linux/arch/mips/math-emu/sp_maddf.c

266 lines
6.4 KiB
C

/*
* IEEE754 floating point arithmetic
* single precision: MADDF.f (Fused Multiply Add)
* MADDF.fmt: FPR[fd] = FPR[fd] + (FPR[fs] x FPR[ft])
*
* MIPS floating point support
* Copyright (C) 2015 Imagination Technologies, Ltd.
* Author: Markos Chandras <markos.chandras@imgtec.com>
*
* This program is free software; you can distribute it and/or modify it
* under the terms of the GNU General Public License as published by the
* Free Software Foundation; version 2 of the License.
*/
#include "ieee754sp.h"
static union ieee754sp _sp_maddf(union ieee754sp z, union ieee754sp x,
union ieee754sp y, enum maddf_flags flags)
{
int re;
int rs;
unsigned int rm;
u64 rm64;
u64 zm64;
int s;
COMPXSP;
COMPYSP;
COMPZSP;
EXPLODEXSP;
EXPLODEYSP;
EXPLODEZSP;
FLUSHXSP;
FLUSHYSP;
FLUSHZSP;
ieee754_clearcx();
/*
* Handle the cases when at least one of x, y or z is a NaN.
* Order of precedence is sNaN, qNaN and z, x, y.
*/
if (zc == IEEE754_CLASS_SNAN)
return ieee754sp_nanxcpt(z);
if (xc == IEEE754_CLASS_SNAN)
return ieee754sp_nanxcpt(x);
if (yc == IEEE754_CLASS_SNAN)
return ieee754sp_nanxcpt(y);
if (zc == IEEE754_CLASS_QNAN)
return z;
if (xc == IEEE754_CLASS_QNAN)
return x;
if (yc == IEEE754_CLASS_QNAN)
return y;
if (zc == IEEE754_CLASS_DNORM)
SPDNORMZ;
/* ZERO z cases are handled separately below */
switch (CLPAIR(xc, yc)) {
/*
* Infinity handling
*/
case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_ZERO):
case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_INF):
ieee754_setcx(IEEE754_INVALID_OPERATION);
return ieee754sp_indef();
case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_INF):
case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_INF):
case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_NORM):
case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_DNORM):
case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_INF):
if ((zc == IEEE754_CLASS_INF) &&
((!(flags & MADDF_NEGATE_PRODUCT) && (zs != (xs ^ ys))) ||
((flags & MADDF_NEGATE_PRODUCT) && (zs == (xs ^ ys))))) {
/*
* Cases of addition of infinities with opposite signs
* or subtraction of infinities with same signs.
*/
ieee754_setcx(IEEE754_INVALID_OPERATION);
return ieee754sp_indef();
}
/*
* z is here either not an infinity, or an infinity having the
* same sign as product (x*y) (in case of MADDF.D instruction)
* or product -(x*y) (in MSUBF.D case). The result must be an
* infinity, and its sign is determined only by the value of
* (flags & MADDF_NEGATE_PRODUCT) and the signs of x and y.
*/
if (flags & MADDF_NEGATE_PRODUCT)
return ieee754sp_inf(1 ^ (xs ^ ys));
else
return ieee754sp_inf(xs ^ ys);
case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_ZERO):
case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_NORM):
case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_DNORM):
case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_ZERO):
case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_ZERO):
if (zc == IEEE754_CLASS_INF)
return ieee754sp_inf(zs);
if (zc == IEEE754_CLASS_ZERO) {
/* Handle cases +0 + (-0) and similar ones. */
if ((!(flags & MADDF_NEGATE_PRODUCT)
&& (zs == (xs ^ ys))) ||
((flags & MADDF_NEGATE_PRODUCT)
&& (zs != (xs ^ ys))))
/*
* Cases of addition of zeros of equal signs
* or subtraction of zeroes of opposite signs.
* The sign of the resulting zero is in any
* such case determined only by the sign of z.
*/
return z;
return ieee754sp_zero(ieee754_csr.rm == FPU_CSR_RD);
}
/* x*y is here 0, and z is not 0, so just return z */
return z;
case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_DNORM):
SPDNORMX;
/* fall through */
case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_DNORM):
if (zc == IEEE754_CLASS_INF)
return ieee754sp_inf(zs);
SPDNORMY;
break;
case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_NORM):
if (zc == IEEE754_CLASS_INF)
return ieee754sp_inf(zs);
SPDNORMX;
break;
case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_NORM):
if (zc == IEEE754_CLASS_INF)
return ieee754sp_inf(zs);
/* continue to real computations */
}
/* Finally get to do some computation */
/*
* Do the multiplication bit first
*
* rm = xm * ym, re = xe + ye basically
*
* At this point xm and ym should have been normalized.
*/
/* rm = xm * ym, re = xe+ye basically */
assert(xm & SP_HIDDEN_BIT);
assert(ym & SP_HIDDEN_BIT);
re = xe + ye;
rs = xs ^ ys;
if (flags & MADDF_NEGATE_PRODUCT)
rs ^= 1;
/* Multiple 24 bit xm and ym to give 48 bit results */
rm64 = (uint64_t)xm * ym;
/* Shunt to top of word */
rm64 = rm64 << 16;
/* Put explicit bit at bit 62 if necessary */
if ((int64_t) rm64 < 0) {
rm64 = rm64 >> 1;
re++;
}
assert(rm64 & (1 << 62));
if (zc == IEEE754_CLASS_ZERO) {
/*
* Move explicit bit from bit 62 to bit 26 since the
* ieee754sp_format code expects the mantissa to be
* 27 bits wide (24 + 3 rounding bits).
*/
rm = XSPSRS64(rm64, (62 - 26));
return ieee754sp_format(rs, re, rm);
}
/* Move explicit bit from bit 23 to bit 62 */
zm64 = (uint64_t)zm << (62 - 23);
assert(zm64 & (1 << 62));
/* Make the exponents the same */
if (ze > re) {
/*
* Have to shift r fraction right to align.
*/
s = ze - re;
rm64 = XSPSRS64(rm64, s);
re += s;
} else if (re > ze) {
/*
* Have to shift z fraction right to align.
*/
s = re - ze;
zm64 = XSPSRS64(zm64, s);
ze += s;
}
assert(ze == re);
assert(ze <= SP_EMAX);
/* Do the addition */
if (zs == rs) {
/*
* Generate 64 bit result by adding two 63 bit numbers
* leaving result in zm64, zs and ze.
*/
zm64 = zm64 + rm64;
if ((int64_t)zm64 < 0) { /* carry out */
zm64 = XSPSRS1(zm64);
ze++;
}
} else {
if (zm64 >= rm64) {
zm64 = zm64 - rm64;
} else {
zm64 = rm64 - zm64;
zs = rs;
}
if (zm64 == 0)
return ieee754sp_zero(ieee754_csr.rm == FPU_CSR_RD);
/*
* Put explicit bit at bit 62 if necessary.
*/
while ((zm64 >> 62) == 0) {
zm64 <<= 1;
ze--;
}
}
/*
* Move explicit bit from bit 62 to bit 26 since the
* ieee754sp_format code expects the mantissa to be
* 27 bits wide (24 + 3 rounding bits).
*/
zm = XSPSRS64(zm64, (62 - 26));
return ieee754sp_format(zs, ze, zm);
}
union ieee754sp ieee754sp_maddf(union ieee754sp z, union ieee754sp x,
union ieee754sp y)
{
return _sp_maddf(z, x, y, 0);
}
union ieee754sp ieee754sp_msubf(union ieee754sp z, union ieee754sp x,
union ieee754sp y)
{
return _sp_maddf(z, x, y, MADDF_NEGATE_PRODUCT);
}