Update.cvs/libc-ud-970414cvs/libc-970430cvs/libc-970429cvs/libc-970428cvs/libc-970427cvs/libc-970426cvs/libc-970425cvs/libc-970424cvs/libc-970423cvs/libc-970422cvs/libc-970421cvs/libc-970420cvs/libc-970419cvs/libc-970418cvs/libc-970417cvs/libc-970416cvs/libc-970415

1997-04-15 03:14  Ulrich Drepper  <drepper@cygnus.com>

	* wcsmbs/wcsstr.c: Add weak alias wcswcs for Unix98 compliance.

	* math/Makefile (libm-calls): Add s_sincos and s_cproj.
	* math/mathcalls.h: Add prototype for sincos.
	* sysdeps/i386/fpu/__math.h: Warn about restriction on arguments to
	inlined sin and cos function.
	Define functions new in ISO C 9X when __USE_ISOC9X is defined.
	Add optimized version of finite.
	Rewrite sincos function to handle too large arguments.
	* sysdeps/libm-i387/s_cos.S: Better code alignment.
	* sysdeps/libm-i387/s_cosl.S: Likewise.
	* sysdeps/libm-i387/s_sin.S: Likewise.
	* sysdeps/libm-i387/s_sinl.S: Likewise.
	* sysdeps/libm-i387/s_finite.S: Yet better code.
	* sysdeps/libm-i387/s_finitef.S: Likewise.
	* sysdeps/libm-i387/s_sincos.S: New file.
	* sysdeps/libm-i387/s_sincosf.S: New file.
	* sysdeps/libm-i387/s_sincosl.S: New file.
	* sysdeps/libm-ieee754/s_sincos.c: New file.
	* sysdeps/libm-ieee754/s_sincosf.c: New file.
	* sysdeps/libm-ieee754/s_sincosl.c: New file.

	* sysdeps/libm-ieee754/s_ccosh.c: Change to use sincos function.
	* sysdeps/libm-ieee754/s_ccoshf.c: Likewise.
	* sysdeps/libm-ieee754/s_ccoshl.c: Likewise.
	* sysdeps/libm-ieee754/s_cexp.c: Likewise.
	* sysdeps/libm-ieee754/s_cexpf.c: Likewise.
	* sysdeps/libm-ieee754/s_cexpl.c: Likewise.
	* sysdeps/libm-ieee754/s_csinh.c: Likewise.
	* sysdeps/libm-ieee754/s_csinhf.c: Likewise.
	* sysdeps/libm-ieee754/s_csinhl.c: Likewise.
	* sysdeps/libm-ieee754/s_ctan.c: Likewise.
	* sysdeps/libm-ieee754/s_ctanf.c: Likewise.
	* sysdeps/libm-ieee754/s_ctanl.c: Likewise.
	* sysdeps/libm-ieee754/s_ctanh.c: Likewise.
	* sysdeps/libm-ieee754/s_ctanhf.c: Likewise.
	* sysdeps/libm-ieee754/s_ctanhl.c: Likewise.

	* sysdeps/libm-ieee754/s_cosl.c: Correct check for range.
	* sysdeps/libm-ieee754/s_sinl.c: Likewise.

	* sysdeps/libm-ieee754/s_roundtol.c: Fix typos.
	* sysdeps/libm-ieee754/s_roundtoll.c: Likewise.

	* nis/Makefile (services): Mention nisplus before compat.
	* nis/nis_call.c (__do_niscall): Change type of variable server_len
	to prevent warning.
	* nis/nss_compat/compat-grp.c (getgrent_next_file_plusgroup): Add
	cast to prevent warning.
	* nis/nss_compat/compat-pwd.c (getpwent_next_file_plususer): Likewise.
	* nis/nss_compat/compat-spwd.c (getspent_next_file_plususer): Likewise.
	* nis/nss_nisplus/nisplus-hosts.c (_nss_nisplus_parse_hostent): Change
	type of variable `i' to prevent warning.
	* nis/nss_nisplus/nisplus-network.c (_nss_nisplus_parse_netent):
	Likewise.
	* nis/nss_nisplus/nisplus-proto.c (_nss_nisplus_parse_protoent):
	Likewise.
	* nis/nss_nisplus/nisplus-rpc.c (_nss_nisplus_parse_rpcent): Likewise.
	* nis/nss_nisplus/nisplus-service.c (_nss_nisplus_parse_servent):
	Likewise.

	* sysdeps/generic/Dist: Remove old math implementation files.
	* sysdeps/generic/mathimpl.h: Removed.
	* sysdeps/generic/trig.h: Removed.
	* sysdeps/generic/sincos.c: Removed.
	* sysdeps/generic/asincos.c: Removed.
	* sysdeps/generic/exp__E.c: Removed.
	* sysdeps/generic/log__L.c: Removed.

1997-04-13 17:18  Wolfram Gloger  <wmglo@dent.med.uni-muenchen.de>

	* malloc/malloc.c (fREe): Small optimization.  Before returning,
	add fast check whether the heap containing top can go away.

1997-04-11 16:46  Wolfram Gloger  <wmglo@dent.med.uni-muenchen.de>

	* malloc/malloc.c (mALLOc, rEALLOc, cALLOc, mEMALIGn): When
	failing to allocate a new chunk, also try the main_arena.  It may
	be that we are only running out of mmapped regions, and other
	arenas cannot be enlarged sufficiently.
	(new_heap, malloc_extend_top): Handle top_pad more consistently,
	i.e. only in new_heap().

1997-04-13 21:10  Philip Blundell  <pjb27@cl.cam.ac.uk>

	* sysdeps/posix/Makefile: Don't try to build and run mk-stdiolim
	if cross-compiling.

	* configure.in: Recognise "...linuxaout" OS names and turn off ELF.

	* configure.in: Add ARM support.
	* sysdeps/arm/Implies: New file.
	* sysdeps/arm/__longjmp.S: New file.
	* sysdeps/arm/bsd-_setjmp: New file.
	* sysdeps/arm/bsd-setjmp: New file.
	* sysdeps/arm/bytesex.h: New file.
	* sysdeps/arm/fpu_control.h: New file.
	* sysdeps/arm/jmp_buf.h: New file.
	* sysdeps/arm/setjmp.h: New file.
	* sysdeps/arm/sysdep.h: New file.
	* sysdeps/unix/arm/config.h: New file.

1997-04-14 04:03  Ulrich Drepper  <drepper@cygnus.com>

	* sysdeps/libm-ieee754/s_cproj.c: New file.
	* sysdeps/libm-ieee754/s_cprojf.c: New file.
	* sysdeps/libm-ieee754/s_cprojl.c: New file.

7 files changed, 0 insertions, 861 deletions

diff --git a/sysdeps/generic/Dist b/sysdeps/generic/Dist
index 2431f7b272..e82624a61c 100644
--- a/sysdeps/generic/Dist
+++ b/sysdeps/generic/Dist
@@ -1,8 +1,2 @@
 make_siglist.c signame.c signame.h
 det_endian.c
-mathimpl.h
-trig.h
-sincos.c
-asincos.c
-exp__E.c
-log__L.c
diff --git a/sysdeps/generic/asincos.c b/sysdeps/generic/asincos.c
deleted file mode 100644
index c746b1652a..0000000000
--- a/sysdeps/generic/asincos.c
+++ /dev/null
@@ -1,169 +0,0 @@
-/*
- * Copyright (c) 1985, 1993
- *	The Regents of the University of California.  All rights reserved.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions
- * are met:
- * 1. Redistributions of source code must retain the above copyright
- *    notice, this list of conditions and the following disclaimer.
- * 2. Redistributions in binary form must reproduce the above copyright
- *    notice, this list of conditions and the following disclaimer in the
- *    documentation and/or other materials provided with the distribution.
- * 3. All advertising materials mentioning features or use of this software
- *    must display the following acknowledgement:
- *	This product includes software developed by the University of
- *	California, Berkeley and its contributors.
- * 4. Neither the name of the University nor the names of its contributors
- *    may be used to endorse or promote products derived from this software
- *    without specific prior written permission.
- *
- * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
- * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
- * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
- * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
- * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
- * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
- * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
- * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
- * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
- * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
- * SUCH DAMAGE.
- */
-
-#ifndef lint
-static char sccsid[] = "@(#)asincos.c	8.1 (Berkeley) 6/4/93";
-#endif /* not lint */
-
-/* ASIN(X)
- * RETURNS ARC SINE OF X
- * DOUBLE PRECISION (IEEE DOUBLE 53 bits, VAX D FORMAT 56 bits)
- * CODED IN C BY K.C. NG, 4/16/85, REVISED ON 6/10/85.
- *
- * Required system supported functions:
- *	copysign(x,y)
- *	sqrt(x)
- *
- * Required kernel function:
- *	atan2(y,x) 
- *
- * Method :                  
- *	asin(x) = atan2(x,sqrt(1-x*x)); for better accuracy, 1-x*x is 
- *		  computed as follows
- *			1-x*x                     if x <  0.5, 
- *			2*(1-|x|)-(1-|x|)*(1-|x|) if x >= 0.5.
- *
- * Special cases:
- *	if x is NaN, return x itself;
- *	if |x|>1, return NaN.
- *
- * Accuracy:
- * 1)  If atan2() uses machine PI, then
- * 
- *	asin(x) returns (PI/pi) * (the exact arc sine of x) nearly rounded;
- *	and PI is the exact pi rounded to machine precision (see atan2 for
- *      details):
- *
- *	in decimal:
- *		pi = 3.141592653589793 23846264338327 ..... 
- *    53 bits   PI = 3.141592653589793 115997963 ..... ,
- *    56 bits   PI = 3.141592653589793 227020265 ..... ,  
- *
- *	in hexadecimal:
- *		pi = 3.243F6A8885A308D313198A2E....
- *    53 bits   PI = 3.243F6A8885A30  =  2 * 1.921FB54442D18	error=.276ulps
- *    56 bits   PI = 3.243F6A8885A308 =  4 * .C90FDAA22168C2    error=.206ulps
- *	
- *	In a test run with more than 200,000 random arguments on a VAX, the 
- *	maximum observed error in ulps (units in the last place) was
- *	2.06 ulps.      (comparing against (PI/pi)*(exact asin(x)));
- *
- * 2)  If atan2() uses true pi, then
- *
- *	asin(x) returns the exact asin(x) with error below about 2 ulps.
- *
- *	In a test run with more than 1,024,000 random arguments on a VAX, the 
- *	maximum observed error in ulps (units in the last place) was
- *      1.99 ulps.
- */
-
-double asin(x)
-double x;
-{
-	double s,t,copysign(),atan2(),sqrt(),one=1.0;
-#if !defined(vax)&&!defined(tahoe)
-	if(x!=x) return(x);	/* x is NaN */
-#endif	/* !defined(vax)&&!defined(tahoe) */
-	s=copysign(x,one);
-	if(s <= 0.5)
-	    return(atan2(x,sqrt(one-x*x)));
-	else 
-	    { t=one-s; s=t+t; return(atan2(x,sqrt(s-t*t))); }
-
-}
-
-/* ACOS(X)
- * RETURNS ARC COS OF X
- * DOUBLE PRECISION (IEEE DOUBLE 53 bits, VAX D FORMAT 56 bits)
- * CODED IN C BY K.C. NG, 4/16/85, REVISED ON 6/10/85.
- *
- * Required system supported functions:
- *	copysign(x,y)
- *	sqrt(x)
- *
- * Required kernel function:
- *	atan2(y,x) 
- *
- * Method :                  
- *			      ________
- *                           / 1 - x
- *	acos(x) = 2*atan2(  / -------- , 1 ) .
- *                        \/   1 + x
- *
- * Special cases:
- *	if x is NaN, return x itself;
- *	if |x|>1, return NaN.
- *
- * Accuracy:
- * 1)  If atan2() uses machine PI, then
- * 
- *	acos(x) returns (PI/pi) * (the exact arc cosine of x) nearly rounded;
- *	and PI is the exact pi rounded to machine precision (see atan2 for
- *      details):
- *
- *	in decimal:
- *		pi = 3.141592653589793 23846264338327 ..... 
- *    53 bits   PI = 3.141592653589793 115997963 ..... ,
- *    56 bits   PI = 3.141592653589793 227020265 ..... ,  
- *
- *	in hexadecimal:
- *		pi = 3.243F6A8885A308D313198A2E....
- *    53 bits   PI = 3.243F6A8885A30  =  2 * 1.921FB54442D18	error=.276ulps
- *    56 bits   PI = 3.243F6A8885A308 =  4 * .C90FDAA22168C2    error=.206ulps
- *	
- *	In a test run with more than 200,000 random arguments on a VAX, the 
- *	maximum observed error in ulps (units in the last place) was
- *	2.07 ulps.      (comparing against (PI/pi)*(exact acos(x)));
- *
- * 2)  If atan2() uses true pi, then
- *
- *	acos(x) returns the exact acos(x) with error below about 2 ulps.
- *
- *	In a test run with more than 1,024,000 random arguments on a VAX, the 
- *	maximum observed error in ulps (units in the last place) was
- *	2.15 ulps.
- */
-
-double acos(x)
-double x;
-{
-	double t,copysign(),atan2(),sqrt(),one=1.0;
-#if !defined(vax)&&!defined(tahoe)
-	if(x!=x) return(x);
-#endif	/* !defined(vax)&&!defined(tahoe) */
-	if( x != -1.0)
-	    t=atan2(sqrt((one-x)/(one+x)),one);
-	else
-	    t=atan2(one,0.0);	/* t = PI/2 */
-	return(t+t);
-}
diff --git a/sysdeps/generic/exp__E.c b/sysdeps/generic/exp__E.c
deleted file mode 100644
index ab972477a0..0000000000
--- a/sysdeps/generic/exp__E.c
+++ /dev/null
@@ -1,136 +0,0 @@
-/*
- * Copyright (c) 1985, 1993
- *	The Regents of the University of California.  All rights reserved.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions
- * are met:
- * 1. Redistributions of source code must retain the above copyright
- *    notice, this list of conditions and the following disclaimer.
- * 2. Redistributions in binary form must reproduce the above copyright
- *    notice, this list of conditions and the following disclaimer in the
- *    documentation and/or other materials provided with the distribution.
- * 3. All advertising materials mentioning features or use of this software
- *    must display the following acknowledgement:
- *	This product includes software developed by the University of
- *	California, Berkeley and its contributors.
- * 4. Neither the name of the University nor the names of its contributors
- *    may be used to endorse or promote products derived from this software
- *    without specific prior written permission.
- *
- * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
- * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
- * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
- * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
- * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
- * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
- * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
- * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
- * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
- * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
- * SUCH DAMAGE.
- */
-
-#ifndef lint
-static char sccsid[] = "@(#)exp__E.c	8.1 (Berkeley) 6/4/93";
-#endif /* not lint */
-
-/* exp__E(x,c)
- * ASSUMPTION: c << x  SO THAT  fl(x+c)=x.
- * (c is the correction term for x)
- * exp__E RETURNS
- *
- *			 /  exp(x+c) - 1 - x ,  1E-19 < |x| < .3465736
- *       exp__E(x,c) = 	| 		     
- *			 \  0 ,  |x| < 1E-19.
- *
- * DOUBLE PRECISION (IEEE 53 bits, VAX D FORMAT 56 BITS)
- * KERNEL FUNCTION OF EXP, EXPM1, POW FUNCTIONS
- * CODED IN C BY K.C. NG, 1/31/85;
- * REVISED BY K.C. NG on 3/16/85, 4/16/85.
- *
- * Required system supported function:
- *	copysign(x,y)	
- *
- * Method:
- *	1. Rational approximation. Let r=x+c.
- *	   Based on
- *                                   2 * sinh(r/2)     
- *                exp(r) - 1 =   ----------------------   ,
- *                               cosh(r/2) - sinh(r/2)
- *	   exp__E(r) is computed using
- *                   x*x            (x/2)*W - ( Q - ( 2*P  + x*P ) )
- *                   --- + (c + x*[---------------------------------- + c ])
- *                    2                          1 - W
- * 	   where  P := p1*x^2 + p2*x^4,
- *	          Q := q1*x^2 + q2*x^4 (for 56 bits precision, add q3*x^6)
- *	          W := x/2-(Q-x*P),
- *
- *	   (See the listing below for the values of p1,p2,q1,q2,q3. The poly-
- *	    nomials P and Q may be regarded as the approximations to sinh
- *	    and cosh :
- *		sinh(r/2) =  r/2 + r * P  ,  cosh(r/2) =  1 + Q . )
- *
- *         The coefficients were obtained by a special Remez algorithm.
- *
- * Approximation error:
- *
- *   |	exp(x) - 1			   |        2**(-57),  (IEEE double)
- *   | ------------  -  (exp__E(x,0)+x)/x  |  <= 
- *   |	     x			           |	    2**(-69).  (VAX D)
- *
- * Constants:
- * The hexadecimal values are the intended ones for the following constants.
- * The decimal values may be used, provided that the compiler will convert
- * from decimal to binary accurately enough to produce the hexadecimal values
- * shown.
- */
-
-#include "mathimpl.h"
-
-vc(p1, 1.5150724356786683059E-2 ,3abe,3d78,066a,67e1,  -6, .F83ABE67E1066A)
-vc(p2, 6.3112487873718332688E-5 ,5b42,3984,0173,48cd, -13, .845B4248CD0173)
-vc(q1, 1.1363478204690669916E-1 ,b95a,3ee8,ec45,44a2,  -3, .E8B95A44A2EC45)
-vc(q2, 1.2624568129896839182E-3 ,7905,3ba5,f5e7,72e4,  -9, .A5790572E4F5E7)
-vc(q3, 1.5021856115869022674E-6 ,9eb4,36c9,c395,604a, -19, .C99EB4604AC395)
-
-ic(p1, 1.3887401997267371720E-2,  -7, 1.C70FF8B3CC2CF)
-ic(p2, 3.3044019718331897649E-5, -15, 1.15317DF4526C4)
-ic(q1, 1.1110813732786649355E-1,  -4, 1.C719538248597)
-ic(q2, 9.9176615021572857300E-4, -10, 1.03FC4CB8C98E8)
-
-#ifdef vccast
-#define       p1    vccast(p1)
-#define       p2    vccast(p2)
-#define       q1    vccast(q1)
-#define       q2    vccast(q2)
-#define       q3    vccast(q3)
-#endif
-
-double __exp__E(x,c)
-double x,c;
-{
-	const static double zero=0.0, one=1.0, half=1.0/2.0, small=1.0E-19;
-	double z,p,q,xp,xh,w;
-	if(copysign(x,one)>small) {
-           z = x*x  ;
-	   p = z*( p1 +z* p2 );
-#if defined(vax)||defined(tahoe)
-           q = z*( q1 +z*( q2 +z* q3 ));
-#else	/* defined(vax)||defined(tahoe) */
-           q = z*( q1 +z*  q2 );
-#endif	/* defined(vax)||defined(tahoe) */
-           xp= x*p     ; 
-	   xh= x*half  ;
-           w = xh-(q-xp)  ;
-	   p = p+p;
-	   c += x*((xh*w-(q-(p+xp)))/(one-w)+c);
-	   return(z*half+c);
-	}
-	/* end of |x| > small */
-
-	else {
-	    if(x!=zero) one+small;	/* raise the inexact flag */
-	    return(copysign(zero,x));
-	}
-}
diff --git a/sysdeps/generic/log__L.c b/sysdeps/generic/log__L.c
deleted file mode 100644
index c00158fa51..0000000000
--- a/sysdeps/generic/log__L.c
+++ /dev/null
@@ -1,110 +0,0 @@
-/*
- * Copyright (c) 1985, 1993
- *	The Regents of the University of California.  All rights reserved.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions
- * are met:
- * 1. Redistributions of source code must retain the above copyright
- *    notice, this list of conditions and the following disclaimer.
- * 2. Redistributions in binary form must reproduce the above copyright
- *    notice, this list of conditions and the following disclaimer in the
- *    documentation and/or other materials provided with the distribution.
- * 3. All advertising materials mentioning features or use of this software
- *    must display the following acknowledgement:
- *	This product includes software developed by the University of
- *	California, Berkeley and its contributors.
- * 4. Neither the name of the University nor the names of its contributors
- *    may be used to endorse or promote products derived from this software
- *    without specific prior written permission.
- *
- * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
- * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
- * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
- * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
- * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
- * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
- * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
- * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
- * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
- * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
- * SUCH DAMAGE.
- */
-
-#ifndef lint
-static char sccsid[] = "@(#)log__L.c	8.1 (Berkeley) 6/4/93";
-#endif /* not lint */
-
-/* log__L(Z)
- *		LOG(1+X) - 2S			       X
- * RETURN      ---------------  WHERE Z = S*S,  S = ------- , 0 <= Z <= .0294...
- *		      S				     2 + X
- *		     
- * DOUBLE PRECISION (VAX D FORMAT 56 bits or IEEE DOUBLE 53 BITS)
- * KERNEL FUNCTION FOR LOG; TO BE USED IN LOG1P, LOG, AND POW FUNCTIONS
- * CODED IN C BY K.C. NG, 1/19/85; 
- * REVISED BY K.C. Ng, 2/3/85, 4/16/85.
- *
- * Method :
- *	1. Polynomial approximation: let s = x/(2+x). 
- *	   Based on log(1+x) = log(1+s) - log(1-s)
- *		 = 2s + 2/3 s**3 + 2/5 s**5 + .....,
- *
- *	   (log(1+x) - 2s)/s is computed by
- *
- *	       z*(L1 + z*(L2 + z*(... (L7 + z*L8)...)))
- *
- *	   where z=s*s. (See the listing below for Lk's values.) The 
- *	   coefficients are obtained by a special Remez algorithm. 
- *
- * Accuracy:
- *	Assuming no rounding error, the maximum magnitude of the approximation 
- *	error (absolute) is 2**(-58.49) for IEEE double, and 2**(-63.63)
- *	for VAX D format.
- *
- * Constants:
- * The hexadecimal values are the intended ones for the following constants.
- * The decimal values may be used, provided that the compiler will convert
- * from decimal to binary accurately enough to produce the hexadecimal values
- * shown.
- */
-
-#include "mathimpl.h"
-
-vc(L1, 6.6666666666666703212E-1 ,aaaa,402a,aac5,aaaa,  0, .AAAAAAAAAAAAC5)
-vc(L2, 3.9999999999970461961E-1 ,cccc,3fcc,2684,cccc, -1, .CCCCCCCCCC2684)
-vc(L3, 2.8571428579395698188E-1 ,4924,3f92,5782,92f8, -1, .92492492F85782)
-vc(L4, 2.2222221233634724402E-1 ,8e38,3f63,af2c,39b7, -2, .E38E3839B7AF2C)
-vc(L5, 1.8181879517064680057E-1 ,2eb4,3f3a,655e,cc39, -2, .BA2EB4CC39655E)
-vc(L6, 1.5382888777946145467E-1 ,8551,3f1d,781d,e8c5, -2, .9D8551E8C5781D)
-vc(L7, 1.3338356561139403517E-1 ,95b3,3f08,cd92,907f, -2, .8895B3907FCD92)
-vc(L8, 1.2500000000000000000E-1 ,0000,3f00,0000,0000, -2, .80000000000000)
-
-ic(L1, 6.6666666666667340202E-1, -1, 1.5555555555592)
-ic(L2, 3.9999999999416702146E-1, -2, 1.999999997FF24)
-ic(L3, 2.8571428742008753154E-1, -2, 1.24924941E07B4)
-ic(L4, 2.2222198607186277597E-1, -3, 1.C71C52150BEA6)
-ic(L5, 1.8183562745289935658E-1, -3, 1.74663CC94342F)
-ic(L6, 1.5314087275331442206E-1, -3, 1.39A1EC014045B)
-ic(L7, 1.4795612545334174692E-1, -3, 1.2F039F0085122)
-
-#ifdef vccast
-#define	L1	vccast(L1)
-#define	L2	vccast(L2)
-#define	L3	vccast(L3)
-#define	L4	vccast(L4)
-#define	L5	vccast(L5)
-#define	L6	vccast(L6)
-#define	L7	vccast(L7)
-#define	L8	vccast(L8)
-#endif
-
-double __log__L(z)
-double z;
-{
-#if defined(vax)||defined(tahoe)
-    return(z*(L1+z*(L2+z*(L3+z*(L4+z*(L5+z*(L6+z*(L7+z*L8))))))));
-#else	/* defined(vax)||defined(tahoe) */
-    return(z*(L1+z*(L2+z*(L3+z*(L4+z*(L5+z*(L6+z*L7)))))));
-#endif	/* defined(vax)||defined(tahoe) */
-}
diff --git a/sysdeps/generic/mathimpl.h b/sysdeps/generic/mathimpl.h
deleted file mode 100644
index a2c963d288..0000000000
--- a/sysdeps/generic/mathimpl.h
+++ /dev/null
@@ -1,127 +0,0 @@
-/* This part here added by roland@prep.ai.mit.edu for the GNU C library.  */
-   
-#include <math.h>		/* Done first so we can #undef.  */
-#include <endian.h>
-#if	__BYTE_ORDER == __LITTLE_ENDIAN
-#undef national
-#define national
-#endif
-
-#undef	isinf
-#define	isinf	__isinf
-#undef	isnan
-#define	isnan	__isnan
-#undef	infnan
-#define	infnan	__infnan
-#undef	copysign
-#define	copysign	__copysign
-#undef	scalb
-#define	scalb	__scalb
-#undef	drem
-#define	drem	__drem
-#undef	logb
-#define	logb	__logb
-#undef	__finite
-#undef	finite
-#define	finite	__finite
-#undef	expm1
-#define	expm1	__expm1
-
-/*
- * Copyright (c) 1988, 1993
- *	The Regents of the University of California.  All rights reserved.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions
- * are met:
- * 1. Redistributions of source code must retain the above copyright
- *    notice, this list of conditions and the following disclaimer.
- * 2. Redistributions in binary form must reproduce the above copyright
- *    notice, this list of conditions and the following disclaimer in the
- *    documentation and/or other materials provided with the distribution.
- * 3. All advertising materials mentioning features or use of this software
- *    must display the following acknowledgement:
- *	This product includes software developed by the University of
- *	California, Berkeley and its contributors.
- * 4. Neither the name of the University nor the names of its contributors
- *    may be used to endorse or promote products derived from this software
- *    without specific prior written permission.
- *
- * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
- * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
- * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
- * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
- * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
- * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
- * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
- * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
- * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
- * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
- * SUCH DAMAGE.
- *
- *