1
/* -*- Mode: C; tab-width: 8; indent-tabs-mode: nil; c-basic-offset: 4 -*-
3
* ***** BEGIN LICENSE BLOCK *****
4
* Version: MPL 1.1/GPL 2.0/LGPL 2.1
6
* The contents of this file are subject to the Mozilla Public License Version
7
* 1.1 (the "License"); you may not use this file except in compliance with
8
* the License. You may obtain a copy of the License at
9
* http://www.mozilla.org/MPL/
11
* Software distributed under the License is distributed on an "AS IS" basis,
12
* WITHOUT WARRANTY OF ANY KIND, either express or implied. See the License
13
* for the specific language governing rights and limitations under the
16
* The Original Code is Mozilla Communicator client code, released
19
* The Initial Developer of the Original Code is
20
* Sun Microsystems, Inc.
21
* Portions created by the Initial Developer are Copyright (C) 1998
22
* the Initial Developer. All Rights Reserved.
26
* Alternatively, the contents of this file may be used under the terms of
27
* either of the GNU General Public License Version 2 or later (the "GPL"),
28
* or the GNU Lesser General Public License Version 2.1 or later (the "LGPL"),
29
* in which case the provisions of the GPL or the LGPL are applicable instead
30
* of those above. If you wish to allow use of your version of this file only
31
* under the terms of either the GPL or the LGPL, and not to allow others to
32
* use your version of this file under the terms of the MPL, indicate your
33
* decision by deleting the provisions above and replace them with the notice
34
* and other provisions required by the GPL or the LGPL. If you do not delete
35
* the provisions above, a recipient may use your version of this file under
36
* the terms of any one of the MPL, the GPL or the LGPL.
38
* ***** END LICENSE BLOCK ***** */
39
/* @(#)e_sqrt.c 1.3 95/01/18 */
41
* ====================================================
42
* Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.
44
* Developed at SunSoft, a Sun Microsystems, Inc. business.
45
* Permission to use, copy, modify, and distribute this
46
* software is freely granted, provided that this notice
48
* ====================================================
52
* Return correctly rounded sqrt.
53
* ------------------------------------------
54
* | Use the hardware sqrt if you have one |
55
* ------------------------------------------
57
* Bit by bit method using integer arithmetic. (Slow, but portable)
59
* Scale x to y in [1,4) with even powers of 2:
60
* find an integer k such that 1 <= (y=x*2^(2k)) < 4, then
61
* sqrt(y) = 2^k * sqrt(x)
62
* 2. Bit by bit computation
63
* Let q = sqrt(y) truncated to i bit after binary point (q = 1),
66
* s = 2*q , and y = 2 * ( y - q ). (1)
69
* To compute q from q , one checks whether
76
* If (2) is false, then q = q ; otherwise q = q + 2 .
79
* With some algebric manipulation, it is not difficult to see
80
* that (2) is equivalent to
85
* The advantage of (3) is that s and y can be computed by
87
* the following recurrence formula:
95
* s = s + 2 , y = y - s - 2 (5)
98
* One may easily use induction to prove (4) and (5).
99
* Note. Since the left hand side of (3) contain only i+2 bits,
100
* it does not necessary to do a full (53-bit) comparison
103
* After generating the 53 bits result, we compute one more bit.
104
* Together with the remainder, we can decide whether the
105
* result is exact, bigger than 1/2ulp, or less than 1/2ulp
106
* (it will never equal to 1/2ulp).
107
* The rounding mode can be detected by checking whether
108
* huge + tiny is equal to huge, and whether huge - tiny is
109
* equal to huge for some floating point number "huge" and "tiny".
112
* sqrt(+-0) = +-0 ... exact
114
* sqrt(-ve) = NaN ... with invalid signal
115
* sqrt(NaN) = NaN ... with invalid signal for signaling NaN
117
* Other methods : see the appended file at the end of the program below.
123
#if defined(_MSC_VER)
124
/* Microsoft Compiler */
125
#pragma warning( disable : 4723 ) /* disables potential divide by 0 warning */
129
static const double one = 1.0, tiny=1.0e-300;
131
static double one = 1.0, tiny=1.0e-300;
135
double __ieee754_sqrt(double x)
137
double __ieee754_sqrt(x)
143
int sign = (int)0x80000000;
144
unsigned r,t1,s1,ix1,q1;
148
ix0 = __HI(u); /* high word of x */
149
ix1 = __LO(u); /* low word of x */
151
/* take care of Inf and NaN */
152
if((ix0&0x7ff00000)==0x7ff00000) {
153
return x*x+x; /* sqrt(NaN)=NaN, sqrt(+inf)=+inf
156
/* take care of zero */
158
if(((ix0&(~sign))|ix1)==0) return x;/* sqrt(+-0) = +-0 */
160
return (x-x)/(x-x); /* sqrt(-ve) = sNaN */
164
if(m==0) { /* subnormal x */
167
ix0 |= (ix1>>11); ix1 <<= 21;
169
for(i=0;(ix0&0x00100000)==0;i++) ix0<<=1;
171
ix0 |= (ix1>>(32-i));
174
m -= 1023; /* unbias exponent */
175
ix0 = (ix0&0x000fffff)|0x00100000;
176
if(m&1){ /* odd m, double x to make it even */
177
ix0 += ix0 + ((ix1&sign)>>31);
180
m >>= 1; /* m = [m/2] */
182
/* generate sqrt(x) bit by bit */
183
ix0 += ix0 + ((ix1&sign)>>31);
185
q = q1 = s0 = s1 = 0; /* [q,q1] = sqrt(x) */
186
r = 0x00200000; /* r = moving bit from right to left */
195
ix0 += ix0 + ((ix1&sign)>>31);
204
if((t<ix0)||((t==ix0)&&(t1<=ix1))) {
206
if(((int)(t1&sign)==sign)&&(s1&sign)==0) s0 += 1;
208
if (ix1 < t1) ix0 -= 1;
212
ix0 += ix0 + ((ix1&sign)>>31);
217
/* use floating add to find out rounding direction */
219
z = one-tiny; /* trigger inexact flag */
222
if (q1==(unsigned)0xffffffff) { q1=0; q += 1;}
224
if (q1==(unsigned)0xfffffffe) q+=1;
230
ix0 = (q>>1)+0x3fe00000;
232
if ((q&1)==1) ix1 |= sign;
242
Other methods (use floating-point arithmetic)
244
(This is a copy of a drafted paper by Prof W. Kahan
245
and K.C. Ng, written in May, 1986)
247
Two algorithms are given here to implement sqrt(x)
248
(IEEE double precision arithmetic) in software.
249
Both supply sqrt(x) correctly rounded. The first algorithm (in
250
Section A) uses newton iterations and involves four divisions.
251
The second one uses reciproot iterations to avoid division, but
252
requires more multiplications. Both algorithms need the ability
253
to chop results of arithmetic operations instead of round them,
254
and the INEXACT flag to indicate when an arithmetic operation
255
is executed exactly with no roundoff error, all part of the
256
standard (IEEE 754-1985). The ability to perform shift, add,
257
subtract and logical AND operations upon 32-bit words is needed
258
too, though not part of the standard.
260
A. sqrt(x) by Newton Iteration
262
(1) Initial approximation
264
Let x0 and x1 be the leading and the trailing 32-bit words of
265
a floating point number x (in IEEE double format) respectively
268
------------------------------------------------------
270
------------------------------------------------------
271
msb lsb msb lsb ...order
274
------------------------ ------------------------
275
x0: |s| e | f1 | x1: | f2 |
276
------------------------ ------------------------
278
By performing shifts and subtracts on x0 and x1 (both regarded
279
as integers), we obtain an 8-bit approximation of sqrt(x) as
282
k := (x0>>1) + 0x1ff80000;
283
y0 := k - T1[31&(k>>15)]. ... y ~ sqrt(x) to 8 bits
284
Here k is a 32-bit integer and T1[] is an integer array containing
285
correction terms. Now magically the floating value of y (y's
286
leading 32-bit word is y0, the value of its trailing word is 0)
287
approximates sqrt(x) to almost 8-bit.
291
0, 1024, 3062, 5746, 9193, 13348, 18162, 23592,
292
29598, 36145, 43202, 50740, 58733, 67158, 75992, 85215,
293
83599, 71378, 60428, 50647, 41945, 34246, 27478, 21581,
294
16499, 12183, 8588, 5674, 3403, 1742, 661, 130,};
296
(2) Iterative refinement
298
Apply Heron's rule three times to y, we have y approximates
299
sqrt(x) to within 1 ulp (Unit in the Last Place):
301
y := (y+x/y)/2 ... almost 17 sig. bits
302
y := (y+x/y)/2 ... almost 35 sig. bits
303
y := y-(y-x/y)/2 ... within 1 ulp
307
Another way to improve y to within 1 ulp is:
309
y := (y+x/y) ... almost 17 sig. bits to 2*sqrt(x)
310
y := y - 0x00100006 ... almost 18 sig. bits to sqrt(x)
314
y := y + 2* ---------- ...within 1 ulp
319
This formula has one division fewer than the one above; however,
320
it requires more multiplications and additions. Also x must be
321
scaled in advance to avoid spurious overflow in evaluating the
322
expression 3y*y+x. Hence it is not recommended uless division
323
is slow. If division is very slow, then one should use the
324
reciproot algorithm given in section B.
328
By twiddling y's last bit it is possible to force y to be
329
correctly rounded according to the prevailing rounding mode
330
as follows. Let r and i be copies of the rounding mode and
331
inexact flag before entering the square root program. Also we
332
use the expression y+-ulp for the next representable floating
333
numbers (up and down) of y. Note that y+-ulp = either fixed
334
point y+-1, or multiply y by nextafter(1,+-inf) in chopped
337
I := FALSE; ... reset INEXACT flag I
338
R := RZ; ... set rounding mode to round-toward-zero
339
z := x/y; ... chopped quotient, possibly inexact
340
If(not I) then { ... if the quotient is exact
342
I := i; ... restore inexact flag
343
R := r; ... restore rounded mode
346
z := z - ulp; ... special rounding
349
i := TRUE; ... sqrt(x) is inexact
350
If (r=RN) then z=z+ulp ... rounded-to-nearest
351
If (r=RP) then { ... round-toward-+inf
354
y := y+z; ... chopped sum
355
y0:=y0-0x00100000; ... y := y/2 is correctly rounded.
356
I := i; ... restore inexact flag
357
R := r; ... restore rounded mode
362
Square root of +inf, +-0, or NaN is itself;
363
Square root of a negative number is NaN with invalid signal.
366
B. sqrt(x) by Reciproot Iteration
368
(1) Initial approximation
370
Let x0 and x1 be the leading and the trailing 32-bit words of
371
a floating point number x (in IEEE double format) respectively
372
(see section A). By performing shifs and subtracts on x0 and y0,
373
we obtain a 7.8-bit approximation of 1/sqrt(x) as follows.
375
k := 0x5fe80000 - (x0>>1);
376
y0:= k - T2[63&(k>>14)]. ... y ~ 1/sqrt(x) to 7.8 bits
378
Here k is a 32-bit integer and T2[] is an integer array
379
containing correction terms. Now magically the floating
380
value of y (y's leading 32-bit word is y0, the value of
381
its trailing word y1 is set to zero) approximates 1/sqrt(x)
386
0x1500, 0x2ef8, 0x4d67, 0x6b02, 0x87be, 0xa395, 0xbe7a, 0xd866,
387
0xf14a, 0x1091b,0x11fcd,0x13552,0x14999,0x15c98,0x16e34,0x17e5f,
388
0x18d03,0x19a01,0x1a545,0x1ae8a,0x1b5c4,0x1bb01,0x1bfde,0x1c28d,
389
0x1c2de,0x1c0db,0x1ba73,0x1b11c,0x1a4b5,0x1953d,0x18266,0x16be0,
390
0x1683e,0x179d8,0x18a4d,0x19992,0x1a789,0x1b445,0x1bf61,0x1c989,
391
0x1d16d,0x1d77b,0x1dddf,0x1e2ad,0x1e5bf,0x1e6e8,0x1e654,0x1e3cd,
392
0x1df2a,0x1d635,0x1cb16,0x1be2c,0x1ae4e,0x19bde,0x1868e,0x16e2e,
393
0x1527f,0x1334a,0x11051,0xe951, 0xbe01, 0x8e0d, 0x5924, 0x1edd,};
395
(2) Iterative refinement
397
Apply Reciproot iteration three times to y and multiply the
398
result by x to get an approximation z that matches sqrt(x)
399
to about 1 ulp. To be exact, we will have
400
-1ulp < sqrt(x)-z<1.0625ulp.
402
... set rounding mode to Round-to-nearest
403
y := y*(1.5-0.5*x*y*y) ... almost 15 sig. bits to 1/sqrt(x)
404
y := y*((1.5-2^-30)+0.5*x*y*y)... about 29 sig. bits to 1/sqrt(x)
405
... special arrangement for better accuracy
406
z := x*y ... 29 bits to sqrt(x), with z*y<1
407
z := z + 0.5*z*(1-z*y) ... about 1 ulp to sqrt(x)
409
Remark 2. The constant 1.5-2^-30 is chosen to bias the error so that
410
(a) the term z*y in the final iteration is always less than 1;
411
(b) the error in the final result is biased upward so that
412
-1 ulp < sqrt(x) - z < 1.0625 ulp
413
instead of |sqrt(x)-z|<1.03125ulp.
417
By twiddling y's last bit it is possible to force y to be
418
correctly rounded according to the prevailing rounding mode
419
as follows. Let r and i be copies of the rounding mode and
420
inexact flag before entering the square root program. Also we
421
use the expression y+-ulp for the next representable floating
422
numbers (up and down) of y. Note that y+-ulp = either fixed
423
point y+-1, or multiply y by nextafter(1,+-inf) in chopped
426
R := RZ; ... set rounding mode to round-toward-zero
428
case RN: ... round-to-nearest
429
if(x<= z*(z-ulp)...chopped) z = z - ulp; else
430
if(x<= z*(z+ulp)...chopped) z = z; else z = z+ulp;
432
case RZ:case RM: ... round-to-zero or round-to--inf
433
R:=RP; ... reset rounding mod to round-to-+inf
434
if(x<z*z ... rounded up) z = z - ulp; else
435
if(x>=(z+ulp)*(z+ulp) ...rounded up) z = z+ulp;
437
case RP: ... round-to-+inf
438
if(x>(z+ulp)*(z+ulp)...chopped) z = z+2*ulp; else
439
if(x>z*z ...chopped) z = z+ulp;
443
Remark 3. The above comparisons can be done in fixed point. For
444
example, to compare x and w=z*z chopped, it suffices to compare
445
x1 and w1 (the trailing parts of x and w), regarding them as
446
two's complement integers.
448
...Is z an exact square root?
449
To determine whether z is an exact square root of x, let z1 be the
450
trailing part of z, and also let x0 and x1 be the leading and
453
If ((z1&0x03ffffff)!=0) ... not exact if trailing 26 bits of z!=0
454
I := 1; ... Raise Inexact flag: z is not exact
456
j := 1 - [(x0>>20)&1] ... j = logb(x) mod 2
457
k := z1 >> 26; ... get z's 25-th and 26-th
459
I := i or (k&j) or ((k&(j+j+1))!=(x1&3));
461
R:= r ... restore rounded mode
464
If multiplication is cheaper then the foregoing red tape, the
465
Inexact flag can be evaluated by
470
Note that z*z can overwrite I; this value must be sensed if it is
473
Remark 4. If z*z = x exactly, then bit 25 to bit 0 of z1 must be
481
Further more, bit 27 and 26 of z1, bit 0 and 1 of x1, and the odd
482
or even of logb(x) have the following relations:
484
-------------------------------------------------
485
bit 27,26 of z1 bit 1,0 of x1 logb(x)
486
-------------------------------------------------
492
-------------------------------------------------
494
(4) Special cases (see (4) of Section A).