8:55 p.m.
"Stephen R. Anderson" <sra(a)bloch.ling.yale.edu> writes:
bacon(a)aerodyne.com said:
> I was looking at this very same problem last night. It turns out
> that the prototype for matherr has changed slightly. You can get
> xemacs to build without any problem by simply changing the variable
> on line 934 in floatfns.c from exception to __exception. Everything
> else goes smoothly after that. Apparently there used to be a typedef
> for exception which has disappeared from the include files.
This implies, unfortunately, that all three of the potential
culprits you suggest are involved in one way or another. I will
leave it to those who are more familiar with the way these things
interact in xemacs code to suggest a patch.
Fred's answer doesn't satisfy me. Correct user code should not be
forced to use __exception -- __foo looks like a system-dependent thing
to me.
--
Hrvoje Niksic <hniksic(a)srce.hr> | Student at FER Zagreb, Croatia
--------------------------------+--------------------------------
"Manic depression is cool... your body can make its own drugs."
8:11 p.m.
New subject: [FAILURE] XEmacs 21.0-beta38 on MkLinux
hniksic(a)srce.hr said:
Fred's answer doesn't satisfy me. Correct user code should
not be
forced to use __exception -- __foo looks like a system-dependent
thing to me.
Don't blame Fred...at least he found OUR problem. He didn't claim to have
found the correct solution from the larger XEmacs point of view. As he says in
a later note to me,
bacon(a)aerodyne.com said:
I really don't know who's at fault. Xemacs may not be to
blame. We
may have a nonstandard version of math.h in the most recent glibc
distribution. Errors do creep into these things. You are free to
pass along the information, but you might also want to find out if
this change to math.h was intentional or accidental.
I enclose a copy of the most recent /usr/include/math.h, from
ftp://ftp.linuxppc.org/linuxppc/redhat/RPMS/ppc/glibc-devel-0.961212-1f.p...
--Steve Anderson
/* Declarations for math functions.
Copyright (C) 1991, 92, 93, 95, 96 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 Library General Public License as
published by the Free Software Foundation; either version 2 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
Library General Public License for more details.
You should have received a copy of the GNU Library General Public
License along with the GNU C Library; see the file COPYING.LIB. If not,
write to the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
Boston, MA 02111-1307, USA. */
/*
* ISO C Standard: 4.5 MATHEMATICS <math.h>
*/
#ifndef _MATH_H
#define _MATH_H 1
#include <features.h>
__BEGIN_DECLS
#define __need_Emath
#include <errno.h>
/* Get machine-dependent HUGE_VAL value (returned on overflow).
On all IEEE754 machines, this is +Infinity. */
#include <huge_val.h>
/* Get machine-dependent NAN value (returned for some domain errors). */
#ifdef __USE_GNU
#include <nan.h>
#endif
/* The file <mathcalls.h> contains the prototypes for all the actual
math functions. These macros are used for those prototypes, so
we can easily declare each function as both `name' and `__name',
and can declare the float versions `namef' and `__namef'. */
#define __MATHCALL(function,suffix, args) \
__MATHDECL (_Mdouble_, function,suffix, args)
#define __MATHDECL(type, function,suffix, args) \
__MATHDECL_1(type, function,suffix, args); \
__MATHDECL_1(type, __##function,suffix, args)
#define __MATHDECL_1(type, function,suffix, args) \
extern type __MATH_PRECNAME(function,suffix) args
#define _Mdouble_ double
#define __MATH_PRECNAME(name,r) name##r
#include <mathcalls.h>
#undef _Mdouble_
#undef __MATH_PRECNAME
#ifdef __USE_MISC
/* Include the file of declarations again, this time using `float'
instead of `double' and appending f to each function name. */
#ifndef _Mfloat_
#define _Mfloat_ float
#endif
#define _Mdouble_ _Mfloat_
#define __MATH_PRECNAME(name,r) name##f##r
#include <mathcalls.h>
#undef _Mdouble_
#undef __MATH_PRECNAME
/* Include the file of declarations again, this time using `long double'
instead of `double' and appending l to each function name. */
#ifndef _Mlong_double_
#define _Mlong_double_ long double
#endif
#define _Mdouble_ _Mlong_double_
#define __MATH_PRECNAME(name,r) name##l##r
#include <mathcalls.h>
#undef _Mdouble_
#undef __MATH_PRECNAME
#endif /* Use misc. */
#ifdef __USE_MISC
/* Support for various different standard error handling behaviors. */
typedef enum { _IEEE_ = -1, _SVID_, _XOPEN_, _POSIX_ } _LIB_VERSION_TYPE;
/* This variable can be changed at run-time to any of the values above to
affect floating point error handling behavior (it may also be necessary
to change the hardware FPU exception settings). */
extern _LIB_VERSION_TYPE _LIB_VERSION;
#endif
#ifdef __USE_SVID
/* In SVID error handling, `matherr' is called with this description
of the exceptional condition. */
struct __exception
{
int type;
char *name;
double arg1;
double arg2;
double retval;
};
extern int __matherr __P ((struct __exception *));
extern int matherr __P ((struct __exception *));
#define X_TLOSS 1.41484755040568800000e+16
/* Types of exceptions in the `type' field. */
#define DOMAIN 1
#define SING 2
#define OVERFLOW 3
#define UNDERFLOW 4
#define TLOSS 5
#define PLOSS 6
/* SVID mode specifies returning this large value instead of infinity. */
#define HUGE FLT_MAX
#include <float.h> /* Defines FLT_MAX. */
#endif
#ifdef __USE_BSD
/* Some useful constants. */
#define M_E _Mldbl(2.7182818284590452354) /* e */
#define M_LOG2E _Mldbl(1.4426950408889634074) /* log 2e */
#define M_LOG10E _Mldbl(0.43429448190325182765) /* log 10e */
#define M_LN2 _Mldbl(0.69314718055994530942) /* log e2 */
#define M_LN10 _Mldbl(2.30258509299404568402) /* log e10 */
#define M_PI _Mldbl(3.14159265358979323846) /* pi */
#define M_PI_2 _Mldbl(1.57079632679489661923) /* pi/2 */
#define M_PI_4 _Mldbl(0.78539816339744830962) /* pi/4 */
#define M_1_PI _Mldbl(0.31830988618379067154) /* 1/pi */
#define M_2_PI _Mldbl(0.63661977236758134308) /* 2/pi */
#define M_2_SQRTPI _Mldbl(1.12837916709551257390) /* 2/sqrt(pi) */
#define M_SQRT2 _Mldbl(1.41421356237309504880) /* sqrt(2) */
#define M_SQRT1_2 _Mldbl(0.70710678118654752440) /* 1/sqrt(2) */
/* Our constants might specify more precision than `double' can represent.
Use `long double' constants in standard and GNU C, where they are
supported and the cast to `double'. */
#if __STDC__ - 0 || __GNUC__ - 0
#define _Mldbl(x) x##L
#else /* Traditional C. */
#define _Mldbl(x) x
#endif /* Standard or GNU C. */
#endif
/* Get machine-dependent inline versions (if there are any). */
#if defined (__NO_MATH_INLINES) || defined (__OPTIMIZE__)
#include <__math.h>
#endif
__END_DECLS
#endif /* math.h */
3:59 a.m.
New subject: [FAILURE] XEmacs 21.0-beta38 on MkLinux
>>>> "Stephen" == Stephen R Anderson
<sra(a)bloch.ling.yale.edu> writes:
Stephen> hniksic(a)srce.hr said:
> Fred's answer doesn't satisfy me. Correct user code
should not be
> forced to use __exception -- __foo looks like a system-dependent
> thing to me.
Stephen> Don't blame Fred...at least he found OUR problem. He didn't claim to
have
Stephen> found the correct solution from the larger XEmacs point of view. As he says in
Stephen> a later note to me,
The source will soon have this comment:
/* In C++, it is impossible to determine what type matherr expects
without some more configure magic.
We shouldn't be using matherr anyways - it's a non-standard SYSVism. */
The unilateral change to __exception must be a mistake in glibc.
It will likely be fixed in any released version of glibc. Someone who
understands floating point error handling should look at this and
likely expunge it. My guess is that a signal handler for SIGFPE is
all a modern system needs to have. FLOAT_CATCH_SIGILL should be nuked
too.
Martin
9697
days inactive
9698
days old
2 comments
3 participants
participants (3)
-
Hrvoje Niksic -
Martin Buchholz -
Stephen R. Anderson