1 /* e_rem_pio2f.c -- float version of e_rem_pio2.c 2 * Conversion to float by Ian Lance Taylor, Cygnus Support, ian@cygnus.com. 3 */ 4 5 /* 6 * ==================================================== 7 * Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved. 8 * 9 * Developed at SunPro, a Sun Microsystems, Inc. business. 10 * Permission to use, copy, modify, and distribute this 11 * software is freely granted, provided that this notice 12 * is preserved. 13 * ==================================================== 14 */ 15 16 #if defined(LIBM_SCCS) && !defined(lint) 17 static char rcsid[] = "$NetBSD: e_rem_pio2f.c,v 1.5 1995/05/10 20:46:03 jtc Exp $"; 18 #endif 19 20 /* __ieee754_rem_pio2f(x,y) 21 * 22 * return the remainder of x rem pi/2 in y[0]+y[1] 23 * use __kernel_rem_pio2f() 24 */ 25 26 #include "math.h" 27 #include "math_private.h" 28 29 /* 30 * Table of constants for 2/pi, 396 Hex digits (476 decimal) of 2/pi 31 */ 32 static const int32_t two_over_pi[] = { 33 0xA2, 0xF9, 0x83, 0x6E, 0x4E, 0x44, 0x15, 0x29, 0xFC, 34 0x27, 0x57, 0xD1, 0xF5, 0x34, 0xDD, 0xC0, 0xDB, 0x62, 35 0x95, 0x99, 0x3C, 0x43, 0x90, 0x41, 0xFE, 0x51, 0x63, 36 0xAB, 0xDE, 0xBB, 0xC5, 0x61, 0xB7, 0x24, 0x6E, 0x3A, 37 0x42, 0x4D, 0xD2, 0xE0, 0x06, 0x49, 0x2E, 0xEA, 0x09, 38 0xD1, 0x92, 0x1C, 0xFE, 0x1D, 0xEB, 0x1C, 0xB1, 0x29, 39 0xA7, 0x3E, 0xE8, 0x82, 0x35, 0xF5, 0x2E, 0xBB, 0x44, 40 0x84, 0xE9, 0x9C, 0x70, 0x26, 0xB4, 0x5F, 0x7E, 0x41, 41 0x39, 0x91, 0xD6, 0x39, 0x83, 0x53, 0x39, 0xF4, 0x9C, 42 0x84, 0x5F, 0x8B, 0xBD, 0xF9, 0x28, 0x3B, 0x1F, 0xF8, 43 0x97, 0xFF, 0xDE, 0x05, 0x98, 0x0F, 0xEF, 0x2F, 0x11, 44 0x8B, 0x5A, 0x0A, 0x6D, 0x1F, 0x6D, 0x36, 0x7E, 0xCF, 45 0x27, 0xCB, 0x09, 0xB7, 0x4F, 0x46, 0x3F, 0x66, 0x9E, 46 0x5F, 0xEA, 0x2D, 0x75, 0x27, 0xBA, 0xC7, 0xEB, 0xE5, 47 0xF1, 0x7B, 0x3D, 0x07, 0x39, 0xF7, 0x8A, 0x52, 0x92, 48 0xEA, 0x6B, 0xFB, 0x5F, 0xB1, 0x1F, 0x8D, 0x5D, 0x08, 49 0x56, 0x03, 0x30, 0x46, 0xFC, 0x7B, 0x6B, 0xAB, 0xF0, 50 0xCF, 0xBC, 0x20, 0x9A, 0xF4, 0x36, 0x1D, 0xA9, 0xE3, 51 0x91, 0x61, 0x5E, 0xE6, 0x1B, 0x08, 0x65, 0x99, 0x85, 52 0x5F, 0x14, 0xA0, 0x68, 0x40, 0x8D, 0xFF, 0xD8, 0x80, 53 0x4D, 0x73, 0x27, 0x31, 0x06, 0x06, 0x15, 0x56, 0xCA, 54 0x73, 0xA8, 0xC9, 0x60, 0xE2, 0x7B, 0xC0, 0x8C, 0x6B, 55 }; 56 57 /* This array is like the one in e_rem_pio2.c, but the numbers are 58 single precision and the last 8 bits are forced to 0. */ 59 static const int32_t npio2_hw[] = { 60 0x3fc90f00, 0x40490f00, 0x4096cb00, 0x40c90f00, 0x40fb5300, 0x4116cb00, 61 0x412fed00, 0x41490f00, 0x41623100, 0x417b5300, 0x418a3a00, 0x4196cb00, 62 0x41a35c00, 0x41afed00, 0x41bc7e00, 0x41c90f00, 0x41d5a000, 0x41e23100, 63 0x41eec200, 0x41fb5300, 0x4203f200, 0x420a3a00, 0x42108300, 0x4216cb00, 64 0x421d1400, 0x42235c00, 0x4229a500, 0x422fed00, 0x42363600, 0x423c7e00, 65 0x4242c700, 0x42490f00 66 }; 67 68 /* 69 * invpio2: 24 bits of 2/pi 70 * pio2_1: first 17 bit of pi/2 71 * pio2_1t: pi/2 - pio2_1 72 * pio2_2: second 17 bit of pi/2 73 * pio2_2t: pi/2 - (pio2_1+pio2_2) 74 * pio2_3: third 17 bit of pi/2 75 * pio2_3t: pi/2 - (pio2_1+pio2_2+pio2_3) 76 */ 77 78 static const float 79 zero = 0.0000000000e+00, /* 0x00000000 */ 80 half = 5.0000000000e-01, /* 0x3f000000 */ 81 two8 = 2.5600000000e+02, /* 0x43800000 */ 82 invpio2 = 6.3661980629e-01, /* 0x3f22f984 */ 83 pio2_1 = 1.5707855225e+00, /* 0x3fc90f80 */ 84 pio2_1t = 1.0804334124e-05, /* 0x37354443 */ 85 pio2_2 = 1.0804273188e-05, /* 0x37354400 */ 86 pio2_2t = 6.0770999344e-11, /* 0x2e85a308 */ 87 pio2_3 = 6.0770943833e-11, /* 0x2e85a300 */ 88 pio2_3t = 6.1232342629e-17; /* 0x248d3132 */ 89 90 int32_t 91 __ieee754_rem_pio2f(float x, float *y) 92 { 93 float z,w,t,r,fn; 94 float tx[3]; 95 int32_t e0,i,j,nx,n,ix,hx; 96 97 GET_FLOAT_WORD(hx,x); 98 ix = hx&0x7fffffff; 99 if(ix<=0x3f490fd8) /* |x| ~<= pi/4 , no need for reduction */ 100 {y[0] = x; y[1] = 0; return 0;} 101 if(ix<0x4016cbe4) { /* |x| < 3pi/4, special case with n=+-1 */ 102 if(hx>0) { 103 z = x - pio2_1; 104 if((ix&0xfffffff0)!=0x3fc90fd0) { /* 24+24 bit pi OK */ 105 y[0] = z - pio2_1t; 106 y[1] = (z-y[0])-pio2_1t; 107 } else { /* near pi/2, use 24+24+24 bit pi */ 108 z -= pio2_2; 109 y[0] = z - pio2_2t; 110 y[1] = (z-y[0])-pio2_2t; 111 } 112 return 1; 113 } else { /* negative x */ 114 z = x + pio2_1; 115 if((ix&0xfffffff0)!=0x3fc90fd0) { /* 24+24 bit pi OK */ 116 y[0] = z + pio2_1t; 117 y[1] = (z-y[0])+pio2_1t; 118 } else { /* near pi/2, use 24+24+24 bit pi */ 119 z += pio2_2; 120 y[0] = z + pio2_2t; 121 y[1] = (z-y[0])+pio2_2t; 122 } 123 return -1; 124 } 125 } 126 if(ix<=0x43490f80) { /* |x| ~<= 2^7*(pi/2), medium size */ 127 t = fabsf(x); 128 n = (int32_t) (t*invpio2+half); 129 fn = (float)n; 130 r = t-fn*pio2_1; 131 w = fn*pio2_1t; /* 1st round good to 40 bit */ 132 if(n<32&&(ix&0xffffff00)!=npio2_hw[n-1]) { 133 y[0] = r-w; /* quick check no cancellation */ 134 } else { 135 u_int32_t high; 136 j = ix>>23; 137 y[0] = r-w; 138 GET_FLOAT_WORD(high,y[0]); 139 i = j-((high>>23)&0xff); 140 if(i>8) { /* 2nd iteration needed, good to 57 */ 141 t = r; 142 w = fn*pio2_2; 143 r = t-w; 144 w = fn*pio2_2t-((t-r)-w); 145 y[0] = r-w; 146 GET_FLOAT_WORD(high,y[0]); 147 i = j-((high>>23)&0xff); 148 if(i>25) { /* 3rd iteration need, 74 bits acc */ 149 t = r; /* will cover all possible cases */ 150 w = fn*pio2_3; 151 r = t-w; 152 w = fn*pio2_3t-((t-r)-w); 153 y[0] = r-w; 154 } 155 } 156 } 157 y[1] = (r-y[0])-w; 158 if(hx<0) {y[0] = -y[0]; y[1] = -y[1]; return -n;} 159 else return n; 160 } 161 /* 162 * all other (large) arguments 163 */ 164 if(ix>=0x7f800000) { /* x is inf or NaN */ 165 y[0]=y[1]=x-x; return 0; 166 } 167 /* set z = scalbn(|x|,ilogb(x)-7) */ 168 e0 = (ix>>23)-134; /* e0 = ilogb(z)-7; */ 169 SET_FLOAT_WORD(z, ix - ((int32_t)(e0<<23))); 170 for(i=0;i<2;i++) { 171 tx[i] = (float)((int32_t)(z)); 172 z = (z-tx[i])*two8; 173 } 174 tx[2] = z; 175 nx = 3; 176 while(tx[nx-1]==zero) nx--; /* skip zero term */ 177 n = __kernel_rem_pio2f(tx,y,e0,nx,2,two_over_pi); 178 if(hx<0) {y[0] = -y[0]; y[1] = -y[1]; return -n;} 179 return n; 180 } 181