bn_mp_set_double.c

Branch :

Show log
Commit
Author : Francois Perrad
Date : 2018-12-01 08:35:22
Hash : 4d6a3cdc
Message : handle the last value of res

bn_mp_set_double.c

#include "tommath_private.h"
#ifdef BN_MP_SET_DOUBLE_C
/* LibTomMath, multiple-precision integer library -- Tom St Denis
 *
 * LibTomMath is a library that provides multiple-precision
 * integer arithmetic as well as number theoretic functionality.
 *
 * The library was designed directly after the MPI library by
 * Michael Fromberger but has been written from scratch with
 * additional optimizations in place.
 *
 * The library is free for all purposes without any express
 * guarantee it works.
 */

#if defined(__STDC_IEC_559__) || defined(__GCC_IEC_559)
int mp_set_double(mp_int *a, double b)
{
   uint64_t frac;
   int exp, res;
   union {
      double   dbl;
      uint64_t bits;
   } cast;
   cast.dbl = b;

   exp = (int)((unsigned)(cast.bits >> 52) & 0x7FFU);
   frac = (cast.bits & ((1ULL << 52) - 1ULL)) | (1ULL << 52);

   if (exp == 0x7FF) { /* +-inf, NaN */
      return MP_VAL;
   }
   exp -= 1023 + 52;

   res = mp_set_long_long(a, frac);
   if (res != MP_OKAY) {
      return res;
   }

   res = (exp < 0) ? mp_div_2d(a, -exp, a, NULL) : mp_mul_2d(a, exp, a);
   if (res != MP_OKAY) {
      return res;
   }

   if (((cast.bits >> 63) != 0ULL) && (mp_iszero(a) == MP_NO)) {
      SIGN(a) = MP_NEG;
   }

   return MP_OKAY;
}
#else
#  warning "mp_set_double implementation is only available on platforms with IEEE754 floating point format"
#endif
#endif

/* ref:         $Format:%D$ */
/* git commit:  $Format:%H$ */
/* commit time: $Format:%ai$ */

kc3-lang/libtommath/bn_mp_set_double.c

Commit