USItype)(sh)), \ + "=&r" ((USItype)(sl)) \ + : "%rJ" ((USItype)(ah)), \ + "rJ" ((USItype)(bh)), \ + "%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" \ + : "=r" ((USItype)(sh)), \ + "=&r" ((USItype)(sl)) \ + : "rJ" ((USItype)(ah)), \ + "rJ" ((USItype)(bh)), \ + "rJ" ((USItype)(al)), \ + "rJ" ((USItype)(bl))) +#define count_leading_zeros(count, x) \ + do { \ + USItype __cbtmp; \ + __asm__ ("ff1 %0,%1" \ + : "=r" (__cbtmp) \ + : "r" ((USItype)(x))); \ + (count) = __cbtmp ^ 31; \ + } while (0) +#define COUNT_LEADING_ZEROS_0 63 /* sic */ +#if defined (__m88110__) +#define umul_ppmm(wh, wl, u, v) \ + do { \ + union {UDItype __ll; \ + struct {USItype __h, __l;} __i; \ + } __x; \ + __asm__ ("mulu.d %0,%1,%2" : "=r" (__x.__ll) : "r" (u), "r" (v)); \ + (wh) = __x.__i.__h; \ + (wl) = __x.__i.__l; \ + } while (0) +#define udiv_qrnnd(q, r, n1, n0, d) \ + ({union {UDItype __ll; \ + struct {USItype __h, __l;} __i; \ + } __x, __q; \ + __x.__i.__h = (n1); __x.__i.__l = (n0); \ + __asm__ ("divu.d %0,%1,%2" \ + : "=r" (__q.__ll) : "r" (__x.__ll), "r" (d)); \ + (r) = (n0) - __q.__l * (d); (q) = __q.__l; }) +#define UMUL_TIME 5 +#define UDIV_TIME 25 +#else +#define UMUL_TIME 17 +#define UDIV_TIME 150 +#endif /* __m88110__ */ +#endif /* __m88000__ */ + +/*************************************** + ************** MIPS ***************** + ***************************************/ +#if defined (__mips__) && W_TYPE_SIZE == 32 +#if (__GNUC__ >= 5) || (__GNUC__ == 4 && __GNUC_MINOR__ >= 4) +#define umul_ppmm(w1, w0, u, v) \ + do { \ + UDItype _r; \ + _r = (UDItype) u * v; \ + (w1) = _r >> 32; \ + (w0) = (USItype) _r; \ + } while (0) +#elif __GNUC__ > 2 || __GNUC_MINOR__ >= 7 +#define umul_ppmm(w1, w0, u, v) \ + __asm__ ("multu %2,%3" \ + : "=l" ((USItype)(w0)), \ + "=h" ((USItype)(w1)) \ + : "d" ((USItype)(u)), \ + "d" ((USItype)(v))) +#else +#define umul_ppmm(w1, w0, u, v) \ + __asm__ ("multu %2,%3 \n" \ + "mflo %0 \n" \ + "mfhi %1" \ + : "=d" ((USItype)(w0)), \ + "=d" ((USItype)(w1)) \ + : "d" ((USItype)(u)), \ + "d" ((USItype)(v))) +#endif +#define UMUL_TIME 10 +#define UDIV_TIME 100 +#endif /* __mips__ */ + +/*************************************** + ************** MIPS/64 ************** + ***************************************/ +#if (defined (__mips) && __mips >= 3) && W_TYPE_SIZE == 64 +#if (__GNUC__ >= 5) || (__GNUC__ == 4 && __GNUC_MINOR__ >= 4) +typedef unsigned int UTItype __attribute__ ((mode (TI))); +#define umul_ppmm(w1, w0, u, v) \ + do { \ + UTItype _r; \ + _r = (UTItype) u * v; \ + (w1) = _r >> 64; \ + (w0) = (UDItype) _r; \ + } while (0) +#elif __GNUC__ > 2 || __GNUC_MINOR__ >= 7 +#define umul_ppmm(w1, w0, u, v) \ + __asm__ ("dmultu %2,%3" \ + : "=l" ((UDItype)(w0)), \ + "=h" ((UDItype)(w1)) \ + : "d" ((UDItype)(u)), \ + "d" ((UDItype)(v))) +#else +#define umul_ppmm(w1, w0, u, v) \ + __asm__ ("dmultu %2,%3 \n" \ + "mflo %0 \n" \ + "mfhi %1" \ + : "=d" ((UDItype)(w0)), \ + "=d" ((UDItype)(w1)) \ + : "d" ((UDItype)(u)), \ + "d" ((UDItype)(v))) +#endif +#define UMUL_TIME 20 +#define UDIV_TIME 140 +#endif /* __mips__ */ + + +/*************************************** + ************** 32000 **************** + ***************************************/ +#if defined (__ns32000__) && W_TYPE_SIZE == 32 +#define umul_ppmm(w1, w0, u, v) \ + ({union {UDItype __ll; \ + struct {USItype __l, __h;} __i; \ + } __xx; \ + __asm__ ("meid %2,%0" \ + : "=g" (__xx.__ll) \ + : "%0" ((USItype)(u)), \ + "g" ((USItype)(v))); \ + (w1) = __xx.__i.__h; (w0) = __xx.__i.__l;}) +#define __umulsidi3(u, v) \ + ({UDItype __w; \ + __asm__ ("meid %2,%0" \ + : "=g" (__w) \ + : "%0" ((USItype)(u)), \ + "g" ((USItype)(v))); \ + __w; }) +#define udiv_qrnnd(q, r, n1, n0, d) \ + ({union {UDItype __ll; \ + struct {USItype __l, __h;} __i; \ + } __xx; \ + __xx.__i.__h = (n1); __xx.__i.__l = (n0); \ + __asm__ ("deid %2,%0" \ + : "=g" (__xx.__ll) \ + : "0" (__xx.__ll), \ + "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))); \ + } while (0) +#endif /* __ns32000__ */ + + +/*************************************** + ************** PPC ****************** + ***************************************/ +#if (defined (_ARCH_PPC) || defined (_IBMR2)) && 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) \ + __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))); \ + 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))); \ + } 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) \ + __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))); \ + 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) \ + __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))); \ + 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))); \ + } while (0) +#define count_leading_zeros(count, x) \ + __asm__ ("{cntlz|cntlzw} %0,%1" \ + : "=r" ((USItype)(count)) \ + : "r" ((USItype)(x))) +#define COUNT_LEADING_ZEROS_0 32 +#if defined (_ARCH_PPC) +#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)); \ + (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)); \ + (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) +#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)) +#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))) +#define UDIV_TIME 100 +#endif +#endif /* Power architecture variants. */ + +/* Powerpc 64 bit support taken from gmp-4.1.2. */ +/* We should test _IBMR2 here when we add assembly support for the system + vendor compilers. */ +#if 0 /* Not yet enabled becuase we don't have hardware for a test. */ +#if (defined (_ARCH_PPC) || defined (__powerpc__)) && 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" (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" (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" (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" (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" (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" (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" (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" (sh), "=&r" (sl) \ + : "r" (ah), "r" (bh), "rI" (al), "r" (bl)); \ + } while (0) +#define count_leading_zeros(count, 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" (ph) : "%r" (m0), "r" (m1)); \ + (pl) = __m0 * __m1; \ + } while (0) +#define UMUL_TIME 15 +#define smul_ppmm(ph, pl, m0, m1) \ + do { \ + DItype __m0 = (m0), __m1 = (m1); \ + __asm__ ("mulhd %0,%1,%2" : "=r" (ph) : "%r" (m0), "r" (m1)); \ + (pl) = __m0 * __m1; \ + } while (0) +#define SMUL_TIME 14 /* ??? */ +#define UDIV_TIME 120 /* ??? */ +#endif /* 64-bit PowerPC. */ +#endif /* if 0 */ + +/*************************************** + ************** PYR ****************** + ***************************************/ +#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__ */ + + +/*************************************** + ************** RT/ROMP ************** + ***************************************/ +#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" \ + : "=r" ((USItype)(sh)), \ + "=&r" ((USItype)(sl)) \ + : "%0" ((USItype)(ah)), \ + "r" ((USItype)(bh)), \ + "%1" ((USItype)(al)), \ + "r" ((USItype)(bl))) +#define sub_ddmmss(sh, sl, ah, al, bh, bl) \ + __asm__ ("s %1,%5\n" \ + "se %0,%3" \ + : "=r" ((USItype)(sh)), \ + "=&r" ((USItype)(sl)) \ + : "0" ((USItype)(ah)), \ + "r" ((USItype)(bh)), \ + "1" ((USItype)(al)), \ + "r" ((USItype)(bl))) +#define umul_ppmm(ph, pl, m0, m1) \ + 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" \ + : "=r" ((USItype)(ph)), \ + "=r" ((USItype)(pl)) \ + : "%r" (__m0), \ + "r" (__m1) \ + : "r2"); \ + (ph) += ((((SItype) __m0 >> 31) & __m1) \ + + (((SItype) __m1 >> 31) & __m0)); \ + } while (0) +#define UMUL_TIME 20 +#define UDIV_TIME 200 +#define count_leading_zeros(count, x) \ + do { \ + if ((x) >= 0x10000) \ + __asm__ ("clz %0,%1" \ + : "=r" ((USItype)(count)) \ + : "r" ((USItype)(x) >> 16)); \ + else \ + { \ + __asm__ ("clz %0,%1" \ + : "=r" ((USItype)(count)) \ + : "r" ((USItype)(x))); \ + (count) += 16; \ + } \ + } while (0) +#endif /* RT/ROMP */ + + +/*************************************** + ************** SH2 ****************** + ***************************************/ +#if (defined (__sh2__) || defined(__sh3__) || defined(__SH4__) ) \ + && W_TYPE_SIZE == 32 +#define umul_ppmm(w1, w0, u, v) \ + __asm__ ( \ + "dmulu.l %2,%3\n" \ + "sts macl,%1\n" \ + "sts mach,%0" \ + : "=r" ((USItype)(w1)), \ + "=r" ((USItype)(w0)) \ + : "r" ((USItype)(u)), \ + "r" ((USItype)(v)) \ + : "macl", "mach") +#define UMUL_TIME 5 +#endif + +/*************************************** + ************** SPARC **************** + ***************************************/ +#if defined (__sparc__) && W_TYPE_SIZE == 32 +#define add_ssaaaa(sh, sl, ah, al, bh, bl) \ + __asm__ ("addcc %r4,%5,%1\n" \ + "addx %r2,%3,%0" \ + : "=r" ((USItype)(sh)), \ + "=&r" ((USItype)(sl)) \ + : "%rJ" ((USItype)(ah)), \ + "rI" ((USItype)(bh)), \ + "%rJ" ((USItype)(al)), \ + "rI" ((USItype)(bl)) \ + __CLOBBER_CC) +#define sub_ddmmss(sh, sl, ah, al, bh, bl) \ + __asm__ ("subcc %r4,%5,%1\n" \ + "subx %r2,%3,%0" \ + : "=r" ((USItype)(sh)), \ + "=&r" ((USItype)(sl)) \ + : "rJ" ((USItype)(ah)), \ + "rI" ((USItype)(bh)), \ + "rJ" ((USItype)(al)), \ + "rI" ((USItype)(bl)) \ + __CLOBBER_CC) +#if defined (__sparc_v8__) +/* Don't match immediate range because, 1) it is not often useful, + 2) the 'I' flag thinks of the range as a 13 bit signed interval, + while we want to match a 13 bit interval, sign extended to 32 bits, + but INTERPRETED AS UNSIGNED. */ +#define umul_ppmm(w1, w0, u, v) \ + __asm__ ("umul %2,%3,%1;rd %%y,%0" \ + : "=r" ((USItype)(w1)), \ + "=r" ((USItype)(w0)) \ + : "r" ((USItype)(u)), \ + "r" ((USItype)(v))) +#define UMUL_TIME 5 +#ifndef SUPERSPARC /* SuperSPARC's udiv only handles 53 bit dividends */ +#define udiv_qrnnd(q, r, n1, n0, d) \ + do { \ + USItype __q; \ + __asm__ ("mov %1,%%y;nop;nop;nop;udiv %2,%3,%0" \ + : "=r" ((USItype)(__q)) \ + : "r" ((USItype)(n1)), \ + "r" ((USItype)(n0)), \ + "r" ((USItype)(d))); \ + (r) = (n0) - __q * (d); \ + (q) = __q; \ + } while (0) +#define UDIV_TIME 25 +#endif /* SUPERSPARC */ +#else /* ! __sparc_v8__ */ +#if defined (__sparclite__) +/* This has hardware multiply but not divide. It also has two additional + instructions scan (ffs from high bit) and divscc. */ +#define umul_ppmm(w1, w0, u, v) \ + __asm__ ("umul %2,%3,%1;rd %%y,%0" \ + : "=r" ((USItype)(w1)), \ + "=r" ((USItype)(w0)) \ + : "r" ((USItype)(u)), \ + "r" ((USItype)(v))) +#define UMUL_TIME 5 +#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" \ + "1: ! End of inline udiv_qrnnd" \ + : "=r" ((USItype)(q)), \ + "=r" ((USItype)(r)) \ + : "r" ((USItype)(n1)), \ + "r" ((USItype)(n0)), \ + "rI" ((USItype)(d)) \ + : "%g1" __AND_CLOBBER_CC) +#define UDIV_TIME 37 +#define count_leading_zeros(count, x) \ + __asm__ ("scan %1,0,%0" \ + : "=r" ((USItype)(x)) \ + : "r" ((USItype)(count))) +/* Early sparclites return 63 for an argument of 0, but they warn that future + implementations might change this. Therefore, leave COUNT_LEADING_ZEROS_0 + undefined. */ +#endif /* __sparclite__ */ +#endif /* __sparc_v8__ */ +/* Default to sparc v7 versions of umul_ppmm and udiv_qrnnd. */ +#ifndef umul_ppmm +#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,%%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" ((USItype)(w1)), \ + "=r" ((USItype)(w0)) \ + : "%rI" ((USItype)(u)), \ + "r" ((USItype)(v)) \ + : "%g1", "%g2" __AND_CLOBBER_CC) +#define UMUL_TIME 39 /* 39 instructions */ +#endif +#ifndef udiv_qrnnd +#ifndef LONGLONG_STANDALONE +#define udiv_qrnnd(q, r, n1, n0, d) \ + do { USItype __r; \ + (q) = __udiv_qrnnd (&__r, (n1), (n0), (d)); \ + (r) = __r; \ + } while (0) +extern USItype __udiv_qrnnd (); +#define UDIV_TIME 140 +#endif /* LONGLONG_STANDALONE */ +#endif /* udiv_qrnnd */ +#endif /* __sparc__ */ + + +/*************************************** + ************** VAX ****************** + ***************************************/ +#if defined (__vax__) && W_TYPE_SIZE == 32 +#define add_ssaaaa(sh, sl, ah, al, bh, bl) \ + __asm__ ("addl2 %5,%1\n" \ + "adwc %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__ ("subl2 %5,%1\n" \ + "sbwc %3,%0" \ + : "=g" ((USItype)(sh)), \ + "=&g" ((USItype)(sl)) \ + : "0" ((USItype)(ah)), \ + "g" ((USItype)(bh)), \ + "1" ((USItype)(al)), \ + "g" ((USItype)(bl))) +#define umul_ppmm(xh, xl, m0, m1) \ + do { \ + union {UDItype __ll; \ + struct {USItype __l, __h;} __i; \ + } __xx; \ + USItype __m0 = (m0), __m1 = (m1); \ + __asm__ ("emul %1,%2,$0,%0" \ + : "=g" (__xx.__ll) \ + : "g" (__m0), \ + "g" (__m1)); \ + (xh) = __xx.__i.__h; (xl) = __xx.__i.__l; \ + (xh) += ((((SItype) __m0 >> 31) & __m1) \ + + (((SItype) __m1 >> 31) & __m0)); \ + } while (0) +#define sdiv_qrnnd(q, r, n1, n0, d) \ + do { \ + union {DItype __ll; \ + struct {SItype __l, __h;} __i; \ + } __xx; \ + __xx.__i.__h = n1; __xx.__i.__l = n0; \ + __asm__ ("ediv %3,%2,%0,%1" \ + : "=g" (q), "=g" (r) \ + : "g" (__xx.__ll), "g" (d)); \ + } while (0) +#endif /* __vax__ */ + + +/*************************************** + ************** Z8000 **************** + ***************************************/ +#if defined (__z8000__) && W_TYPE_SIZE == 16 +#define add_ssaaaa(sh, sl, ah, al, bh, bl) \ + __asm__ ("add %H1,%H5\n\tadc %H0,%H3" \ + : "=r" ((unsigned int)(sh)), \ + "=&r" ((unsigned int)(sl)) \ + : "%0" ((unsigned int)(ah)), \ + "r" ((unsigned int)(bh)), \ + "%1" ((unsigned int)(al)), \ + "rQR" ((unsigned int)(bl))) +#define sub_ddmmss(sh, sl, ah, al, bh, bl) \ + __asm__ ("sub %H1,%H5\n\tsbc %H0,%H3" \ + : "=r" ((unsigned int)(sh)), \ + "=&r" ((unsigned int)(sl)) \ + : "0" ((unsigned int)(ah)), \ + "r" ((unsigned int)(bh)), \ + "1" ((unsigned int)(al)), \ + "rQR" ((unsigned int)(bl))) +#define umul_ppmm(xh, xl, m0, m1) \ + do { \ + union {long int __ll; \ + struct {unsigned int __h, __l;} __i; \ + } __xx; \ + unsigned int __m0 = (m0), __m1 = (m1); \ + __asm__ ("mult %S0,%H3" \ + : "=r" (__xx.__i.__h), \ + "=r" (__xx.__i.__l) \ + : "%1" (__m0), \ + "rQR" (__m1)); \ + (xh) = __xx.__i.__h; (xl) = __xx.__i.__l; \ + (xh) += ((((signed int) __m0 >> 15) & __m1) \ + + (((signed int) __m1 >> 15) & __m0)); \ + } while (0) +#endif /* __z8000__ */ + +#endif /* __GNUC__ */ +#endif /* !__riscos__ */ + + +/*************************************** + *********** Generic Versions ******** + ***************************************/ +#if !defined (umul_ppmm) && defined (__umulsidi3) +#define umul_ppmm(ph, pl, m0, m1) \ + { \ + UDWtype __ll = __umulsidi3 (m0, m1); \ + ph = (UWtype) (__ll >> W_TYPE_SIZE); \ + pl = (UWtype) __ll; \ + } +#endif + +#if !defined (__umulsidi3) +#define __umulsidi3(u, v) \ + ({UWtype __hi, __lo; \ + umul_ppmm (__hi, __lo, u, v); \ + ((UDWtype) __hi << W_TYPE_SIZE) | __lo; }) +#endif + +/* If this machine has no inline assembler, use C macros. */ + +#if !defined (add_ssaaaa) +#define add_ssaaaa(sh, sl, ah, al, bh, bl) \ + do { \ + UWtype __x; \ + __x = (al) + (bl); \ + (sh) = (ah) + (bh) + (__x < (al)); \ + (sl) = __x; \ + } while (0) +#endif + +#if !defined (sub_ddmmss) +#define sub_ddmmss(sh, sl, ah, al, bh, bl) \ + do { \ + UWtype __x; \ + __x = (al) - (bl); \ + (sh) = (ah) - (bh) - (__x > (al)); \ + (sl) = __x; \ + } while (0) +#endif + +#if !defined (umul_ppmm) +#define umul_ppmm(w1, w0, u, v) \ + do { \ + UWtype __x0, __x1, __x2, __x3; \ + UHWtype __ul, __vl, __uh, __vh; \ + UWtype __u = (u), __v = (v); \ + \ + __ul = __ll_lowpart (__u); \ + __uh = __ll_highpart (__u); \ + __vl = __ll_lowpart (__v); \ + __vh = __ll_highpart (__v); \ + \ + __x0 = (UWtype) __ul * __vl; \ + __x1 = (UWtype) __ul * __vh; \ + __x2 = (UWtype) __uh * __vl; \ + __x3 = (UWtype) __uh * __vh; \ + \ + __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. */ \ + \ + (w1) = __x3 + __ll_highpart (__x1); \ + (w0) = (__ll_lowpart (__x1) << W_TYPE_SIZE/2) + __ll_lowpart (__x0);\ + } while (0) +#endif + +#if !defined (umul_ppmm) +#define smul_ppmm(w1, w0, u, v) \ + do { \ + UWtype __w1; \ + UWtype __m0 = (u), __m1 = (v); \ + umul_ppmm (__w1, w0, __m0, __m1); \ + (w1) = __w1 - (-(__m0 >> (W_TYPE_SIZE - 1)) & __m1) \ + - (-(__m1 >> (W_TYPE_SIZE - 1)) & __m0); \ + } while (0) +#endif + +/* Define this unconditionally, so it can be used for debugging. */ +#define __udiv_qrnnd_c(q, r, n1, n0, d) \ + do { \ + UWtype __d1, __d0, __q1, __q0, __r1, __r0, __m; \ + __d1 = __ll_highpart (d); \ + __d0 = __ll_lowpart (d); \ + \ + __r1 = (n1) % __d1; \ + __q1 = (n1) / __d1; \ + __m = (UWtype) __q1 * __d0; \ + __r1 = __r1 * __ll_B | __ll_highpart (n0); \ + if (__r1 < __m) \ + { \ + __q1--, __r1 += (d); \ + if (__r1 >= (d)) /* i.e. we didn't get carry when adding to __r1 */\ + if (__r1 < __m) \ + __q1--, __r1 += (d); \ + } \ + __r1 -= __m; \ + \ + __r0 = __r1 % __d1; \ + __q0 = __r1 / __d1; \ + __m = (UWtype) __q0 * __d0; \ + __r0 = __r0 * __ll_B | __ll_lowpart (n0); \ + if (__r0 < __m) \ + { \ + __q0--, __r0 += (d); \ + if (__r0 >= (d)) \ + if (__r0 < __m) \ + __q0--, __r0 += (d); \ + } \ + __r0 -= __m; \ + \ + (q) = (UWtype) __q1 * __ll_B | __q0; \ + (r) = __r0; \ + } while (0) + +/* If the processor has no udiv_qrnnd but sdiv_qrnnd, go through + __udiv_w_sdiv (defined in libgcc or elsewhere). */ +#if !defined (udiv_qrnnd) && defined (sdiv_qrnnd) +#define udiv_qrnnd(q, r, nh, nl, d) \ + do { \ + UWtype __r; \ + (q) = __MPN(udiv_w_sdiv) (&__r, nh, nl, d); \ + (r) = __r; \ + } while (0) +#endif + +/* If udiv_qrnnd was not defined for this processor, use __udiv_qrnnd_c. */ +#if !defined (udiv_qrnnd) +#define UDIV_NEEDS_NORMALIZATION 1 +#define udiv_qrnnd __udiv_qrnnd_c +#endif + +#if !defined (count_leading_zeros) +extern +#ifdef __STDC__ +const +#endif +unsigned char _gcry_clz_tab[]; +#define MPI_INTERNAL_NEED_CLZ_TAB 1 +#define count_leading_zeros(count, x) \ + do { \ + UWtype __xr = (x); \ + UWtype __a; \ + \ + if (W_TYPE_SIZE <= 32) \ + { \ + __a = __xr < ((UWtype) 1 << 2*__BITS4) \ + ? (__xr < ((UWtype) 1 << __BITS4) ? 0 : __BITS4) \ + : (__xr < ((UWtype) 1 << 3*__BITS4) ? 2*__BITS4 : 3*__BITS4);\ + } \ + else \ + { \ + for (__a = W_TYPE_SIZE - 8; __a > 0; __a -= 8) \ + if (((__xr >> __a) & 0xff) != 0) \ + break; \ + } \ + \ + (count) = W_TYPE_SIZE - (_gcry_clz_tab[__xr >> __a] + __a); \ + } while (0) +/* This version gives a well-defined value for zero. */ +#define COUNT_LEADING_ZEROS_0 W_TYPE_SIZE +#endif + +#if !defined (count_trailing_zeros) +/* Define count_trailing_zeros using count_leading_zeros. The latter might be + defined in asm, but if it is not, the C version above is good enough. */ +#define count_trailing_zeros(count, x) \ + do { \ + UWtype __ctz_x = (x); \ + UWtype __ctz_c; \ + count_leading_zeros (__ctz_c, __ctz_x & -__ctz_x); \ + (count) = W_TYPE_SIZE - 1 - __ctz_c; \ + } while (0) +#endif + +#ifndef UDIV_NEEDS_NORMALIZATION +#define UDIV_NEEDS_NORMALIZATION 0 +#endif diff --git a/libgcrypt-1.4.6/mpi/mpi-bit.c b/libgcrypt-1.4.6/mpi/mpi-bit.c index 6c59fc0..32c820c 100644 --- a/libgcrypt-1.4.6/mpi/mpi-bit.c +++ b/libgcrypt-1.4.6/mpi/mpi-bit.c @@ -1,365 +1,365 @@ -/* mpi-bit.c - MPI bit level functions
- * Copyright (C) 1998, 1999, 2001, 2002, 2006 Free Software Foundation, Inc.
- *
- * This file is part of Libgcrypt.
- *
- * Libgcrypt is free software; you can redistribute it and/or modify
- * it under the terms of the GNU Lesser General Public License as
- * published by the Free Software Foundation; either version 2.1 of
- * the License, or (at your option) any later version.
- *
- * Libgcrypt is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU Lesser General Public License for more details.
- *
- * You should have received a copy of the GNU Lesser General Public
- * License along with this program; if not, write to the Free Software
- * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA
- */
-
-#include <config.h>
-#include <stdio.h>
-#include <stdlib.h>
-#include "mpi-internal.h"
-#include "longlong.h"
-
-
-#ifdef MPI_INTERNAL_NEED_CLZ_TAB
-#ifdef __STDC__
-const
-#endif
-unsigned char
-_gcry_clz_tab[] =
-{
- 0,1,2,2,3,3,3,3,4,4,4,4,4,4,4,4,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,
- 6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,
- 7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,
- 7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,
- 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,
- 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,
- 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,
- 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,
-};
-#endif
-
-
-#define A_LIMB_1 ((mpi_limb_t)1)
-
-
-/****************
- * Sometimes we have MSL (most significant limbs) which are 0;
- * this is for some reasons not good, so this function removes them.
- */
-void
-_gcry_mpi_normalize( gcry_mpi_t a )
-{
- if( mpi_is_opaque(a) )
- return;
-
- for( ; a->nlimbs && !a->d[a->nlimbs-1]; a->nlimbs-- )
- ;
-}
-
-
-
-/****************
- * Return the number of bits in A.
- */
-unsigned int
-gcry_mpi_get_nbits( gcry_mpi_t a )
-{
- unsigned n;
-
- if( mpi_is_opaque(a) ) {
- return a->sign; /* which holds the number of bits */
- }
-
- _gcry_mpi_normalize( a );
- if( a->nlimbs ) {
- mpi_limb_t alimb = a->d[a->nlimbs-1];
- if( alimb )
- count_leading_zeros( n, alimb );
- else
- n = BITS_PER_MPI_LIMB;
- n = BITS_PER_MPI_LIMB - n + (a->nlimbs-1) * BITS_PER_MPI_LIMB;
- }
- else
- n = 0;
- return n;
-}
-
-
-/****************
- * Test whether bit N is set.
- */
-int
-gcry_mpi_test_bit( gcry_mpi_t a, unsigned int n )
-{
- unsigned int limbno, bitno;
- mpi_limb_t limb;
-
- limbno = n / BITS_PER_MPI_LIMB;
- bitno = n % BITS_PER_MPI_LIMB;
-
- if( limbno >= a->nlimbs )
- return 0; /* too far left: this is a 0 */
- limb = a->d[limbno];
- return (limb & (A_LIMB_1 << bitno))? 1: 0;
-}
-
-
-/****************
- * Set bit N of A.
- */
-void
-gcry_mpi_set_bit( gcry_mpi_t a, unsigned int n )
-{
- unsigned int limbno, bitno;
-
- limbno = n / BITS_PER_MPI_LIMB;
- bitno = n % BITS_PER_MPI_LIMB;
-
- if ( limbno >= a->nlimbs )
- {
- mpi_resize (a, limbno+1 );
- a->nlimbs = limbno+1;
- }
- a->d[limbno] |= (A_LIMB_1<<bitno);
-}
-
-/****************
- * Set bit N of A. and clear all bits above
- */
-void
-gcry_mpi_set_highbit( gcry_mpi_t a, unsigned int n )
-{
- unsigned int limbno, bitno;
-
- limbno = n / BITS_PER_MPI_LIMB;
- bitno = n % BITS_PER_MPI_LIMB;
-
- if ( limbno >= a->nlimbs )
- {
- mpi_resize (a, limbno+1 );
- a->nlimbs = limbno+1;
- }
- a->d[limbno] |= (A_LIMB_1<<bitno);
- for ( bitno++; bitno < BITS_PER_MPI_LIMB; bitno++ )
- a->d[limbno] &= ~(A_LIMB_1 << bitno);
- a->nlimbs = limbno+1;
-}
-
-/****************
- * clear bit N of A and all bits above
- */
-void
-gcry_mpi_clear_highbit( gcry_mpi_t a, unsigned int n )
-{
- unsigned int limbno, bitno;
-
- limbno = n / BITS_PER_MPI_LIMB;
- bitno = n % BITS_PER_MPI_LIMB;
-
- if( limbno >= a->nlimbs )
- return; /* not allocated, therefore no need to clear bits
- :-) */
-
- for( ; bitno < BITS_PER_MPI_LIMB; bitno++ )
- a->d[limbno] &= ~(A_LIMB_1 << bitno);
- a->nlimbs = limbno+1;
-}
-
-/****************
- * Clear bit N of A.
- */
-void
-gcry_mpi_clear_bit( gcry_mpi_t a, unsigned int n )
-{
- unsigned int limbno, bitno;
-
- limbno = n / BITS_PER_MPI_LIMB;
- bitno = n % BITS_PER_MPI_LIMB;
-
- if( limbno >= a->nlimbs )
- return; /* don't need to clear this bit, it's to far to left */
- a->d[limbno] &= ~(A_LIMB_1 << bitno);
-}
-
-
-/****************
- * Shift A by COUNT limbs to the right
- * This is used only within the MPI library
- */
-void
-_gcry_mpi_rshift_limbs( gcry_mpi_t a, unsigned int count )
-{
- mpi_ptr_t ap = a->d;
- mpi_size_t n = a->nlimbs;
- unsigned int i;
-
- if( count >= n ) {
- a->nlimbs = 0;
- return;
- }
-
- for( i = 0; i < n - count; i++ )
- ap[i] = ap[i+count];
- ap[i] = 0;
- a->nlimbs -= count;
-}
-
-
-/*
- * Shift A by N bits to the right.
- */
-void
-gcry_mpi_rshift ( gcry_mpi_t x, gcry_mpi_t a, unsigned int n )
-{
- mpi_size_t xsize;
- unsigned int i;
- unsigned int nlimbs = (n/BITS_PER_MPI_LIMB);
- unsigned int nbits = (n%BITS_PER_MPI_LIMB);
-
- if ( x == a )
- {
- /* In-place operation. */
- if ( nlimbs >= x->nlimbs )
- {
- x->nlimbs = 0;
- return;
- }
-
- if (nlimbs)
- {
- for (i=0; i < x->nlimbs - nlimbs; i++ )
- x->d[i] = x->d[i+nlimbs];
- x->d[i] = 0;
- x->nlimbs -= nlimbs;
-
- }
- if ( x->nlimbs && nbits )
- _gcry_mpih_rshift ( x->d, x->d, x->nlimbs, nbits );
- }
- else if ( nlimbs )
- {
- /* Copy and shift by more or equal bits than in a limb. */
- xsize = a->nlimbs;
- x->sign = a->sign;
- RESIZE_IF_NEEDED (x, xsize);
- x->nlimbs = xsize;
- for (i=0; i < a->nlimbs; i++ )
- x->d[i] = a->d[i];
- x->nlimbs = i;
-
- if ( nlimbs >= x->nlimbs )
- {
- x->nlimbs = 0;
- return;
- }
-
- if (nlimbs)
- {
- for (i=0; i < x->nlimbs - nlimbs; i++ )
- x->d[i] = x->d[i+nlimbs];
- x->d[i] = 0;
- x->nlimbs -= nlimbs;
- }
-
- if ( x->nlimbs && nbits )
- _gcry_mpih_rshift ( x->d, x->d, x->nlimbs, nbits );
- }
- else
- {
- /* Copy and shift by less than bits in a limb. */
- xsize = a->nlimbs;
- x->sign = a->sign;
- RESIZE_IF_NEEDED (x, xsize);
- x->nlimbs = xsize;
-
- if ( xsize )
- {
- if (nbits )
- _gcry_mpih_rshift (x->d, a->d, x->nlimbs, nbits );
- else
- {
- /* The rshift helper function is not specified for
- NBITS==0, thus we do a plain copy here. */
- for (i=0; i < x->nlimbs; i++ )
- x->d[i] = a->d[i];
- }
- }
- }
- MPN_NORMALIZE (x->d, x->nlimbs);
-}
-
-
-/****************
- * Shift A by COUNT limbs to the left
- * This is used only within the MPI library
- */
-void
-_gcry_mpi_lshift_limbs (gcry_mpi_t a, unsigned int count)
-{
- mpi_ptr_t ap;
- int n = a->nlimbs;
- int i;
-
- if (!count || !n)
- return;
-
- RESIZE_IF_NEEDED (a, n+count);
-
- ap = a->d;
- for (i = n-1; i >= 0; i--)
- ap[i+count] = ap[i];
- for (i=0; i < count; i++ )
- ap[i] = 0;
- a->nlimbs += count;
-}
-
-
-/*
- * Shift A by N bits to the left.
- */
-void
-gcry_mpi_lshift ( gcry_mpi_t x, gcry_mpi_t a, unsigned int n )
-{
- unsigned int nlimbs = (n/BITS_PER_MPI_LIMB);
- unsigned int nbits = (n%BITS_PER_MPI_LIMB);
-
- if (x == a && !n)
- return; /* In-place shift with an amount of zero. */
-
- if ( x != a )
- {
- /* Copy A to X. */
- unsigned int alimbs = a->nlimbs;
- int asign = a->sign;
- mpi_ptr_t xp, ap;
-
- RESIZE_IF_NEEDED (x, alimbs+nlimbs+1);
- xp = x->d;
- ap = a->d;
- MPN_COPY (xp, ap, alimbs);
- x->nlimbs = alimbs;
- x->flags = a->flags;
- x->sign = asign;
- }
-
- if (nlimbs && !nbits)
- {
- /* Shift a full number of limbs. */
- _gcry_mpi_lshift_limbs (x, nlimbs);
- }
- else if (n)
- {
- /* We use a very dump approach: Shift left by the number of
- limbs plus one and than fix it up by an rshift. */
- _gcry_mpi_lshift_limbs (x, nlimbs+1);
- gcry_mpi_rshift (x, x, BITS_PER_MPI_LIMB - nbits);
- }
-
- MPN_NORMALIZE (x->d, x->nlimbs);
-}
-
+/* mpi-bit.c - MPI bit level fucntions + * Copyright (C) 1998, 1999, 2001, 2002, 2006 Free Software Foundation, Inc. + * + * This file is part of Libgcrypt. + * + * Libgcrypt is free software; you can redistribute it and/or modify + * it under the terms of the GNU Lesser General Public License as + * published by the Free Software Foundation; either version 2.1 of + * the License, or (at your option) any later version. + * + * Libgcrypt is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA + */ + +#include <config.h> +#include <stdio.h> +#include <stdlib.h> +#include "mpi-internal.h" +#include "longlong.h" + + +#ifdef MPI_INTERNAL_NEED_CLZ_TAB +#ifdef __STDC__ +const +#endif +unsigned char +_gcry_clz_tab[] = +{ + 0,1,2,2,3,3,3,3,4,4,4,4,4,4,4,4,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5, + 6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6,6, + 7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7, + 7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7, + 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, + 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, + 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, + 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, +}; +#endif + + +#define A_LIMB_1 ((mpi_limb_t)1) + + +/**************** + * Sometimes we have MSL (most significant limbs) which are 0; + * this is for some reasons not good, so this function removes them. + */ +void +_gcry_mpi_normalize( gcry_mpi_t a ) +{ + if( mpi_is_opaque(a) ) + return; + + for( ; a->nlimbs && !a->d[a->nlimbs-1]; a->nlimbs-- ) + ; +} + + + +/**************** + * Return the number of bits in A. + */ +unsigned int +gcry_mpi_get_nbits( gcry_mpi_t a ) +{ + unsigned n; + + if( mpi_is_opaque(a) ) { + return a->sign; /* which holds the number of bits */ + } + + _gcry_mpi_normalize( a ); + if( a->nlimbs ) { + mpi_limb_t alimb = a->d[a->nlimbs-1]; + if( alimb ) + count_leading_zeros( n, alimb ); + else + n = BITS_PER_MPI_LIMB; + n = BITS_PER_MPI_LIMB - n + (a->nlimbs-1) * BITS_PER_MPI_LIMB; + } + else + n = 0; + return n; +} + + +/**************** + * Test whether bit N is set. + */ +int +gcry_mpi_test_bit( gcry_mpi_t a, unsigned int n ) +{ + unsigned int limbno, bitno; + mpi_limb_t limb; + + limbno = n / BITS_PER_MPI_LIMB; + bitno = n % BITS_PER_MPI_LIMB; + + if( limbno >= a->nlimbs ) + return 0; /* too far left: this is a 0 */ + limb = a->d[limbno]; + return (limb & (A_LIMB_1 << bitno))? 1: 0; +} + + +/**************** + * Set bit N of A. + */ +void +gcry_mpi_set_bit( gcry_mpi_t a, unsigned int n ) +{ + unsigned int limbno, bitno; + + limbno = n / BITS_PER_MPI_LIMB; + bitno = n % BITS_PER_MPI_LIMB; + + if ( limbno >= a->nlimbs ) + { + mpi_resize (a, limbno+1 ); + a->nlimbs = limbno+1; + } + a->d[limbno] |= (A_LIMB_1<<bitno); +} + +/**************** + * Set bit N of A. and clear all bits above + */ +void +gcry_mpi_set_highbit( gcry_mpi_t a, unsigned int n ) +{ + unsigned int limbno, bitno; + + limbno = n / BITS_PER_MPI_LIMB; + bitno = n % BITS_PER_MPI_LIMB; + + if ( limbno >= a->nlimbs ) + { + mpi_resize (a, limbno+1 ); + a->nlimbs = limbno+1; + } + a->d[limbno] |= (A_LIMB_1<<bitno); + for ( bitno++; bitno < BITS_PER_MPI_LIMB; bitno++ ) + a->d[limbno] &= ~(A_LIMB_1 << bitno); + a->nlimbs = limbno+1; +} + +/**************** + * clear bit N of A and all bits above + */ +void +gcry_mpi_clear_highbit( gcry_mpi_t a, unsigned int n ) +{ + unsigned int limbno, bitno; + + limbno = n / BITS_PER_MPI_LIMB; + bitno = n % BITS_PER_MPI_LIMB; + + if( limbno >= a->nlimbs ) + return; /* not allocated, therefore no need to clear bits + :-) */ + + for( ; bitno < BITS_PER_MPI_LIMB; bitno++ ) + a->d[limbno] &= ~(A_LIMB_1 << bitno); + a->nlimbs = limbno+1; +} + +/**************** + * Clear bit N of A. + */ +void +gcry_mpi_clear_bit( gcry_mpi_t a, unsigned int n ) +{ + unsigned int limbno, bitno; + + limbno = n / BITS_PER_MPI_LIMB; + bitno = n % BITS_PER_MPI_LIMB; + + if( limbno >= a->nlimbs ) + return; /* don't need to clear this bit, it's to far to left */ + a->d[limbno] &= ~(A_LIMB_1 << bitno); +} + + +/**************** + * Shift A by COUNT limbs to the right + * This is used only within the MPI library + */ +void +_gcry_mpi_rshift_limbs( gcry_mpi_t a, unsigned int count ) +{ + mpi_ptr_t ap = a->d; + mpi_size_t n = a->nlimbs; + unsigned int i; + + if( count >= n ) { + a->nlimbs = 0; + return; + } + + for( i = 0; i < n - count; i++ ) + ap[i] = ap[i+count]; + ap[i] = 0; + a->nlimbs -= count; +} + + +/* + * Shift A by N bits to the right. + */ +void +gcry_mpi_rshift ( gcry_mpi_t x, gcry_mpi_t a, unsigned int n ) +{ + mpi_size_t xsize; + unsigned int i; + unsigned int nlimbs = (n/BITS_PER_MPI_LIMB); + unsigned int nbits = (n%BITS_PER_MPI_LIMB); + + if ( x == a ) + { + /* In-place operation. */ + if ( nlimbs >= x->nlimbs ) + { + x->nlimbs = 0; + return; + } + + if (nlimbs) + { + for (i=0; i < x->nlimbs - nlimbs; i++ ) + x->d[i] = x->d[i+nlimbs]; + x->d[i] = 0; + x->nlimbs -= nlimbs; + + } + if ( x->nlimbs && nbits ) + _gcry_mpih_rshift ( x->d, x->d, x->nlimbs, nbits ); + } + else if ( nlimbs ) + { + /* Copy and shift by more or equal bits than in a limb. */ + xsize = a->nlimbs; + x->sign = a->sign; + RESIZE_IF_NEEDED (x, xsize); + x->nlimbs = xsize; + for (i=0; i < a->nlimbs; i++ ) + x->d[i] = a->d[i]; + x->nlimbs = i; + + if ( nlimbs >= x->nlimbs ) + { + x->nlimbs = 0; + return; + } + + if (nlimbs) + { + for (i=0; i < x->nlimbs - nlimbs; i++ ) + x->d[i] = x->d[i+nlimbs]; + x->d[i] = 0; + x->nlimbs -= nlimbs; + } + + if ( x->nlimbs && nbits ) + _gcry_mpih_rshift ( x->d, x->d, x->nlimbs, nbits ); + } + else + { + /* Copy and shift by less than bits in a limb. */ + xsize = a->nlimbs; + x->sign = a->sign; + RESIZE_IF_NEEDED (x, xsize); + x->nlimbs = xsize; + + if ( xsize ) + { + if (nbits ) + _gcry_mpih_rshift (x->d, a->d, x->nlimbs, nbits ); + else + { + /* The rshift helper function is not specified for + NBITS==0, thus we do a plain copy here. */ + for (i=0; i < x->nlimbs; i++ ) + x->d[i] = a->d[i]; + } + } + } + MPN_NORMALIZE (x->d, x->nlimbs); +} + + +/**************** + * Shift A by COUNT limbs to the left + * This is used only within the MPI library + */ +void +_gcry_mpi_lshift_limbs (gcry_mpi_t a, unsigned int count) +{ + mpi_ptr_t ap; + int n = a->nlimbs; + int i; + + if (!count || !n) + return; + + RESIZE_IF_NEEDED (a, n+count); + + ap = a->d; + for (i = n-1; i >= 0; i--) + ap[i+count] = ap[i]; + for (i=0; i < count; i++ ) + ap[i] = 0; + a->nlimbs += count; +} + + +/* + * Shift A by N bits to the left. + */ +void +gcry_mpi_lshift ( gcry_mpi_t x, gcry_mpi_t a, unsigned int n ) +{ + unsigned int nlimbs = (n/BITS_PER_MPI_LIMB); + unsigned int nbits = (n%BITS_PER_MPI_LIMB); + + if (x == a && !n) + return; /* In-place shift with an amount of zero. */ + + if ( x != a ) + { + /* Copy A to X. */ + unsigned int alimbs = a->nlimbs; + int asign = a->sign; + mpi_ptr_t xp, ap; + + RESIZE_IF_NEEDED (x, alimbs+nlimbs+1); + xp = x->d; + ap = a->d; + MPN_COPY (xp, ap, alimbs); + x->nlimbs = alimbs; + x->flags = a->flags; + x->sign = asign; + } + + if (nlimbs && !nbits) + { + /* Shift a full number of limbs. */ + _gcry_mpi_lshift_limbs (x, nlimbs); + } + else if (n) + { + /* We use a very dump approach: Shift left by the number of + limbs plus one and than fix it up by an rshift. */ + _gcry_mpi_lshift_limbs (x, nlimbs+1); + gcry_mpi_rshift (x, x, BITS_PER_MPI_LIMB - nbits); + } + + MPN_NORMALIZE (x->d, x->nlimbs); +} + diff --git a/libgcrypt-1.4.6/mpi/mpi-div.c b/libgcrypt-1.4.6/mpi/mpi-div.c index e94e531..0d8a2d1 100644 --- a/libgcrypt-1.4.6/mpi/mpi-div.c +++ b/libgcrypt-1.4.6/mpi/mpi-div.c @@ -1,358 +1,358 @@ -/* mpi-div.c - MPI functions
- * Copyright (C) 1994, 1996, 1998, 2001, 2002,
- * 2003 Free Software Foundation, Inc.
- *
- * This file is part of Libgcrypt.
- *
- * Libgcrypt is free software; you can redistribute it and/or modify
- * it under the terms of the GNU Lesser General Public License as
- * published by the Free Software Foundation; either version 2.1 of
- * the License, or (at your option) any later version.
- *
- * Libgcrypt is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU Lesser General Public License for more details.
- *
- * You should have received a copy of the GNU Lesser General Public
- * License along with this program; if not, write to the Free Software
- * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA
- *
- * Note: This code is heavily based on the GNU MP Library.
- * Actually it's the same code with only minor changes in the
- * way the data is stored; this is to support the abstraction
- * of an optional secure memory allocation which may be used
- * to avoid revealing of sensitive data due to paging etc.
- */
-
-#include <config.h>
-#include <stdio.h>
-#include <stdlib.h>
-#include "mpi-internal.h"
-#include "longlong.h"
-#include "g10lib.h"
-
-
-void
-_gcry_mpi_fdiv_r( gcry_mpi_t rem, gcry_mpi_t dividend, gcry_mpi_t divisor )
-{
- int divisor_sign = divisor->sign;
- gcry_mpi_t temp_divisor = NULL;
-
- /* We need the original value of the divisor after the remainder has been
- * preliminary calculated. We have to copy it to temporary space if it's
- * the same variable as REM. */
- if( rem == divisor ) {
- temp_divisor = mpi_copy( divisor );
- divisor = temp_divisor;
- }
-
- _gcry_mpi_tdiv_r( rem, dividend, divisor );
-
- if( ((divisor_sign?1:0) ^ (dividend->sign?1:0)) && rem->nlimbs )
- gcry_mpi_add( rem, rem, divisor);
-
- if( temp_divisor )
- mpi_free(temp_divisor);
-}
-
-
-
-/****************
- * Division rounding the quotient towards -infinity.
- * The remainder gets the same sign as the denominator.
- * rem is optional
- */
-
-ulong
-_gcry_mpi_fdiv_r_ui( gcry_mpi_t rem, gcry_mpi_t dividend, ulong divisor )
-{
- mpi_limb_t rlimb;
-
- rlimb = _gcry_mpih_mod_1( dividend->d, dividend->nlimbs, divisor );
- if( rlimb && dividend->sign )
- rlimb = divisor - rlimb;
-
- if( rem ) {
- rem->d[0] = rlimb;
- rem->nlimbs = rlimb? 1:0;
- }
- return rlimb;
-}
-
-
-void
-_gcry_mpi_fdiv_q( gcry_mpi_t quot, gcry_mpi_t dividend, gcry_mpi_t divisor )
-{
- gcry_mpi_t tmp = mpi_alloc( mpi_get_nlimbs(quot) );
- _gcry_mpi_fdiv_qr( quot, tmp, dividend, divisor);
- mpi_free(tmp);
-}
-
-void
-_gcry_mpi_fdiv_qr( gcry_mpi_t quot, gcry_mpi_t rem, gcry_mpi_t dividend, gcry_mpi_t divisor )
-{
- int divisor_sign = divisor->sign;
- gcry_mpi_t temp_divisor = NULL;
-
- if( quot == divisor || rem == divisor ) {
- temp_divisor = mpi_copy( divisor );
- divisor = temp_divisor;
- }
-
- _gcry_mpi_tdiv_qr( quot, rem, dividend, divisor );
-
- if( (divisor_sign ^ dividend->sign) && rem->nlimbs ) {
- gcry_mpi_sub_ui( quot, quot, 1 );
- gcry_mpi_add( rem, rem, divisor);
- }
-
- if( temp_divisor )
- mpi_free(temp_divisor);
-}
-
-
-/* If den == quot, den needs temporary storage.
- * If den == rem, den needs temporary storage.
- * If num == quot, num needs temporary storage.
- * If den has temporary storage, it can be normalized while being copied,
- * i.e no extra storage should be allocated.
- */
-
-void
-_gcry_mpi_tdiv_r( gcry_mpi_t rem, gcry_mpi_t num, gcry_mpi_t den)
-{
- _gcry_mpi_tdiv_qr(NULL, rem, num, den );
-}
-
-void
-_gcry_mpi_tdiv_qr( gcry_mpi_t quot, gcry_mpi_t rem, gcry_mpi_t num, gcry_mpi_t den)
-{
- mpi_ptr_t np, dp;
- mpi_ptr_t qp, rp;
- mpi_size_t nsize = num->nlimbs;
- mpi_size_t dsize = den->nlimbs;
- mpi_size_t qsize, rsize;
- mpi_size_t sign_remainder = num->sign;
- mpi_size_t sign_quotient = num->sign ^ den->sign;
- unsigned normalization_steps;
- mpi_limb_t q_limb;
- mpi_ptr_t marker[5];
- unsigned int marker_nlimbs[5];
- int markidx=0;
-
- /* Ensure space is enough for quotient and remainder.
- * We need space for an extra limb in the remainder, because it's
- * up-shifted (normalized) below. */
- rsize = nsize + 1;
- mpi_resize( rem, rsize);
-
- qsize = rsize - dsize; /* qsize cannot be bigger than this. */
- if( qsize <= 0 ) {
- if( num != rem ) {
- rem->nlimbs = num->nlimbs;
- rem->sign = num->sign;
- MPN_COPY(rem->d, num->d, nsize);
- }
- if( quot ) {
- /* This needs to follow the assignment to rem, in case the
- * numerator and quotient are the same. */
- quot->nlimbs = 0;
- quot->sign = 0;
- }
- return;
- }
-
- if( quot )
- mpi_resize( quot, qsize);
-
- /* Read pointers here, when reallocation is finished. */
- np = num->d;
- dp = den->d;
- rp = rem->d;
-
- /* Optimize division by a single-limb divisor. */
- if( dsize == 1 ) {
- mpi_limb_t rlimb;
- if( quot ) {
- qp = quot->d;
- rlimb = _gcry_mpih_divmod_1( qp, np, nsize, dp[0] );
- qsize -= qp[qsize - 1] == 0;
- quot->nlimbs = qsize;
- quot->sign = sign_quotient;
- }
- else
- rlimb = _gcry_mpih_mod_1( np, nsize, dp[0] );
- rp[0] = rlimb;
- rsize = rlimb != 0?1:0;
- rem->nlimbs = rsize;
- rem->sign = sign_remainder;
- return;
- }
-
-
- if( quot ) {
- qp = quot->d;
- /* Make sure QP and NP point to different objects. Otherwise the
- * numerator would be gradually overwritten by the quotient limbs. */
- if(qp == np) { /* Copy NP object to temporary space. */
- marker_nlimbs[markidx] = nsize;
- np = marker[markidx++] = mpi_alloc_limb_space(nsize,
- mpi_is_secure(quot));
- MPN_COPY(np, qp, nsize);
- }
- }
- else /* Put quotient at top of remainder. */
- qp = rp + dsize;
-
- count_leading_zeros( normalization_steps, dp[dsize - 1] );
-
- /* Normalize the denominator, i.e. make its most significant bit set by
- * shifting it NORMALIZATION_STEPS bits to the left. Also shift the
- * numerator the same number of steps (to keep the quotient the same!).
- */
- if( normalization_steps ) {
- mpi_ptr_t tp;
- mpi_limb_t nlimb;
-
- /* Shift up the denominator setting the most significant bit of
- * the most significant word. Use temporary storage not to clobber
- * the original contents of the denominator. */
- marker_nlimbs[markidx] = dsize;
- tp = marker[markidx++] = mpi_alloc_limb_space(dsize,mpi_is_secure(den));
- _gcry_mpih_lshift( tp, dp, dsize, normalization_steps );
- dp = tp;
-
- /* Shift up the numerator, possibly introducing a new most
- * significant word. Move the shifted numerator in the remainder
- * meanwhile. */
- nlimb = _gcry_mpih_lshift(rp, np, nsize, normalization_steps);
- if( nlimb ) {
- rp[nsize] = nlimb;
- rsize = nsize + 1;
- }
- else
- rsize = nsize;
- }
- else {
- /* The denominator is already normalized, as required. Copy it to
- * temporary space if it overlaps with the quotient or remainder. */
- if( dp == rp || (quot && (dp == qp))) {
- mpi_ptr_t tp;
-
- marker_nlimbs[markidx] = dsize;
- tp = marker[markidx++] = mpi_alloc_limb_space(dsize,
- mpi_is_secure(den));
- MPN_COPY( tp, dp, dsize );
- dp = tp;
- }
-
- /* Move the numerator to the remainder. */
- if( rp != np )
- MPN_COPY(rp, np, nsize);
-
- rsize = nsize;
- }
-
- q_limb = _gcry_mpih_divrem( qp, 0, rp, rsize, dp, dsize );
-
- if( quot ) {
- qsize = rsize - dsize;
- if(q_limb) {
- qp[qsize] = q_limb;
- qsize += 1;
- }
-
- quot->nlimbs = qsize;
- quot->sign = sign_quotient;
- }
-
- rsize = dsize;
- MPN_NORMALIZE (rp, rsize);
-
- if( normalization_steps && rsize ) {
- _gcry_mpih_rshift(rp, rp, rsize, normalization_steps);
- rsize -= rp[rsize - 1] == 0?1:0;
- }
-
- rem->nlimbs = rsize;
- rem->sign = sign_remainder;
- while( markidx )
- {
- markidx--;
- _gcry_mpi_free_limb_space (marker[markidx], marker_nlimbs[markidx]);
- }
-}
-
-void
-_gcry_mpi_tdiv_q_2exp( gcry_mpi_t w, gcry_mpi_t u, unsigned int count )
-{
- mpi_size_t usize, wsize;
- mpi_size_t limb_cnt;
-
- usize = u->nlimbs;
- limb_cnt = count / BITS_PER_MPI_LIMB;
- wsize = usize - limb_cnt;
- if( limb_cnt >= usize )
- w->nlimbs = 0;
- else {
- mpi_ptr_t wp;
- mpi_ptr_t up;
-
- RESIZE_IF_NEEDED( w, wsize );
- wp = w->d;
- up = u->d;
-
- count %= BITS_PER_MPI_LIMB;
- if( count ) {
- _gcry_mpih_rshift( wp, up + limb_cnt, wsize, count );
- wsize -= !wp[wsize - 1];
- }
- else {
- MPN_COPY_INCR( wp, up + limb_cnt, wsize);
- }
-
- w->nlimbs = wsize;
- }
-}
-
-/****************
- * Check whether dividend is divisible by divisor
- * (note: divisor must fit into a limb)
- */
-int
-_gcry_mpi_divisible_ui(gcry_mpi_t dividend, ulong divisor )
-{
- return !_gcry_mpih_mod_1( dividend->d, dividend->nlimbs, divisor );
-}
-
-
-void
-gcry_mpi_div (gcry_mpi_t quot, gcry_mpi_t rem, gcry_mpi_t dividend, gcry_mpi_t divisor, int round)
-{
- if (!round)
- {
- if (!rem)
- {
- gcry_mpi_t tmp = mpi_alloc (mpi_get_nlimbs(quot));
- _gcry_mpi_tdiv_qr (quot, tmp, dividend, divisor);
- mpi_free (tmp);
- }
- else
- _gcry_mpi_tdiv_qr (quot, rem, dividend, divisor);
- }
- else if (round < 0)
- {
- if (!rem)
- _gcry_mpi_fdiv_q (quot, dividend, divisor);
- else if (!quot)
- _gcry_mpi_fdiv_r (rem, dividend, divisor);
- else
- _gcry_mpi_fdiv_qr (quot, rem, dividend, divisor);
- }
- else
- log_bug ("mpi rounding to ceiling not yet implemented\n");
-}
-
-
-
+/* mpi-div.c - MPI functions + * Copyright (C) 1994, 1996, 1998, 2001, 2002, + * 2003 Free Software Foundation, Inc. + * + * This file is part of Libgcrypt. + * + * Libgcrypt is free software; you can redistribute it and/or modify + * it under the terms of the GNU Lesser General Public License as + * published by the Free Software Foundation; either version 2.1 of + * the License, or (at your option) any later version. + * + * Libgcrypt is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA + * + * Note: This code is heavily based on the GNU MP Library. + * Actually it's the same code with only minor changes in the + * way the data is stored; this is to support the abstraction + * of an optional secure memory allocation which may be used + * to avoid revealing of sensitive data due to paging etc. + */ + +#include <config.h> +#include <stdio.h> +#include <stdlib.h> +#include "mpi-internal.h" +#include "longlong.h" +#include "g10lib.h" + + +void +_gcry_mpi_fdiv_r( gcry_mpi_t rem, gcry_mpi_t dividend, gcry_mpi_t divisor ) +{ + int divisor_sign = divisor->sign; + gcry_mpi_t temp_divisor = NULL; + + /* We need the original value of the divisor after the remainder has been + * preliminary calculated. We have to copy it to temporary space if it's + * the same variable as REM. */ + if( rem == divisor ) { + temp_divisor = mpi_copy( divisor ); + divisor = temp_divisor; + } + + _gcry_mpi_tdiv_r( rem, dividend, divisor ); + + if( ((divisor_sign?1:0) ^ (dividend->sign?1:0)) && rem->nlimbs ) + gcry_mpi_add( rem, rem, divisor); + + if( temp_divisor ) + mpi_free(temp_divisor); +} + + + +/**************** + * Division rounding the quotient towards -infinity. + * The remainder gets the same sign as the denominator. + * rem is optional + */ + +ulong +_gcry_mpi_fdiv_r_ui( gcry_mpi_t rem, gcry_mpi_t dividend, ulong divisor ) +{ + mpi_limb_t rlimb; + + rlimb = _gcry_mpih_mod_1( dividend->d, dividend->nlimbs, divisor ); + if( rlimb && dividend->sign ) + rlimb = divisor - rlimb; + + if( rem ) { + rem->d[0] = rlimb; + rem->nlimbs = rlimb? 1:0; + } + return rlimb; +} + + +void +_gcry_mpi_fdiv_q( gcry_mpi_t quot, gcry_mpi_t dividend, gcry_mpi_t divisor ) +{ + gcry_mpi_t tmp = mpi_alloc( mpi_get_nlimbs(quot) ); + _gcry_mpi_fdiv_qr( quot, tmp, dividend, divisor); + mpi_free(tmp); +} + +void +_gcry_mpi_fdiv_qr( gcry_mpi_t quot, gcry_mpi_t rem, gcry_mpi_t dividend, gcry_mpi_t divisor ) +{ + int divisor_sign = divisor->sign; + gcry_mpi_t temp_divisor = NULL; + + if( quot == divisor || rem == divisor ) { + temp_divisor = mpi_copy( divisor ); + divisor = temp_divisor; + } + + _gcry_mpi_tdiv_qr( quot, rem, dividend, divisor ); + + if( (divisor_sign ^ dividend->sign) && rem->nlimbs ) { + gcry_mpi_sub_ui( quot, quot, 1 ); + gcry_mpi_add( rem, rem, divisor); + } + + if( temp_divisor ) + mpi_free(temp_divisor); +} + + +/* If den == quot, den needs temporary storage. + * If den == rem, den needs temporary storage. + * If num == quot, num needs temporary storage. + * If den has temporary storage, it can be normalized while being copied, + * i.e no extra storage should be allocated. + */ + +void +_gcry_mpi_tdiv_r( gcry_mpi_t rem, gcry_mpi_t num, gcry_mpi_t den) +{ + _gcry_mpi_tdiv_qr(NULL, rem, num, den ); +} + +void +_gcry_mpi_tdiv_qr( gcry_mpi_t quot, gcry_mpi_t rem, gcry_mpi_t num, gcry_mpi_t den) +{ + mpi_ptr_t np, dp; + mpi_ptr_t qp, rp; + mpi_size_t nsize = num->nlimbs; + mpi_size_t dsize = den->nlimbs; + mpi_size_t qsize, rsize; + mpi_size_t sign_remainder = num->sign; + mpi_size_t sign_quotient = num->sign ^ den->sign; + unsigned normalization_steps; + mpi_limb_t q_limb; + mpi_ptr_t marker[5]; + unsigned int marker_nlimbs[5]; + int markidx=0; + + /* Ensure space is enough for quotient and remainder. + * We need space for an extra limb in the remainder, because it's + * up-shifted (normalized) below. */ + rsize = nsize + 1; + mpi_resize( rem, rsize); + + qsize = rsize - dsize; /* qsize cannot be bigger than this. */ + if( qsize <= 0 ) { + if( num != rem ) { + rem->nlimbs = num->nlimbs; + rem->sign = num->sign; + MPN_COPY(rem->d, num->d, nsize); + } + if( quot ) { + /* This needs to follow the assignment to rem, in case the + * numerator and quotient are the same. */ + quot->nlimbs = 0; + quot->sign = 0; + } + return; + } + + if( quot ) + mpi_resize( quot, qsize); + + /* Read pointers here, when reallocation is finished. */ + np = num->d; + dp = den->d; + rp = rem->d; + + /* Optimize division by a single-limb divisor. */ + if( dsize == 1 ) { + mpi_limb_t rlimb; + if( quot ) { + qp = quot->d; + rlimb = _gcry_mpih_divmod_1( qp, np, nsize, dp[0] ); + qsize -= qp[qsize - 1] == 0; + quot->nlimbs = qsize; + quot->sign = sign_quotient; + } + else + rlimb = _gcry_mpih_mod_1( np, nsize, dp[0] ); + rp[0] = rlimb; + rsize = rlimb != 0?1:0; + rem->nlimbs = rsize; + rem->sign = sign_remainder; + return; + } + + + if( quot ) { + qp = quot->d; + /* Make sure QP and NP point to different objects. Otherwise the + * numerator would be gradually overwritten by the quotient limbs. */ + if(qp == np) { /* Copy NP object to temporary space. */ + marker_nlimbs[markidx] = nsize; + np = marker[markidx++] = mpi_alloc_limb_space(nsize, + mpi_is_secure(quot)); + MPN_COPY(np, qp, nsize); + } + } + else /* Put quotient at top of remainder. */ + qp = rp + dsize; + + count_leading_zeros( normalization_steps, dp[dsize - 1] ); + + /* Normalize the denominator, i.e. make its most significant bit set by + * shifting it NORMALIZATION_STEPS bits to the left. Also shift the + * numerator the same number of steps (to keep the quotient the same!). + */ + if( normalization_steps ) { + mpi_ptr_t tp; + mpi_limb_t nlimb; + + /* Shift up the denominator setting the most significant bit of + * the most significant word. Use temporary storage not to clobber + * the original contents of the denominator. */ + marker_nlimbs[markidx] = dsize; + tp = marker[markidx++] = mpi_alloc_limb_space(dsize,mpi_is_secure(den)); + _gcry_mpih_lshift( tp, dp, dsize, normalization_steps ); + dp = tp; + + /* Shift up the numerator, possibly introducing a new most + * significant word. Move the shifted numerator in the remainder + * meanwhile. */ + nlimb = _gcry_mpih_lshift(rp, np, nsize, normalization_steps); + if( nlimb ) { + rp[nsize] = nlimb; + rsize = nsize + 1; + } + else + rsize = nsize; + } + else { + /* The denominator is already normalized, as required. Copy it to + * temporary space if it overlaps with the quotient or remainder. */ + if( dp == rp || (quot && (dp == qp))) { + mpi_ptr_t tp; + + marker_nlimbs[markidx] = dsize; + tp = marker[markidx++] = mpi_alloc_limb_space(dsize, + mpi_is_secure(den)); + MPN_COPY( tp, dp, dsize ); + dp = tp; + } + + /* Move the numerator to the remainder. */ + if( rp != np ) + MPN_COPY(rp, np, nsize); + + rsize = nsize; + } + + q_limb = _gcry_mpih_divrem( qp, 0, rp, rsize, dp, dsize ); + + if( quot ) { + qsize = rsize - dsize; + if(q_limb) { + qp[qsize] = q_limb; + qsize += 1; + } + + quot->nlimbs = qsize; + quot->sign = sign_quotient; + } + + rsize = dsize; + MPN_NORMALIZE (rp, rsize); + + if( normalization_steps && rsize ) { + _gcry_mpih_rshift(rp, rp, rsize, normalization_steps); + rsize -= rp[rsize - 1] == 0?1:0; + } + + rem->nlimbs = rsize; + rem->sign = sign_remainder; + while( markidx ) + { + markidx--; + _gcry_mpi_free_limb_space (marker[markidx], marker_nlimbs[markidx]); + } +} + +void +_gcry_mpi_tdiv_q_2exp( gcry_mpi_t w, gcry_mpi_t u, unsigned int count ) +{ + mpi_size_t usize, wsize; + mpi_size_t limb_cnt; + + usize = u->nlimbs; + limb_cnt = count / BITS_PER_MPI_LIMB; + wsize = usize - limb_cnt; + if( limb_cnt >= usize ) + w->nlimbs = 0; + else { + mpi_ptr_t wp; + mpi_ptr_t up; + + RESIZE_IF_NEEDED( w, wsize ); + wp = w->d; + up = u->d; + + count %= BITS_PER_MPI_LIMB; + if( count ) { + _gcry_mpih_rshift( wp, up + limb_cnt, wsize, count ); + wsize -= !wp[wsize - 1]; + } + else { + MPN_COPY_INCR( wp, up + limb_cnt, wsize); + } + + w->nlimbs = wsize; + } +} + +/**************** + * Check whether dividend is divisible by divisor + * (note: divisor must fit into a limb) + */ +int +_gcry_mpi_divisible_ui(gcry_mpi_t dividend, ulong divisor ) +{ + return !_gcry_mpih_mod_1( dividend->d, dividend->nlimbs, divisor ); +} + + +void +gcry_mpi_div (gcry_mpi_t quot, gcry_mpi_t rem, gcry_mpi_t dividend, gcry_mpi_t divisor, int round) +{ + if (!round) + { + if (!rem) + { + gcry_mpi_t tmp = mpi_alloc (mpi_get_nlimbs(quot)); + _gcry_mpi_tdiv_qr (quot, tmp, dividend, divisor); + mpi_free (tmp); + } + else + _gcry_mpi_tdiv_qr (quot, rem, dividend, divisor); + } + else if (round < 0) + { + if (!rem) + _gcry_mpi_fdiv_q (quot, dividend, divisor); + else if (!quot) + _gcry_mpi_fdiv_r (rem, dividend, divisor); + else + _gcry_mpi_fdiv_qr (quot, rem, dividend, divisor); + } + else + log_bug ("mpi rounding to ceiling not yet implemented\n"); +} + + + diff --git a/libgcrypt-1.4.6/mpi/mpi-scan.c b/libgcrypt-1.4.6/mpi/mpi-scan.c index 3ee09d2..90699cd 100644 --- a/libgcrypt-1.4.6/mpi/mpi-scan.c +++ b/libgcrypt-1.4.6/mpi/mpi-scan.c @@ -1,132 +1,132 @@ -/* mpi-scan.c - MPI functions
- * Copyright (C) 1998, 2001, 2002, 2003 Free Software Foundation, Inc.
- *
- * This file is part of Libgcrypt.
- *
- * Libgcrypt is free software; you can redistribute it and/or modify
- * it under the terms of the GNU Lesser General Public License as
- * published by the Free Software Foundation; either version 2.1 of
- * the License, or (at your option) any later version.
- *
- * Libgcrypt is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU Lesser General Public License for more details.
- *
- * You should have received a copy of the GNU Lesser General Public
- * License along with this program; if not, write to the Free Software
- * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA
- */
-
-#include <config.h>
-#include <stdio.h>
-#include <stdlib.h>
-#include "mpi-internal.h"
-#include "longlong.h"
-
-/****************
- * Scan through an mpi and return byte for byte. a -1 is returned to indicate
- * the end of the mpi. Scanning is done from the lsb to the msb, returned
- * values are in the range of 0 .. 255.
- *
- * FIXME: This code is VERY ugly!
- */
-int
-_gcry_mpi_getbyte( gcry_mpi_t a, unsigned idx )
-{
- int i, j;
- unsigned n;
- mpi_ptr_t ap;
- mpi_limb_t limb;
-
- ap = a->d;
- for(n=0,i=0; i < a->nlimbs; i++ ) {
- limb = ap[i];
- for( j=0; j < BYTES_PER_MPI_LIMB; j++, n++ )
- if( n == idx )
- return (limb >> j*8) & 0xff;
- }
- return -1;
-}
-
-
-/****************
- * Put a value at position IDX into A. idx counts from lsb to msb
- */
-void
-_gcry_mpi_putbyte( gcry_mpi_t a, unsigned idx, int xc )
-{
- int i, j;
- unsigned n;
- mpi_ptr_t ap;
- mpi_limb_t limb, c;
-
- c = xc & 0xff;
- ap = a->d;
- for(n=0,i=0; i < a->alloced; i++ ) {
- limb = ap[i];
- for( j=0; j < BYTES_PER_MPI_LIMB; j++, n++ )
- if( n == idx ) {
- #if BYTES_PER_MPI_LIMB == 4
- if( j == 0 )
- limb = (limb & 0xffffff00) | c;
- else if( j == 1 )
- limb = (limb & 0xffff00ff) | (c<<8);
- else if( j == 2 )
- limb = (limb & 0xff00ffff) | (c<<16);
- else
- limb = (limb & 0x00ffffff) | (c<<24);
- #elif BYTES_PER_MPI_LIMB == 8
- if( j == 0 )
- limb = (limb & 0xffffffffffffff00) | c;
- else if( j == 1 )
- limb = (limb & 0xffffffffffff00ff) | (c<<8);
- else if( j == 2 )
- limb = (limb & 0xffffffffff00ffff) | (c<<16);
- else if( j == 3 )
- limb = (limb & 0xffffffff00ffffff) | (c<<24);
- else if( j == 4 )
- limb = (limb & 0xffffff00ffffffff) | (c<<32);
- else if( j == 5 )
- limb = (limb & 0xffff00ffffffffff) | (c<<40);
- else if( j == 6 )
- limb = (limb & 0xff00ffffffffffff) | (c<<48);
- else
- limb = (limb & 0x00ffffffffffffff) | (c<<56);
- #else
- #error please enhance this function, its ugly - i know.
- #endif
- if( a->nlimbs <= i )
- a->nlimbs = i+1;
- ap[i] = limb;
- return;
- }
- }
- abort(); /* index out of range */
-}
-
-
-/****************
- * Count the number of zerobits at the low end of A
- */
-unsigned
-_gcry_mpi_trailing_zeros( gcry_mpi_t a )
-{
- unsigned n, count = 0;
-
- for(n=0; n < a->nlimbs; n++ ) {
- if( a->d[n] ) {
- unsigned nn;
- mpi_limb_t alimb = a->d[n];
-
- count_trailing_zeros( nn, alimb );
- count += nn;
- break;
- }
- count += BITS_PER_MPI_LIMB;
- }
- return count;
-
-}
-
-
+/* mpi-scan.c - MPI functions + * Copyright (C) 1998, 2001, 2002, 2003 Free Software Foundation, Inc. + * + * This file is part of Libgcrypt. + * + * Libgcrypt is free software; you can redistribute it and/or modify + * it under the terms of the GNU Lesser General Public License as + * published by the Free Software Foundation; either version 2.1 of + * the License, or (at your option) any later version. + * + * Libgcrypt is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA + */ + +#include <config.h> +#include <stdio.h> +#include <stdlib.h> +#include "mpi-internal.h" +#include "longlong.h" + +/**************** + * Scan through an mpi and return byte for byte. a -1 is returned to indicate + * the end of the mpi. Scanning is done from the lsb to the msb, returned + * values are in the range of 0 .. 255. + * + * FIXME: This code is VERY ugly! + */ +int +_gcry_mpi_getbyte( gcry_mpi_t a, unsigned idx ) +{ + int i, j; + unsigned n; + mpi_ptr_t ap; + mpi_limb_t limb; + + ap = a->d; + for(n=0,i=0; i < a->nlimbs; i++ ) { + limb = ap[i]; + for( j=0; j < BYTES_PER_MPI_LIMB; j++, n++ ) + if( n == idx ) + return (limb >> j*8) & 0xff; + } + return -1; +} + + +/**************** + * Put a value at position IDX into A. idx counts from lsb to msb + */ +void +_gcry_mpi_putbyte( gcry_mpi_t a, unsigned idx, int xc ) +{ + int i, j; + unsigned n; + mpi_ptr_t ap; + mpi_limb_t limb, c; + + c = xc & 0xff; + ap = a->d; + for(n=0,i=0; i < a->alloced; i++ ) { + limb = ap[i]; + for( j=0; j < BYTES_PER_MPI_LIMB; j++, n++ ) + if( n == idx ) { + #if BYTES_PER_MPI_LIMB == 4 + if( j == 0 ) + limb = (limb & 0xffffff00) | c; + else if( j == 1 ) + limb = (limb & 0xffff00ff) | (c<<8); + else if( j == 2 ) + limb = (limb & 0xff00ffff) | (c<<16); + else + limb = (limb & 0x00ffffff) | (c<<24); + #elif BYTES_PER_MPI_LIMB == 8 + if( j == 0 ) + limb = (limb & 0xffffffffffffff00) | c; + else if( j == 1 ) + limb = (limb & 0xffffffffffff00ff) | (c<<8); + else if( j == 2 ) + limb = (limb & 0xffffffffff00ffff) | (c<<16); + else if( j == 3 ) + limb = (limb & 0xffffffff00ffffff) | (c<<24); + else if( j == 4 ) + limb = (limb & 0xffffff00ffffffff) | (c<<32); + else if( j == 5 ) + limb = (limb & 0xffff00ffffffffff) | (c<<40); + else if( j == 6 ) + limb = (limb & 0xff00ffffffffffff) | (c<<48); + else + limb = (limb & 0x00ffffffffffffff) | (c<<56); + #else + #error please enhance this function, its ugly - i know. + #endif + if( a->nlimbs <= i ) + a->nlimbs = i+1; + ap[i] = limb; + return; + } + } + abort(); /* index out of range */ +} + + +/**************** + * Count the number of zerobits at the low end of A + */ +unsigned +_gcry_mpi_trailing_zeros( gcry_mpi_t a ) +{ + unsigned n, count = 0; + + for(n=0; n < a->nlimbs; n++ ) { + if( a->d[n] ) { + unsigned nn; + mpi_limb_t alimb = a->d[n]; + + count_trailing_zeros( nn, alimb ); + count += nn; + break; + } + count += BITS_PER_MPI_LIMB; + } + return count; + +} + + diff --git a/libgcrypt-1.4.6/mpi/mpicoder.c b/libgcrypt-1.4.6/mpi/mpicoder.c index b2d542c..8f0c76f 100644 --- a/libgcrypt-1.4.6/mpi/mpicoder.c +++ b/libgcrypt-1.4.6/mpi/mpicoder.c @@ -1,751 +1,752 @@ -/* mpicoder.c - Coder for the external representation of MPIs
- * Copyright (C) 1998, 1999, 2000, 2001, 2002, 2003
- * 2008 Free Software Foundation, Inc.
- *
- * This file is part of Libgcrypt.
- *
- * Libgcrypt is free software; you can redistribute it and/or modify
- * it under the terms of the GNU Lesser General Public License as
- * published by the Free Software Foundation; either version 2.1 of
- * the License, or (at your option) any later version.
- *
- * Libgcrypt is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU Lesser General Public License for more details.
- *
- * You should have received a copy of the GNU Lesser General Public
- * License along with this program; if not, see <http://www.gnu.org/licenses/>.
- */
-
-#include <config.h>
-#include <stdio.h>
-#include <string.h>
-#include <stdlib.h>
-
-#include "mpi-internal.h"
-#include "g10lib.h"
-
-#define MAX_EXTERN_MPI_BITS 16384
-
-/* Helper used to scan PGP style MPIs. Returns NULL on failure. */
-static gcry_mpi_t
-mpi_read_from_buffer (const unsigned char *buffer, unsigned *ret_nread,
- int secure)
-{
- int i, j;
- unsigned int nbits, nbytes, nlimbs, nread=0;
- mpi_limb_t a;
- gcry_mpi_t val = MPI_NULL;
-
- if ( *ret_nread < 2 )
- goto leave;
- nbits = buffer[0] << 8 | buffer[1];
- if ( nbits > MAX_EXTERN_MPI_BITS )
- {
-/* log_debug ("mpi too large (%u bits)\n", nbits); */
- goto leave;
- }
- buffer += 2;
- nread = 2;
-
- nbytes = (nbits+7) / 8;
- nlimbs = (nbytes+BYTES_PER_MPI_LIMB-1) / BYTES_PER_MPI_LIMB;
- val = secure? mpi_alloc_secure (nlimbs) : mpi_alloc (nlimbs);
- i = BYTES_PER_MPI_LIMB - nbytes % BYTES_PER_MPI_LIMB;
- i %= BYTES_PER_MPI_LIMB;
- j= val->nlimbs = nlimbs;
- val->sign = 0;
- for ( ; j > 0; j-- )
- {
- a = 0;
- for (; i < BYTES_PER_MPI_LIMB; i++ )
- {
- if ( ++nread > *ret_nread )
- {
-/* log_debug ("mpi larger than buffer"); */
- mpi_free (val);
- val = NULL;
- goto leave;
- }
- a <<= 8;
- a |= *buffer++;
- }
- i = 0;
- val->d[j-1] = a;
- }
-
- leave:
- *ret_nread = nread;
- return val;
-}
-
-
-/****************
- * Fill the mpi VAL from the hex string in STR.
- */
-static int
-mpi_fromstr (gcry_mpi_t val, const char *str)
-{
- int sign = 0;
- int prepend_zero = 0;
- int i, j, c, c1, c2;
- unsigned int nbits, nbytes, nlimbs;
- mpi_limb_t a;
-
- if ( *str == '-' )
- {
- sign = 1;
- str++;
- }
-
- /* Skip optional hex prefix. */
- if ( *str == '0' && str[1] == 'x' )
- str += 2;
-
- nbits = 4 * strlen (str);
- if ((nbits % 8))
- prepend_zero = 1;
-
- nbytes = (nbits+7) / 8;
- nlimbs = (nbytes+BYTES_PER_MPI_LIMB-1) / BYTES_PER_MPI_LIMB;
-
- if ( val->alloced < nlimbs )
- mpi_resize (val, nlimbs);
-
- i = BYTES_PER_MPI_LIMB - (nbytes % BYTES_PER_MPI_LIMB);
- i %= BYTES_PER_MPI_LIMB;
- j = val->nlimbs = nlimbs;
- val->sign = sign;
- for (; j > 0; j--)
- {
- a = 0;
- for (; i < BYTES_PER_MPI_LIMB; i++)
- {
- if (prepend_zero)
- {
- c1 = '0';
- prepend_zero = 0;
- }
- else
- c1 = *str++;
-
- if (!c1)
- {
- mpi_clear (val);
- return 1; /* Error. */
- }
- c2 = *str++;
- if (!c2)
- {
- mpi_clear (val);
- return 1; /* Error. */
- }
- if ( c1 >= '0' && c1 <= '9' )
- c = c1 - '0';
- else if ( c1 >= 'a' && c1 <= 'f' )
- c = c1 - 'a' + 10;
- else if ( c1 >= 'A' && c1 <= 'F' )
- c = c1 - 'A' + 10;
- else
- {
- mpi_clear (val);
- return 1; /* Error. */
- }
- c <<= 4;
- if ( c2 >= '0' && c2 <= '9' )
- c |= c2 - '0';
- else if( c2 >= 'a' && c2 <= 'f' )
- c |= c2 - 'a' + 10;
- else if( c2 >= 'A' && c2 <= 'F' )
- c |= c2 - 'A' + 10;
- else
- {
- mpi_clear(val);
- return 1; /* Error. */
- }
- a <<= 8;
- a |= c;
- }
- i = 0;
- val->d[j-1] = a;
- }
-
- return 0; /* Okay. */
-}
-
-
-/* Dump the value of A in a format suitable for debugging to
- Libgcrypt's logging stream. Note that one leading space but no
- trailing space or linefeed will be printed. It is okay to pass
- NULL for A. */
-void
-gcry_mpi_dump (const gcry_mpi_t a)
-{
- int i;
-
- log_printf (" ");
- if (!a)
- log_printf ("[MPI_NULL]");
- else
- {
- if (a->sign)
- log_printf ( "-");
-#if BYTES_PER_MPI_LIMB == 2
-# define X "4"
-#elif BYTES_PER_MPI_LIMB == 4
-# define X "8"
-#elif BYTES_PER_MPI_LIMB == 8
-# define X "16"
-#elif BYTES_PER_MPI_LIMB == 16
-# define X "32"
-#else
-# error please define the format here
-#endif
- for (i=a->nlimbs; i > 0 ; i-- )
- {
- log_printf (i != a->nlimbs? "%0" X "lX":"%lX", (ulong)a->d[i-1]);
- }
-#undef X
- if (!a->nlimbs)
- log_printf ("0");
- }
-}
-
-/* Convience function used internally. */
-void
-_gcry_log_mpidump (const char *text, gcry_mpi_t a)
-{
- log_printf ("%s:", text);
- gcry_mpi_dump (a);
- log_printf ("\n");
-}
-
-
-/* Return an allocated buffer with the MPI (msb first). NBYTES
- receives the length of this buffer. Caller must free the return
- string. This function returns an allocated buffer with NBYTES set
- to zero if the value of A is zero. If sign is not NULL, it will be
- set to the sign of the A. On error NULL is returned and ERRNO set
- appropriately. */
-static unsigned char *
-do_get_buffer (gcry_mpi_t a, unsigned int *nbytes, int *sign, int force_secure)
-{
- unsigned char *p, *buffer;
- mpi_limb_t alimb;
- int i;
- size_t n;
-
- if (sign)
- *sign = a->sign;
-
- *nbytes = a->nlimbs * BYTES_PER_MPI_LIMB;
- n = *nbytes? *nbytes:1; /* Allocate at least one byte. */
- p = buffer = (force_secure || mpi_is_secure(a))? gcry_malloc_secure (n)
- : gcry_malloc (n);
- if (!buffer)
- return NULL;
-
- for (i=a->nlimbs-1; i >= 0; i--)
- {
- alimb = a->d[i];
-#if BYTES_PER_MPI_LIMB == 4
- *p++ = alimb >> 24;
- *p++ = alimb >> 16;
- *p++ = alimb >> 8;
- *p++ = alimb ;
-#elif BYTES_PER_MPI_LIMB == 8
- *p++ = alimb >> 56;
- *p++ = alimb >> 48;
- *p++ = alimb >> 40;
- *p++ = alimb >> 32;
- *p++ = alimb >> 24;
- *p++ = alimb >> 16;
- *p++ = alimb >> 8;
- *p++ = alimb ;
-#else
-# error please implement for this limb size.
-#endif
- }
-
- /* This is sub-optimal but we need to do the shift operation because
- the caller has to free the returned buffer. */
- for (p=buffer; !*p && *nbytes; p++, --*nbytes)
- ;
- if (p != buffer)
- memmove (buffer,p, *nbytes);
- return buffer;
-}
-
-
-byte *
-_gcry_mpi_get_buffer (gcry_mpi_t a, unsigned int *nbytes, int *sign)
-{
- return do_get_buffer (a, nbytes, sign, 0);
-}
-
-byte *
-_gcry_mpi_get_secure_buffer (gcry_mpi_t a, unsigned *nbytes, int *sign)
-{
- return do_get_buffer (a, nbytes, sign, 1);
-}
-
-
-/*
- * Use the NBYTES at BUFFER_ARG to update A. Set the sign of a to
- * SIGN.
- */
-void
-_gcry_mpi_set_buffer (gcry_mpi_t a, const void *buffer_arg,
- unsigned int nbytes, int sign)
-{
- const unsigned char *buffer = (const unsigned char*)buffer_arg;
- const unsigned char *p;
- mpi_limb_t alimb;
- int nlimbs;
- int i;
-
- nlimbs = (nbytes + BYTES_PER_MPI_LIMB - 1) / BYTES_PER_MPI_LIMB;
- RESIZE_IF_NEEDED(a, nlimbs);
- a->sign = sign;
-
- for (i=0, p = buffer+nbytes-1; p >= buffer+BYTES_PER_MPI_LIMB; )
- {
-#if BYTES_PER_MPI_LIMB == 4
- alimb = *p-- ;
- alimb |= *p-- << 8 ;
- alimb |= *p-- << 16 ;
- alimb |= *p-- << 24 ;
-#elif BYTES_PER_MPI_LIMB == 8
- alimb = (mpi_limb_t)*p-- ;
- alimb |= (mpi_limb_t)*p-- << 8 ;
- alimb |= (mpi_limb_t)*p-- << 16 ;
- alimb |= (mpi_limb_t)*p-- << 24 ;
- alimb |= (mpi_limb_t)*p-- << 32 ;
- alimb |= (mpi_limb_t)*p-- << 40 ;
- alimb |= (mpi_limb_t)*p-- << 48 ;
- alimb |= (mpi_limb_t)*p-- << 56 ;
-#else
-# error please implement for this limb size.
-#endif
- a->d[i++] = alimb;
- }
- if ( p >= buffer )
- {
-#if BYTES_PER_MPI_LIMB == 4
- alimb = *p--;
- if (p >= buffer)
- alimb |= *p-- << 8;
- if (p >= buffer)
- alimb |= *p-- << 16;
- if (p >= buffer)
- alimb |= *p-- << 24;
-#elif BYTES_PER_MPI_LIMB == 8
- alimb = (mpi_limb_t)*p--;
- if (p >= buffer)
- alimb |= (mpi_limb_t)*p-- << 8;
- if (p >= buffer)
- alimb |= (mpi_limb_t)*p-- << 16;
- if (p >= buffer)
- alimb |= (mpi_limb_t)*p-- << 24;
- if (p >= buffer)
- alimb |= (mpi_limb_t)*p-- << 32;
- if (p >= buffer)
- alimb |= (mpi_limb_t)*p-- << 40;
- if (p >= buffer)
- alimb |= (mpi_limb_t)*p-- << 48;
- if (p >= buffer)
- alimb |= (mpi_limb_t)*p-- << 56;
-#else
-# error please implement for this limb size.
-#endif
- a->d[i++] = alimb;
- }
- a->nlimbs = i;
- gcry_assert (i == nlimbs);
-}
-
-
-/* Convert the external representation of an integer stored in BUFFER
- with a length of BUFLEN into a newly create MPI returned in
- RET_MPI. If NBYTES is not NULL, it will receive the number of
- bytes actually scanned after a successful operation. */
-gcry_error_t
-gcry_mpi_scan (struct gcry_mpi **ret_mpi, enum gcry_mpi_format format,
- const void *buffer_arg, size_t buflen, size_t *nscanned)
-{
- const unsigned char *buffer = (const unsigned char*)buffer_arg;
- struct gcry_mpi *a = NULL;
- unsigned int len;
- int secure = (buffer && gcry_is_secure (buffer));
-
- if (format == GCRYMPI_FMT_SSH)
- len = 0;
- else
- len = buflen;
-
- if (format == GCRYMPI_FMT_STD)
- {
- const unsigned char *s = buffer;
-
- a = secure? mpi_alloc_secure ((len+BYTES_PER_MPI_LIMB-1)
- /BYTES_PER_MPI_LIMB)
- : mpi_alloc ((len+BYTES_PER_MPI_LIMB-1)/BYTES_PER_MPI_LIMB);
- if (len)
- {
- a->sign = !!(*s & 0x80);
- if (a->sign)
- {
- /* FIXME: we have to convert from 2compl to magnitude format */
- mpi_free (a);
- return gcry_error (GPG_ERR_INTERNAL);
- }
- else
- _gcry_mpi_set_buffer (a, s, len, 0);
- }
- if (ret_mpi)
- {
- mpi_normalize ( a );
- *ret_mpi = a;
- }
- else
- mpi_free(a);
- return 0;
- }
- else if (format == GCRYMPI_FMT_USG)
- {
- a = secure? mpi_alloc_secure ((len+BYTES_PER_MPI_LIMB-1)
- /BYTES_PER_MPI_LIMB)
- : mpi_alloc ((len+BYTES_PER_MPI_LIMB-1)/BYTES_PER_MPI_LIMB);
-
- if (len)
- _gcry_mpi_set_buffer (a, buffer, len, 0);
- if (ret_mpi)
- {
- mpi_normalize ( a );
- *ret_mpi = a;
- }
- else
- mpi_free(a);
- return 0;
- }
- else if (format == GCRYMPI_FMT_PGP)
- {
- a = mpi_read_from_buffer (buffer, &len, secure);
- if (nscanned)
- *nscanned = len;
- if (ret_mpi && a)
- {
- mpi_normalize (a);
- *ret_mpi = a;
- }
- else if (a)
- {
- mpi_free(a);
- a = NULL;
- }
- return a? 0 : gcry_error (GPG_ERR_INV_OBJ);
- }
- else if (format == GCRYMPI_FMT_SSH)
- {
- const unsigned char *s = buffer;
- size_t n;
-
- /* This test is not strictly necessary and an assert (!len)
- would be sufficient. We keep this test in case we later
- allow the BUFLEN argument to act as a sanitiy check. Same
- below. */
- if (len && len < 4)
- return gcry_error (GPG_ERR_TOO_SHORT);
-
- n = (s[0] << 24 | s[1] << 16 | s[2] << 8 | s[3]);
- s += 4;
- if (len)
- len -= 4;
- if (len && n > len)
- return gcry_error (GPG_ERR_TOO_LARGE);
-
- a = secure? mpi_alloc_secure ((n+BYTES_PER_MPI_LIMB-1)
- /BYTES_PER_MPI_LIMB)
- : mpi_alloc ((n+BYTES_PER_MPI_LIMB-1)/BYTES_PER_MPI_LIMB);
- if (n)
- {
- a->sign = !!(*s & 0x80);
- if (a->sign)
- {
- /* FIXME: we have to convert from 2compl to magnitude format */
- mpi_free(a);
- return gcry_error (GPG_ERR_INTERNAL);
- }
- else
- _gcry_mpi_set_buffer( a, s, n, 0 );
- }
- if (nscanned)
- *nscanned = n+4;
- if (ret_mpi)
- {
- mpi_normalize ( a );
- *ret_mpi = a;
- }
- else
- mpi_free(a);
- return 0;
- }
- else if (format == GCRYMPI_FMT_HEX)
- {
- /* We can only handle C strings for now. */
- if (buflen)
- return gcry_error (GPG_ERR_INV_ARG);
-
- a = secure? mpi_alloc_secure (0) : mpi_alloc(0);
- if (mpi_fromstr (a, (const char *)buffer))
- {
- mpi_free (a);
- return gcry_error (GPG_ERR_INV_OBJ);
- }
- if (ret_mpi)
- {
- mpi_normalize ( a );
- *ret_mpi = a;
- }
- else
- mpi_free(a);
- return 0;
- }
- else
- return gcry_error (GPG_ERR_INV_ARG);
-}
-
-
-/* Convert the big integer A into the external representation
- described by FORMAT and store it in the provided BUFFER which has
- been allocated by the user with a size of BUFLEN bytes. NWRITTEN
- receives the actual length of the external representation unless it
- has been passed as NULL. BUFFER may be NULL to query the required
- length. */
-gcry_error_t
-gcry_mpi_print (enum gcry_mpi_format format,
- unsigned char *buffer, size_t buflen,
- size_t *nwritten, struct gcry_mpi *a)
-{
- unsigned int nbits = mpi_get_nbits (a);
- size_t len;
- size_t dummy_nwritten;
-
- if (!nwritten)
- nwritten = &dummy_nwritten;
-
- len = buflen;
- *nwritten = 0;
- if (format == GCRYMPI_FMT_STD)
- {
- unsigned char *tmp;
- int extra = 0;
- unsigned int n;
-
- if (a->sign)
- return gcry_error (GPG_ERR_INTERNAL); /* Can't handle it yet. */
-
- tmp = _gcry_mpi_get_buffer (a, &n, NULL);
- if (!tmp)
- return gpg_error_from_syserror ();
- if (n && (*tmp & 0x80))
- {
- n++;
- extra=1;
- }
-
- if (buffer && n > len)
- {
- /* The provided buffer is too short. */
- gcry_free (tmp);
- return gcry_error (GPG_ERR_TOO_SHORT);
- }
- if (buffer)
- {
- unsigned char *s = buffer;
-
- if (extra)
- *s++ = 0;
- memcpy (s, tmp, n-extra);
- }
- gcry_free(tmp);
- *nwritten = n;
- return 0;
- }
- else if (format == GCRYMPI_FMT_USG)
- {
- unsigned int n = (nbits + 7)/8;
-
- /* Note: We ignore the sign for this format. */
- /* FIXME: for performance reasons we should put this into
- mpi_aprint because we can then use the buffer directly. */
- if (buffer && n > len)
- return gcry_error (GPG_ERR_TOO_SHORT);
- if (buffer)
- {
- unsigned char *tmp;
-
- tmp = _gcry_mpi_get_buffer (a, &n, NULL);
- if (!tmp)
- return gpg_error_from_syserror ();
- memcpy (buffer, tmp, n);
- gcry_free (tmp);
- }
- *nwritten = n;
- return 0;
- }
- else if (format == GCRYMPI_FMT_PGP)
- {
- unsigned int n = (nbits + 7)/8;
-
- /* The PGP format can only handle unsigned integers. */
- if( a->sign )
- return gcry_error (GPG_ERR_INV_ARG);
-
- if (buffer && n+2 > len)
- return gcry_error (GPG_ERR_TOO_SHORT);
-
- if (buffer)
- {
- unsigned char *tmp;
- unsigned char *s = buffer;
-
- s[0] = nbits >> 8;
- s[1] = nbits;
-
- tmp = _gcry_mpi_get_buffer (a, &n, NULL);
- if (!tmp)
- return gpg_error_from_syserror ();
- memcpy (s+2, tmp, n);
- gcry_free (tmp);
- }
- *nwritten = n+2;
- return 0;
- }
- else if (format == GCRYMPI_FMT_SSH)
- {
- unsigned char *tmp;
- int extra = 0;
- unsigned int n;
-
- if (a->sign)
- return gcry_error (GPG_ERR_INTERNAL); /* Can't handle it yet. */
-
- tmp = _gcry_mpi_get_buffer (a, &n, NULL);
- if (!tmp)
- return gpg_error_from_syserror ();
- if (n && (*tmp & 0x80))
- {
- n++;
- extra=1;
- }
-
- if (buffer && n+4 > len)
- {
- gcry_free(tmp);
- return gcry_error (GPG_ERR_TOO_SHORT);
- }
-
- if (buffer)
- {
- unsigned char *s = buffer;
-
- *s++ = n >> 24;
- *s++ = n >> 16;
- *s++ = n >> 8;
- *s++ = n;
- if (extra)
- *s++ = 0;
-
- memcpy (s, tmp, n-extra);
- }
- gcry_free (tmp);
- *nwritten = 4+n;
- return 0;
- }
- else if (format == GCRYMPI_FMT_HEX)
- {
- unsigned char *tmp;
- int i;
- int extra = 0;
- unsigned int n = 0;
-
- tmp = _gcry_mpi_get_buffer (a, &n, NULL);
- if (!tmp)
- return gpg_error_from_syserror ();
- if (!n || (*tmp & 0x80))
- extra = 2;
-
- if (buffer && 2*n + extra + !!a->sign + 1 > len)
- {
- gcry_free(tmp);
- return gcry_error (GPG_ERR_TOO_SHORT);
- }
- if (buffer)
- {
- unsigned char *s = buffer;
-
- if (a->sign)
- *s++ = '-';
- if (extra)
- {
- *s++ = '0';
- *s++ = '0';
- }
-
- for (i=0; i < n; i++)
- {
- unsigned int c = tmp[i];
-
- *s++ = (c >> 4) < 10? '0'+(c>>4) : 'A'+(c>>4)-10 ;
- c &= 15;
- *s++ = c < 10? '0'+c : 'A'+c-10 ;
- }
- *s++ = 0;
- *nwritten = s - buffer;
- }
- else
- {
- *nwritten = 2*n + extra + !!a->sign + 1;
- }
- gcry_free (tmp);
- return 0;
- }
- else
- return gcry_error (GPG_ERR_INV_ARG);
-}
-
-
-/*
- * Like gcry_mpi_print but this function allocates the buffer itself.
- * The caller has to supply the address of a pointer. NWRITTEN may be
- * NULL.
- */
-gcry_error_t
-gcry_mpi_aprint (enum gcry_mpi_format format,
- unsigned char **buffer, size_t *nwritten,
- struct gcry_mpi *a)
-{
- size_t n;
- gcry_error_t rc;
-
- *buffer = NULL;
- rc = gcry_mpi_print (format, NULL, 0, &n, a);
- if (rc)
- return rc;
-
- *buffer = mpi_is_secure(a) ? gcry_malloc_secure (n) : gcry_malloc (n);
- if (!*buffer)
- return gpg_error_from_syserror ();
- rc = gcry_mpi_print( format, *buffer, n, &n, a );
- if (rc)
- {
- gcry_free(*buffer);
- *buffer = NULL;
- }
- else if (nwritten)
- *nwritten = n;
- return rc;
-}
-
+/* mpicoder.c - Coder for the external representation of MPIs + * Copyright (C) 1998, 1999, 2000, 2001, 2002, 2003 + * 2008 Free Software Foundation, Inc. + * + * This file is part of Libgcrypt. + * + * Libgcrypt is free software; you can redistribute it and/or modify + * it under the terms of the GNU Lesser General Public License as + * published by the Free Software Foundation; either version 2.1 of + * the License, or (at your option) any later version. + * + * Libgcrypt is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with this program; if not, see <http://www.gnu.org/licenses/>. + */ + +#include <config.h> +#include <stdio.h> +#include <string.h> +#include <stdlib.h> + +#include "mpi-internal.h" +#include "g10lib.h" + +#define MAX_EXTERN_MPI_BITS 16384 + +/* Helper used to scan PGP style MPIs. Returns NULL on failure. */ +static gcry_mpi_t +mpi_read_from_buffer (const unsigned char *buffer, unsigned *ret_nread, + int secure) +{ + int i, j; + unsigned int nbits, nbytes, nlimbs, nread=0; + mpi_limb_t a; + gcry_mpi_t val = MPI_NULL; + + if ( *ret_nread < 2 ) + goto leave; + nbits = buffer[0] << 8 | buffer[1]; + if ( nbits > MAX_EXTERN_MPI_BITS ) + { +/* log_debug ("mpi too large (%u bits)\n", nbits); */ + goto leave; + } + else if( !nbits ) + { +/* log_debug ("an mpi of size 0 is not allowed\n"); */ + goto leave; + } + buffer += 2; + nread = 2; + + nbytes = (nbits+7) / 8; + nlimbs = (nbytes+BYTES_PER_MPI_LIMB-1) / BYTES_PER_MPI_LIMB; + val = secure? mpi_alloc_secure (nlimbs) : mpi_alloc (nlimbs); + i = BYTES_PER_MPI_LIMB - nbytes % BYTES_PER_MPI_LIMB; + i %= BYTES_PER_MPI_LIMB; + j= val->nlimbs = nlimbs; + val->sign = 0; + for ( ; j > 0; j-- ) + { + a = 0; + for (; i < BYTES_PER_MPI_LIMB; i++ ) + { + if ( ++nread > *ret_nread ) + { +/* log_debug ("mpi larger than buffer"); */ + mpi_free (val); + val = NULL; + goto leave; + } + a <<= 8; + a |= *buffer++; + } + i = 0; + val->d[j-1] = a; + } + + leave: + *ret_nread = nread; + return val; +} + + +/**************** + * Fill the mpi VAL from the hex string in STR. + */ +static int +mpi_fromstr (gcry_mpi_t val, const char *str) +{ + int sign = 0; + int prepend_zero = 0; + int i, j, c, c1, c2; + unsigned int nbits, nbytes, nlimbs; + mpi_limb_t a; + + if ( *str == '-' ) + { + sign = 1; + str++; + } + + /* Skip optional hex prefix. */ + if ( *str == '0' && str[1] == 'x' ) + str += 2; + + nbits = 4 * strlen (str); + if ((nbits % 8)) + prepend_zero = 1; + + nbytes = (nbits+7) / 8; + nlimbs = (nbytes+BYTES_PER_MPI_LIMB-1) / BYTES_PER_MPI_LIMB; + + if ( val->alloced < nlimbs ) + mpi_resize (val, nlimbs); + + i = BYTES_PER_MPI_LIMB - (nbytes % BYTES_PER_MPI_LIMB); + i %= BYTES_PER_MPI_LIMB; + j = val->nlimbs = nlimbs; + val->sign = sign; + for (; j > 0; j--) + { + a = 0; + for (; i < BYTES_PER_MPI_LIMB; i++) + { + if (prepend_zero) + { + c1 = '0'; + prepend_zero = 0; + } + else + c1 = *str++; + + if (!c1) + { + mpi_clear (val); + return 1; /* Error. */ + } + c2 = *str++; + if (!c2) + { + mpi_clear (val); + return 1; /* Error. */ + } + if ( c1 >= '0' && c1 <= '9' ) + c = c1 - '0'; + else if ( c1 >= 'a' && c1 <= 'f' ) + c = c1 - 'a' + 10; + else if ( c1 >= 'A' && c1 <= 'F' ) + c = c1 - 'A' + 10; + else + { + mpi_clear (val); + return 1; /* Error. */ + } + c <<= 4; + if ( c2 >= '0' && c2 <= '9' ) + c |= c2 - '0'; + else if( c2 >= 'a' && c2 <= 'f' ) + c |= c2 - 'a' + 10; + else if( c2 >= 'A' && c2 <= 'F' ) + c |= c2 - 'A' + 10; + else + { + mpi_clear(val); + return 1; /* Error. */ + } + a <<= 8; + a |= c; + } + i = 0; + val->d[j-1] = a; + } + + return 0; /* Okay. */ +} + + +/* Dump the value of A in a format suitable for debugging to + Libgcrypt's logging stream. Note that one leading space but no + trailing space or linefeed will be printed. It is okay to pass + NULL for A. */ +void +gcry_mpi_dump (const gcry_mpi_t a) +{ + int i; + + log_printf (" "); + if (!a) + log_printf ("[MPI_NULL]"); + else + { + if (a->sign) + log_printf ( "-"); +#if BYTES_PER_MPI_LIMB == 2 +# define X "4" +#elif BYTES_PER_MPI_LIMB == 4 +# define X "8" +#elif BYTES_PER_MPI_LIMB == 8 +# define X "16" +#elif BYTES_PER_MPI_LIMB == 16 +# define X "32" +#else +# error please define the format here +#endif + for (i=a->nlimbs; i > 0 ; i-- ) + { + log_printf (i != a->nlimbs? "%0" X "lX":"%lX", (ulong)a->d[i-1]); + } +#undef X + if (!a->nlimbs) + log_printf ("0"); + } +} + +/* Convience function used internally. */ +void +_gcry_log_mpidump (const char *text, gcry_mpi_t a) +{ + log_printf ("%s:", text); + gcry_mpi_dump (a); + log_printf ("\n"); +} + + +/* Return an allocated buffer with the MPI (msb first). NBYTES + receives the length of this buffer. Caller must free the return + string. This function returns an allocated buffer with NBYTES set + to zero if the value of A is zero. If sign is not NULL, it will be + set to the sign of the A. On error NULL is returned and ERRNO set + appropriately. */ +static unsigned char * +do_get_buffer (gcry_mpi_t a, unsigned int *nbytes, int *sign, int force_secure) +{ + unsigned char *p, *buffer; + mpi_limb_t alimb; + int i; + size_t n; + + if (sign) + *sign = a->sign; + + *nbytes = a->nlimbs * BYTES_PER_MPI_LIMB; + n = *nbytes? *nbytes:1; /* Allocate at least one byte. */ + p = buffer = (force_secure || mpi_is_secure(a))? gcry_malloc_secure (n) + : gcry_malloc (n); + if (!buffer) + return NULL; + + for (i=a->nlimbs-1; i >= 0; i--) + { + alimb = a->d[i]; +#if BYTES_PER_MPI_LIMB == 4 + *p++ = alimb >> 24; + *p++ = alimb >> 16; + *p++ = alimb >> 8; + *p++ = alimb ; +#elif BYTES_PER_MPI_LIMB == 8 + *p++ = alimb >> 56; + *p++ = alimb >> 48; + *p++ = alimb >> 40; + *p++ = alimb >> 32; + *p++ = alimb >> 24; + *p++ = alimb >> 16; + *p++ = alimb >> 8; + *p++ = alimb ; +#else +# error please implement for this limb size. +#endif + } + + /* This is sub-optimal but we need to do the shift operation because + the caller has to free the returned buffer. */ + for (p=buffer; !*p && *nbytes; p++, --*nbytes) + ; + if (p != buffer) + memmove (buffer,p, *nbytes); + return buffer; +} + + +byte * +_gcry_mpi_get_buffer (gcry_mpi_t a, unsigned int *nbytes, int *sign) +{ + return do_get_buffer (a, nbytes, sign, 0); +} + +byte * +_gcry_mpi_get_secure_buffer (gcry_mpi_t a, unsigned *nbytes, int *sign) +{ + return do_get_buffer (a, nbytes, sign, 1); +} + + +/* + * Use the NBYTES at BUFFER_ARG to update A. Set the sign of a to + * SIGN. + */ +void +_gcry_mpi_set_buffer (gcry_mpi_t a, const void *buffer_arg, + unsigned int nbytes, int sign) +{ + const unsigned char *buffer = (const unsigned char*)buffer_arg; + const unsigned char *p; + mpi_limb_t alimb; + int nlimbs; + int i; + + nlimbs = (nbytes + BYTES_PER_MPI_LIMB - 1) / BYTES_PER_MPI_LIMB; + RESIZE_IF_NEEDED(a, nlimbs); + a->sign = sign; + + for (i=0, p = buffer+nbytes-1; p >= buffer+BYTES_PER_MPI_LIMB; ) + { +#if BYTES_PER_MPI_LIMB == 4 + alimb = *p-- ; + alimb |= *p-- << 8 ; + alimb |= *p-- << 16 ; + alimb |= *p-- << 24 ; +#elif BYTES_PER_MPI_LIMB == 8 + alimb = (mpi_limb_t)*p-- ; + alimb |= (mpi_limb_t)*p-- << 8 ; + alimb |= (mpi_limb_t)*p-- << 16 ; + alimb |= (mpi_limb_t)*p-- << 24 ; + alimb |= (mpi_limb_t)*p-- << 32 ; + alimb |= (mpi_limb_t)*p-- << 40 ; + alimb |= (mpi_limb_t)*p-- << 48 ; + alimb |= (mpi_limb_t)*p-- << 56 ; +#else +# error please implement for this limb size. +#endif + a->d[i++] = alimb; + } + if ( p >= buffer ) + { +#if BYTES_PER_MPI_LIMB == 4 + alimb = *p--; + if (p >= buffer) + alimb |= *p-- << 8; + if (p >= buffer) + alimb |= *p-- << 16; + if (p >= buffer) + alimb |= *p-- << 24; +#elif BYTES_PER_MPI_LIMB == 8 + alimb = (mpi_limb_t)*p--; + if (p >= buffer) + alimb |= (mpi_limb_t)*p-- << 8; + if (p >= buffer) + alimb |= (mpi_limb_t)*p-- << 16; + if (p >= buffer) + alimb |= (mpi_limb_t)*p-- << 24; + if (p >= buffer) + alimb |= (mpi_limb_t)*p-- << 32; + if (p >= buffer) + alimb |= (mpi_limb_t)*p-- << 40; + if (p >= buffer) + alimb |= (mpi_limb_t)*p-- << 48; + if (p >= buffer) + alimb |= (mpi_limb_t)*p-- << 56; +#else +# error please implement for this limb size. +#endif + a->d[i++] = alimb; + } + a->nlimbs = i; + gcry_assert (i == nlimbs); +} + + +/* Convert the external representation of an integer stored in BUFFER + with a length of BUFLEN into a newly create MPI returned in + RET_MPI. If NBYTES is not NULL, it will receive the number of + bytes actually scanned after a successful operation. */ +gcry_error_t +gcry_mpi_scan (struct gcry_mpi **ret_mpi, enum gcry_mpi_format format, + const void *buffer_arg, size_t buflen, size_t *nscanned) +{ + const unsigned char *buffer = (const unsigned char*)buffer_arg; + struct gcry_mpi *a = NULL; + unsigned int len; + int secure = (buffer && gcry_is_secure (buffer)); + + if (format == GCRYMPI_FMT_SSH) + len = 0; + else + len = buflen; + + if (format == GCRYMPI_FMT_STD) + { + const unsigned char *s = buffer; + + a = secure? mpi_alloc_secure ((len+BYTES_PER_MPI_LIMB-1) + /BYTES_PER_MPI_LIMB) + : mpi_alloc ((len+BYTES_PER_MPI_LIMB-1)/BYTES_PER_MPI_LIMB); + if (len) + { + a->sign = !!(*s & 0x80); + if (a->sign) + { + /* FIXME: we have to convert from 2compl to magnitude format */ + mpi_free (a); + return gcry_error (GPG_ERR_INTERNAL); + } + else + _gcry_mpi_set_buffer (a, s, len, 0); + } + if (ret_mpi) + { + mpi_normalize ( a ); + *ret_mpi = a; + } + else + mpi_free(a); + return 0; + } + else if (format == GCRYMPI_FMT_USG) + { + a = secure? mpi_alloc_secure ((len+BYTES_PER_MPI_LIMB-1) + /BYTES_PER_MPI_LIMB) + : mpi_alloc ((len+BYTES_PER_MPI_LIMB-1)/BYTES_PER_MPI_LIMB); + + if (len) + _gcry_mpi_set_buffer (a, buffer, len, 0); + if (ret_mpi) + { + mpi_normalize ( a ); + *ret_mpi = a; + } + else + mpi_free(a); + return 0; + } + else if (format == GCRYMPI_FMT_PGP) + { + a = mpi_read_from_buffer (buffer, &len, secure); + if (nscanned) + *nscanned = len; + if (ret_mpi && a) + { + mpi_normalize (a); + *ret_mpi = a; + } + else if (a) + { + mpi_free(a); + a = NULL; + } + return a? 0 : gcry_error (GPG_ERR_INV_OBJ); + } + else if (format == GCRYMPI_FMT_SSH) + { + const unsigned char *s = buffer; + size_t n; + + if (len && len < 4) + return gcry_error (GPG_ERR_TOO_SHORT); + + n = (s[0] << 24 | s[1] << 16 | s[2] << 8 | s[3]); + s += 4; + if (len) + len -= 4; + if (len && n > len) + return gcry_error (GPG_ERR_TOO_LARGE); + + a = secure? mpi_alloc_secure ((n+BYTES_PER_MPI_LIMB-1) + /BYTES_PER_MPI_LIMB) + : mpi_alloc ((n+BYTES_PER_MPI_LIMB-1)/BYTES_PER_MPI_LIMB); + if (n) + { + a->sign = !!(*s & 0x80); + if (a->sign) + { + /* FIXME: we have to convert from 2compl to magnitude format */ + mpi_free(a); + return gcry_error (GPG_ERR_INTERNAL); + } + else + _gcry_mpi_set_buffer( a, s, n, 0 ); + } + if (nscanned) + *nscanned = n+4; + if (ret_mpi) + { + mpi_normalize ( a ); + *ret_mpi = a; + } + else + mpi_free(a); + return 0; + } + else if (format == GCRYMPI_FMT_HEX) + { + /* We can only handle C strings for now. */ + if (buflen) + return gcry_error (GPG_ERR_INV_ARG); + + a = secure? mpi_alloc_secure (0) : mpi_alloc(0); + if (mpi_fromstr (a, (const char *)buffer)) + { + mpi_free (a); + return gcry_error (GPG_ERR_INV_OBJ); + } + if (ret_mpi) + { + mpi_normalize ( a ); + *ret_mpi = a; + } + else + mpi_free(a); + return 0; + } + else + return gcry_error (GPG_ERR_INV_ARG); +} + + +/* Convert the big integer A into the external representation + described by FORMAT and store it in the provided BUFFER which has + been allocated by the user with a size of BUFLEN bytes. NWRITTEN + receives the actual length of the external representation unless it + has been passed as NULL. BUFFER may be NULL to query the required + length. */ +gcry_error_t +gcry_mpi_print (enum gcry_mpi_format format, + unsigned char *buffer, size_t buflen, + size_t *nwritten, struct gcry_mpi *a) +{ + unsigned int nbits = mpi_get_nbits (a); + size_t len; + size_t dummy_nwritten; + + if (!nwritten) + nwritten = &dummy_nwritten; + + len = buflen; + *nwritten = 0; + if (format == GCRYMPI_FMT_STD) + { + unsigned char *tmp; + int extra = 0; + unsigned int n; + + if (a->sign) + return gcry_error (GPG_ERR_INTERNAL); /* Can't handle it yet. */ + + tmp = _gcry_mpi_get_buffer (a, &n, NULL); + if (!tmp) + return gpg_error_from_syserror (); + if (n && (*tmp & 0x80)) + { + n++; + extra=1; + } + + if (buffer && n > len) + { + /* The provided buffer is too short. */ + gcry_free (tmp); + return gcry_error (GPG_ERR_TOO_SHORT); + } + if (buffer) + { + unsigned char *s = buffer; + + if (extra) + *s++ = 0; + memcpy (s, tmp, n-extra); + } + gcry_free(tmp); + *nwritten = n; + return 0; + } + else if (format == GCRYMPI_FMT_USG) + { + unsigned int n = (nbits + 7)/8; + + /* Note: We ignore the sign for this format. */ + /* FIXME: for performance reasons we should put this into + mpi_aprint because we can then use the buffer directly. */ + if (buffer && n > len) + return gcry_error (GPG_ERR_TOO_SHORT); + if (buffer) + { + unsigned char *tmp; + + tmp = _gcry_mpi_get_buffer (a, &n, NULL); + if (!tmp) + return gpg_error_from_syserror (); + memcpy (buffer, tmp, n); + gcry_free (tmp); + } + *nwritten = n; + return 0; + } + else if (format == GCRYMPI_FMT_PGP) + { + unsigned int n = (nbits + 7)/8; + + /* The PGP format can only handle unsigned integers. */ + if( a->sign ) + return gcry_error (GPG_ERR_INV_ARG); + + if (buffer && n+2 > len) + return gcry_error (GPG_ERR_TOO_SHORT); + + if (buffer) + { + unsigned char *tmp; + unsigned char *s = buffer; + + s[0] = nbits >> 8; + s[1] = nbits; + + tmp = _gcry_mpi_get_buffer (a, &n, NULL); + if (!tmp) + return gpg_error_from_syserror (); + memcpy (s+2, tmp, n); + gcry_free (tmp); + } + *nwritten = n+2; + return 0; + } + else if (format == GCRYMPI_FMT_SSH) + { + unsigned char *tmp; + int extra = 0; + unsigned int n; + + if (a->sign) + return gcry_error (GPG_ERR_INTERNAL); /* Can't handle it yet. */ + + tmp = _gcry_mpi_get_buffer (a, &n, NULL); + if (!tmp) + return gpg_error_from_syserror (); + if (n && (*tmp & 0x80)) + { + n++; + extra=1; + } + + if (buffer && n+4 > len) + { + gcry_free(tmp); + return gcry_error (GPG_ERR_TOO_SHORT); + } + + if (buffer) + { + unsigned char *s = buffer; + + *s++ = n >> 24; + *s++ = n >> 16; + *s++ = n >> 8; + *s++ = n; + if (extra) + *s++ = 0; + + memcpy (s, tmp, n-extra); + } + gcry_free (tmp); + *nwritten = 4+n; + return 0; + } + else if (format == GCRYMPI_FMT_HEX) + { + unsigned char *tmp; + int i; + int extra = 0; + unsigned int n = 0; + + tmp = _gcry_mpi_get_buffer (a, &n, NULL); + if (!tmp) + return gpg_error_from_syserror (); + if (!n || (*tmp & 0x80)) + extra = 2; + + if (buffer && 2*n + extra + !!a->sign + 1 > len) + { + gcry_free(tmp); + return gcry_error (GPG_ERR_TOO_SHORT); + } + if (buffer) + { + unsigned char *s = buffer; + + if (a->sign) + *s++ = '-'; + if (extra) + { + *s++ = '0'; + *s++ = '0'; + } + + for (i=0; i < n; i++) + { + unsigned int c = tmp[i]; + + *s++ = (c >> 4) < 10? '0'+(c>>4) : 'A'+(c>>4)-10 ; + c &= 15; + *s++ = c < 10? '0'+c : 'A'+c-10 ; + } + *s++ = 0; + *nwritten = s - buffer; + } + else + { + *nwritten = 2*n + extra + !!a->sign + 1; + } + gcry_free (tmp); + return 0; + } + else + return gcry_error (GPG_ERR_INV_ARG); +} + + +/* + * Like gcry_mpi_print but this function allocates the buffer itself. + * The caller has to supply the address of a pointer. NWRITTEN may be + * NULL. + */ +gcry_error_t +gcry_mpi_aprint (enum gcry_mpi_format format, + unsigned char **buffer, size_t *nwritten, + struct gcry_mpi *a) +{ + size_t n; + gcry_error_t rc; + + *buffer = NULL; + rc = gcry_mpi_print (format, NULL, 0, &n, a); + if (rc) + return rc; + + *buffer = mpi_is_secure(a) ? gcry_malloc_secure (n) : gcry_malloc (n); + if (!*buffer) + return gpg_error_from_syserror (); + rc = gcry_mpi_print( format, *buffer, n, &n, a ); + if (rc) + { + gcry_free(*buffer); + *buffer = NULL; + } + else if (nwritten) + *nwritten = n; + return rc; +} + diff --git a/libgcrypt-1.4.6/mpi/mpiutil.c b/libgcrypt-1.4.6/mpi/mpiutil.c index f406cbc..4dc5211 100644 --- a/libgcrypt-1.4.6/mpi/mpiutil.c +++ b/libgcrypt-1.4.6/mpi/mpiutil.c @@ -1,461 +1,462 @@ -/* mpiutil.ac - Utility functions for MPI
- * Copyright (C) 1998, 2000, 2001, 2002, 2003,
- * 2007 Free Software Foundation, Inc.
- *
- * This file is part of Libgcrypt.
- *
- * Libgcrypt is free software; you can redistribute it and/or modify
- * it under the terms of the GNU Lesser General Public License as
- * published by the Free Software Foundation; either version 2.1 of
- * the License, or (at your option) any later version.
- *
- * Libgcrypt is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU Lesser General Public License for more details.
- *
- * You should have received a copy of the GNU Lesser General Public
- * License along with this program; if not, see <http://www.gnu.org/licenses/>.
- */
-
-#include <config.h>
-#include <stdio.h>
-#include <stdlib.h>
-#include <string.h>
-
-#include "g10lib.h"
-#include "mpi-internal.h"
-#include "mod-source-info.h"
-
-
-const char *
-_gcry_mpi_get_hw_config (void)
-{
- return mod_source_info + 1;
-}
-
-
-/****************
- * Note: It was a bad idea to use the number of limbs to allocate
- * because on a alpha the limbs are large but we normally need
- * integers of n bits - So we should change this to bits (or bytes).
- *
- * But mpi_alloc is used in a lot of places :-(. New code
- * should use mpi_new.
- */
-gcry_mpi_t
-_gcry_mpi_alloc( unsigned nlimbs )
-{
- gcry_mpi_t a;
-
- a = gcry_xmalloc( sizeof *a );
- a->d = nlimbs? mpi_alloc_limb_space( nlimbs, 0 ) : NULL;
- a->alloced = nlimbs;
- a->nlimbs = 0;
- a->sign = 0;
- a->flags = 0;
- return a;
-}
-
-void
-_gcry_mpi_m_check( gcry_mpi_t a )
-{
- _gcry_check_heap(a);
- _gcry_check_heap(a->d);
-}
-
-gcry_mpi_t
-_gcry_mpi_alloc_secure( unsigned nlimbs )
-{
- gcry_mpi_t a;
-
- a = gcry_xmalloc( sizeof *a );
- a->d = nlimbs? mpi_alloc_limb_space( nlimbs, 1 ) : NULL;
- a->alloced = nlimbs;
- a->flags = 1;
- a->nlimbs = 0;
- a->sign = 0;
- return a;
-}
-
-
-
-mpi_ptr_t
-_gcry_mpi_alloc_limb_space( unsigned int nlimbs, int secure )
-{
- mpi_ptr_t p;
- size_t len;
-
- len = (nlimbs ? nlimbs : 1) * sizeof (mpi_limb_t);
- p = secure ? gcry_xmalloc_secure (len) : gcry_xmalloc (len);
- if (! nlimbs)
- *p = 0;
-
- return p;
-}
-
-void
-_gcry_mpi_free_limb_space( mpi_ptr_t a, unsigned int nlimbs)
-{
- if (a)
- {
- size_t len = nlimbs * sizeof(mpi_limb_t);
-
- /* If we have information on the number of allocated limbs, we
- better wipe that space out. This is a failsafe feature if
- secure memory has been disabled or was not properly
- implemented in user provided allocation functions. */
- if (len)
- wipememory (a, len);
- gcry_free(a);
- }
-}
-
-
-void
-_gcry_mpi_assign_limb_space( gcry_mpi_t a, mpi_ptr_t ap, unsigned int nlimbs )
-{
- _gcry_mpi_free_limb_space (a->d, a->alloced);
- a->d = ap;
- a->alloced = nlimbs;
-}
-
-
-
-/****************
- * Resize the array of A to NLIMBS. The additional space is cleared
- * (set to 0).
- */
-void
-_gcry_mpi_resize (gcry_mpi_t a, unsigned nlimbs)
-{
- size_t i;
-
- if (nlimbs <= a->alloced)
- {
- /* We only need to clear the new space (this is a nop if the
- limb space is already of the correct size. */
- for (i=a->nlimbs; i < a->alloced; i++)
- a->d[i] = 0;
- return;
- }
-
- /* Actually resize the limb space. */
- if (a->d)
- {
- a->d = gcry_xrealloc (a->d, nlimbs * sizeof (mpi_limb_t));
- for (i=a->alloced; i < nlimbs; i++)
- a->d[i] = 0;
- }
- else
- {
- if (a->flags & 1)
- /* Secure memory is wanted. */
- a->d = gcry_xcalloc_secure (nlimbs , sizeof (mpi_limb_t));
- else
- /* Standard memory. */
- a->d = gcry_xcalloc (nlimbs , sizeof (mpi_limb_t));
- }
- a->alloced = nlimbs;
-}
-
-void
-_gcry_mpi_clear( gcry_mpi_t a )
-{
- a->nlimbs = 0;
- a->flags = 0;
-}
-
-
-void
-_gcry_mpi_free( gcry_mpi_t a )
-{
- if (!a )
- return;
- if ((a->flags & 4))
- gcry_free( a->d );
- else
- {
- _gcry_mpi_free_limb_space(a->d, a->alloced);
- }
- if ((a->flags & ~7))
- log_bug("invalid flag value in mpi\n");
- gcry_free(a);
-}
-
-static void
-mpi_set_secure( gcry_mpi_t a )
-{
- mpi_ptr_t ap, bp;
-
- if ( (a->flags & 1) )
- return;
- a->flags |= 1;
- ap = a->d;
- if (!a->nlimbs)
- {
- gcry_assert (!ap);
- return;
- }
- bp = mpi_alloc_limb_space (a->nlimbs, 1);
- MPN_COPY( bp, ap, a->nlimbs );
- a->d = bp;
- _gcry_mpi_free_limb_space (ap, a->alloced);
-}
-
-
-gcry_mpi_t
-gcry_mpi_set_opaque( gcry_mpi_t a, void *p, unsigned int nbits )
-{
- if (!a)
- a = mpi_alloc(0);
-
- if( a->flags & 4 )
- gcry_free( a->d );
- else
- _gcry_mpi_free_limb_space (a->d, a->alloced);
-
- a->d = p;
- a->alloced = 0;
- a->nlimbs = 0;
- a->sign = nbits;
- a->flags = 4;
- return a;
-}
-
-
-void *
-gcry_mpi_get_opaque( gcry_mpi_t a, unsigned int *nbits )
-{
- if( !(a->flags & 4) )
- log_bug("mpi_get_opaque on normal mpi\n");
- if( nbits )
- *nbits = a->sign;
- return a->d;
-}
-
-
-/****************
- * Note: This copy function should not interpret the MPI
- * but copy it transparently.
- */
-gcry_mpi_t
-gcry_mpi_copy( gcry_mpi_t a )
-{
- int i;
- gcry_mpi_t b;
-
- if( a && (a->flags & 4) ) {
- void *p = gcry_is_secure(a->d)? gcry_xmalloc_secure( (a->sign+7)/8 )
- : gcry_xmalloc( (a->sign+7)/8 );
- memcpy( p, a->d, (a->sign+7)/8 );
- b = gcry_mpi_set_opaque( NULL, p, a->sign );
- }
- else if( a ) {
- b = mpi_is_secure(a)? mpi_alloc_secure( a->nlimbs )
- : mpi_alloc( a->nlimbs );
- b->nlimbs = a->nlimbs;
- b->sign = a->sign;
- b->flags = a->flags;
- for(i=0; i < b->nlimbs; i++ )
- b->d[i] = a->d[i];
- }
- else
- b = NULL;
- return b;
-}
-
-
-/****************
- * This function allocates an MPI which is optimized to hold
- * a value as large as the one given in the argument and allocates it
- * with the same flags as A.
- */
-gcry_mpi_t
-_gcry_mpi_alloc_like( gcry_mpi_t a )
-{
- gcry_mpi_t b;
-
- if( a && (a->flags & 4) ) {
- int n = (a->sign+7)/8;
- void *p = gcry_is_secure(a->d)? gcry_malloc_secure( n )
- : gcry_malloc( n );
- memcpy( p, a->d, n );
- b = gcry_mpi_set_opaque( NULL, p, a->sign );
- }
- else if( a ) {
- b = mpi_is_secure(a)? mpi_alloc_secure( a->nlimbs )
- : mpi_alloc( a->nlimbs );
- b->nlimbs = 0;
- b->sign = 0;
- b->flags = a->flags;
- }
- else
- b = NULL;
- return b;
-}
-
-
-gcry_mpi_t
-gcry_mpi_set( gcry_mpi_t w, gcry_mpi_t u)
-{
- mpi_ptr_t wp, up;
- mpi_size_t usize = u->nlimbs;
- int usign = u->sign;
-
- if (!w)
- w = _gcry_mpi_alloc( mpi_get_nlimbs(u) );
- RESIZE_IF_NEEDED(w, usize);
- wp = w->d;
- up = u->d;
- MPN_COPY( wp, up, usize );
- w->nlimbs = usize;
- w->flags = u->flags;
- w->sign = usign;
- return w;
-}
-
-
-gcry_mpi_t
-gcry_mpi_set_ui( gcry_mpi_t w, unsigned long u)
-{
- if (!w)
- w = _gcry_mpi_alloc (1);
- /* FIXME: If U is 0 we have no need to resize and thus possible
- allocating the the limbs. */
- RESIZE_IF_NEEDED(w, 1);
- w->d[0] = u;
- w->nlimbs = u? 1:0;
- w->sign = 0;
- w->flags = 0;
- return w;
-}
-
-gcry_err_code_t
-_gcry_mpi_get_ui (gcry_mpi_t w, unsigned long *u)
-{
- gcry_err_code_t err = GPG_ERR_NO_ERROR;
- unsigned long x = 0;
-
- if (w->nlimbs > 1)
- err = GPG_ERR_TOO_LARGE;
- else if (w->nlimbs == 1)
- x = w->d[0];
- else
- x = 0;
-
- if (! err)
- *u = x;
-
- return err;
-}
-
-gcry_error_t
-gcry_mpi_get_ui (gcry_mpi_t w, unsigned long *u)
-{
- gcry_err_code_t err = GPG_ERR_NO_ERROR;
-
- err = _gcry_mpi_get_ui (w, u);
-
- return gcry_error (err);
-}
-
-gcry_mpi_t
-_gcry_mpi_alloc_set_ui( unsigned long u)
-{
- gcry_mpi_t w = mpi_alloc(1);
- w->d[0] = u;
- w->nlimbs = u? 1:0;
- w->sign = 0;
- return w;
-}
-
-void
-gcry_mpi_swap( gcry_mpi_t a, gcry_mpi_t b)
-{
- struct gcry_mpi tmp;
-
- tmp = *a; *a = *b; *b = tmp;
-}
-
-
-gcry_mpi_t
-gcry_mpi_new( unsigned int nbits )
-{
- return _gcry_mpi_alloc ( (nbits+BITS_PER_MPI_LIMB-1)
- / BITS_PER_MPI_LIMB );
-}
-
-
-gcry_mpi_t
-gcry_mpi_snew( unsigned int nbits )
-{
- return _gcry_mpi_alloc_secure ( (nbits+BITS_PER_MPI_LIMB-1)
- / BITS_PER_MPI_LIMB );
-}
-
-void
-gcry_mpi_release( gcry_mpi_t a )
-{
- _gcry_mpi_free( a );
-}
-
-void
-gcry_mpi_randomize( gcry_mpi_t w,
- unsigned int nbits, enum gcry_random_level level )
-{
- unsigned char *p;
- size_t nbytes = (nbits+7)/8;
-
- if (level == GCRY_WEAK_RANDOM)
- {
- p = mpi_is_secure(w) ? gcry_xmalloc_secure (nbytes)
- : gcry_xmalloc (nbytes);
- gcry_create_nonce (p, nbytes);
- }
- else
- {
- p = mpi_is_secure(w) ? gcry_random_bytes_secure (nbytes, level)
- : gcry_random_bytes (nbytes, level);
- }
- _gcry_mpi_set_buffer( w, p, nbytes, 0 );
- gcry_free (p);
-}
-
-
-void
-gcry_mpi_set_flag( gcry_mpi_t a, enum gcry_mpi_flag flag )
-{
- switch( flag ) {
- case GCRYMPI_FLAG_SECURE: mpi_set_secure(a); break;
- case GCRYMPI_FLAG_OPAQUE:
- default: log_bug("invalid flag value\n");
- }
-}
-
-void
-gcry_mpi_clear_flag( gcry_mpi_t a, enum gcry_mpi_flag flag )
-{
- (void)a; /* Not yet used. */
-
- switch (flag)
- {
- case GCRYMPI_FLAG_SECURE:
- case GCRYMPI_FLAG_OPAQUE:
- default: log_bug("invalid flag value\n");
- }
-}
-
-int
-gcry_mpi_get_flag( gcry_mpi_t a, enum gcry_mpi_flag flag )
-{
- switch (flag)
- {
- case GCRYMPI_FLAG_SECURE: return (a->flags & 1);
- case GCRYMPI_FLAG_OPAQUE: return (a->flags & 4);
- default: log_bug("invalid flag value\n");
- }
- /*NOTREACHED*/
- return 0;
-}
-
+/* mpiutil.ac - Utility functions for MPI + * Copyright (C) 1998, 2000, 2001, 2002, 2003, + * 2007 Free Software Foundation, Inc. + * + * This file is part of Libgcrypt. + * + * Libgcrypt is free software; you can redistribute it and/or modify + * it under the terms of the GNU Lesser General Public License as + * published by the Free Software Foundation; either version 2.1 of + * the License, or (at your option) any later version. + * + * Libgcrypt is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with this program; if not, see <http://www.gnu.org/licenses/>. + */ + +#include <config.h> +#include <stdio.h> +#include <stdlib.h> +#include <string.h> + +#include "g10lib.h" +#include "mpi-internal.h" +#include "memory.h" +#include "mod-source-info.h" + + +const char * +_gcry_mpi_get_hw_config (void) +{ + return mod_source_info + 1; +} + + +/**************** + * Note: It was a bad idea to use the number of limbs to allocate + * because on a alpha the limbs are large but we normally need + * integers of n bits - So we should change this to bits (or bytes). + * + * But mpi_alloc is used in a lot of places :-(. New code + * should use mpi_new. + */ +gcry_mpi_t +_gcry_mpi_alloc( unsigned nlimbs ) +{ + gcry_mpi_t a; + + a = gcry_xmalloc( sizeof *a ); + a->d = nlimbs? mpi_alloc_limb_space( nlimbs, 0 ) : NULL; + a->alloced = nlimbs; + a->nlimbs = 0; + a->sign = 0; + a->flags = 0; + return a; +} + +void +_gcry_mpi_m_check( gcry_mpi_t a ) +{ + _gcry_check_heap(a); + _gcry_check_heap(a->d); +} + +gcry_mpi_t +_gcry_mpi_alloc_secure( unsigned nlimbs ) +{ + gcry_mpi_t a; + + a = gcry_xmalloc( sizeof *a ); + a->d = nlimbs? mpi_alloc_limb_space( nlimbs, 1 ) : NULL; + a->alloced = nlimbs; + a->flags = 1; + a->nlimbs = 0; + a->sign = 0; + return a; +} + + + +mpi_ptr_t +_gcry_mpi_alloc_limb_space( unsigned int nlimbs, int secure ) +{ + mpi_ptr_t p; + size_t len; + + len = (nlimbs ? nlimbs : 1) * sizeof (mpi_limb_t); + p = secure ? gcry_xmalloc_secure (len) : gcry_xmalloc (len); + if (! nlimbs) + *p = 0; + + return p; +} + +void +_gcry_mpi_free_limb_space( mpi_ptr_t a, unsigned int nlimbs) +{ + if (a) + { + size_t len = nlimbs * sizeof(mpi_limb_t); + + /* If we have information on the number of allocated limbs, we + better wipe that space out. This is a failsafe feature if + secure memory has been disabled or was not properly + implemented in user provided allocation functions. */ + if (len) + wipememory (a, len); + gcry_free(a); + } +} + + +void +_gcry_mpi_assign_limb_space( gcry_mpi_t a, mpi_ptr_t ap, unsigned int nlimbs ) +{ + _gcry_mpi_free_limb_space (a->d, a->alloced); + a->d = ap; + a->alloced = nlimbs; +} + + + +/**************** + * Resize the array of A to NLIMBS. The additional space is cleared + * (set to 0). + */ +void +_gcry_mpi_resize (gcry_mpi_t a, unsigned nlimbs) +{ + size_t i; + + if (nlimbs <= a->alloced) + { + /* We only need to clear the new space (this is a nop if the + limb space is already of the correct size. */ + for (i=a->nlimbs; i < a->alloced; i++) + a->d[i] = 0; + return; + } + + /* Actually resize the limb space. */ + if (a->d) + { + a->d = gcry_xrealloc (a->d, nlimbs * sizeof (mpi_limb_t)); + for (i=a->alloced; i < nlimbs; i++) + a->d[i] = 0; + } + else + { + if (a->flags & 1) + /* Secure memory is wanted. */ + a->d = gcry_xcalloc_secure (nlimbs , sizeof (mpi_limb_t)); + else + /* Standard memory. */ + a->d = gcry_xcalloc (nlimbs , sizeof (mpi_limb_t)); + } + a->alloced = nlimbs; +} + +void +_gcry_mpi_clear( gcry_mpi_t a ) +{ + a->nlimbs = 0; + a->flags = 0; +} + + +void +_gcry_mpi_free( gcry_mpi_t a ) +{ + if (!a ) + return; + if ((a->flags & 4)) + gcry_free( a->d ); + else + { + _gcry_mpi_free_limb_space(a->d, a->alloced); + } + if ((a->flags & ~7)) + log_bug("invalid flag value in mpi\n"); + gcry_free(a); +} + +static void +mpi_set_secure( gcry_mpi_t a ) +{ + mpi_ptr_t ap, bp; + + if ( (a->flags & 1) ) + return; + a->flags |= 1; + ap = a->d; + if (!a->nlimbs) + { + gcry_assert (!ap); + return; + } + bp = mpi_alloc_limb_space (a->nlimbs, 1); + MPN_COPY( bp, ap, a->nlimbs ); + a->d = bp; + _gcry_mpi_free_limb_space (ap, a->alloced); +} + + +gcry_mpi_t +gcry_mpi_set_opaque( gcry_mpi_t a, void *p, unsigned int nbits ) +{ + if (!a) + a = mpi_alloc(0); + + if( a->flags & 4 ) + gcry_free( a->d ); + else + _gcry_mpi_free_limb_space (a->d, a->alloced); + + a->d = p; + a->alloced = 0; + a->nlimbs = 0; + a->sign = nbits; + a->flags = 4; + return a; +} + + +void * +gcry_mpi_get_opaque( gcry_mpi_t a, unsigned int *nbits ) +{ + if( !(a->flags & 4) ) + log_bug("mpi_get_opaque on normal mpi\n"); + if( nbits ) + *nbits = a->sign; + return a->d; +} + + +/**************** + * Note: This copy function should not interpret the MPI + * but copy it transparently. + */ +gcry_mpi_t +gcry_mpi_copy( gcry_mpi_t a ) +{ + int i; + gcry_mpi_t b; + + if( a && (a->flags & 4) ) { + void *p = gcry_is_secure(a->d)? gcry_xmalloc_secure( (a->sign+7)/8 ) + : gcry_xmalloc( (a->sign+7)/8 ); + memcpy( p, a->d, (a->sign+7)/8 ); + b = gcry_mpi_set_opaque( NULL, p, a->sign ); + } + else if( a ) { + b = mpi_is_secure(a)? mpi_alloc_secure( a->nlimbs ) + : mpi_alloc( a->nlimbs ); + b->nlimbs = a->nlimbs; + b->sign = a->sign; + b->flags = a->flags; + for(i=0; i < b->nlimbs; i++ ) + b->d[i] = a->d[i]; + } + else + b = NULL; + return b; +} + + +/**************** + * This function allocates an MPI which is optimized to hold + * a value as large as the one given in the argument and allocates it + * with the same flags as A. + */ +gcry_mpi_t +_gcry_mpi_alloc_like( gcry_mpi_t a ) +{ + gcry_mpi_t b; + + if( a && (a->flags & 4) ) { + int n = (a->sign+7)/8; + void *p = gcry_is_secure(a->d)? gcry_malloc_secure( n ) + : gcry_malloc( n ); + memcpy( p, a->d, n ); + b = gcry_mpi_set_opaque( NULL, p, a->sign ); + } + else if( a ) { + b = mpi_is_secure(a)? mpi_alloc_secure( a->nlimbs ) + : mpi_alloc( a->nlimbs ); + b->nlimbs = 0; + b->sign = 0; + b->flags = a->flags; + } + else + b = NULL; + return b; +} + + +gcry_mpi_t +gcry_mpi_set( gcry_mpi_t w, gcry_mpi_t u) +{ + mpi_ptr_t wp, up; + mpi_size_t usize = u->nlimbs; + int usign = u->sign; + + if (!w) + w = _gcry_mpi_alloc( mpi_get_nlimbs(u) ); + RESIZE_IF_NEEDED(w, usize); + wp = w->d; + up = u->d; + MPN_COPY( wp, up, usize ); + w->nlimbs = usize; + w->flags = u->flags; + w->sign = usign; + return w; +} + + +gcry_mpi_t +gcry_mpi_set_ui( gcry_mpi_t w, unsigned long u) +{ + if (!w) + w = _gcry_mpi_alloc (1); + /* FIXME: If U is 0 we have no need to resize and thus possible + allocating the the limbs. */ + RESIZE_IF_NEEDED(w, 1); + w->d[0] = u; + w->nlimbs = u? 1:0; + w->sign = 0; + w->flags = 0; + return w; +} + +gcry_err_code_t +_gcry_mpi_get_ui (gcry_mpi_t w, unsigned long *u) +{ + gcry_err_code_t err = GPG_ERR_NO_ERROR; + unsigned long x = 0; + + if (w->nlimbs > 1) + err = GPG_ERR_TOO_LARGE; + else if (w->nlimbs == 1) + x = w->d[0]; + else + x = 0; + + if (! err) + *u = x; + + return err; +} + +gcry_error_t +gcry_mpi_get_ui (gcry_mpi_t w, unsigned long *u) +{ + gcry_err_code_t err = GPG_ERR_NO_ERROR; + + err = _gcry_mpi_get_ui (w, u); + + return gcry_error (err); +} + +gcry_mpi_t +_gcry_mpi_alloc_set_ui( unsigned long u) +{ + gcry_mpi_t w = mpi_alloc(1); + w->d[0] = u; + w->nlimbs = u? 1:0; + w->sign = 0; + return w; +} + +void +gcry_mpi_swap( gcry_mpi_t a, gcry_mpi_t b) +{ + struct gcry_mpi tmp; + + tmp = *a; *a = *b; *b = tmp; +} + + +gcry_mpi_t +gcry_mpi_new( unsigned int nbits ) +{ + return _gcry_mpi_alloc ( (nbits+BITS_PER_MPI_LIMB-1) + / BITS_PER_MPI_LIMB ); +} + + +gcry_mpi_t +gcry_mpi_snew( unsigned int nbits ) +{ + return _gcry_mpi_alloc_secure ( (nbits+BITS_PER_MPI_LIMB-1) + / BITS_PER_MPI_LIMB ); +} + +void +gcry_mpi_release( gcry_mpi_t a ) +{ + _gcry_mpi_free( a ); +} + +void +gcry_mpi_randomize( gcry_mpi_t w, + unsigned int nbits, enum gcry_random_level level ) +{ + unsigned char *p; + size_t nbytes = (nbits+7)/8; + + if (level == GCRY_WEAK_RANDOM) + { + p = mpi_is_secure(w) ? gcry_xmalloc_secure (nbytes) + : gcry_xmalloc (nbytes); + gcry_create_nonce (p, nbytes); + } + else + { + p = mpi_is_secure(w) ? gcry_random_bytes_secure (nbytes, level) + : gcry_random_bytes (nbytes, level); + } + _gcry_mpi_set_buffer( w, p, nbytes, 0 ); + gcry_free (p); +} + + +void +gcry_mpi_set_flag( gcry_mpi_t a, enum gcry_mpi_flag flag ) +{ + switch( flag ) { + case GCRYMPI_FLAG_SECURE: mpi_set_secure(a); break; + case GCRYMPI_FLAG_OPAQUE: + default: log_bug("invalid flag value\n"); + } +} + +void +gcry_mpi_clear_flag( gcry_mpi_t a, enum gcry_mpi_flag flag ) +{ + (void)a; /* Not yet used. */ + + switch (flag) + { + case GCRYMPI_FLAG_SECURE: + case GCRYMPI_FLAG_OPAQUE: + default: log_bug("invalid flag value\n"); + } +} + +int +gcry_mpi_get_flag( gcry_mpi_t a, enum gcry_mpi_flag flag ) +{ + switch (flag) + { + case GCRYMPI_FLAG_SECURE: return (a->flags & 1); + case GCRYMPI_FLAG_OPAQUE: return (a->flags & 4); + default: log_bug("invalid flag value\n"); + } + /*NOTREACHED*/ + return 0; +} + |