Use internal exp function instead of wrapper.

This commit is contained in:
Ulrich Drepper 1997-04-11 10:27:30 +00:00
parent 79aa242491
commit 45b0751eae

View File

@ -40,18 +40,24 @@ __complex__ float_type
s(__cexp) (__complex__ float_type x)
{
__complex__ float_type retval;
float_type sin_ix, cos_ix;
if (m81(__finite) (__real__ x))
{
if (m81(__finite) (__imag__ x))
{
float_type exp_val = s(__exp) (__real__ x);
float_type exp_val = m81(__ieee754_exp) (__real__ x);
__asm ("fsincos%.x %2,%1:%0" : "=f" (sin_ix), "=f" (cos_ix)
: "f" (__imag__ x));
__real__ retval = exp_val * cos_ix;
__imag__ retval = exp_val * sin_ix;
__real__ retval = __imag__ retval = exp_val;
if (m81(__finite) (exp_val))
{
float_type sin_ix, cos_ix;
__asm ("fsincos%.x %2,%1:%0" : "=f" (sin_ix), "=f" (cos_ix)
: "f" (__imag__ x));
__real__ retval *= cos_ix;
__imag__ retval *= sin_ix;
}
else
goto fix_sign;
}
else
/* If the imaginary part is +-inf or NaN and the real part is
@ -62,16 +68,41 @@ s(__cexp) (__complex__ float_type x)
{
if (m81(__finite) (__imag__ x))
{
if (m81(__signbit) (__real__ x) == 0 && __imag__ x == 0.0)
retval = huge_val;
float_type value = m81(__signbit) (__real__ x) ? 0.0 : huge_val;
if (__imag__ x == 0.0)
{
__real__ retval = value;
__imag__ retval = __imag__ x;
}
else
{
float_type value = m81(__signbit) (__real__ x) ? 0.0 : huge_val;
float_type remainder, pi_2;
int quadrant;
__real__ retval = value;
__imag__ retval = value;
__asm ("fsincos%.x %2,%1:%0" : "=f" (sin_ix), "=f" (cos_ix)
: "f" (__imag__ x));
__real__ retval = value * cos_ix;
__imag__ retval = value * sin_ix;
fix_sign:
__asm ("fmovecr %#0,%0\n\tfscale%.w %#-1,%0" : "=f" (pi_2));
__asm ("fmod%.x %2,%0\n\tfmove%.l %/fpsr,%1"
: "=f" (remainder), "=dm" (quadrant)
: "f" (pi_2), "0" (__imag__ x));
quadrant = (quadrant >> 16) & 0x83;
if (quadrant & 0x80)
quadrant ^= 0x83;
switch (quadrant)
{
default:
break;
case 1:
__real__ retval = -__real__ retval;
break;
case 2:
__real__ retval = -__real__ retval;
case 3:
__imag__ retval = -__imag__ retval;
break;
}
}
}
else if (m81(__signbit) (__real__ x) == 0)
@ -80,7 +111,10 @@ s(__cexp) (__complex__ float_type x)
__imag__ retval = 0.0/0.0;
}
else
retval = 0.0;
{
__real__ retval = 0.0;
__imag__ retval = s(__copysign) (0.0, __imag__ x);
}
}
else
/* If the real part is NaN the result is NaN + iNaN. */