gcc/libdecnumber/decimal64.c
Ben Elliston c01731362e Makefile.in (clean): Remove stray reference to libcpp.a.
* Makefile.in (clean): Remove stray reference to libcpp.a.

	* decimal128.h, decContext.c, decRound.c, decimal32.c,
	decNumber.c, decContext.h, decimal64.c, decimal32.h, decNumber.h,
	decimal64.h, decUtility.c, decLibrary.c, configure.ac,
	decNumberLocal.h, decUtility.h, decDPD.h, decimal128.c: Update FSF
	office address.

From-SVN: r108024
2005-12-05 10:50:48 +11:00

329 lines
12 KiB
C

/* Decimal 64-bit format module for the decNumber C Library
Copyright (C) 2005 Free Software Foundation, Inc.
Contributed by IBM Corporation. Author Mike Cowlishaw.
This file is part of GCC.
GCC is free software; you can redistribute it and/or modify it under
the terms of the GNU General Public License as published by the Free
Software Foundation; either version 2, or (at your option) any later
version.
GCC 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 General Public License
for more details.
You should have received a copy of the GNU General Public License
along with GCC; see the file COPYING. If not, write to the Free
Software Foundation, 51 Franklin Street, Fifth Floor, Boston, MA
02110-1301, USA. */
/* ------------------------------------------------------------------ */
/* This module comprises the routines for decimal64 format numbers. */
/* Conversions are supplied to and from decNumber and String. */
/* */
/* No arithmetic routines are included; decNumber provides these. */
/* */
/* Error handling is the same as decNumber (qv.). */
/* ------------------------------------------------------------------ */
#include <string.h> /* [for memset/memcpy] */
#include <stdio.h> /* [for printf] */
#define DECNUMDIGITS 16 /* we need decNumbers with space for 16 */
#include "config.h"
#include "decNumber.h" /* base number library */
#include "decNumberLocal.h" /* decNumber local types, etc. */
#include "decimal64.h" /* our primary include */
#include "decUtility.h" /* utility routines */
#if DECTRACE || DECCHECK
void decimal64Show (decimal64 *); /* for debug */
void decNumberShow (decNumber *); /* .. */
#endif
/* Useful macro */
/* Clear a structure (e.g., a decNumber) */
#define DEC_clear(d) memset(d, 0, sizeof(*d))
/* ------------------------------------------------------------------ */
/* decimal64FromNumber -- convert decNumber to decimal64 */
/* */
/* ds is the target decimal64 */
/* dn is the source number (assumed valid) */
/* set is the context, used only for reporting errors */
/* */
/* The set argument is used only for status reporting and for the */
/* rounding mode (used if the coefficient is more than DECIMAL64_Pmax */
/* digits or an overflow is detected). If the exponent is out of the */
/* valid range then Overflow or Underflow will be raised. */
/* After Underflow a subnormal result is possible. */
/* */
/* DEC_Clamped is set if the number has to be 'folded down' to fit, */
/* by reducing its exponent and multiplying the coefficient by a */
/* power of ten, or if the exponent on a zero had to be clamped. */
/* ------------------------------------------------------------------ */
decimal64 *
decimal64FromNumber (decimal64 * d64, decNumber * dn, decContext * set)
{
uInt status = 0; /* status accumulator */
Int pad = 0; /* coefficient pad digits */
decNumber dw; /* work */
decContext dc; /* .. */
uByte isneg = dn->bits & DECNEG; /* non-0 if original sign set */
uInt comb, exp; /* work */
/* If the number is finite, and has too many digits, or the exponent */
/* could be out of range then we reduce the number under the */
/* appropriate constraints */
if (!(dn->bits & DECSPECIAL))
{ /* not a special value */
Int ae = dn->exponent + dn->digits - 1; /* adjusted exponent */
if (dn->digits > DECIMAL64_Pmax /* too many digits */
|| ae > DECIMAL64_Emax /* likely overflow */
|| ae < DECIMAL64_Emin)
{ /* likely underflow */
decContextDefault (&dc, DEC_INIT_DECIMAL64); /* [no traps] */
dc.round = set->round; /* use supplied rounding */
decNumberPlus (&dw, dn, &dc); /* (round and check) */
/* [this changes -0 to 0, but it will be restored below] */
status |= dc.status; /* save status */
dn = &dw; /* use the work number */
}
/* [this could have pushed number to Infinity or zero, so this */
/* rounding must be done before we generate the decimal64] */
}
DEC_clear (d64); /* clean the target */
if (dn->bits & DECSPECIAL)
{ /* a special value */
uByte top; /* work */
if (dn->bits & DECINF)
top = DECIMAL_Inf;
else
{ /* sNaN or qNaN */
if ((*dn->lsu != 0 || dn->digits > 1) /* non-zero coefficient */
&& (dn->digits < DECIMAL64_Pmax))
{ /* coefficient fits */
decDensePackCoeff (dn, d64->bytes, sizeof (d64->bytes), 0);
}
if (dn->bits & DECNAN)
top = DECIMAL_NaN;
else
top = DECIMAL_sNaN;
}
d64->bytes[0] = top;
}
else if (decNumberIsZero (dn))
{ /* a zero */
/* set and clamp exponent */
if (dn->exponent < -DECIMAL64_Bias)
{
exp = 0;
status |= DEC_Clamped;
}
else
{
exp = dn->exponent + DECIMAL64_Bias; /* bias exponent */
if (exp > DECIMAL64_Ehigh)
{ /* top clamp */
exp = DECIMAL64_Ehigh;
status |= DEC_Clamped;
}
}
comb = (exp >> 5) & 0x18; /* combination field */
d64->bytes[0] = (uByte) (comb << 2);
exp &= 0xff; /* remaining exponent bits */
decimal64SetExpCon (d64, exp);
}
else
{ /* non-zero finite number */
uInt msd; /* work */
/* we have a dn that fits, but it may need to be padded */
exp = (uInt) (dn->exponent + DECIMAL64_Bias); /* bias exponent */
if (exp > DECIMAL64_Ehigh)
{ /* fold-down case */
pad = exp - DECIMAL64_Ehigh;
exp = DECIMAL64_Ehigh; /* [to maximum] */
status |= DEC_Clamped;
}
decDensePackCoeff (dn, d64->bytes, sizeof (d64->bytes), pad);
/* save and clear the top digit */
msd = ((unsigned) d64->bytes[1] >> 2) & 0x0f;
d64->bytes[1] &= 0x03;
/* create the combination field */
if (msd >= 8)
comb = 0x18 | (msd & 0x01) | ((exp >> 7) & 0x06);
else
comb = (msd & 0x07) | ((exp >> 5) & 0x18);
d64->bytes[0] = (uByte) (comb << 2);
exp &= 0xff; /* remaining exponent bits */
decimal64SetExpCon (d64, exp);
}
if (isneg)
decimal64SetSign (d64, 1);
if (status != 0)
decContextSetStatus (set, status); /* pass on status */
/*decimal64Show(d64); */
return d64;
}
/* ------------------------------------------------------------------ */
/* decimal64ToNumber -- convert decimal64 to decNumber */
/* d64 is the source decimal64 */
/* dn is the target number, with appropriate space */
/* No error is possible. */
/* ------------------------------------------------------------------ */
decNumber *
decimal64ToNumber (decimal64 * d64, decNumber * dn)
{
uInt msd; /* coefficient MSD */
decimal64 wk; /* working copy, if needed */
uInt top = d64->bytes[0] & 0x7f; /* top byte, less sign bit */
decNumberZero (dn); /* clean target */
/* set the sign if negative */
if (decimal64Sign (d64))
dn->bits = DECNEG;
if (top >= 0x78)
{ /* is a special */
if ((top & 0x7c) == (DECIMAL_Inf & 0x7c))
dn->bits |= DECINF;
else if ((top & 0x7e) == (DECIMAL_NaN & 0x7e))
dn->bits |= DECNAN;
else
dn->bits |= DECSNAN;
msd = 0; /* no top digit */
}
else
{ /* have a finite number */
uInt comb = top >> 2; /* combination field */
uInt exp; /* exponent */
if (comb >= 0x18)
{
msd = 8 + (comb & 0x01);
exp = (comb & 0x06) << 7; /* MSBs */
}
else
{
msd = comb & 0x07;
exp = (comb & 0x18) << 5;
}
dn->exponent = exp + decimal64ExpCon (d64) - DECIMAL64_Bias; /* remove bias */
}
/* get the coefficient, unless infinite */
if (!(dn->bits & DECINF))
{
Int bunches = DECIMAL64_Pmax / 3; /* coefficient full bunches to convert */
Int odd = 0; /* assume MSD is 0 (no odd bunch) */
if (msd != 0)
{ /* coefficient has leading non-0 digit */
/* make a copy of the decimal64, with an extra bunch which has */
/* the top digit ready for conversion */
wk = *d64; /* take a copy */
wk.bytes[0] = 0; /* clear all but coecon */
wk.bytes[1] &= 0x03; /* .. */
wk.bytes[1] |= (msd << 2); /* and prefix MSD */
odd++; /* indicate the extra */
d64 = &wk; /* use the work copy */
}
decDenseUnpackCoeff (d64->bytes, sizeof (d64->bytes), dn, bunches, odd);
}
return dn;
}
/* ------------------------------------------------------------------ */
/* to-scientific-string -- conversion to numeric string */
/* to-engineering-string -- conversion to numeric string */
/* */
/* decimal64ToString(d64, string); */
/* decimal64ToEngString(d64, string); */
/* */
/* d64 is the decimal64 format number to convert */
/* string is the string where the result will be laid out */
/* */
/* string must be at least 24 characters */
/* */
/* No error is possible, and no status can be set. */
/* ------------------------------------------------------------------ */
char *
decimal64ToString (decimal64 * d64, char *string)
{
decNumber dn; /* work */
decimal64ToNumber (d64, &dn);
decNumberToString (&dn, string);
return string;
}
char *
decimal64ToEngString (decimal64 * d64, char *string)
{
decNumber dn; /* work */
decimal64ToNumber (d64, &dn);
decNumberToEngString (&dn, string);
return string;
}
/* ------------------------------------------------------------------ */
/* to-number -- conversion from numeric string */
/* */
/* decimal64FromString(result, string, set); */
/* */
/* result is the decimal64 format number which gets the result of */
/* the conversion */
/* *string is the character string which should contain a valid */
/* number (which may be a special value) */
/* set is the context */
/* */
/* The context is supplied to this routine is used for error handling */
/* (setting of status and traps) and for the rounding mode, only. */
/* If an error occurs, the result will be a valid decimal64 NaN. */
/* ------------------------------------------------------------------ */
decimal64 *
decimal64FromString (decimal64 * result, char *string, decContext * set)
{
decContext dc; /* work */
decNumber dn; /* .. */
decContextDefault (&dc, DEC_INIT_DECIMAL64); /* no traps, please */
dc.round = set->round; /* use supplied rounding */
decNumberFromString (&dn, string, &dc); /* will round if needed */
decimal64FromNumber (result, &dn, &dc);
if (dc.status != 0)
{ /* something happened */
decContextSetStatus (set, dc.status); /* .. pass it on */
}
return result;
}
#if DECTRACE || DECCHECK
/* ------------------------------------------------------------------ */
/* decimal64Show -- display a single in hexadecimal [debug aid] */
/* d64 -- the number to show */
/* ------------------------------------------------------------------ */
/* Also shows sign/cob/expconfields extracted */
void
decimal64Show (decimal64 * d64)
{
char buf[DECIMAL64_Bytes * 2 + 1];
Int i, j;
j = 0;
for (i = 0; i < DECIMAL64_Bytes; i++)
{
sprintf (&buf[j], "%02x", d64->bytes[i]);
j = j + 2;
}
printf (" D64> %s [S:%d Cb:%02x E:%d]\n", buf,
decimal64Sign (d64), decimal64Comb (d64), decimal64ExpCon (d64));
}
#endif