softfloat.c

說明FPU的演算法以及浮點運算式中的加減乘除

                }
                else {
                    increment = ( roundingMode == float_round_up ) && zSig1;
                }
            }
            if ( increment ) {
                ++zSig0;
                zSig0 &=
                    ~ ( ( (bits64) ( zSig1<<1 ) == 0 ) & roundNearestEven );
                if ( (sbits64) zSig0 < 0 ) zExp = 1;
            }
            return packFloatx80( zSign, zExp, zSig0 );
        }
    }
    if ( zSig1 ) float_exception_flags |= float_flag_inexact;
    if ( increment ) {
        ++zSig0;
        if ( zSig0 == 0 ) {
            ++zExp;
            zSig0 = LIT64( 0x8000000000000000 );
        }
        else {
            zSig0 &= ~ ( ( (bits64) ( zSig1<<1 ) == 0 ) & roundNearestEven );
        }
    }
    else {
        if ( zSig0 == 0 ) zExp = 0;
    }
    return packFloatx80( zSign, zExp, zSig0 );

}

/*----------------------------------------------------------------------------
| Takes an abstract floating-point value having sign `zSign', exponent
| `zExp', and significand formed by the concatenation of `zSig0' and `zSig1',
| and returns the proper extended double-precision floating-point value
| corresponding to the abstract input.  This routine is just like
| `roundAndPackFloatx80' except that the input significand does not have to be
| normalized.
*----------------------------------------------------------------------------*/

static floatx80
 normalizeRoundAndPackFloatx80(
     int8 roundingPrecision, flag zSign, int32 zExp, bits64 zSig0, bits64 zSig1
 )
{
    int8 shiftCount;

    if ( zSig0 == 0 ) {
        zSig0 = zSig1;
        zSig1 = 0;
        zExp -= 64;
    }
    shiftCount = countLeadingZeros64( zSig0 );
    shortShift128Left( zSig0, zSig1, shiftCount, &zSig0, &zSig1 );
    zExp -= shiftCount;
    return
        roundAndPackFloatx80( roundingPrecision, zSign, zExp, zSig0, zSig1 );

}

#endif

#ifdef FLOAT128

/*----------------------------------------------------------------------------
| Returns the least-significant 64 fraction bits of the quadruple-precision
| floating-point value `a'.
*----------------------------------------------------------------------------*/

INLINE bits64 extractFloat128Frac1( float128 a )
{

    return a.low;

}

/*----------------------------------------------------------------------------
| Returns the most-significant 48 fraction bits of the quadruple-precision
| floating-point value `a'.
*----------------------------------------------------------------------------*/

INLINE bits64 extractFloat128Frac0( float128 a )
{

    return a.high & LIT64( 0x0000FFFFFFFFFFFF );

}

/*----------------------------------------------------------------------------
| Returns the exponent bits of the quadruple-precision floating-point value
| `a'.
*----------------------------------------------------------------------------*/

INLINE int32 extractFloat128Exp( float128 a )
{

    return ( a.high>>48 ) & 0x7FFF;

}

/*----------------------------------------------------------------------------
| Returns the sign bit of the quadruple-precision floating-point value `a'.
*----------------------------------------------------------------------------*/

INLINE flag extractFloat128Sign( float128 a )
{

    return a.high>>63;

}

/*----------------------------------------------------------------------------
| Normalizes the subnormal quadruple-precision floating-point value
| represented by the denormalized significand formed by the concatenation of
| `aSig0' and `aSig1'.  The normalized exponent is stored at the location
| pointed to by `zExpPtr'.  The most significant 49 bits of the normalized
| significand are stored at the location pointed to by `zSig0Ptr', and the
| least significant 64 bits of the normalized significand are stored at the
| location pointed to by `zSig1Ptr'.
*----------------------------------------------------------------------------*/

static void
 normalizeFloat128Subnormal(
     bits64 aSig0,
     bits64 aSig1,
     int32 *zExpPtr,
     bits64 *zSig0Ptr,
     bits64 *zSig1Ptr
 )
{
    int8 shiftCount;

    if ( aSig0 == 0 ) {
        shiftCount = countLeadingZeros64( aSig1 ) - 15;
        if ( shiftCount < 0 ) {
            *zSig0Ptr = aSig1>>( - shiftCount );
            *zSig1Ptr = aSig1<<( shiftCount & 63 );
        }
        else {
            *zSig0Ptr = aSig1<<shiftCount;
            *zSig1Ptr = 0;
        }
        *zExpPtr = - shiftCount - 63;
    }
    else {
        shiftCount = countLeadingZeros64( aSig0 ) - 15;
        shortShift128Left( aSig0, aSig1, shiftCount, zSig0Ptr, zSig1Ptr );
        *zExpPtr = 1 - shiftCount;
    }

}

/*----------------------------------------------------------------------------
| Packs the sign `zSign', the exponent `zExp', and the significand formed
| by the concatenation of `zSig0' and `zSig1' into a quadruple-precision
| floating-point value, returning the result.  After being shifted into the
| proper positions, the three fields `zSign', `zExp', and `zSig0' are simply
| added together to form the most significant 32 bits of the result.  This
| means that any integer portion of `zSig0' will be added into the exponent.
| Since a properly normalized significand will have an integer portion equal
| to 1, the `zExp' input should be 1 less than the desired result exponent
| whenever `zSig0' and `zSig1' concatenated form a complete, normalized
| significand.
*----------------------------------------------------------------------------*/

INLINE float128
 packFloat128( flag zSign, int32 zExp, bits64 zSig0, bits64 zSig1 )
{
    float128 z;

    z.low = zSig1;
    z.high = ( ( (bits64) zSign )<<63 ) + ( ( (bits64) zExp )<<48 ) + zSig0;
    return z;

}

/*----------------------------------------------------------------------------
| Takes an abstract floating-point value having sign `zSign', exponent `zExp',
| and extended significand formed by the concatenation of `zSig0', `zSig1',
| and `zSig2', and returns the proper quadruple-precision floating-point value
| corresponding to the abstract input.  Ordinarily, the abstract value is
| simply rounded and packed into the quadruple-precision format, with the
| inexact exception raised if the abstract input cannot be represented
| exactly.  However, if the abstract value is too large, the overflow and
| inexact exceptions are raised and an infinity or maximal finite value is
| returned.  If the abstract value is too small, the input value is rounded to
| a subnormal number, and the underflow and inexact exceptions are raised if
| the abstract input cannot be represented exactly as a subnormal quadruple-
| precision floating-point number.
|     The input significand must be normalized or smaller.  If the input
| significand is not normalized, `zExp' must be 0; in that case, the result
| returned is a subnormal number, and it must not require rounding.  In the
| usual case that the input significand is normalized, `zExp' must be 1 less
| than the ``true'' floating-point exponent.  The handling of underflow and
| overflow follows the IEC/IEEE Standard for Binary Floating-Point Arithmetic.
*----------------------------------------------------------------------------*/

static float128
 roundAndPackFloat128(
     flag zSign, int32 zExp, bits64 zSig0, bits64 zSig1, bits64 zSig2 )
{
    int8 roundingMode;
    flag roundNearestEven, increment, isTiny;

    roundingMode = float_rounding_mode;
    roundNearestEven = ( roundingMode == float_round_nearest_even );
    increment = ( (sbits64) zSig2 < 0 );
    if ( ! roundNearestEven ) {
        if ( roundingMode == float_round_to_zero ) {
            increment = 0;
        }
        else {
            if ( zSign ) {
                increment = ( roundingMode == float_round_down ) && zSig2;
            }
            else {
                increment = ( roundingMode == float_round_up ) && zSig2;
            }
        }
    }
    if ( 0x7FFD <= (bits32) zExp ) {
        if (    ( 0x7FFD < zExp )
             || (    ( zExp == 0x7FFD )
                  && eq128(
                         LIT64( 0x0001FFFFFFFFFFFF ),
                         LIT64( 0xFFFFFFFFFFFFFFFF ),
                         zSig0,
                         zSig1
                     )
                  && increment
                )
           ) {
            float_raise( float_flag_overflow | float_flag_inexact );
            if (    ( roundingMode == float_round_to_zero )
                 || ( zSign && ( roundingMode == float_round_up ) )
                 || ( ! zSign && ( roundingMode == float_round_down ) )
               ) {
                return
                    packFloat128(
                        zSign,
                        0x7FFE,
                        LIT64( 0x0000FFFFFFFFFFFF ),
                        LIT64( 0xFFFFFFFFFFFFFFFF )
                    );
            }
            return packFloat128( zSign, 0x7FFF, 0, 0 );
        }
        if ( zExp < 0 ) {
            isTiny =
                   ( float_detect_tininess == float_tininess_before_rounding )
                || ( zExp < -1 )
                || ! increment
                || lt128(
                       zSig0,
                       zSig1,
                       LIT64( 0x0001FFFFFFFFFFFF ),
                       LIT64( 0xFFFFFFFFFFFFFFFF )
                   );
            shift128ExtraRightJamming(
                zSig0, zSig1, zSig2, - zExp, &zSig0, &zSig1, &zSig2 );
            zExp = 0;
            if ( isTiny && zSig2 ) float_raise( float_flag_underflow );
            if ( roundNearestEven ) {
                increment = ( (sbits64) zSig2 < 0 );
            }
            else {
                if ( zSign ) {
                    increment = ( roundingMode == float_round_down ) && zSig2;
                }
                else {
                    increment = ( roundingMode == float_round_up ) && zSig2;
                }
            }
        }
    }
    if ( zSig2 ) float_exception_flags |= float_flag_inexact;
    if ( increment ) {
        add128( zSig0, zSig1, 0, 1, &zSig0, &zSig1 );
        zSig1 &= ~ ( ( zSig2 + zSig2 == 0 ) & roundNearestEven );
    }
    else {
        if ( ( zSig0 | zSig1 ) == 0 ) zExp = 0;
    }
    return packFloat128( zSign, zExp, zSig0, zSig1 );

}

/*----------------------------------------------------------------------------
| Takes an abstract floating-point value having sign `zSign', exponent `zExp',
| and significand formed by the concatenation of `zSig0' and `zSig1', and
| returns the proper quadruple-precision floating-point value corresponding
| to the abstract input.  This routine is just like `roundAndPackFloat128'
| except that the input significand has fewer bits and does not have to be
| normalized.  In all cases, `zExp' must be 1 less than the ``true'' floating-
| point exponent.
*----------------------------------------------------------------------------*/

static float128
 normalizeRoundAndPackFloat128(
     flag zSign, int32 zExp, bits64 zSig0, bits64 zSig1 )
{
    int8 shiftCount;
    bits64 zSig2;

    if ( zSig0 == 0 ) {
        zSig0 = zSig1;
        zSig1 = 0;
        zExp -= 64;
    }
    shiftCount = countLeadingZeros64( zSig0 ) - 15;
    if ( 0 <= shiftCount ) {
        zSig2 = 0;
        shortShift128Left( zSig0, zSig1, shiftCount, &zSig0, &zSig1 );
    }
    else {
        shift128ExtraRightJamming(
            zSig0, zSig1, 0, - shiftCount, &zSig0, &zSig1, &zSig2 );
    }
    zExp -= shiftCount;
    return roundAndPackFloat128( zSign, zExp, zSig0, zSig1, zSig2 );

}

#endif

/*----------------------------------------------------------------------------
| Returns the result of converting the 32-bit two's complement integer `a'
| to the single-precision floating-point format.  The conversion is performed
| according to the IEC/IEEE Standard for Binary Floating-Point Arithmetic.
*----------------------------------------------------------------------------*/

float32 int32_to_float32( int32 a )
{
    flag zSign;

    if ( a == 0 ) return 0;
    if ( a == (sbits32) 0x80000000 ) return packFloat32( 1, 0x9E, 0 );
    zSign = ( a < 0 );
    return normalizeRoundAndPackFloat32( zSign, 0x9C, zSign ? - a : a );

}

/*----------------------------------------------------------------------------
| Returns the result of converting the 32-bit two's complement integer `a'
| to the double-precision floating-point format.  The conversion is performed
| according to the IEC/IEEE Standard for Binary Floating-Point Arithmetic.
*----------------------------------------------------------------------------*/

float64 int32_to_float64( int32 a )
{
    flag zSign;
    uint32 absA;
    int8 shiftCount;
    bits64 zSig;