| 1 | /* Include file for internal GNU MP types and definitions. |
| 2 | |
| 3 | Copyright (C) 1991-2021 Free Software Foundation, Inc. |
| 4 | |
| 5 | This file is part of the GNU MP Library. |
| 6 | |
| 7 | The GNU MP Library is free software; you can redistribute it and/or modify |
| 8 | it under the terms of the GNU Lesser General Public License as published by |
| 9 | the Free Software Foundation; either version 2.1 of the License, or (at your |
| 10 | option) any later version. |
| 11 | |
| 12 | The GNU MP Library is distributed in the hope that it will be useful, but |
| 13 | WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY |
| 14 | or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public |
| 15 | License for more details. |
| 16 | |
| 17 | You should have received a copy of the GNU Lesser General Public License |
| 18 | along with the GNU MP Library; see the file COPYING.LIB. If not, see |
| 19 | <https://www.gnu.org/licenses/>. */ |
| 20 | |
| 21 | /* When using gcc, make sure to use its builtin alloca. */ |
| 22 | #if ! defined (alloca) && defined (__GNUC__) |
| 23 | #define alloca __builtin_alloca |
| 24 | #define HAVE_ALLOCA |
| 25 | #endif |
| 26 | |
| 27 | /* When using cc, do whatever necessary to allow use of alloca. For many |
| 28 | machines, this means including alloca.h. IBM's compilers need a #pragma |
| 29 | in "each module that needs to use alloca". */ |
| 30 | #if ! defined (alloca) |
| 31 | /* We need lots of variants for MIPS, to cover all versions and perversions |
| 32 | of OSes for MIPS. */ |
| 33 | #if defined (__mips) || defined (MIPSEL) || defined (MIPSEB) \ |
| 34 | || defined (_MIPSEL) || defined (_MIPSEB) || defined (__sgi) \ |
| 35 | || defined (__alpha) || defined (__sparc) || defined (sparc) \ |
| 36 | || defined (__ksr__) |
| 37 | #include <alloca.h> |
| 38 | #define HAVE_ALLOCA |
| 39 | #endif |
| 40 | #if defined (_IBMR2) |
| 41 | #pragma alloca |
| 42 | #define HAVE_ALLOCA |
| 43 | #endif |
| 44 | #if defined (__DECC) |
| 45 | #define alloca(x) __ALLOCA(x) |
| 46 | #define HAVE_ALLOCA |
| 47 | #endif |
| 48 | #endif |
| 49 | |
| 50 | #if (! defined (alloca) && ! defined (HAVE_ALLOCA)) \ |
| 51 | || defined (USE_STACK_ALLOC) |
| 52 | #include "stack-alloc.h" |
| 53 | #else |
| 54 | #define TMP_DECL(m) |
| 55 | #define TMP_ALLOC(x) alloca(x) |
| 56 | #define TMP_MARK(m) |
| 57 | #define TMP_FREE(m) |
| 58 | #endif |
| 59 | |
| 60 | #ifndef NULL |
| 61 | #define NULL ((void *) 0) |
| 62 | #endif |
| 63 | |
| 64 | #if ! defined (__GNUC__) |
| 65 | #define inline /* Empty */ |
| 66 | #endif |
| 67 | |
| 68 | /* Get MAX/MIN macros. */ |
| 69 | #include <sys/param.h> |
| 70 | |
| 71 | /* Field access macros. */ |
| 72 | #define SIZ(x) ((x)->_mp_size) |
| 73 | #define PTR(x) ((x)->_mp_d) |
| 74 | #define EXP(x) ((x)->_mp_exp) |
| 75 | #define PREC(x) ((x)->_mp_prec) |
| 76 | #define ALLOC(x) ((x)->_mp_alloc) |
| 77 | |
| 78 | #include "gmp-mparam.h" |
| 79 | /* #include "longlong.h" */ |
| 80 | |
| 81 | #if defined (__STDC__) || defined (__cplusplus) |
| 82 | void *malloc (size_t); |
| 83 | void *realloc (void *, size_t); |
| 84 | void free (void *); |
| 85 | |
| 86 | extern void * (*_mp_allocate_func) (size_t); |
| 87 | extern void * (*_mp_reallocate_func) (void *, size_t, size_t); |
| 88 | extern void (*_mp_free_func) (void *, size_t); |
| 89 | |
| 90 | void *_mp_default_allocate (size_t); |
| 91 | void *_mp_default_reallocate (void *, size_t, size_t); |
| 92 | void _mp_default_free (void *, size_t); |
| 93 | |
| 94 | #else |
| 95 | |
| 96 | #define const /* Empty */ |
| 97 | #define signed /* Empty */ |
| 98 | |
| 99 | void *malloc (); |
| 100 | void *realloc (); |
| 101 | void free (); |
| 102 | |
| 103 | extern void * (*_mp_allocate_func) (); |
| 104 | extern void * (*_mp_reallocate_func) (); |
| 105 | extern void (*_mp_free_func) (); |
| 106 | |
| 107 | void *_mp_default_allocate (); |
| 108 | void *_mp_default_reallocate (); |
| 109 | void _mp_default_free (); |
| 110 | #endif |
| 111 | |
| 112 | /* Copy NLIMBS *limbs* from SRC to DST. */ |
| 113 | #define MPN_COPY_INCR(DST, SRC, NLIMBS) \ |
| 114 | do { \ |
| 115 | mp_size_t __i; \ |
| 116 | for (__i = 0; __i < (NLIMBS); __i++) \ |
| 117 | (DST)[__i] = (SRC)[__i]; \ |
| 118 | } while (0) |
| 119 | #define MPN_COPY_DECR(DST, SRC, NLIMBS) \ |
| 120 | do { \ |
| 121 | mp_size_t __i; \ |
| 122 | for (__i = (NLIMBS) - 1; __i >= 0; __i--) \ |
| 123 | (DST)[__i] = (SRC)[__i]; \ |
| 124 | } while (0) |
| 125 | #define MPN_COPY MPN_COPY_INCR |
| 126 | |
| 127 | /* Zero NLIMBS *limbs* AT DST. */ |
| 128 | #define MPN_ZERO(DST, NLIMBS) \ |
| 129 | do { \ |
| 130 | mp_size_t __i; \ |
| 131 | for (__i = 0; __i < (NLIMBS); __i++) \ |
| 132 | (DST)[__i] = 0; \ |
| 133 | } while (0) |
| 134 | |
| 135 | #define MPN_NORMALIZE(DST, NLIMBS) \ |
| 136 | do { \ |
| 137 | while (NLIMBS > 0) \ |
| 138 | { \ |
| 139 | if ((DST)[(NLIMBS) - 1] != 0) \ |
| 140 | break; \ |
| 141 | NLIMBS--; \ |
| 142 | } \ |
| 143 | } while (0) |
| 144 | #define MPN_NORMALIZE_NOT_ZERO(DST, NLIMBS) \ |
| 145 | do { \ |
| 146 | while (1) \ |
| 147 | { \ |
| 148 | if ((DST)[(NLIMBS) - 1] != 0) \ |
| 149 | break; \ |
| 150 | NLIMBS--; \ |
| 151 | } \ |
| 152 | } while (0) |
| 153 | |
| 154 | /* Initialize the MP_INT X with space for NLIMBS limbs. |
| 155 | X should be a temporary variable, and it will be automatically |
| 156 | cleared out when the running function returns. |
| 157 | We use __x here to make it possible to accept both mpz_ptr and mpz_t |
| 158 | arguments. */ |
| 159 | #define MPZ_TMP_INIT(X, NLIMBS) \ |
| 160 | do { \ |
| 161 | mpz_ptr __x = (X); \ |
| 162 | __x->_mp_alloc = (NLIMBS); \ |
| 163 | __x->_mp_d = (mp_ptr) TMP_ALLOC ((NLIMBS) * BYTES_PER_MP_LIMB); \ |
| 164 | } while (0) |
| 165 | |
| 166 | #define MPN_MUL_N_RECURSE(prodp, up, vp, size, tspace) \ |
| 167 | do { \ |
| 168 | if ((size) < KARATSUBA_THRESHOLD) \ |
| 169 | impn_mul_n_basecase (prodp, up, vp, size); \ |
| 170 | else \ |
| 171 | impn_mul_n (prodp, up, vp, size, tspace); \ |
| 172 | } while (0); |
| 173 | #define MPN_SQR_N_RECURSE(prodp, up, size, tspace) \ |
| 174 | do { \ |
| 175 | if ((size) < KARATSUBA_THRESHOLD) \ |
| 176 | impn_sqr_n_basecase (prodp, up, size); \ |
| 177 | else \ |
| 178 | impn_sqr_n (prodp, up, size, tspace); \ |
| 179 | } while (0); |
| 180 | |
| 181 | /* Structure for conversion between internal binary format and |
| 182 | strings in base 2..36. */ |
| 183 | struct bases |
| 184 | { |
| 185 | /* Number of digits in the conversion base that always fits in an mp_limb_t. |
| 186 | For example, for base 10 on a machine where a mp_limb_t has 32 bits this |
| 187 | is 9, since 10**9 is the largest number that fits into a mp_limb_t. */ |
| 188 | int chars_per_limb; |
| 189 | |
| 190 | /* log(2)/log(conversion_base) */ |
| 191 | float chars_per_bit_exactly; |
| 192 | |
| 193 | /* base**chars_per_limb, i.e. the biggest number that fits a word, built by |
| 194 | factors of base. Exception: For 2, 4, 8, etc, big_base is log2(base), |
| 195 | i.e. the number of bits used to represent each digit in the base. */ |
| 196 | mp_limb_t big_base; |
| 197 | |
| 198 | /* A BITS_PER_MP_LIMB bit approximation to 1/big_base, represented as a |
| 199 | fixed-point number. Instead of dividing by big_base an application can |
| 200 | choose to multiply by big_base_inverted. */ |
| 201 | mp_limb_t big_base_inverted; |
| 202 | }; |
| 203 | |
| 204 | extern const struct bases __mp_bases[]; |
| 205 | extern mp_size_t __gmp_default_fp_limb_precision; |
| 206 | |
| 207 | /* Divide the two-limb number in (NH,,NL) by D, with DI being the largest |
| 208 | limb not larger than (2**(2*BITS_PER_MP_LIMB))/D - (2**BITS_PER_MP_LIMB). |
| 209 | If this would yield overflow, DI should be the largest possible number |
| 210 | (i.e., only ones). For correct operation, the most significant bit of D |
| 211 | has to be set. Put the quotient in Q and the remainder in R. */ |
| 212 | #define udiv_qrnnd_preinv(q, r, nh, nl, d, di) \ |
| 213 | do { \ |
| 214 | mp_limb_t _ql __attribute__ ((unused)); \ |
| 215 | mp_limb_t _q, _r; \ |
| 216 | mp_limb_t _xh, _xl; \ |
| 217 | umul_ppmm (_q, _ql, (nh), (di)); \ |
| 218 | _q += (nh); /* DI is 2**BITS_PER_MP_LIMB too small */\ |
| 219 | umul_ppmm (_xh, _xl, _q, (d)); \ |
| 220 | sub_ddmmss (_xh, _r, (nh), (nl), _xh, _xl); \ |
| 221 | if (_xh != 0) \ |
| 222 | { \ |
| 223 | sub_ddmmss (_xh, _r, _xh, _r, 0, (d)); \ |
| 224 | _q += 1; \ |
| 225 | if (_xh != 0) \ |
| 226 | { \ |
| 227 | sub_ddmmss (_xh, _r, _xh, _r, 0, (d)); \ |
| 228 | _q += 1; \ |
| 229 | } \ |
| 230 | } \ |
| 231 | if (_r >= (d)) \ |
| 232 | { \ |
| 233 | _r -= (d); \ |
| 234 | _q += 1; \ |
| 235 | } \ |
| 236 | (r) = _r; \ |
| 237 | (q) = _q; \ |
| 238 | } while (0) |
| 239 | /* Like udiv_qrnnd_preinv, but for any value D. DNORM is D shifted left |
| 240 | so that its most significant bit is set. LGUP is ceil(log2(D)). */ |
| 241 | #define udiv_qrnnd_preinv2gen(q, r, nh, nl, d, di, dnorm, lgup) \ |
| 242 | do { \ |
| 243 | mp_limb_t n2, n10, n1, nadj, q1; \ |
| 244 | mp_limb_t _xh, _xl; \ |
| 245 | n2 = ((nh) << (BITS_PER_MP_LIMB - (lgup))) + ((nl) >> 1 >> (l - 1));\ |
| 246 | n10 = (nl) << (BITS_PER_MP_LIMB - (lgup)); \ |
| 247 | n1 = ((mp_limb_signed_t) n10 >> (BITS_PER_MP_LIMB - 1)); \ |
| 248 | nadj = n10 + (n1 & (dnorm)); \ |
| 249 | umul_ppmm (_xh, _xl, di, n2 - n1); \ |
| 250 | add_ssaaaa (_xh, _xl, _xh, _xl, 0, nadj); \ |
| 251 | q1 = ~(n2 + _xh); \ |
| 252 | umul_ppmm (_xh, _xl, q1, d); \ |
| 253 | add_ssaaaa (_xh, _xl, _xh, _xl, nh, nl); \ |
| 254 | _xh -= (d); \ |
| 255 | (r) = _xl + ((d) & _xh); \ |
| 256 | (q) = _xh - q1; \ |
| 257 | } while (0) |
| 258 | /* Exactly like udiv_qrnnd_preinv, but branch-free. It is not clear which |
| 259 | version to use. */ |
| 260 | #define udiv_qrnnd_preinv2norm(q, r, nh, nl, d, di) \ |
| 261 | do { \ |
| 262 | mp_limb_t n2, n10, n1, nadj, q1; \ |
| 263 | mp_limb_t _xh, _xl; \ |
| 264 | n2 = (nh); \ |
| 265 | n10 = (nl); \ |
| 266 | n1 = ((mp_limb_signed_t) n10 >> (BITS_PER_MP_LIMB - 1)); \ |
| 267 | nadj = n10 + (n1 & (d)); \ |
| 268 | umul_ppmm (_xh, _xl, di, n2 - n1); \ |
| 269 | add_ssaaaa (_xh, _xl, _xh, _xl, 0, nadj); \ |
| 270 | q1 = ~(n2 + _xh); \ |
| 271 | umul_ppmm (_xh, _xl, q1, d); \ |
| 272 | add_ssaaaa (_xh, _xl, _xh, _xl, nh, nl); \ |
| 273 | _xh -= (d); \ |
| 274 | (r) = _xl + ((d) & _xh); \ |
| 275 | (q) = _xh - q1; \ |
| 276 | } while (0) |
| 277 | |
| 278 | #if defined (__GNUC__) |
| 279 | /* Define stuff for longlong.h. */ |
| 280 | typedef unsigned int UQItype __attribute__ ((mode (QI))); |
| 281 | typedef int SItype __attribute__ ((mode (SI))); |
| 282 | typedef unsigned int USItype __attribute__ ((mode (SI))); |
| 283 | typedef int DItype __attribute__ ((mode (DI))); |
| 284 | typedef unsigned int UDItype __attribute__ ((mode (DI))); |
| 285 | #else |
| 286 | typedef unsigned char UQItype; |
| 287 | typedef long SItype; |
| 288 | typedef unsigned long USItype; |
| 289 | #endif |
| 290 | |
| 291 | typedef mp_limb_t UWtype; |
| 292 | typedef unsigned int UHWtype; |
| 293 | #define W_TYPE_SIZE BITS_PER_MP_LIMB |
| 294 | |
| 295 | /* Internal mpn calls */ |
| 296 | #define impn_mul_n_basecase __MPN(impn_mul_n_basecase) |
| 297 | #define impn_mul_n __MPN(impn_mul_n) |
| 298 | #define impn_sqr_n_basecase __MPN(impn_sqr_n_basecase) |
| 299 | #define impn_sqr_n __MPN(impn_sqr_n) |
| 300 | |
| 301 | #ifndef _PROTO |
| 302 | #if defined (__STDC__) || defined (__cplusplus) |
| 303 | #define _PROTO(x) x |
| 304 | #else |
| 305 | #define _PROTO(x) () |
| 306 | #endif |
| 307 | #endif |
| 308 | |
| 309 | /* Prototypes for internal mpn calls. */ |
| 310 | extern void impn_mul_n_basecase _PROTO ((mp_ptr prodp, mp_srcptr up, |
| 311 | mp_srcptr vp, mp_size_t size)) |
| 312 | attribute_hidden; |
| 313 | extern void impn_mul_n _PROTO ((mp_ptr prodp, mp_srcptr up, mp_srcptr vp, |
| 314 | mp_size_t size, mp_ptr tspace)) |
| 315 | attribute_hidden; |
| 316 | extern void impn_sqr_n_basecase _PROTO ((mp_ptr prodp, mp_srcptr up, |
| 317 | mp_size_t size)) |
| 318 | attribute_hidden; |
| 319 | extern void impn_sqr_n _PROTO ((mp_ptr prodp, mp_srcptr up, mp_size_t size, |
| 320 | mp_ptr tspace)) |
| 321 | attribute_hidden; |
| 322 | |