* [PATCH v3] New generic cosf
@ 2017-12-07 17:07 Paul Clarke
2017-12-07 17:59 ` H.J. Lu
0 siblings, 1 reply; 4+ messages in thread
From: Paul Clarke @ 2017-12-07 17:07 UTC (permalink / raw)
To: libc-alpha, H.J. Lu, Joseph Myers, Matt Turner, Andreas Schwab
The same logic used in s_cosf.S version for x86 and powerpc
is used to create a generic s_cosf.c, so there is no performance
improvement in x86_64 and powerpc64.
2017-12-06 Paul A. Clarke <pc@us.ibm.com>
* sysdeps/ieee754/flt-32/s_cosf.c: New implementation.
---
Changes in v3:
- Instead of changing the "PI_2_{lo|hi}" variable names and associated
comments because they were defined as negative values, I made them
positive and changed their use in the code accordingly.
Changes in v2:
- use integers instead of floats in 2 places
- fixed the performance numbers to "before" and "after"
--
For s390, this is the improvement noted.
Before patch:
"cosf": {
"": {
"duration": 1.00388e+10,
"iterations": 1.34624e+08,
"max": 1655.32,
"min": 4.243,
"mean": 74.5692
}
}
After patch:
"cosf": {
"": {
"duration": 9.90318e+09,
"iterations": 4.9282e+08,
"max": 120.234,
"min": 6.584,
"mean": 20.0949
}
}
Tested on s390x, x86_64 and powerpc64le and powerpc32.
sysdeps/ieee754/flt-32/s_cosf.c | 254 +++++++++++++++++++++++++++++++++-------
1 file changed, 214 insertions(+), 40 deletions(-)
diff --git a/sysdeps/ieee754/flt-32/s_cosf.c b/sysdeps/ieee754/flt-32/s_cosf.c
index 5ed0bca..ac6d044 100644
--- a/sysdeps/ieee754/flt-32/s_cosf.c
+++ b/sysdeps/ieee754/flt-32/s_cosf.c
@@ -1,21 +1,20 @@
-/* s_cosf.c -- float version of s_cos.c.
- * Conversion to float by Ian Lance Taylor, Cygnus Support, ian@cygnus.com.
- */
-
-/*
- * ====================================================
- * Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.
- *
- * Developed at SunPro, a Sun Microsystems, Inc. business.
- * Permission to use, copy, modify, and distribute this
- * software is freely granted, provided that this notice
- * is preserved.
- * ====================================================
- */
-
-#if defined(LIBM_SCCS) && !defined(lint)
-static char rcsid[] = "$NetBSD: s_cosf.c,v 1.4 1995/05/10 20:47:03 jtc Exp $";
-#endif
+/* Compute cosine of argument.
+ Copyright (C) 2017 Free Software Foundation, Inc.
+ This file is part of the GNU C Library.
+
+ The GNU C Library is free software; you can redistribute it and/or
+ modify it under the terms of the GNU Lesser General Public
+ License as published by the Free Software Foundation; either
+ version 2.1 of the License, or (at your option) any later version.
+
+ The GNU C Library is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ Lesser General Public License for more details.
+
+ You should have received a copy of the GNU Lesser General Public
+ License along with the GNU C Library; if not, see
+ <http://www.gnu.org/licenses/>. */
#include <errno.h>
#include <math.h>
@@ -28,35 +27,210 @@ static char rcsid[] = "$NetBSD: s_cosf.c,v 1.4 1995/05/10 20:47:03 jtc Exp $";
# define COSF_FUNC COSF
#endif
-float COSF_FUNC(float x)
+/* Chebyshev constants for cos, range -PI/4 - PI/4. */
+static const double C0 = -0x1.ffffffffe98aep-2;
+static const double C1 = 0x1.55555545c50c7p-5;
+static const double C2 = -0x1.6c16b348b6874p-10;
+static const double C3 = 0x1.a00eb9ac43ccp-16;
+static const double C4 = -0x1.23c97dd8844d7p-22;
+
+/* Chebyshev constants for sin, range -PI/4 - PI/4. */
+static const double S0 = -0x1.5555555551cd9p-3;
+static const double S1 = 0x1.1111110c2688bp-7;
+static const double S2 = -0x1.a019f8b4bd1f9p-13;
+static const double S3 = 0x1.71d7264e6b5b4p-19;
+static const double S4 = -0x1.a947e1674b58ap-26;
+
+/* Chebyshev constants for cos, range 2^-27 - 2^-5. */
+static const double CC0 = -0x1.fffffff5cc6fdp-2;
+static const double CC1 = 0x1.55514b178dac5p-5;
+
+/* PI/2 with 98 bits of accuracy. */
+static const double PI_2_hi = 0x1.921fb544p+0;
+static const double PI_2_lo = 0x1.0b4611a626332p-34;
+
+static const double inv_PI_4 = 0x1.45f306dc9c883p+0; /* 4/PI. */
+
+#define FLOAT_EXPONENT_SHIFT 23
+#define FLOAT_EXPONENT_BIAS 127
+
+static const double pio2_table[] = {
+ 0 * M_PI_2,
+ 1 * M_PI_2,
+ 2 * M_PI_2,
+ 3 * M_PI_2,
+ 4 * M_PI_2,
+ 5 * M_PI_2
+};
+
+static const double invpio4_table[] = {
+ 0x0p+0,
+ 0x1.45f306cp+0,
+ 0x1.c9c882ap-28,
+ 0x1.4fe13a8p-58,
+ 0x1.f47d4dp-85,
+ 0x1.bb81b6cp-112,
+ 0x1.4acc9ep-142,
+ 0x1.0e4107cp-169
+};
+
+static const double ones[] = { 1.0, -1.0 };
+
+/* Compute the cosine value using Chebyshev polynomials where
+ THETA is the range reduced absolute value of the input
+ and it is less than Pi/4,
+ N is calculated as trunc(|x|/(Pi/4)) + 1 and it is used to decide
+ whether a sine or cosine approximation is more accurate and
+ the sign of the result. */
+static inline float
+reduced (double theta, unsigned int n)
{
- float y[2],z=0.0;
- int32_t n,ix;
+ double sign, cx;
+ const double theta2 = theta * theta;
- GET_FLOAT_WORD(ix,x);
+ /* Determine positive or negative primary interval. */
+ n += 2;
+ sign = ones[(n >> 2) & 1];
- /* |x| ~< pi/4 */
- ix &= 0x7fffffff;
- if(ix <= 0x3f490fd8) return __kernel_cosf(x,z);
+ /* Are we in the primary interval of sin or cos? */
+ if ((n & 2) == 0)
+ {
+ /* Here cosf() is calculated using sin Chebyshev polynomial:
+ x+x^3*(S0+x^2*(S1+x^2*(S2+x^2*(S3+x^2*S4)))). */
+ cx = S3 + theta2 * S4;
+ cx = S2 + theta2 * cx;
+ cx = S1 + theta2 * cx;
+ cx = S0 + theta2 * cx;
+ cx = theta + theta * theta2 * cx;
+ }
+ else
+ {
+ /* Here cosf() is calculated using cos Chebyshev polynomial:
+ 1.0+x^2*(C0+x^2*(C1+x^2*(C2+x^2*(C3+x^2*C4)))). */
+ cx = C3 + theta2 * C4;
+ cx = C2 + theta2 * cx;
+ cx = C1 + theta2 * cx;
+ cx = C0 + theta2 * cx;
+ cx = 1. + theta2 * cx;
+ }
+ return sign * cx;
+}
- /* cos(Inf or NaN) is NaN */
- else if (ix>=0x7f800000) {
- if (ix == 0x7f800000)
- __set_errno (EDOM);
- return x-x;
+float
+COSF_FUNC (float x)
+{
+ double theta = x;
+ double abstheta = fabs (theta);
+ if (isless (abstheta, M_PI_4))
+ {
+ double cx;
+ if (abstheta >= 0x1p-5)
+ {
+ const double theta2 = theta * theta;
+ /* Chebyshev polynomial of the form for cos:
+ * 1 + x^2 (C0 + x^2 (C1 + x^2 (C2 + x^2 (C3 + x^2 * C4)))). */
+ cx = C3 + theta2 * C4;
+ cx = C2 + theta2 * cx;
+ cx = C1 + theta2 * cx;
+ cx = C0 + theta2 * cx;
+ cx = 1. + theta2 * cx;
+ return cx;
}
-
- /* argument reduction needed */
- else {
- n = __ieee754_rem_pio2f(x,y);
- switch(n&3) {
- case 0: return __kernel_cosf(y[0],y[1]);
- case 1: return -__kernel_sinf(y[0],y[1],1);
- case 2: return -__kernel_cosf(y[0],y[1]);
- default:
- return __kernel_sinf(y[0],y[1],1);
+ else if (abstheta >= 0x1p-27)
+ {
+ /* A simpler Chebyshev approximation is close enough for this range:
+ * 1 + x^2 (CC0 + x^3 * CC1). */
+ const double theta2 = theta * theta;
+ cx = CC0 + theta * theta2 * CC1;
+ cx = 1.0 + theta2 * cx;
+ return cx;
+ }
+ else
+ {
+ /* For small enough |theta|, this is close enough. */
+ return 1.0 - abstheta;
+ }
+ }
+ else /* |theta| >= Pi/4. */
+ {
+ if (isless (abstheta, 9 * M_PI_4))
+ {
+ /* There are cases where FE_UPWARD rounding mode can
+ produce a result of abstheta * inv_PI_4 == 9,
+ where abstheta < 9pi/4, so the domain for
+ pio2_table must go to 5 (9 / 2 + 1). */
+ unsigned int n = (abstheta * inv_PI_4) + 1;
+ theta = abstheta - pio2_table[n / 2];
+ return reduced (theta, n);
+ }
+ else if (isless (abstheta, INFINITY))
+ {
+ if (abstheta < 0x1p+23)
+ {
+ unsigned int n = ((unsigned int) (abstheta * inv_PI_4)) + 1;
+ double x = n / 2;
+ theta = (abstheta - x * PI_2_hi) - x * PI_2_lo;
+ /* Argument reduction needed. */
+ return reduced (theta, n);
+ }
+ else /* |theta| >= 2^23. */
+ {
+ x = fabsf (x);
+ int exponent;
+ GET_FLOAT_WORD (exponent, x);
+ exponent = (exponent >> FLOAT_EXPONENT_SHIFT)
+ - FLOAT_EXPONENT_BIAS;
+ exponent += 3;
+ exponent /= 28;
+ double a = invpio4_table[exponent] * x;
+ double b = invpio4_table[exponent + 1] * x;
+ double c = invpio4_table[exponent + 2] * x;
+ double d = invpio4_table[exponent + 3] * x;
+ uint64_t l = a;
+ l &= ~0x7;
+ a -= l;
+ double e = a + b;
+ l = e;
+ e = a - l;
+ if (l & 1)
+ {
+ e -= 1.0;
+ e += b;
+ e += c;
+ e += d;
+ e *= M_PI_4;
+ return reduced (e, l + 1);
+ }
+ else
+ {
+ e += b;
+ e += c;
+ e += d;
+ if (e <= 1.0)
+ {
+ e *= M_PI_4;
+ return reduced (e, l + 1);
+ }
+ else
+ {
+ l++;
+ e -= 2.0;
+ e *= M_PI_4;
+ return reduced (e, l + 1);
+ }
+ }
}
}
+ else
+ {
+ int32_t ix;
+ GET_FLOAT_WORD (ix, abstheta);
+ /* cos(Inf or NaN) is NaN. */
+ if (ix == 0x7f800000) /* Inf. */
+ __set_errno (EDOM);
+ return x - x;
+ }
+ }
}
#ifndef COSF
--
1.8.3.1
Regards,
PC
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH v3] New generic cosf
2017-12-07 17:07 [PATCH v3] New generic cosf Paul Clarke
@ 2017-12-07 17:59 ` H.J. Lu
2017-12-07 18:15 ` Paul Clarke
0 siblings, 1 reply; 4+ messages in thread
From: H.J. Lu @ 2017-12-07 17:59 UTC (permalink / raw)
To: Paul Clarke; +Cc: GNU C Library, Joseph Myers, Matt Turner, Andreas Schwab
On Thu, Dec 7, 2017 at 9:07 AM, Paul Clarke <pc@us.ibm.com> wrote:
> The same logic used in s_cosf.S version for x86 and powerpc
> is used to create a generic s_cosf.c, so there is no performance
> improvement in x86_64 and powerpc64.
>
> 2017-12-06 Paul A. Clarke <pc@us.ibm.com>
>
> * sysdeps/ieee754/flt-32/s_cosf.c: New implementation.
Please take a look at sysdeps/ieee754/flt-32/math_config.h to see how
to share constants between s_cosf.c and s_sinf.c.
--
H.J.
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH v3] New generic cosf
2017-12-07 17:59 ` H.J. Lu
@ 2017-12-07 18:15 ` Paul Clarke
2017-12-11 19:49 ` Tulio Magno Quites Machado Filho
0 siblings, 1 reply; 4+ messages in thread
From: Paul Clarke @ 2017-12-07 18:15 UTC (permalink / raw)
To: H.J. Lu; +Cc: GNU C Library, Joseph Myers, Matt Turner, Andreas Schwab
On 12/07/2017 11:59 AM, H.J. Lu wrote:
> On Thu, Dec 7, 2017 at 9:07 AM, Paul Clarke <pc@us.ibm.com> wrote:
>> The same logic used in s_cosf.S version for x86 and powerpc
>> is used to create a generic s_cosf.c, so there is no performance
>> improvement in x86_64 and powerpc64.
>>
>> 2017-12-06 Paul A. Clarke <pc@us.ibm.com>
>>
>> * sysdeps/ieee754/flt-32/s_cosf.c: New implementation.
>
> Please take a look at sysdeps/ieee754/flt-32/math_config.h to see how
> to share constants between s_cosf.c and s_sinf.c.
I interpreted Joseph's response to your earlier suggestion of the same as it to be acceptable to common these tables and constants into a future submission of a combined sinf/cosf/sincosf.
Raji is looking into that approach.
As for cosf, I figured committing a good cosf now is better than waiting for a perfect sinf/cosf/sincosf. Unless there are strong objections?
(FYI, for some reason, I'm not receiving my own posts back to my inbox, so I have been unable to reply to them. This seems like a recent change in behavior.)
PC
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH v3] New generic cosf
2017-12-07 18:15 ` Paul Clarke
@ 2017-12-11 19:49 ` Tulio Magno Quites Machado Filho
0 siblings, 0 replies; 4+ messages in thread
From: Tulio Magno Quites Machado Filho @ 2017-12-11 19:49 UTC (permalink / raw)
To: Paul Clarke, H.J. Lu
Cc: GNU C Library, Joseph Myers, Matt Turner, Andreas Schwab
Paul Clarke <pc@us.ibm.com> writes:
> On 12/07/2017 11:59 AM, H.J. Lu wrote:
>> On Thu, Dec 7, 2017 at 9:07 AM, Paul Clarke <pc@us.ibm.com> wrote:
>>> The same logic used in s_cosf.S version for x86 and powerpc
>>> is used to create a generic s_cosf.c, so there is no performance
>>> improvement in x86_64 and powerpc64.
>>>
>>> 2017-12-06 Paul A. Clarke <pc@us.ibm.com>
>>>
>>> * sysdeps/ieee754/flt-32/s_cosf.c: New implementation.
>>
>> Please take a look at sysdeps/ieee754/flt-32/math_config.h to see how
>> to share constants between s_cosf.c and s_sinf.c.
>
> I interpreted Joseph's response to your earlier suggestion of the same as it to be acceptable to common these tables and constants into a future submission of a combined sinf/cosf/sincosf.
>
> Raji is looking into that approach.
>
> As for cosf, I figured committing a good cosf now is better than waiting for a perfect sinf/cosf/sincosf. Unless there are strong objections?
Pushed as f4b2aea6e133.
--
Tulio Magno
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2017-12-11 19:49 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2017-12-07 17:07 [PATCH v3] New generic cosf Paul Clarke
2017-12-07 17:59 ` H.J. Lu
2017-12-07 18:15 ` Paul Clarke
2017-12-11 19:49 ` Tulio Magno Quites Machado Filho
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).