1
0
mirror of https://git.savannah.gnu.org/git/gnulib.git synced 2025-08-16 01:22:18 +03:00
Files
gnulib/tests/nan.h
Paul Eggert 16f6a8d8d8 Port to PGI 16.10 x86-64
This patch fixes one real bug in gl_anylinked_list2.h, along with
some minor glitches that are not bugs.  It does not silence PGI’s
thousands of bogus warnings when compiling test-intprops.c.
Fortunately, the warnings do not cause a failure.
* lib/c-ctype.h (_C_CTYPE_LOWER_A_THRU_F_N, _C_CTYPE_LOWER_N):
Rename parameter to avoid PGI warning about ‘#define f(n) 'n'’.
My goodness, PGI goes back a long ways - this predates C89!
* lib/gl_anylinked_list2.h (ASYNCSAFE): Fix bug caught by PGI.
For example, ASYNCSAFE (const void *) should expand to
‘const void *volatile’, not to ‘volatile const void *’.
* lib/spawn.in.h (POSIX_SPAWN_USEVFORK): Don't define if already defined.
* lib/verify.h (verify) [!__GNUC__]:
Use shorter albeit meaningless string to bypass silly compiler limits.
* tests/infinity.h (Infinityf, Infinityd, Infinityl) [__PGI]:
* tests/nan.h (NaNf, NaNd, NaNl):
Use static functions to avoid misguided compiler diagnostics.
Is there some reason we don’t use static functions on all platforms?
2017-01-30 18:21:35 -08:00

86 lines
2.5 KiB
C

/* Macros for not-a-number.
Copyright (C) 2007-2017 Free Software Foundation, Inc.
This program is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 3 of the License, or
(at your option) any later version.
This program 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 General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program. If not, see <http://www.gnu.org/licenses/>. */
/* IBM z/OS supports both hexadecimal and IEEE floating-point formats. The
former does not support NaN and its isnan() implementation returns zero
for all values. */
#if defined __MVS__ && defined __IBMC__ && !defined __BFP__
# error "NaN is not supported with IBM's hexadecimal floating-point format; please re-compile with -qfloat=ieee"
#endif
/* NaNf () returns a 'float' not-a-number. */
/* The Compaq (ex-DEC) C 6.4 compiler and the Microsoft MSVC 9 compiler choke
on the expression 0.0 / 0.0. The IBM XL C compiler on z/OS complains.
PGI 16.10 complains. */
#if (defined __DECC || defined _MSC_VER \
|| (defined __MVS__ && defined __IBMC__) \
|| defined __PGI)
static float
NaNf ()
{
static float zero = 0.0f;
return zero / zero;
}
#else
# define NaNf() (0.0f / 0.0f)
#endif
/* NaNd () returns a 'double' not-a-number. */
/* The Compaq (ex-DEC) C 6.4 compiler and the Microsoft MSVC 9 compiler choke
on the expression 0.0 / 0.0. The IBM XL C compiler on z/OS complains.
PGI 16.10 complains. */
#if (defined __DECC || defined _MSC_VER \
|| (defined __MVS__ && defined __IBMC__) \
|| defined __PGI)
static double
NaNd ()
{
static double zero = 0.0;
return zero / zero;
}
#else
# define NaNd() (0.0 / 0.0)
#endif
/* NaNl () returns a 'long double' not-a-number. */
/* On Irix 6.5, gcc 3.4.3 can't compute compile-time NaN, and needs the
runtime type conversion.
The Microsoft MSVC 9 compiler chokes on the expression 0.0L / 0.0L.
The IBM XL C compiler on z/OS complains.
PGI 16.10 complains. */
#ifdef __sgi
static long double NaNl ()
{
double zero = 0.0;
return zero / zero;
}
#elif defined _MSC_VER || (defined __MVS__ && defined __IBMC__) || defined __PGI
static long double
NaNl ()
{
static long double zero = 0.0L;
return zero / zero;
}
#else
# define NaNl() (0.0L / 0.0L)
#endif