crypto/bn/bn_local.h

   1 /*
   2  * Copyright 1995-2022 The OpenSSL Project Authors. All Rights Reserved.
   3  *
   4  * Licensed under the Apache License 2.0 (the "License").  You may not use
   5  * this file except in compliance with the License.  You can obtain a copy
   6  * in the file LICENSE in the source distribution or at
   7  * https://www.openssl.org/source/license.html
   8  */
   9
  10 #ifndef OSSL_CRYPTO_BN_LOCAL_H
  11 # define OSSL_CRYPTO_BN_LOCAL_H
  12
  13 /*
  14  * The EDK2 build doesn't use bn_conf.h; it sets THIRTY_TWO_BIT or
  15  * SIXTY_FOUR_BIT in its own environment since it doesn't re-run our
  16  * Configure script and needs to support both 32-bit and 64-bit.
  17  */
  18 # include <openssl/opensslconf.h>
  19
  20 # if !defined(OPENSSL_SYS_UEFI)
  21 #  include "crypto/bn_conf.h"
  22 # endif
  23
  24 # include "crypto/bn.h"
  25 # include "internal/cryptlib.h"
  26 # include "internal/numbers.h"
  27
  28 /*
  29  * These preprocessor symbols control various aspects of the bignum headers
  30  * and library code. They're not defined by any "normal" configuration, as
  31  * they are intended for development and testing purposes. NB: defining
  32  * them can be useful for debugging application code as well as openssl
  33  * itself. BN_DEBUG - turn on various debugging alterations to the bignum
  34  * code BN_RAND_DEBUG - uses random poisoning of unused words to trip up
  35  * mismanagement of bignum internals. Enable BN_RAND_DEBUG is known to
  36  * break some of the OpenSSL tests.
  37  */
  38 # if defined(BN_RAND_DEBUG) && !defined(BN_DEBUG)
  39 #  define BN_DEBUG
  40 # endif
  41 # if defined(BN_RAND_DEBUG)
  42 #  include <openssl/rand.h>
  43 # endif
  44
  45 /*
  46  * This should limit the stack usage due to alloca to about 4K.
  47  * BN_SOFT_LIMIT is a soft limit equivalent to 2*OPENSSL_RSA_MAX_MODULUS_BITS.
  48  * Beyond that size bn_mul_mont is no longer used, and the constant time
  49  * assembler code is disabled, due to the blatant alloca and bn_mul_mont usage.
  50  * Note that bn_mul_mont does an alloca that is hidden away in assembly.
  51  * It is not recommended to do computations with numbers exceeding this limit,
  52  * since the result will be highly version dependent:
  53  * While the current OpenSSL version will use non-optimized, but safe code,
  54  * previous versions will use optimized code, that may crash due to unexpected
  55  * stack overflow, and future versions may very well turn this into a hard
  56  * limit.
  57  * Note however, that it is possible to override the size limit using
  58  * "./config -DBN_SOFT_LIMIT=<limit>" if necessary, and the O/S specific
  59  * stack limit is known and taken into consideration.
  60  */
  61 # ifndef BN_SOFT_LIMIT
  62 #  define BN_SOFT_LIMIT         (4096 / BN_BYTES)
  63 # endif
  64
  65 # ifndef OPENSSL_SMALL_FOOTPRINT
  66 #  define BN_MUL_COMBA
  67 #  define BN_SQR_COMBA
  68 #  define BN_RECURSION
  69 # endif
  70
  71 /*
  72  * This next option uses the C libraries (2 word)/(1 word) function. If it is
  73  * not defined, I use my C version (which is slower). The reason for this
  74  * flag is that when the particular C compiler library routine is used, and
  75  * the library is linked with a different compiler, the library is missing.
  76  * This mostly happens when the library is built with gcc and then linked
  77  * using normal cc.  This would be a common occurrence because gcc normally
  78  * produces code that is 2 times faster than system compilers for the big
  79  * number stuff. For machines with only one compiler (or shared libraries),
  80  * this should be on.  Again this in only really a problem on machines using
  81  * "long long's", are 32bit, and are not using my assembler code.
  82  */
  83 # if defined(OPENSSL_SYS_MSDOS) || defined(OPENSSL_SYS_WINDOWS) || \
  84     defined(OPENSSL_SYS_WIN32) || defined(linux)
  85 #  define BN_DIV2W
  86 # endif
  87
  88 /*
  89  * 64-bit processor with LP64 ABI
  90  */
  91 # ifdef SIXTY_FOUR_BIT_LONG
  92 #  define BN_ULLONG       unsigned long long
  93 #  define BN_BITS4        32
  94 #  define BN_MASK2        (0xffffffffffffffffL)
  95 #  define BN_MASK2l       (0xffffffffL)
  96 #  define BN_MASK2h       (0xffffffff00000000L)
  97 #  define BN_MASK2h1      (0xffffffff80000000L)
  98 #  define BN_DEC_CONV     (10000000000000000000UL)
  99 #  define BN_DEC_NUM      19
 100 #  define BN_DEC_FMT1     "%lu"
 101 #  define BN_DEC_FMT2     "%019lu"
 102 # endif
 103
 104 /*
 105  * 64-bit processor other than LP64 ABI
 106  */
 107 # ifdef SIXTY_FOUR_BIT
 108 #  undef BN_LLONG
 109 #  undef BN_ULLONG
 110 #  define BN_BITS4        32
 111 #  define BN_MASK2        (0xffffffffffffffffLL)
 112 #  define BN_MASK2l       (0xffffffffL)
 113 #  define BN_MASK2h       (0xffffffff00000000LL)
 114 #  define BN_MASK2h1      (0xffffffff80000000LL)
 115 #  define BN_DEC_CONV     (10000000000000000000ULL)
 116 #  define BN_DEC_NUM      19
 117 #  define BN_DEC_FMT1     "%llu"
 118 #  define BN_DEC_FMT2     "%019llu"
 119 # endif
 120
 121 # ifdef THIRTY_TWO_BIT
 122 #  ifdef BN_LLONG
 123 #   if defined(_WIN32) && !defined(__GNUC__)
 124 #    define BN_ULLONG     unsigned __int64
 125 #   else
 126 #    define BN_ULLONG     unsigned long long
 127 #   endif
 128 #  endif
 129 #  define BN_BITS4        16
 130 #  define BN_MASK2        (0xffffffffL)
 131 #  define BN_MASK2l       (0xffff)
 132 #  define BN_MASK2h1      (0xffff8000L)
 133 #  define BN_MASK2h       (0xffff0000L)
 134 #  define BN_DEC_CONV     (1000000000L)
 135 #  define BN_DEC_NUM      9
 136 #  define BN_DEC_FMT1     "%u"
 137 #  define BN_DEC_FMT2     "%09u"
 138 # endif
 139
 140
 141 /*-
 142  * Bignum consistency macros
 143  * There is one "API" macro, bn_fix_top(), for stripping leading zeroes from
 144  * bignum data after direct manipulations on the data. There is also an
 145  * "internal" macro, bn_check_top(), for verifying that there are no leading
 146  * zeroes. Unfortunately, some auditing is required due to the fact that
 147  * bn_fix_top() has become an overabused duct-tape because bignum data is
 148  * occasionally passed around in an inconsistent state. So the following
 149  * changes have been made to sort this out;
 150  * - bn_fix_top()s implementation has been moved to bn_correct_top()
 151  * - if BN_DEBUG isn't defined, bn_fix_top() maps to bn_correct_top(), and
 152  *   bn_check_top() is as before.
 153  * - if BN_DEBUG *is* defined;
 154  *   - bn_check_top() tries to pollute unused words even if the bignum 'top' is
 155  *     consistent. (ed: only if BN_RAND_DEBUG is defined)
 156  *   - bn_fix_top() maps to bn_check_top() rather than "fixing" anything.
 157  * The idea is to have debug builds flag up inconsistent bignums when they
 158  * occur. If that occurs in a bn_fix_top(), we examine the code in question; if
 159  * the use of bn_fix_top() was appropriate (ie. it follows directly after code
 160  * that manipulates the bignum) it is converted to bn_correct_top(), and if it
 161  * was not appropriate, we convert it permanently to bn_check_top() and track
 162  * down the cause of the bug. Eventually, no internal code should be using the
 163  * bn_fix_top() macro. External applications and libraries should try this with
 164  * their own code too, both in terms of building against the openssl headers
 165  * with BN_DEBUG defined *and* linking with a version of OpenSSL built with it
 166  * defined. This not only improves external code, it provides more test
 167  * coverage for openssl's own code.
 168  */
 169
 170 # ifdef BN_DEBUG
 171 /*
 172  * The new BN_FLG_FIXED_TOP flag marks vectors that were not treated with
 173  * bn_correct_top, in other words such vectors are permitted to have zeros
 174  * in most significant limbs. Such vectors are used internally to achieve
 175  * execution time invariance for critical operations with private keys.
 176  * It's BN_DEBUG-only flag, because user application is not supposed to
 177  * observe it anyway. Moreover, optimizing compiler would actually remove
 178  * all operations manipulating the bit in question in non-BN_DEBUG build.
 179  */
 180 #  define BN_FLG_FIXED_TOP 0x10000
 181 #  ifdef BN_RAND_DEBUG
 182 #   define bn_pollute(a) \
 183         do { \
 184             const BIGNUM *_bnum1 = (a); \
 185             if (_bnum1->top < _bnum1->dmax) { \
 186                 unsigned char _tmp_char; \
 187                 /* We cast away const without the compiler knowing, any \
 188                  * *genuinely* constant variables that aren't mutable \
 189                  * wouldn't be constructed with top!=dmax. */ \
 190                 BN_ULONG *_not_const; \
 191                 memcpy(&_not_const, &_bnum1->d, sizeof(_not_const)); \
 192                 (void)RAND_bytes(&_tmp_char, 1); /* Debug only - safe to ignore error return */\
 193                 memset(_not_const + _bnum1->top, _tmp_char, \
 194                        sizeof(*_not_const) * (_bnum1->dmax - _bnum1->top)); \
 195             } \
 196         } while(0)
 197 #  else
 198 #   define bn_pollute(a)
 199 #  endif
 200 #  define bn_check_top(a) \
 201         do { \
 202                 const BIGNUM *_bnum2 = (a); \
 203                 if (_bnum2 != NULL) { \
 204                         int _top = _bnum2->top; \
 205                         (void)ossl_assert((_top == 0 && !_bnum2->neg) || \
 206                                   (_top && ((_bnum2->flags & BN_FLG_FIXED_TOP) \
 207                                             || _bnum2->d[_top - 1] != 0))); \
 208                         bn_pollute(_bnum2); \
 209                 } \
 210         } while(0)
 211
 212 #  define bn_fix_top(a)           bn_check_top(a)
 213
 214 #  define bn_check_size(bn, bits) bn_wcheck_size(bn, ((bits+BN_BITS2-1))/BN_BITS2)
 215 #  define bn_wcheck_size(bn, words) \
 216         do { \
 217                 const BIGNUM *_bnum2 = (bn); \
 218                 assert((words) <= (_bnum2)->dmax && \
 219                        (words) >= (_bnum2)->top); \
 220                 /* avoid unused variable warning with NDEBUG */ \
 221                 (void)(_bnum2); \
 222         } while(0)
 223
 224 # else                          /* !BN_DEBUG */
 225
 226 #  define BN_FLG_FIXED_TOP 0
 227 #  define bn_pollute(a)
 228 #  define bn_check_top(a)
 229 #  define bn_fix_top(a)           bn_correct_top(a)
 230 #  define bn_check_size(bn, bits)
 231 #  define bn_wcheck_size(bn, words)
 232
 233 # endif
 234
 235 BN_ULONG bn_mul_add_words(BN_ULONG *rp, const BN_ULONG *ap, int num,
 236                           BN_ULONG w);
 237 BN_ULONG bn_mul_words(BN_ULONG *rp, const BN_ULONG *ap, int num, BN_ULONG w);
 238 void bn_sqr_words(BN_ULONG *rp, const BN_ULONG *ap, int num);
 239 BN_ULONG bn_div_words(BN_ULONG h, BN_ULONG l, BN_ULONG d);
 240 BN_ULONG bn_add_words(BN_ULONG *rp, const BN_ULONG *ap, const BN_ULONG *bp,
 241                       int num);
 242 BN_ULONG bn_sub_words(BN_ULONG *rp, const BN_ULONG *ap, const BN_ULONG *bp,
 243                       int num);
 244
 245 struct bignum_st {
 246     BN_ULONG *d;                /*
 247                                  * Pointer to an array of 'BN_BITS2' bit
 248                                  * chunks. These chunks are organised in
 249                                  * a least significant chunk first order.
 250                                  */
 251     int top;                    /* Index of last used d +1. */
 252     /* The next are internal book keeping for bn_expand. */
 253     int dmax;                   /* Size of the d array. */
 254     int neg;                    /* one if the number is negative */
 255     int flags;
 256 };
 257
 258 /* Used for montgomery multiplication */
 259 struct bn_mont_ctx_st {
 260     int ri;                     /* number of bits in R */
 261     BIGNUM RR;                  /* used to convert to montgomery form,
 262                                    possibly zero-padded */
 263     BIGNUM N;                   /* The modulus */
 264     BIGNUM Ni;                  /* R*(1/R mod N) - N*Ni = 1 (Ni is only
 265                                  * stored for bignum algorithm) */
 266     BN_ULONG n0[2];             /* least significant word(s) of Ni; (type
 267                                  * changed with 0.9.9, was "BN_ULONG n0;"
 268                                  * before) */
 269     int flags;
 270 };
 271
 272 /*
 273  * Used for reciprocal division/mod functions It cannot be shared between
 274  * threads
 275  */
 276 struct bn_recp_ctx_st {
 277     BIGNUM N;                   /* the divisor */
 278     BIGNUM Nr;                  /* the reciprocal */
 279     int num_bits;
 280     int shift;
 281     int flags;
 282 };
 283
 284 /* Used for slow "generation" functions. */
 285 struct bn_gencb_st {
 286     unsigned int ver;           /* To handle binary (in)compatibility */
 287     void *arg;                  /* callback-specific data */
 288     union {
 289         /* if (ver==1) - handles old style callbacks */
 290         void (*cb_1) (int, int, void *);
 291         /* if (ver==2) - new callback style */
 292         int (*cb_2) (int, int, BN_GENCB *);
 293     } cb;
 294 };
 295
 296 /*-
 297  * BN_window_bits_for_exponent_size -- macro for sliding window mod_exp functions
 298  *
 299  *
 300  * For window size 'w' (w >= 2) and a random 'b' bits exponent,
 301  * the number of multiplications is a constant plus on average
 302  *
 303  *    2^(w-1) + (b-w)/(w+1);
 304  *
 305  * here  2^(w-1)  is for precomputing the table (we actually need
 306  * entries only for windows that have the lowest bit set), and
 307  * (b-w)/(w+1)  is an approximation for the expected number of
 308  * w-bit windows, not counting the first one.
 309  *
 310  * Thus we should use
 311  *
 312  *    w >= 6  if        b > 671
 313  *     w = 5  if  671 > b > 239
 314  *     w = 4  if  239 > b >  79
 315  *     w = 3  if   79 > b >  23
 316  *    w <= 2  if   23 > b
 317  *
 318  * (with draws in between).  Very small exponents are often selected
 319  * with low Hamming weight, so we use  w = 1  for b <= 23.
 320  */
 321 # define BN_window_bits_for_exponent_size(b) \
 322                 ((b) > 671 ? 6 : \
 323                  (b) > 239 ? 5 : \
 324                  (b) >  79 ? 4 : \
 325                  (b) >  23 ? 3 : 1)
 326
 327 /*
 328  * BN_mod_exp_mont_consttime is based on the assumption that the L1 data cache
 329  * line width of the target processor is at least the following value.
 330  */
 331 # define MOD_EXP_CTIME_MIN_CACHE_LINE_WIDTH      ( 64 )
 332 # define MOD_EXP_CTIME_MIN_CACHE_LINE_MASK       (MOD_EXP_CTIME_MIN_CACHE_LINE_WIDTH - 1)
 333
 334 /*
 335  * Window sizes optimized for fixed window size modular exponentiation
 336  * algorithm (BN_mod_exp_mont_consttime). To achieve the security goals of
 337  * BN_mode_exp_mont_consttime, the maximum size of the window must not exceed
 338  * log_2(MOD_EXP_CTIME_MIN_CACHE_LINE_WIDTH). Window size thresholds are
 339  * defined for cache line sizes of 32 and 64, cache line sizes where
 340  * log_2(32)=5 and log_2(64)=6 respectively. A window size of 7 should only be
 341  * used on processors that have a 128 byte or greater cache line size.
 342  */
 343 # if MOD_EXP_CTIME_MIN_CACHE_LINE_WIDTH == 64
 344
 345 #  define BN_window_bits_for_ctime_exponent_size(b) \
 346                 ((b) > 937 ? 6 : \
 347                  (b) > 306 ? 5 : \
 348                  (b) >  89 ? 4 : \
 349                  (b) >  22 ? 3 : 1)
 350 #  define BN_MAX_WINDOW_BITS_FOR_CTIME_EXPONENT_SIZE    (6)
 351
 352 # elif MOD_EXP_CTIME_MIN_CACHE_LINE_WIDTH == 32
 353
 354 #  define BN_window_bits_for_ctime_exponent_size(b) \
 355                 ((b) > 306 ? 5 : \
 356                  (b) >  89 ? 4 : \
 357                  (b) >  22 ? 3 : 1)
 358 #  define BN_MAX_WINDOW_BITS_FOR_CTIME_EXPONENT_SIZE    (5)
 359
 360 # endif
 361
 362 /* Pentium pro 16,16,16,32,64 */
 363 /* Alpha       16,16,16,16.64 */
 364 # define BN_MULL_SIZE_NORMAL                     (16)/* 32 */
 365 # define BN_MUL_RECURSIVE_SIZE_NORMAL            (16)/* 32 less than */
 366 # define BN_SQR_RECURSIVE_SIZE_NORMAL            (16)/* 32 */
 367 # define BN_MUL_LOW_RECURSIVE_SIZE_NORMAL        (32)/* 32 */
 368 # define BN_MONT_CTX_SET_SIZE_WORD               (64)/* 32 */
 369
 370 /*
 371  * 2011-02-22 SMS. In various places, a size_t variable or a type cast to
 372  * size_t was used to perform integer-only operations on pointers.  This
 373  * failed on VMS with 64-bit pointers (CC /POINTER_SIZE = 64) because size_t
 374  * is still only 32 bits.  What's needed in these cases is an integer type
 375  * with the same size as a pointer, which size_t is not certain to be. The
 376  * only fix here is VMS-specific.
 377  */
 378 # if defined(OPENSSL_SYS_VMS)
 379 #  if __INITIAL_POINTER_SIZE == 64
 380 #   define PTR_SIZE_INT long long
 381 #  else                         /* __INITIAL_POINTER_SIZE == 64 */
 382 #   define PTR_SIZE_INT int
 383 #  endif                        /* __INITIAL_POINTER_SIZE == 64 [else] */
 384 # elif !defined(PTR_SIZE_INT)   /* defined(OPENSSL_SYS_VMS) */
 385 #  define PTR_SIZE_INT size_t
 386 # endif                         /* defined(OPENSSL_SYS_VMS) [else] */
 387
 388 # if !defined(OPENSSL_NO_ASM) && !defined(OPENSSL_NO_INLINE_ASM) && !defined(PEDANTIC)
 389 /*
 390  * BN_UMULT_HIGH section.
 391  * If the compiler doesn't support 2*N integer type, then you have to
 392  * replace every N*N multiplication with 4 (N/2)*(N/2) accompanied by some
 393  * shifts and additions which unavoidably results in severe performance
 394  * penalties. Of course provided that the hardware is capable of producing
 395  * 2*N result... That's when you normally start considering assembler
 396  * implementation. However! It should be pointed out that some CPUs (e.g.,
 397  * PowerPC, Alpha, and IA-64) provide *separate* instruction calculating
 398  * the upper half of the product placing the result into a general
 399  * purpose register. Now *if* the compiler supports inline assembler,
 400  * then it's not impossible to implement the "bignum" routines (and have
 401  * the compiler optimize 'em) exhibiting "native" performance in C. That's
 402  * what BN_UMULT_HIGH macro is about:-) Note that more recent compilers do
 403  * support 2*64 integer type, which is also used here.
 404  */
 405 #  if defined(__SIZEOF_INT128__) && __SIZEOF_INT128__==16 && \
 406       (defined(SIXTY_FOUR_BIT) || defined(SIXTY_FOUR_BIT_LONG))
 407 #   define BN_UMULT_HIGH(a,b)          (((uint128_t)(a)*(b))>>64)
 408 #   define BN_UMULT_LOHI(low,high,a,b) ({       \
 409         uint128_t ret=(uint128_t)(a)*(b);   \
 410         (high)=ret>>64; (low)=ret;      })
 411 #  elif defined(__alpha) && (defined(SIXTY_FOUR_BIT_LONG) || defined(SIXTY_FOUR_BIT))
 412 #   if defined(__DECC)
 413 #    include <c_asm.h>
 414 #    define BN_UMULT_HIGH(a,b)   (BN_ULONG)asm("umulh %a0,%a1,%v0",(a),(b))
 415 #   elif defined(__GNUC__) && __GNUC__>=2
 416 #    define BN_UMULT_HIGH(a,b)   ({     \
 417         register BN_ULONG ret;          \
 418         asm ("umulh     %1,%2,%0"       \
 419              : "=r"(ret)                \
 420              : "r"(a), "r"(b));         \
 421         ret;                      })
 422 #   endif                       /* compiler */
 423 #  elif defined(_ARCH_PPC64) && defined(SIXTY_FOUR_BIT_LONG)
 424 #   if defined(__GNUC__) && __GNUC__>=2
 425 #    define BN_UMULT_HIGH(a,b)   ({     \
 426         register BN_ULONG ret;          \
 427         asm ("mulhdu    %0,%1,%2"       \
 428              : "=r"(ret)                \
 429              : "r"(a), "r"(b));         \
 430         ret;                      })
 431 #   endif                       /* compiler */
 432 #  elif (defined(__x86_64) || defined(__x86_64__)) && \
 433        (defined(SIXTY_FOUR_BIT_LONG) || defined(SIXTY_FOUR_BIT))
 434 #   if defined(__GNUC__) && __GNUC__>=2
 435 #    define BN_UMULT_HIGH(a,b)   ({     \
 436         register BN_ULONG ret,discard;  \
 437         asm ("mulq      %3"             \
 438              : "=a"(discard),"=d"(ret)  \
 439              : "a"(a), "g"(b)           \
 440              : "cc");                   \
 441         ret;                      })
 442 #    define BN_UMULT_LOHI(low,high,a,b) \
 443         asm ("mulq      %3"             \
 444                 : "=a"(low),"=d"(high)  \
 445                 : "a"(a),"g"(b)         \
 446                 : "cc");
 447 #   endif
 448 #  elif (defined(_M_AMD64) || defined(_M_X64)) && defined(SIXTY_FOUR_BIT)
 449 #   if defined(_MSC_VER) && _MSC_VER>=1400
 450 unsigned __int64 __umulh(unsigned __int64 a, unsigned __int64 b);
 451 unsigned __int64 _umul128(unsigned __int64 a, unsigned __int64 b,
 452                           unsigned __int64 *h);
 453 #    pragma intrinsic(__umulh,_umul128)
 454 #    define BN_UMULT_HIGH(a,b)           __umulh((a),(b))
 455 #    define BN_UMULT_LOHI(low,high,a,b)  ((low)=_umul128((a),(b),&(high)))
 456 #   endif
 457 #  elif defined(__mips) && (defined(SIXTY_FOUR_BIT) || defined(SIXTY_FOUR_BIT_LONG))
 458 #   if defined(__GNUC__) && __GNUC__>=2
 459 #    define BN_UMULT_HIGH(a,b) ({       \
 460         register BN_ULONG ret;          \
 461         asm ("dmultu    %1,%2"          \
 462              : "=h"(ret)                \
 463              : "r"(a), "r"(b) : "l");   \
 464         ret;                    })
 465 #    define BN_UMULT_LOHI(low,high,a,b) \
 466         asm ("dmultu    %2,%3"          \
 467              : "=l"(low),"=h"(high)     \
 468              : "r"(a), "r"(b));
 469 #   endif
 470 #  elif defined(__aarch64__) && defined(SIXTY_FOUR_BIT_LONG)
 471 #   if defined(__GNUC__) && __GNUC__>=2
 472 #    define BN_UMULT_HIGH(a,b)   ({     \
 473         register BN_ULONG ret;          \
 474         asm ("umulh     %0,%1,%2"       \
 475              : "=r"(ret)                \
 476              : "r"(a), "r"(b));         \
 477         ret;                      })
 478 #   endif
 479 #  endif                        /* cpu */
 480 # endif                         /* OPENSSL_NO_ASM */
 481
 482 # ifdef BN_RAND_DEBUG
 483 #  define bn_clear_top2max(a) \
 484         { \
 485         int      ind = (a)->dmax - (a)->top; \
 486         BN_ULONG *ftl = &(a)->d[(a)->top-1]; \
 487         for (; ind != 0; ind--) \
 488                 *(++ftl) = 0x0; \
 489         }
 490 # else
 491 #  define bn_clear_top2max(a)
 492 # endif
 493
 494 # ifdef BN_LLONG
 495 /*******************************************************************
 496  * Using the long long type, has to be twice as wide as BN_ULONG...
 497  */
 498 #  define Lw(t)    (((BN_ULONG)(t))&BN_MASK2)
 499 #  define Hw(t)    (((BN_ULONG)((t)>>BN_BITS2))&BN_MASK2)
 500
 501 #  define mul_add(r,a,w,c) { \
 502         BN_ULLONG t; \
 503         t=(BN_ULLONG)w * (a) + (r) + (c); \
 504         (r)= Lw(t); \
 505         (c)= Hw(t); \
 506         }
 507
 508 #  define mul(r,a,w,c) { \
 509         BN_ULLONG t; \
 510         t=(BN_ULLONG)w * (a) + (c); \
 511         (r)= Lw(t); \
 512         (c)= Hw(t); \
 513         }
 514
 515 #  define sqr(r0,r1,a) { \
 516         BN_ULLONG t; \
 517         t=(BN_ULLONG)(a)*(a); \
 518         (r0)=Lw(t); \
 519         (r1)=Hw(t); \
 520         }
 521
 522 # elif defined(BN_UMULT_LOHI)
 523 #  define mul_add(r,a,w,c) {              \
 524         BN_ULONG high,low,ret,tmp=(a);  \
 525         ret =  (r);                     \
 526         BN_UMULT_LOHI(low,high,w,tmp);  \
 527         ret += (c);                     \
 528         (c) =  (ret<(c))?1:0;           \
 529         (c) += high;                    \
 530         ret += low;                     \
 531         (c) += (ret<low)?1:0;           \
 532         (r) =  ret;                     \
 533         }
 534
 535 #  define mul(r,a,w,c)    {               \
 536         BN_ULONG high,low,ret,ta=(a);   \
 537         BN_UMULT_LOHI(low,high,w,ta);   \
 538         ret =  low + (c);               \
 539         (c) =  high;                    \
 540         (c) += (ret<low)?1:0;           \
 541         (r) =  ret;                     \
 542         }
 543
 544 #  define sqr(r0,r1,a)    {               \
 545         BN_ULONG tmp=(a);               \
 546         BN_UMULT_LOHI(r0,r1,tmp,tmp);   \
 547         }
 548
 549 # elif defined(BN_UMULT_HIGH)
 550 #  define mul_add(r,a,w,c) {              \
 551         BN_ULONG high,low,ret,tmp=(a);  \
 552         ret =  (r);                     \
 553         high=  BN_UMULT_HIGH(w,tmp);    \
 554         ret += (c);                     \
 555         low =  (w) * tmp;               \
 556         (c) =  (ret<(c))?1:0;           \
 557         (c) += high;                    \
 558         ret += low;                     \
 559         (c) += (ret<low)?1:0;           \
 560         (r) =  ret;                     \
 561         }
 562
 563 #  define mul(r,a,w,c)    {               \
 564         BN_ULONG high,low,ret,ta=(a);   \
 565         low =  (w) * ta;                \
 566         high=  BN_UMULT_HIGH(w,ta);     \
 567         ret =  low + (c);               \
 568         (c) =  high;                    \
 569         (c) += (ret<low)?1:0;           \
 570         (r) =  ret;                     \
 571         }
 572
 573 #  define sqr(r0,r1,a)    {               \
 574         BN_ULONG tmp=(a);               \
 575         (r0) = tmp * tmp;               \
 576         (r1) = BN_UMULT_HIGH(tmp,tmp);  \
 577         }
 578
 579 # else
 580 /*************************************************************
 581  * No long long type
 582  */
 583
 584 #  define LBITS(a)        ((a)&BN_MASK2l)
 585 #  define HBITS(a)        (((a)>>BN_BITS4)&BN_MASK2l)
 586 #  define L2HBITS(a)      (((a)<<BN_BITS4)&BN_MASK2)
 587
 588 #  define LLBITS(a)       ((a)&BN_MASKl)
 589 #  define LHBITS(a)       (((a)>>BN_BITS2)&BN_MASKl)
 590 #  define LL2HBITS(a)     ((BN_ULLONG)((a)&BN_MASKl)<<BN_BITS2)
 591
 592 #  define mul64(l,h,bl,bh) \
 593         { \
 594         BN_ULONG m,m1,lt,ht; \
 595  \
 596         lt=l; \
 597         ht=h; \
 598         m =(bh)*(lt); \
 599         lt=(bl)*(lt); \
 600         m1=(bl)*(ht); \
 601         ht =(bh)*(ht); \
 602         m=(m+m1)&BN_MASK2; if (m < m1) ht+=L2HBITS((BN_ULONG)1); \
 603         ht+=HBITS(m); \
 604         m1=L2HBITS(m); \
 605         lt=(lt+m1)&BN_MASK2; if (lt < m1) ht++; \
 606         (l)=lt; \
 607         (h)=ht; \
 608         }
 609
 610 #  define sqr64(lo,ho,in) \
 611         { \
 612         BN_ULONG l,h,m; \
 613  \
 614         h=(in); \
 615         l=LBITS(h); \
 616         h=HBITS(h); \
 617         m =(l)*(h); \
 618         l*=l; \
 619         h*=h; \
 620         h+=(m&BN_MASK2h1)>>(BN_BITS4-1); \
 621         m =(m&BN_MASK2l)<<(BN_BITS4+1); \
 622         l=(l+m)&BN_MASK2; if (l < m) h++; \
 623         (lo)=l; \
 624         (ho)=h; \
 625         }
 626
 627 #  define mul_add(r,a,bl,bh,c) { \
 628         BN_ULONG l,h; \
 629  \
 630         h= (a); \
 631         l=LBITS(h); \
 632         h=HBITS(h); \
 633         mul64(l,h,(bl),(bh)); \
 634  \
 635         /* non-multiply part */ \
 636         l=(l+(c))&BN_MASK2; if (l < (c)) h++; \
 637         (c)=(r); \
 638         l=(l+(c))&BN_MASK2; if (l < (c)) h++; \
 639         (c)=h&BN_MASK2; \
 640         (r)=l; \
 641         }
 642
 643 #  define mul(r,a,bl,bh,c) { \
 644         BN_ULONG l,h; \
 645  \
 646         h= (a); \
 647         l=LBITS(h); \
 648         h=HBITS(h); \
 649         mul64(l,h,(bl),(bh)); \
 650  \
 651         /* non-multiply part */ \
 652         l+=(c); if ((l&BN_MASK2) < (c)) h++; \
 653         (c)=h&BN_MASK2; \
 654         (r)=l&BN_MASK2; \
 655         }
 656 # endif                         /* !BN_LLONG */
 657
 658 void BN_RECP_CTX_init(BN_RECP_CTX *recp);
 659 void BN_MONT_CTX_init(BN_MONT_CTX *ctx);
 660
 661 void bn_init(BIGNUM *a);
 662 void bn_mul_normal(BN_ULONG *r, BN_ULONG *a, int na, BN_ULONG *b, int nb);
 663 void bn_mul_comba8(BN_ULONG *r, BN_ULONG *a, BN_ULONG *b);
 664 void bn_mul_comba4(BN_ULONG *r, BN_ULONG *a, BN_ULONG *b);
 665 void bn_sqr_normal(BN_ULONG *r, const BN_ULONG *a, int n, BN_ULONG *tmp);
 666 void bn_sqr_comba8(BN_ULONG *r, const BN_ULONG *a);
 667 void bn_sqr_comba4(BN_ULONG *r, const BN_ULONG *a);
 668 int bn_cmp_words(const BN_ULONG *a, const BN_ULONG *b, int n);
 669 int bn_cmp_part_words(const BN_ULONG *a, const BN_ULONG *b, int cl, int dl);
 670 void bn_mul_recursive(BN_ULONG *r, BN_ULONG *a, BN_ULONG *b, int n2,
 671                       int dna, int dnb, BN_ULONG *t);
 672 void bn_mul_part_recursive(BN_ULONG *r, BN_ULONG *a, BN_ULONG *b,
 673                            int n, int tna, int tnb, BN_ULONG *t);
 674 void bn_sqr_recursive(BN_ULONG *r, const BN_ULONG *a, int n2, BN_ULONG *t);
 675 void bn_mul_low_normal(BN_ULONG *r, BN_ULONG *a, BN_ULONG *b, int n);
 676 void bn_mul_low_recursive(BN_ULONG *r, BN_ULONG *a, BN_ULONG *b, int n2,
 677                           BN_ULONG *t);
 678 BN_ULONG bn_sub_part_words(BN_ULONG *r, const BN_ULONG *a, const BN_ULONG *b,
 679                            int cl, int dl);
 680 int bn_mul_mont(BN_ULONG *rp, const BN_ULONG *ap, const BN_ULONG *bp,
 681                 const BN_ULONG *np, const BN_ULONG *n0, int num);
 682
 683 BIGNUM *int_bn_mod_inverse(BIGNUM *in,
 684                            const BIGNUM *a, const BIGNUM *n, BN_CTX *ctx,
 685                            int *noinv);
 686
 687 static ossl_inline BIGNUM *bn_expand(BIGNUM *a, int bits)
 688 {
 689     if (bits > (INT_MAX - BN_BITS2 + 1))
 690         return NULL;
 691
 692     if (((bits+BN_BITS2-1)/BN_BITS2) <= (a)->dmax)
 693         return a;
 694
 695     return bn_expand2((a),(bits+BN_BITS2-1)/BN_BITS2);
 696 }
 697
 698 int ossl_bn_check_prime(const BIGNUM *w, int checks, BN_CTX *ctx,
 699                         int do_trial_division, BN_GENCB *cb);
 700
 701 #endif