Copyright 2004, 2006, 2007 Free Software Foundation, Inc.
Contributed by the Arenaire and Cacao projects, INRIA.
This file is part of the MPFR Library.
The MPFR Library 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.
The MPFR Library 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 the MPFR Library; see the file COPYING.LIB. If not, write to
the Free Software Foundation, Inc., 51 Franklin St, Fifth Floor, Boston,
MA 02110-1301, USA. */
#if HAVE_CONFIG_H
# include "config.h"
#endif
INTMAX_MAX, ... macros should only be defined if explicitly requested. */
#if defined __cplusplus
# define __STDC_LIMIT_MACROS
# define __STDC_CONSTANT_MACROS
#endif
#ifdef HAVE_STDINT_H
# include <stdint.h>
#endif
#ifdef HAVE_INTTYPES_H
# include <inttypes.h>
#endif
#include "mpfr-impl.h"
#ifdef _MPFR_H_HAVE_INTMAX_T
intmax_t
mpfr_get_sj (mpfr_srcptr f, mpfr_rnd_t rnd)
{
intmax_t r;
mp_prec_t prec;
mpfr_t x;
if (!mpfr_fits_intmax_p (f, rnd))
{
MPFR_SET_ERANGE ();
return MPFR_IS_NEG (f) ? INTMAX_MIN : INTMAX_MAX;
}
if (MPFR_IS_ZERO (f))
return (intmax_t) 0;
for (r = INTMAX_MIN, prec = 0; r != 0; r /= 2, prec++)
{
}
we chose INTMAX_MIN so that INTMAX_MIN - 1 is always representable in
precision prec; this is useful to detect overflows in GMP_RNDZ (will
be needed later). */
mpfr_init2 (x, prec);
mpfr_rint (x, f, rnd);
MPFR_ASSERTN (MPFR_IS_FP (x));
if (MPFR_NOTZERO (x))
{
mp_limb_t *xp;
int sh, n;
xp = MPFR_MANT (x);
sh = MPFR_GET_EXP (x);
MPFR_ASSERTN ((mp_prec_t) sh <= prec);
if (INTMAX_MIN + INTMAX_MAX != 0
&& MPFR_UNLIKELY ((mp_prec_t) sh == prec))
{
has the same size as intmax_t, we cannot use the code in
the for loop since the operations would be performed in
unsigned arithmetic. */
MPFR_ASSERTN (MPFR_IS_NEG (x) && (mpfr_powerof2_raw (x)));
r = INTMAX_MIN;
}
else if (MPFR_IS_POS (x))
{
for (n = MPFR_LIMB_SIZE (x) - 1; n >= 0; n--)
{
sh -= BITS_PER_MP_LIMB;
When sh >= 0, the cast must be performed before the shift
for the case sizeof(intmax_t) > sizeof(mp_limb_t).
When sh < 0, the cast must be performed after the shift
for the case sizeof(intmax_t) == sizeof(mp_limb_t), as
mp_limb_t is unsigned, therefore not representable as an
intmax_t when the MSB is 1 (this is the case here). */
r += (sh >= 0
? (intmax_t) xp[n] << sh
: (intmax_t) (xp[n] >> (-sh)));
}
}
else
{
for (n = MPFR_LIMB_SIZE (x) - 1; n >= 0; n--)
{
sh -= BITS_PER_MP_LIMB;
r -= (sh >= 0
? (intmax_t) xp[n] << sh
: (intmax_t) (xp[n] >> (-sh)));
}
}
}
mpfr_clear (x);
return r;
}
#endif