glibc/sysdeps/ieee754/ldbl-128ibm/math_ldbl.h
Alan Modra 4ebd120cd9 PowerPC floating point little-endian [2 of 15]
http://sourceware.org/ml/libc-alpha/2013-08/msg00082.html

This patch replaces occurrences of GET_LDOUBLE_* and SET_LDOUBLE_*
macros, and union ieee854_long_double_shape_type in ldbl-128ibm/,
and a stray one in the 32-bit fpu support.  These files have no
significant changes apart from rewriting the long double bit access.

	* sysdeps/ieee754/ldbl-128ibm/math_ldbl.h (ldbl_high): Define.
	* sysdeps/ieee754/ldbl-128ibm/e_acoshl.c (__ieee754_acoshl): Rewrite
	all uses of ieee854 long double macros and unions.
	* sysdeps/ieee754/ldbl-128ibm/e_acosl.c (__ieee754_acosl): Likewise.
	* sysdeps/ieee754/ldbl-128ibm/e_asinl.c (__ieee754_asinl): Likewise.
	* sysdeps/ieee754/ldbl-128ibm/e_atanhl.c (__ieee754_atanhl): Likewise.
	* sysdeps/ieee754/ldbl-128ibm/e_coshl.c (__ieee754_coshl): Likewise.
	* sysdeps/ieee754/ldbl-128ibm/e_log2l.c (__ieee754_log2l): Likewise.
	* sysdeps/ieee754/ldbl-128ibm/e_rem_pio2l.c (__ieee754_rem_pio2l):
	Likewise.
	* sysdeps/ieee754/ldbl-128ibm/e_sinhl.c (__ieee754_sinhl): Likewise.
	* sysdeps/ieee754/ldbl-128ibm/k_cosl.c (__kernel_cosl): Likewise.
	* sysdeps/ieee754/ldbl-128ibm/k_sincosl.c (__kernel_sincosl): Likewise.
	* sysdeps/ieee754/ldbl-128ibm/k_sinl.c (__kernel_sinl): Likewise.
	* sysdeps/ieee754/ldbl-128ibm/s_asinhl.c (__asinhl): Likewise.
	* sysdeps/ieee754/ldbl-128ibm/s_atanl.c (__atanl): Likewise.
	Simplify sign and nan test too.
	* sysdeps/ieee754/ldbl-128ibm/s_cosl.c (__cosl): Likewise.
	* sysdeps/ieee754/ldbl-128ibm/s_fabsl.c (__fabsl): Likewise.
	* sysdeps/ieee754/ldbl-128ibm/s_finitel.c (___finitel): Likewise.
	* sysdeps/ieee754/ldbl-128ibm/s_fpclassifyl.c (___fpclassifyl):
	Likewise.
	* sysdeps/ieee754/ldbl-128ibm/s_isnanl.c (___isnanl): Likewise.
	* sysdeps/ieee754/ldbl-128ibm/s_issignalingl.c (__issignalingl):
	Likewise.
	* sysdeps/ieee754/ldbl-128ibm/s_logbl.c (__logbl): Likewise.
	* sysdeps/ieee754/ldbl-128ibm/s_signbitl.c (___signbitl): Likewise.
	* sysdeps/ieee754/ldbl-128ibm/s_sincosl.c (__sincosl): Likewise.
	* sysdeps/ieee754/ldbl-128ibm/s_sinl.c (__sinl): Likewise.
	* sysdeps/ieee754/ldbl-128ibm/s_tanl.c (__tanl): Likewise.
	* sysdeps/powerpc/powerpc32/power7/fpu/s_logbl.c (__logbl): Likewise.
2013-10-04 10:32:19 +09:30

234 lines
6.2 KiB
C

#ifndef _MATH_PRIVATE_H_
#error "Never use <math_ldbl.h> directly; include <math_private.h> instead."
#endif
#include <sysdeps/ieee754/ldbl-128/math_ldbl.h>
#include <ieee754.h>
#include <stdint.h>
/* To suit our callers we return *hi64 and *lo64 as if they came from
an ieee854 112 bit mantissa, that is, 48 bits in *hi64 (plus one
implicit bit) and 64 bits in *lo64. */
static inline void
ldbl_extract_mantissa (int64_t *hi64, uint64_t *lo64, int *exp, long double x)
{
/* We have 105 bits of mantissa plus one implicit digit. Since
106 bits are representable we use the first implicit digit for
the number before the decimal point and the second implicit bit
as bit 53 of the mantissa. */
uint64_t hi, lo;
union ibm_extended_long_double u;
u.ld = x;
*exp = u.d[0].ieee.exponent - IEEE754_DOUBLE_BIAS;
lo = ((uint64_t) u.d[1].ieee.mantissa0 << 32) | u.d[1].ieee.mantissa1;
hi = ((uint64_t) u.d[0].ieee.mantissa0 << 32) | u.d[0].ieee.mantissa1;
if (u.d[0].ieee.exponent != 0)
{
int ediff;
/* If not a denormal or zero then we have an implicit 53rd bit. */
hi |= (uint64_t) 1 << 52;
if (u.d[1].ieee.exponent != 0)
lo |= (uint64_t) 1 << 52;
else
/* A denormal is to be interpreted as having a biased exponent
of 1. */
lo = lo << 1;
/* We are going to shift 4 bits out of hi later, because we only
want 48 bits in *hi64. That means we want 60 bits in lo, but
we currently only have 53. Shift the value up. */
lo = lo << 7;
/* The lower double is normalized separately from the upper.
We may need to adjust the lower mantissa to reflect this.
The difference between the exponents can be larger than 53
when the low double is much less than 1ULP of the upper
(in which case there are significant bits, all 0's or all
1's, between the two significands). The difference between
the exponents can be less than 53 when the upper double
exponent is nearing its minimum value (in which case the low
double is denormal ie. has an exponent of zero). */
ediff = u.d[0].ieee.exponent - u.d[1].ieee.exponent - 53;
if (ediff > 0)
{
if (ediff < 64)
lo = lo >> ediff;
else
lo = 0;
}
else if (ediff < 0)
lo = lo << -ediff;
if (u.d[0].ieee.negative != u.d[1].ieee.negative
&& lo != 0)
{
hi--;
lo = ((uint64_t) 1 << 60) - lo;
if (hi < (uint64_t) 1 << 52)
{
/* We have a borrow from the hidden bit, so shift left 1. */
hi = (hi << 1) | (lo >> 59);
lo = (((uint64_t) 1 << 60) - 1) & (lo << 1);
*exp = *exp - 1;
}
}
}
else
/* If the larger magnitude double is denormal then the smaller
one must be zero. */
hi = hi << 1;
*lo64 = (hi << 60) | lo;
*hi64 = hi >> 4;
}
static inline long double
ldbl_insert_mantissa (int sign, int exp, int64_t hi64, uint64_t lo64)
{
union ibm_extended_long_double u;
int expnt2;
uint64_t hi, lo;
u.d[0].ieee.negative = sign;
u.d[1].ieee.negative = sign;
u.d[0].ieee.exponent = exp + IEEE754_DOUBLE_BIAS;
u.d[1].ieee.exponent = 0;
expnt2 = exp - 53 + IEEE754_DOUBLE_BIAS;
/* Expect 113 bits (112 bits + hidden) right justified in two longs.
The low order 53 bits (52 + hidden) go into the lower double */
lo = (lo64 >> 7) & (((uint64_t) 1 << 53) - 1);
/* The high order 53 bits (52 + hidden) go into the upper double */
hi = lo64 >> 60;
hi |= hi64 << 4;
if (lo != 0)
{
int lzcount;
/* hidden bit of low double controls rounding of the high double.
If hidden is '1' and either the explicit mantissa is non-zero
or hi is odd, then round up hi and adjust lo (2nd mantissa)
plus change the sign of the low double to compensate. */
if ((lo & ((uint64_t) 1 << 52)) != 0
&& ((hi & 1) != 0 || (lo & (((uint64_t) 1 << 52) - 1)) != 0))
{
hi++;
if ((hi & ((uint64_t) 1 << 53)) != 0)
{
hi = hi >> 1;
u.d[0].ieee.exponent++;
}
u.d[1].ieee.negative = !sign;
lo = ((uint64_t) 1 << 53) - lo;
}
/* Normalize the low double. Shift the mantissa left until
the hidden bit is '1' and adjust the exponent accordingly. */
if (sizeof (lo) == sizeof (long))
lzcount = __builtin_clzl (lo);
else if ((lo >> 32) != 0)
lzcount = __builtin_clzl ((long) (lo >> 32));
else
lzcount = __builtin_clzl ((long) lo) + 32;
lzcount = lzcount - (64 - 53);
lo <<= lzcount;
expnt2 -= lzcount;
if (expnt2 >= 1)
/* Not denormal. */
u.d[1].ieee.exponent = expnt2;
else
{
/* Is denormal. Note that biased exponent of 0 is treated
as if it was 1, hence the extra shift. */
if (expnt2 > -53)
lo >>= 1 - expnt2;
else
lo = 0;
}
}
else
u.d[1].ieee.negative = 0;
u.d[1].ieee.mantissa1 = lo;
u.d[1].ieee.mantissa0 = lo >> 32;
u.d[0].ieee.mantissa1 = hi;
u.d[0].ieee.mantissa0 = hi >> 32;
return u.ld;
}
/* Handy utility functions to pack/unpack/cononicalize and find the nearbyint
of long double implemented as double double. */
static inline long double
default_ldbl_pack (double a, double aa)
{
union ibm_extended_long_double u;
u.d[0].d = a;
u.d[1].d = aa;
return u.ld;
}
static inline void
default_ldbl_unpack (long double l, double *a, double *aa)
{
union ibm_extended_long_double u;
u.ld = l;
*a = u.d[0].d;
*aa = u.d[1].d;
}
#ifndef ldbl_pack
# define ldbl_pack default_ldbl_pack
#endif
#ifndef ldbl_unpack
# define ldbl_unpack default_ldbl_unpack
#endif
/* Extract high double. */
#define ldbl_high(x) ((double) x)
/* Convert a finite long double to canonical form.
Does not handle +/-Inf properly. */
static inline void
ldbl_canonicalize (double *a, double *aa)
{
double xh, xl;
xh = *a + *aa;
xl = (*a - xh) + *aa;
*a = xh;
*aa = xl;
}
/* Simple inline nearbyint (double) function.
Only works in the default rounding mode
but is useful in long double rounding functions. */
static inline double
ldbl_nearbyint (double a)
{
double two52 = 0x1p52;
if (__builtin_expect ((__builtin_fabs (a) < two52), 1))
{
if (__builtin_expect ((a > 0.0), 1))
{
a += two52;
a -= two52;
}
else if (__builtin_expect ((a < 0.0), 1))
{
a = two52 - a;
a = -(a - two52);
}
}
return a;
}