From 0cbb65479147ecdaa664e88cc2a5a925f3de502f Mon Sep 17 00:00:00 2001 From: nsz Date: Mon, 19 Mar 2012 23:41:19 +0100 Subject: code cleanup of named constants zero, one, two, half are replaced by const literals The policy was to use the f suffix for float consts (1.0f), but don't use suffix for long double consts (these consts can be exactly represented as double). --- src/math/lgammaf_r.c | 40 ++++++++++++++++++---------------------- 1 file changed, 18 insertions(+), 22 deletions(-) (limited to 'src/math/lgammaf_r.c') diff --git a/src/math/lgammaf_r.c b/src/math/lgammaf_r.c index f1adcf69..976986aa 100644 --- a/src/math/lgammaf_r.c +++ b/src/math/lgammaf_r.c @@ -17,8 +17,6 @@ static const float two23= 8.3886080000e+06, /* 0x4b000000 */ -half= 5.0000000000e-01, /* 0x3f000000 */ -one = 1.0000000000e+00, /* 0x3f800000 */ pi = 3.1415927410e+00, /* 0x40490fdb */ a0 = 7.7215664089e-02, /* 0x3d9e233f */ a1 = 3.2246702909e-01, /* 0x3ea51a66 */ @@ -83,8 +81,6 @@ w4 = -5.9518753551e-04, /* 0xba1c065c */ w5 = 8.3633989561e-04, /* 0x3a5b3dd2 */ w6 = -1.6309292987e-03; /* 0xbad5c4e8 */ -static const float zero = 0.0000000000e+00; - static float sin_pif(float x) { float y,z; @@ -109,7 +105,7 @@ static float sin_pif(float x) n = (int)(y*4.0f); } else { if (ix >= 0x4b800000) { - y = zero; /* y must be even */ + y = 0.0f; /* y must be even */ n = 0; } else { if (ix < 0x4b000000) @@ -125,7 +121,7 @@ static float sin_pif(float x) case 1: case 2: y = __cosdf(pi*(0.5f - y)); break; case 3: - case 4: y = __sindf(pi*(one - y)); break; + case 4: y = __sindf(pi*(1.0f - y)); break; case 5: case 6: y = -__cosdf(pi*(y - 1.5f)); break; default: y = __sindf(pi*(y - 2.0f)); break; @@ -148,7 +144,7 @@ float __lgammaf_r(float x, int *signgamp) if (ix >= 0x7f800000) return x*x; if (ix == 0) - return one/zero; + return 1.0f/0.0f; if (ix < 0x35000000) { /* |x| < 2**-21, return -log(|x|) */ if (hx < 0) { *signgamp = -1; @@ -158,12 +154,12 @@ float __lgammaf_r(float x, int *signgamp) } if (hx < 0) { if (ix >= 0x4b000000) /* |x| >= 2**23, must be -integer */ - return one/zero; + return 1.0f/0.0f; t = sin_pif(x); - if (t == zero) /* -integer */ - return one/zero; + if (t == 0.0f) /* -integer */ + return 1.0f/0.0f; nadj = logf(pi/fabsf(t*x)); - if (t < zero) + if (t < 0.0f) *signgamp = -1; x = -x; } @@ -176,17 +172,17 @@ float __lgammaf_r(float x, int *signgamp) if (ix <= 0x3f666666) { /* lgamma(x) = lgamma(x+1)-log(x) */ r = -logf(x); if (ix >= 0x3f3b4a20) { - y = one - x; + y = 1.0f - x; i = 0; } else if (ix >= 0x3e6d3308) { - y = x - (tc-one); + y = x - (tc-1.0f); i = 1; } else { y = x; i = 2; } } else { - r = zero; + r = 0.0f; if (ix >= 0x3fdda618) { /* [1.7316,2] */ y = 2.0f - x; i = 0; @@ -194,7 +190,7 @@ float __lgammaf_r(float x, int *signgamp) y = x - tc; i = 1; } else { - y = x - one; + y = x - 1.0f; i = 2; } } @@ -217,16 +213,16 @@ float __lgammaf_r(float x, int *signgamp) break; case 2: p1 = y*(u0+y*(u1+y*(u2+y*(u3+y*(u4+y*u5))))); - p2 = one+y*(v1+y*(v2+y*(v3+y*(v4+y*v5)))); + p2 = 1.0f+y*(v1+y*(v2+y*(v3+y*(v4+y*v5)))); r += -0.5f*y + p1/p2; } } else if (ix < 0x41000000) { /* x < 8.0 */ i = (int)x; y = x - (float)i; p = y*(s0+y*(s1+y*(s2+y*(s3+y*(s4+y*(s5+y*s6)))))); - q = one+y*(r1+y*(r2+y*(r3+y*(r4+y*(r5+y*r6))))); - r = half*y+p/q; - z = one; /* lgamma(1+s) = log(s) + lgamma(s) */ + q = 1.0f+y*(r1+y*(r2+y*(r3+y*(r4+y*(r5+y*r6))))); + r = 0.5f*y+p/q; + z = 1.0f; /* lgamma(1+s) = log(s) + lgamma(s) */ switch (i) { case 7: z *= y + 6.0f; /* FALLTHRU */ case 6: z *= y + 5.0f; /* FALLTHRU */ @@ -238,12 +234,12 @@ float __lgammaf_r(float x, int *signgamp) } } else if (ix < 0x5c800000) { /* 8.0 <= x < 2**58 */ t = logf(x); - z = one/x; + z = 1.0f/x; y = z*z; w = w0+z*(w1+y*(w2+y*(w3+y*(w4+y*(w5+y*w6))))); - r = (x-half)*(t-one)+w; + r = (x-0.5f)*(t-1.0f)+w; } else /* 2**58 <= x <= inf */ - r = x*(logf(x)-one); + r = x*(logf(x)-1.0f); if (hx < 0) r = nadj - r; return r; -- cgit v1.2.3-70-g09d2