-# isnanl.m4 serial 2
-dnl Copyright (C) 2007 Free Software Foundation, Inc.
+# isnanl.m4 serial 12
+dnl Copyright (C) 2007-2010 Free Software Foundation, Inc.
dnl This file is free software; the Free Software Foundation
dnl gives unlimited permission to copy and/or distribute it,
dnl with or without modifications, as long as this notice is preserved.
AC_DEFUN([gl_FUNC_ISNANL],
[
- AC_REQUIRE([gt_TYPE_LONGDOUBLE])
+ AC_REQUIRE([gl_MATH_H_DEFAULTS])
ISNANL_LIBM=
- if test $gt_cv_c_long_double = yes; then
- gl_HAVE_ISNANL_NO_LIBM
- if test $gl_cv_func_isnanl_no_libm = no; then
- gl_HAVE_ISNANL_IN_LIBM
- if test $gl_cv_func_isnanl_in_libm = yes; then
- ISNANL_LIBM=-lm
- fi
- fi
- if test $gl_cv_func_isnanl_no_libm = yes \
- || test $gl_cv_func_isnanl_in_libm = yes; then
- save_LIBS="$LIBS"
- LIBS="$LIBS $ISNANL_LIBM"
- gl_FUNC_ISNANL_WORKS
- LIBS="$save_LIBS"
- case "$gl_cv_func_isnanl_works" in
- *yes) gl_func_isnanl=yes ;;
- *) gl_func_isnanl=no; ISNANL_LIBM= ;;
- esac
- else
- gl_func_isnanl=no
- fi
- if test $gl_func_isnanl = yes; then
- AC_DEFINE([HAVE_ISNANL], 1,
- [Define if the isnan(long double) function is available.])
- else
- AC_LIBOBJ([isnanl])
- gl_LONG_DOUBLE_EXPONENT_LOCATION
+ gl_HAVE_ISNANL_NO_LIBM
+ if test $gl_cv_func_isnanl_no_libm = no; then
+ gl_HAVE_ISNANL_IN_LIBM
+ if test $gl_cv_func_isnanl_in_libm = yes; then
+ ISNANL_LIBM=-lm
fi
fi
+ if test $gl_cv_func_isnanl_no_libm = yes \
+ || test $gl_cv_func_isnanl_in_libm = yes; then
+ save_LIBS="$LIBS"
+ LIBS="$LIBS $ISNANL_LIBM"
+ gl_FUNC_ISNANL_WORKS
+ LIBS="$save_LIBS"
+ case "$gl_cv_func_isnanl_works" in
+ *yes) gl_func_isnanl=yes ;;
+ *) gl_func_isnanl=no; ISNANL_LIBM= ;;
+ esac
+ else
+ gl_func_isnanl=no
+ fi
+ if test $gl_func_isnanl != yes; then
+ HAVE_ISNANL=0
+ gl_BUILD_ISNANL
+ fi
AC_SUBST([ISNANL_LIBM])
])
AC_DEFUN([gl_FUNC_ISNANL_NO_LIBM],
[
- AC_REQUIRE([gt_TYPE_LONGDOUBLE])
- if test $gt_cv_c_long_double = yes; then
- gl_HAVE_ISNANL_NO_LIBM
- gl_func_isnanl_no_libm=$gl_cv_func_isnanl_no_libm
- if test $gl_func_isnanl_no_libm = yes; then
- gl_FUNC_ISNANL_WORKS
- case "$gl_cv_func_isnanl_works" in
- *yes) ;;
- *) gl_func_isnanl_no_libm=no ;;
- esac
- fi
- if test $gl_func_isnanl_no_libm = yes; then
- AC_DEFINE([HAVE_ISNANL_IN_LIBC], 1,
- [Define if the isnan(long double) function is available in libc.])
- else
- AC_LIBOBJ([isnanl])
- gl_LONG_DOUBLE_EXPONENT_LOCATION
- fi
+ gl_HAVE_ISNANL_NO_LIBM
+ gl_func_isnanl_no_libm=$gl_cv_func_isnanl_no_libm
+ if test $gl_func_isnanl_no_libm = yes; then
+ gl_FUNC_ISNANL_WORKS
+ case "$gl_cv_func_isnanl_works" in
+ *yes) ;;
+ *) gl_func_isnanl_no_libm=no ;;
+ esac
+ fi
+ if test $gl_func_isnanl_no_libm = yes; then
+ AC_DEFINE([HAVE_ISNANL_IN_LIBC], [1],
+ [Define if the isnan(long double) function is available in libc.])
+ else
+ gl_BUILD_ISNANL
fi
])
+dnl Pull in replacement isnanl definition. It does not need -lm.
+AC_DEFUN([gl_BUILD_ISNANL],
+[
+ AC_LIBOBJ([isnanl])
+ gl_LONG_DOUBLE_EXPONENT_LOCATION
+])
+
dnl Test whether isnanl() can be used without libm.
AC_DEFUN([gl_HAVE_ISNANL_NO_LIBM],
[
[gl_cv_func_isnanl_no_libm],
[
AC_TRY_LINK([#include <math.h>
- #ifdef isnan
+ #if __GNUC__ >= 4
+ # undef isnanl
+ # define isnanl(x) __builtin_isnanl ((long double)(x))
+ #elif defined isnan
# undef isnanl
# define isnanl(x) isnan ((long double)(x))
#endif
save_LIBS="$LIBS"
LIBS="$LIBS -lm"
AC_TRY_LINK([#include <math.h>
- #ifdef isnan
+ #if __GNUC__ >= 4
+ # undef isnanl
+ # define isnanl(x) __builtin_isnanl ((long double)(x))
+ #elif defined isnan
# undef isnanl
# define isnanl(x) isnan ((long double)(x))
#endif
])
dnl Test whether isnanl() recognizes all numbers which are neither finite nor
-dnl infinite. This test fails e.g. on NetBSD/i386.
+dnl infinite. This test fails e.g. on NetBSD/i386 and on glibc/ia64.
+dnl Also, the GCC >= 4.0 built-in __builtin_isnanl does not pass the tests
+dnl - for pseudo-denormals on i686 and x86_64,
+dnl - for pseudo-zeroes, unnormalized numbers, and pseudo-denormals on ia64.
AC_DEFUN([gl_FUNC_ISNANL_WORKS],
[
AC_REQUIRE([AC_PROG_CC])
- AC_REQUIRE([gt_TYPE_LONGDOUBLE])
+ AC_REQUIRE([gl_BIGENDIAN])
AC_REQUIRE([AC_CANONICAL_HOST]) dnl for cross-compiles
AC_CACHE_CHECK([whether isnanl works], [gl_cv_func_isnanl_works],
[
AC_TRY_RUN([
+#include <float.h>
#include <limits.h>
#include <math.h>
-#ifdef isnan
+#if __GNUC__ >= 4
+# undef isnanl
+# define isnanl(x) __builtin_isnanl ((long double)(x))
+#elif defined isnan
# undef isnanl
# define isnanl(x) isnan ((long double)(x))
#endif
#define NWORDS \
((sizeof (long double) + sizeof (unsigned int) - 1) / sizeof (unsigned int))
-typedef union { long double value; unsigned int word[NWORDS]; }
+typedef union { unsigned int word[NWORDS]; long double value; }
memory_long_double;
+/* On Irix 6.5, gcc 3.4.3 can't compute compile-time NaN, and needs the
+ runtime type conversion. */
+#ifdef __sgi
+static long double NaNl ()
+{
+ double zero = 0.0;
+ return zero / zero;
+}
+#else
+# define NaNl() (0.0L / 0.0L)
+#endif
int main ()
{
memory_long_double m;
unsigned int i;
+ if (!isnanl (NaNl ()))
+ return 1;
+
/* The isnanl function should be immune against changes in the sign bit and
in the mantissa bits. The xor operation twiddles a bit that can only be
a sign bit or a mantissa bit (since the exponent never extends to
bit 31). */
- m.value = 0.0L / 0.0L;
+ m.value = NaNl ();
m.word[NWORDS / 2] ^= (unsigned int) 1 << (sizeof (unsigned int) * CHAR_BIT - 1);
for (i = 0; i < NWORDS; i++)
m.word[i] |= 1;
if (!isnanl (m.value))
return 1;
+#if ((defined __ia64 && LDBL_MANT_DIG == 64) || (defined __x86_64__ || defined __amd64__) || (defined __i386 || defined __i386__ || defined _I386 || defined _M_IX86 || defined _X86_))
+/* Representation of an 80-bit 'long double' as an initializer for a sequence
+ of 'unsigned int' words. */
+# ifdef WORDS_BIGENDIAN
+# define LDBL80_WORDS(exponent,manthi,mantlo) \
+ { ((unsigned int) (exponent) << 16) | ((unsigned int) (manthi) >> 16), \
+ ((unsigned int) (manthi) << 16) | (unsigned int) (mantlo) >> 16), \
+ (unsigned int) (mantlo) << 16 \
+ }
+# else
+# define LDBL80_WORDS(exponent,manthi,mantlo) \
+ { mantlo, manthi, exponent }
+# endif
+ { /* Quiet NaN. */
+ static memory_long_double x =
+ { LDBL80_WORDS (0xFFFF, 0xC3333333, 0x00000000) };
+ if (!isnanl (x.value))
+ return 1;
+ }
+ {
+ /* Signalling NaN. */
+ static memory_long_double x =
+ { LDBL80_WORDS (0xFFFF, 0x83333333, 0x00000000) };
+ if (!isnanl (x.value))
+ return 1;
+ }
+ /* The isnanl function should recognize Pseudo-NaNs, Pseudo-Infinities,
+ Pseudo-Zeroes, Unnormalized Numbers, and Pseudo-Denormals, as defined in
+ Intel IA-64 Architecture Software Developer's Manual, Volume 1:
+ Application Architecture.
+ Table 5-2 "Floating-Point Register Encodings"
+ Figure 5-6 "Memory to Floating-Point Register Data Translation"
+ */
+ { /* Pseudo-NaN. */
+ static memory_long_double x =
+ { LDBL80_WORDS (0xFFFF, 0x40000001, 0x00000000) };
+ if (!isnanl (x.value))
+ return 1;
+ }
+ { /* Pseudo-Infinity. */
+ static memory_long_double x =
+ { LDBL80_WORDS (0xFFFF, 0x00000000, 0x00000000) };
+ if (!isnanl (x.value))
+ return 1;
+ }
+ { /* Pseudo-Zero. */
+ static memory_long_double x =
+ { LDBL80_WORDS (0x4004, 0x00000000, 0x00000000) };
+ if (!isnanl (x.value))
+ return 1;
+ }
+ { /* Unnormalized number. */
+ static memory_long_double x =
+ { LDBL80_WORDS (0x4000, 0x63333333, 0x00000000) };
+ if (!isnanl (x.value))
+ return 1;
+ }
+ { /* Pseudo-Denormal. */
+ static memory_long_double x =
+ { LDBL80_WORDS (0x0000, 0x83333333, 0x00000000) };
+ if (!isnanl (x.value))
+ return 1;
+ }
+#endif
+
return 0;
}], [gl_cv_func_isnanl_works=yes], [gl_cv_func_isnanl_works=no],
- [case "$host_os" in
- netbsd*) gl_cv_func_isnanl_works="guessing no";;
- *) gl_cv_func_isnanl_works="guessing yes";;
+ [case "$host_cpu" in
+ # Guess no on ia64, x86_64, i386.
+ ia64 | x86_64 | i*86) gl_cv_func_isnanl_works="guessing no";;
+ *)
+ case "$host_os" in
+ netbsd*) gl_cv_func_isnanl_works="guessing no";;
+ *) gl_cv_func_isnanl_works="guessing yes";;
+ esac
+ ;;
esac
])
])
])
-
-AC_DEFUN([gl_LONG_DOUBLE_EXPONENT_LOCATION],
-[
- AC_CACHE_CHECK([where to find the exponent in a 'long double'],
- [gl_cv_cc_long_double_expbit0],
- [
- AC_TRY_RUN([
-#include <float.h>
-#include <stddef.h>
-#include <stdio.h>
-#include <string.h>
-#define NWORDS \
- ((sizeof (long double) + sizeof (unsigned int) - 1) / sizeof (unsigned int))
-typedef union { long double value; unsigned int word[NWORDS]; }
- memory_long_double;
-static unsigned int ored_words[NWORDS];
-static unsigned int anded_words[NWORDS];
-static void add_to_ored_words (long double x)
-{
- memory_long_double m;
- size_t i;
- /* Clear it first, in case
- sizeof (long double) < sizeof (memory_long_double). */
- memset (&m, 0, sizeof (memory_long_double));
- m.value = x;
- for (i = 0; i < NWORDS; i++)
- {
- ored_words[i] |= m.word[i];
- anded_words[i] &= m.word[i];
- }
-}
-int main ()
-{
- size_t j;
- FILE *fp = fopen ("conftest.out", "w");
- if (fp == NULL)
- return 1;
- for (j = 0; j < NWORDS; j++)
- anded_words[j] = ~ (unsigned int) 0;
- add_to_ored_words (0.25L);
- add_to_ored_words (0.5L);
- add_to_ored_words (1.0L);
- add_to_ored_words (2.0L);
- add_to_ored_words (4.0L);
- /* Remove bits that are common (e.g. if representation of the first mantissa
- bit is explicit). */
- for (j = 0; j < NWORDS; j++)
- ored_words[j] &= ~anded_words[j];
- /* Now find the nonzero word. */
- for (j = 0; j < NWORDS; j++)
- if (ored_words[j] != 0)
- break;
- if (j < NWORDS)
- {
- size_t i;
- for (i = j + 1; i < NWORDS; i++)
- if (ored_words[i] != 0)
- {
- fprintf (fp, "unknown");
- return (fclose (fp) != 0);
- }
- for (i = 0; ; i++)
- if ((ored_words[j] >> i) & 1)
- {
- fprintf (fp, "word %d bit %d", (int) j, (int) i);
- return (fclose (fp) != 0);
- }
- }
- fprintf (fp, "unknown");
- return (fclose (fp) != 0);
-}
- ],
- [gl_cv_cc_long_double_expbit0=`cat conftest.out`],
- [gl_cv_cc_long_double_expbit0="unknown"],
- [
- dnl When cross-compiling, we don't know. It depends on the
- dnl ABI and compiler version. There are too many cases.
- gl_cv_cc_long_double_expbit0="unknown"
- ])
- rm -f conftest.out
- ])
- case "$gl_cv_cc_long_double_expbit0" in
- word*bit*)
- word=`echo "$gl_cv_cc_long_double_expbit0" | sed -e 's/word //' -e 's/ bit.*//'`
- bit=`echo "$gl_cv_cc_long_double_expbit0" | sed -e 's/word.*bit //'`
- AC_DEFINE_UNQUOTED([LDBL_EXPBIT0_WORD], [$word],
- [Define as the word index where to find the exponent of 'long double'.])
- AC_DEFINE_UNQUOTED([LDBL_EXPBIT0_BIT], [$bit],
- [Define as the bit index in the word where to find bit 0 of the exponent of 'long double'.])
- ;;
- esac
-])
projects / gnulib.git / blobdiff