提交 159c7655 编写于 作者: S Szabolcs Nagy

math: cleanup exp2.c exp2f.c and exp2l.c

* old code relied on sign extension on right shift
* exp2l ld64 wrapper was wrong
* use scalbn instead of bithacks
上级 bbbf045c
...@@ -27,11 +27,9 @@ ...@@ -27,11 +27,9 @@
#include "libm.h" #include "libm.h"
#define TBLBITS 8 #define TBLSIZE 256
#define TBLSIZE (1 << TBLBITS)
static const double static const double
huge = 0x1p1000,
redux = 0x1.8p52 / TBLSIZE, redux = 0x1.8p52 / TBLSIZE,
P1 = 0x1.62e42fefa39efp-1, P1 = 0x1.62e42fefa39efp-1,
P2 = 0x1.ebfbdff82c575p-3, P2 = 0x1.ebfbdff82c575p-3,
...@@ -39,8 +37,6 @@ P3 = 0x1.c6b08d704a0a6p-5, ...@@ -39,8 +37,6 @@ P3 = 0x1.c6b08d704a0a6p-5,
P4 = 0x1.3b2ab88f70400p-7, P4 = 0x1.3b2ab88f70400p-7,
P5 = 0x1.5d88003875c74p-10; P5 = 0x1.5d88003875c74p-10;
static const volatile double twom1000 = 0x1p-1000;
static const double tbl[TBLSIZE * 2] = { static const double tbl[TBLSIZE * 2] = {
/* exp2(z + eps) eps */ /* exp2(z + eps) eps */
0x1.6a09e667f3d5dp-1, 0x1.9880p-44, 0x1.6a09e667f3d5dp-1, 0x1.9880p-44,
...@@ -334,25 +330,28 @@ static const double tbl[TBLSIZE * 2] = { ...@@ -334,25 +330,28 @@ static const double tbl[TBLSIZE * 2] = {
*/ */
double exp2(double x) double exp2(double x)
{ {
double r, t, twopk, twopkp1000, z; double r, t, z;
uint32_t hx, ix, lx, i0; uint32_t hx, ix, i0;
int k; union {uint32_t u; int32_t i;} k;
/* Filter out exceptional cases. */ /* Filter out exceptional cases. */
GET_HIGH_WORD(hx, x); GET_HIGH_WORD(hx, x);
ix = hx & 0x7fffffff; ix = hx & 0x7fffffff;
if (ix >= 0x40900000) { /* |x| >= 1024 */ if (ix >= 0x40900000) { /* |x| >= 1024 */
if (ix >= 0x7ff00000) { if (ix >= 0x7ff00000) {
GET_LOW_WORD(lx, x); GET_LOW_WORD(ix, x);
if (((ix & 0xfffff) | lx) != 0 || (hx & 0x80000000) == 0) if (hx == 0xfff00000 && ix == 0) /* -inf */
return x + x; /* x is NaN or +Inf */ return 0;
else return x;
return 0.0; /* x is -Inf */ }
if (x >= 1024) {
STRICT_ASSIGN(double, x, x * 0x1p1023);
return x;
}
if (x <= -1075) {
STRICT_ASSIGN(double, x, 0x1p-1000*0x1p-1000);
return x;
} }
if (x >= 0x1.0p10)
return huge * huge; /* overflow */
if (x <= -0x1.0ccp10)
return twom1000 * twom1000; /* underflow */
} else if (ix < 0x3c900000) { /* |x| < 0x1p-54 */ } else if (ix < 0x3c900000) { /* |x| < 0x1p-54 */
return 1.0 + x; return 1.0 + x;
} }
...@@ -361,24 +360,16 @@ double exp2(double x) ...@@ -361,24 +360,16 @@ double exp2(double x)
STRICT_ASSIGN(double, t, x + redux); STRICT_ASSIGN(double, t, x + redux);
GET_LOW_WORD(i0, t); GET_LOW_WORD(i0, t);
i0 += TBLSIZE / 2; i0 += TBLSIZE / 2;
k = (i0 >> TBLBITS) << 20; k.u = i0 / TBLSIZE * TBLSIZE;
i0 = (i0 & (TBLSIZE - 1)) << 1; k.i /= TBLSIZE;
i0 %= TBLSIZE;
t -= redux; t -= redux;
z = x - t; z = x - t;
/* Compute r = exp2(y) = exp2t[i0] * p(z - eps[i]). */ /* Compute r = exp2(y) = exp2t[i0] * p(z - eps[i]). */
t = tbl[i0]; /* exp2t[i0] */ t = tbl[2*i0]; /* exp2t[i0] */
z -= tbl[i0 + 1]; /* eps[i0] */ z -= tbl[2*i0 + 1]; /* eps[i0] */
if (k >= -1021 << 20)
INSERT_WORDS(twopk, 0x3ff00000 + k, 0);
else
INSERT_WORDS(twopkp1000, 0x3ff00000 + k + (1000 << 20), 0);
r = t + t * z * (P1 + z * (P2 + z * (P3 + z * (P4 + z * P5)))); r = t + t * z * (P1 + z * (P2 + z * (P3 + z * (P4 + z * P5))));
/* Scale by 2**(k>>20). */ return scalbn(r, k.i);
if (k < -1021 << 20)
return r * twopkp1000 * twom1000;
if (k == 1024 << 20)
return r * 2.0 * 0x1p1023;
return r * twopk;
} }
...@@ -27,19 +27,15 @@ ...@@ -27,19 +27,15 @@
#include "libm.h" #include "libm.h"
#define TBLBITS 4 #define TBLSIZE 16
#define TBLSIZE (1 << TBLBITS)
static const float static const float
huge = 0x1p100f,
redux = 0x1.8p23f / TBLSIZE, redux = 0x1.8p23f / TBLSIZE,
P1 = 0x1.62e430p-1f, P1 = 0x1.62e430p-1f,
P2 = 0x1.ebfbe0p-3f, P2 = 0x1.ebfbe0p-3f,
P3 = 0x1.c6b348p-5f, P3 = 0x1.c6b348p-5f,
P4 = 0x1.3b2c9cp-7f; P4 = 0x1.3b2c9cp-7f;
static const volatile float twom100 = 0x1p-100f;
static const double exp2ft[TBLSIZE] = { static const double exp2ft[TBLSIZE] = {
0x1.6a09e667f3bcdp-1, 0x1.6a09e667f3bcdp-1,
0x1.7a11473eb0187p-1, 0x1.7a11473eb0187p-1,
...@@ -89,23 +85,25 @@ float exp2f(float x) ...@@ -89,23 +85,25 @@ float exp2f(float x)
{ {
double tv, twopk, u, z; double tv, twopk, u, z;
float t; float t;
uint32_t hx, ix, i0; uint32_t hx, ix, i0, k;
int32_t k;
/* Filter out exceptional cases. */ /* Filter out exceptional cases. */
GET_FLOAT_WORD(hx, x); GET_FLOAT_WORD(hx, x);
ix = hx & 0x7fffffff; ix = hx & 0x7fffffff;
if (ix >= 0x43000000) { /* |x| >= 128 */ if (ix >= 0x43000000) { /* |x| >= 128 */
if (ix >= 0x7f800000) { if (ix >= 0x7f800000) {
if ((ix & 0x7fffff) != 0 || (hx & 0x80000000) == 0) if (hx == 0xff800000) /* -inf */
return x + x; /* x is NaN or +Inf */ return 0;
else return x;
return 0.0; /* x is -Inf */ }
if (x >= 128) {
STRICT_ASSIGN(float, x, x * 0x1p127);
return x;
}
if (x <= -150) {
STRICT_ASSIGN(float, x, 0x1p-100*0x1p-100);
return x;
} }
if (x >= 0x1.0p7f)
return huge * huge; /* overflow */
if (x <= -0x1.2cp7f)
return twom100 * twom100; /* underflow */
} else if (ix <= 0x33000000) { /* |x| <= 0x1p-25 */ } else if (ix <= 0x33000000) { /* |x| <= 0x1p-25 */
return 1.0f + x; return 1.0f + x;
} }
...@@ -114,7 +112,7 @@ float exp2f(float x) ...@@ -114,7 +112,7 @@ float exp2f(float x)
STRICT_ASSIGN(float, t, x + redux); STRICT_ASSIGN(float, t, x + redux);
GET_FLOAT_WORD(i0, t); GET_FLOAT_WORD(i0, t);
i0 += TBLSIZE / 2; i0 += TBLSIZE / 2;
k = (i0 >> TBLBITS) << 20; k = (i0 / TBLSIZE) << 20;
i0 &= TBLSIZE - 1; i0 &= TBLSIZE - 1;
t -= redux; t -= redux;
z = x - t; z = x - t;
......
...@@ -30,7 +30,7 @@ ...@@ -30,7 +30,7 @@
#if LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024 #if LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024
long double exp2l(long double x) long double exp2l(long double x)
{ {
return exp2l(x); return exp2(x);
} }
#elif LDBL_MANT_DIG == 64 && LDBL_MAX_EXP == 16384 #elif LDBL_MANT_DIG == 64 && LDBL_MAX_EXP == 16384
...@@ -40,10 +40,6 @@ long double exp2l(long double x) ...@@ -40,10 +40,6 @@ long double exp2l(long double x)
#define BIAS (LDBL_MAX_EXP - 1) #define BIAS (LDBL_MAX_EXP - 1)
#define EXPMASK (BIAS + LDBL_MAX_EXP) #define EXPMASK (BIAS + LDBL_MAX_EXP)
static const long double huge = 0x1p10000L;
/* XXX Prevent gcc from erroneously constant folding this. */
static const volatile long double twom10000 = 0x1p-10000L;
static const double static const double
redux = 0x1.8p63 / TBLSIZE, redux = 0x1.8p63 / TBLSIZE,
P1 = 0x1.62e42fefa39efp-1, P1 = 0x1.62e42fefa39efp-1,
...@@ -208,27 +204,28 @@ static const double tbl[TBLSIZE * 2] = { ...@@ -208,27 +204,28 @@ static const double tbl[TBLSIZE * 2] = {
long double exp2l(long double x) long double exp2l(long double x)
{ {
union IEEEl2bits u, v; union IEEEl2bits u, v;
long double r, twopk, twopkp10000, z; long double r, z;
uint32_t hx, ix, i0; uint32_t hx, ix, i0;
int k; union {uint32_t u; int32_t i;} k;
/* Filter out exceptional cases. */ /* Filter out exceptional cases. */
u.e = x; u.e = x;
hx = u.xbits.expsign; hx = u.xbits.expsign;
ix = hx & EXPMASK; ix = hx & EXPMASK;
if (ix >= BIAS + 14) { /* |x| >= 16384 or x is NaN */ if (ix >= BIAS + 14) { /* |x| >= 16384 or x is NaN */
if (ix == BIAS + LDBL_MAX_EXP) { if (ix == EXPMASK) {
if (u.xbits.man != 1ULL << 63 || (hx & 0x8000) == 0) if (u.xbits.man == 1ULL << 63 && hx == 0xffff) /* -inf */
return x + x; /* x is +Inf or NaN */ return 0;
return 0.0; /* x is -Inf */ return x;
}
if (x >= 16384) {
x *= 0x1p16383L;
return x;
} }
if (x >= 16384)
return huge * huge; /* overflow */
if (x <= -16446) if (x <= -16446)
return twom10000 * twom10000; /* underflow */ return 0x1p-10000L*0x1p-10000L;
} else if (ix <= BIAS - 66) { /* |x| < 0x1p-66 */ } else if (ix < BIAS - 64) /* |x| < 0x1p-64 */
return 1.0 + x; return 1 + x;
}
/* /*
* Reduce x, computing z, i0, and k. The low bits of x + redux * Reduce x, computing z, i0, and k. The low bits of x + redux
...@@ -240,38 +237,22 @@ long double exp2l(long double x) ...@@ -240,38 +237,22 @@ long double exp2l(long double x)
* Then the low-order word of x + redux is 0x000abc12, * Then the low-order word of x + redux is 0x000abc12,
* We split this into k = 0xabc and i0 = 0x12 (adjusted to * We split this into k = 0xabc and i0 = 0x12 (adjusted to
* index into the table), then we compute z = 0x0.003456p0. * index into the table), then we compute z = 0x0.003456p0.
*
* XXX If the exponent is negative, the computation of k depends on
* '>>' doing sign extension.
*/ */
u.e = x + redux; u.e = x + redux;
i0 = u.bits.manl + TBLSIZE / 2; i0 = u.bits.manl + TBLSIZE / 2;
k = (int)i0 >> TBLBITS; k.u = i0 / TBLSIZE * TBLSIZE;
i0 = (i0 & (TBLSIZE - 1)) << 1; k.i /= TBLSIZE;
i0 %= TBLSIZE;
u.e -= redux; u.e -= redux;
z = x - u.e; z = x - u.e;
v.xbits.man = 1ULL << 63;
if (k >= LDBL_MIN_EXP) {
v.xbits.expsign = LDBL_MAX_EXP - 1 + k;
twopk = v.e;
} else {
v.xbits.expsign = LDBL_MAX_EXP - 1 + k + 10000;
twopkp10000 = v.e;
}
/* Compute r = exp2l(y) = exp2lt[i0] * p(z). */ /* Compute r = exp2l(y) = exp2lt[i0] * p(z). */
long double t_hi = tbl[i0]; long double t_hi = tbl[2*i0];
long double t_lo = tbl[i0 + 1]; long double t_lo = tbl[2*i0 + 1];
/* XXX This gives > 1 ulp errors outside of FE_TONEAREST mode */ /* XXX This gives > 1 ulp errors outside of FE_TONEAREST mode */
r = t_lo + (t_hi + t_lo) * z * (P1 + z * (P2 + z * (P3 + z * (P4 r = t_lo + (t_hi + t_lo) * z * (P1 + z * (P2 + z * (P3 + z * (P4
+ z * (P5 + z * P6))))) + t_hi; + z * (P5 + z * P6))))) + t_hi;
/* Scale by 2**k. */ return scalbnl(r, k.i);
if (k >= LDBL_MIN_EXP) {
if (k == LDBL_MAX_EXP)
return r * 2.0 * 0x1p16383L;
return r * twopk;
}
return r * twopkp10000 * twom10000;
} }
#endif #endif
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册