[PATCH v4 1/2] float128: Add strtof128, wcstof128, and related functions.

[PATCH v4 1/2] float128: Add strtof128, wcstof128, and related functions.

[Gabriel F. T. Gomes]

From: "Paul E. Murphy" <murphyp@linux.vnet.ibm.com>

Changes since v3:

  - Removed the definition of CHAR (defined before inclusion).
  - Converted the unnecessarily complex macros (_GEN and _DO) into IF_FLOAT128.

Changes since v2:

  - Fixed the condition for defining F128 (no dependency on compiler version).
  - Fixed the condition for defining CHAR (no dependency on __HAVE_FLOAT128).
  - Removed the unnecessary indirections in the macros _GEN and _DO.
  - Fixed the condition for declaring wcstof128 (__USE_GNU, not TS 18661-3).

Changes since v1:

  - Removed NEWS entry to the commit that actually enables strtof128.
  - Removed the check for __STDC_WANT_IEC_60559_TYPES_EXT__ from blocks of code
    that are inside __USE_GNU conditional.
  - Moved the declaration of wcstof128 closer to wcstod, wcstof, and wcstold.
  - Redefined F128 in the tests to use the existing workaround for old.
    compilers (i.e.: __f128 ()).
  - Added descriptive comment before the copyright, as needed.

-- 8< --
The implementations are contained with sysdeps/ieee754/float128 as
they are only built when _Float128 is enabled within libc/m.

2017-05-23  Paul E. Murphy  <murphyp@linux.vnet.ibm.com>

	* include/gmp.h (__mpn_construct_float128): New declaration.
	* include/stdlib.h: Include bits/floatn.h for _Float128 tests.
	(__strtof128_l): New declaration.
	(__strtof128_nan): Likewise.
	(__wcstof128_nan): Likewise.
	(__strtof128_internal): Likewise.
	(____strtof128_l_internal): Likewise.
	* include/wchar.h: Include bits/floatn.h for _Float128 tests.
	(__wcstof128_l): New declaration.
	(__wcstof128_internal): Likewise.

	* stdlib/Makefile (bug-strtod2): Link libm too.

	* stdlib/stdlib.h (strtof128): New declaration.
	(strtof128_l): Likewise.

	* stdlib/tst-strtod-nan-locale-main.c: Updated to use
	tst-strtod.h macros to ensure float128 gets tested too.

	* stdlib/tst-strtod-round-skeleton.c (CHOOSE_f128): New macro.

	* stdlib/tst-strtod.h: Include bits/floatn.h for _Float128
	tests.
	(IF_FLOAT128): New macro.
	(GEN_TEST_STRTOD): Update to optionally include _Float128 in
	the tests.
	(STRTOD_TEST_FOREACH): Likewise.

	* sysdeps/ieee754/float128/Makefile: Insert new strtof128 and
	wcstof128 functions into libc.

	* sysdeps/ieee754/float128/Versions: Add exports for the above
	new functions.

	* sysdeps/ieee754/float128/mpn2float128.c: New file.
	* sysdeps/ieee754/float128/strtod_nan_float128.h: New file.
	* sysdeps/ieee754/float128/strtof128.c: New file.
	* sysdeps/ieee754/float128/strtof128_l.c: New file.
	* sysdeps/ieee754/float128/strtof128_nan.c: New file.
	* sysdeps/ieee754/float128/wcstof128.c: New file.
	* sysdeps/ieee754/float128/wcstof128_l.c: New file.
	* sysdeps/ieee754/float128/wcstof128_nan.c: New fike.
	* wcsmbs/Makefile: (CFLAGS-wcstof128.c): Append strtox-CFLAGS.
	(CFLAGS-wcstof128_l): Likewise.

	* wcsmbs/wchar.h: Include bits/floatn.h for _Float128 tests.
	(wcstof128): New declaration.
	(wcstof128_l): Likewise.
---
 include/gmp.h                                  |  4 ++
 include/stdlib.h                               | 29 ++++++++++++++
 include/wchar.h                                | 13 +++++++
 stdlib/Makefile                                |  1 +
 stdlib/stdlib.h                                | 14 +++++++
 stdlib/tst-strtod-nan-locale-main.c            | 36 ++++++++---------
 stdlib/tst-strtod-round-skeleton.c             |  1 +
 stdlib/tst-strtod.h                            | 36 +++++++++++------
 sysdeps/ieee754/float128/Makefile              |  5 +++
 sysdeps/ieee754/float128/Versions              | 10 +++++
 sysdeps/ieee754/float128/mpn2float128.c        | 54 ++++++++++++++++++++++++++
 sysdeps/ieee754/float128/strtod_nan_float128.h | 33 ++++++++++++++++
 sysdeps/ieee754/float128/strtof128.c           | 34 ++++++++++++++++
 sysdeps/ieee754/float128/strtof128_l.c         | 43 ++++++++++++++++++++
 sysdeps/ieee754/float128/strtof128_nan.c       | 28 +++++++++++++
 sysdeps/ieee754/float128/wcstof128.c           | 30 ++++++++++++++
 sysdeps/ieee754/float128/wcstof128_l.c         | 30 ++++++++++++++
 sysdeps/ieee754/float128/wcstof128_nan.c       | 24 ++++++++++++
 wcsmbs/Makefile                                |  2 +
 wcsmbs/wchar.h                                 | 15 +++++++
 20 files changed, 413 insertions(+), 29 deletions(-)
 create mode 100644 sysdeps/ieee754/float128/mpn2float128.c
 create mode 100644 sysdeps/ieee754/float128/strtod_nan_float128.h
 create mode 100644 sysdeps/ieee754/float128/strtof128.c
 create mode 100644 sysdeps/ieee754/float128/strtof128_l.c
 create mode 100644 sysdeps/ieee754/float128/strtof128_nan.c
 create mode 100644 sysdeps/ieee754/float128/wcstof128.c
 create mode 100644 sysdeps/ieee754/float128/wcstof128_l.c
 create mode 100644 sysdeps/ieee754/float128/wcstof128_nan.c

diff --git a/include/gmp.h b/include/gmp.h
index e6f635e..3590acf 100644
--- a/include/gmp.h
+++ b/include/gmp.h
@@ -31,5 +31,9 @@ extern double __mpn_construct_double (mp_srcptr frac_ptr, int expt,
 extern long double __mpn_construct_long_double (mp_srcptr frac_ptr, int expt,
 						int sign);
 
+#if __HAVE_DISTINCT_FLOAT128
+extern _Float128 __mpn_construct_float128 (mp_srcptr frac_ptr, int expt,
+					   int sign);
+#endif
 
 #endif
diff --git a/include/stdlib.h b/include/stdlib.h
index 5847445..cae9f2c 100644
--- a/include/stdlib.h
+++ b/include/stdlib.h
@@ -219,6 +219,35 @@ libc_hidden_proto (__wcstof_nan)
 libc_hidden_proto (__wcstod_nan)
 libc_hidden_proto (__wcstold_nan)
 
+/* Enable _FloatN bits as needed.  */
+#include <bits/floatn.h>
+
+#if __HAVE_DISTINCT_FLOAT128
+extern __typeof (strtof128_l) __strtof128_l;
+
+libc_hidden_proto (__strtof128_l)
+libc_hidden_proto (strtof128)
+
+extern _Float128 __strtof128_nan (const char *, char **, char)
+     internal_function;
+extern _Float128 __wcstof128_nan (const wchar_t *, wchar_t **, wchar_t)
+     internal_function;
+
+libc_hidden_proto (__strtof128_nan)
+libc_hidden_proto (__wcstof128_nan)
+
+extern _Float128 __strtof128_internal (const char *__restrict __nptr,
+				       char **__restrict __endptr,
+				       int __group);
+libc_hidden_proto (__strtof128_internal)
+
+extern _Float128 ____strtof128_l_internal (const char *__restrict __nptr,
+					   char **__restrict __endptr,
+					   int __group, __locale_t __loc);
+
+libc_hidden_proto (____strtof128_l_internal)
+#endif
+
 extern char *__ecvt (double __value, int __ndigit, int *__restrict __decpt,
 		     int *__restrict __sign);
 extern char *__fcvt (double __value, int __ndigit, int *__restrict __decpt,
diff --git a/include/wchar.h b/include/wchar.h
index a773d56..55de9be 100644
--- a/include/wchar.h
+++ b/include/wchar.h
@@ -4,6 +4,8 @@
 # ifndef _ISOMAC
 # ifdef _WCHAR_H
 
+#include <bits/floatn.h>
+
 extern __typeof (wcscasecmp_l) __wcscasecmp_l;
 extern __typeof (wcsncasecmp_l) __wcsncasecmp_l;
 extern __typeof (wcscoll_l) __wcscoll_l;
@@ -70,6 +72,17 @@ libc_hidden_proto (wcstoll)
 libc_hidden_proto (wcstoul)
 libc_hidden_proto (wcstoull)
 
+#if __HAVE_DISTINCT_FLOAT128
+extern __typeof (wcstof128_l) __wcstof128_l;
+libc_hidden_proto (__wcstof128_l)
+extern _Float128 __wcstof128_internal (const wchar_t *__restrict __nptr,
+				       wchar_t **__restrict __endptr,
+				       int __group) __THROW;
+
+libc_hidden_proto (__wcstof128_internal)
+libc_hidden_proto (wcstof128)
+#endif
+
 libc_hidden_proto (__wcscasecmp_l)
 libc_hidden_proto (__wcsncasecmp_l)
 
diff --git a/stdlib/Makefile b/stdlib/Makefile
index 9b0acce..0314d59 100644
--- a/stdlib/Makefile
+++ b/stdlib/Makefile
@@ -198,6 +198,7 @@ $(objpfx)tst-putenvmod.so: $(objpfx)tst-putenvmod.os $(link-libc-deps)
 libof-tst-putenvmod = extramodules
 
 $(objpfx)bug-getcontext: $(libm)
+$(objpfx)bug-strtod2: $(libm)
 $(objpfx)tst-strtod-round: $(libm)
 $(objpfx)tst-tininess: $(libm)
 $(objpfx)tst-strtod-underflow: $(libm)
diff --git a/stdlib/stdlib.h b/stdlib/stdlib.h
index 0a693c2..6f1e70e 100644
--- a/stdlib/stdlib.h
+++ b/stdlib/stdlib.h
@@ -128,6 +128,13 @@ extern long double strtold (const char *__restrict __nptr,
      __THROW __nonnull ((1));
 #endif
 
+#if __HAVE_FLOAT128 && __GLIBC_USE (IEC_60559_TYPES_EXT)
+/* Likewise for the '_Float128' format  */
+extern _Float128 strtof128 (const char *__restrict __nptr,
+		      char **__restrict __endptr)
+     __THROW __nonnull ((1));
+#endif
+
 /* Convert a string to a long integer.  */
 extern long int strtol (const char *__restrict __nptr,
 			char **__restrict __endptr, int __base)
@@ -236,6 +243,13 @@ extern long double strtold_l (const char *__restrict __nptr,
 			      char **__restrict __endptr,
 			      __locale_t __loc)
      __THROW __nonnull ((1, 3));
+
+# if __HAVE_FLOAT128
+extern _Float128 strtof128_l (const char *__restrict __nptr,
+			      char **__restrict __endptr,
+			      __locale_t __loc)
+     __THROW __nonnull ((1, 3));
+# endif
 #endif /* GNU */
 
 
diff --git a/stdlib/tst-strtod-nan-locale-main.c b/stdlib/tst-strtod-nan-locale-main.c
index 8044dad..8222174 100644
--- a/stdlib/tst-strtod-nan-locale-main.c
+++ b/stdlib/tst-strtod-nan-locale-main.c
@@ -23,6 +23,8 @@
 #include <stdlib.h>
 #include <wchar.h>
 
+#include <stdlib/tst-strtod.h>
+
 #define STR_(X) #X
 #define STR(X) STR_(X)
 #define FNPFXS STR (FNPFX)
@@ -30,20 +32,22 @@
 #define CONCAT(X, Y) CONCAT_ (X, Y)
 #define FNX(FN) CONCAT (FNPFX, FN)
 
-#define TEST(LOC, STR, FN, TYPE)					\
-  do									\
+#define TEST_STRTOD(FSUF, FTYPE, FTOSTR, LSUF, CSUF)			\
+static int								\
+test_strto ## FSUF (const char * loc, CHAR * s)				\
+{									\
+  CHAR *ep;								\
+  FTYPE val = FNX (FSUF) (s, &ep);					\
+  if (isnan (val) && *ep == 0)						\
+    printf ("PASS: %s: " FNPFXS #FSUF " (" SFMT ")\n", loc, s);		\
+  else									\
     {									\
-      CHAR *ep;								\
-      TYPE val = FNX (FN) (STR, &ep);					\
-      if (isnan (val) && *ep == 0)					\
-	printf ("PASS: %s: " FNPFXS #FN " (" SFMT ")\n", LOC, STR);	\
-      else								\
-	{								\
-	  printf ("FAIL: %s: " FNPFXS #FN " (" SFMT ")\n", LOC, STR);	\
-	  result = 1;							\
-	}								\
+      printf ("FAIL: %s: " FNPFXS #FSUF " (" SFMT ")\n", loc, s);	\
+      return 1;							        \
     }									\
-  while (0)
+  return 0;								\
+}
+GEN_TEST_STRTOD_FOREACH (TEST_STRTOD)
 
 static int
 test_one_locale (const char *loc)
@@ -64,13 +68,9 @@ test_one_locale (const char *loc)
       s[4] = L_('A') + i - 10;
       s[5] = L_(')');
       s[6] = 0;
-      TEST (loc, s, f, float);
-      TEST (loc, s, d, double);
-      TEST (loc, s, ld, long double);
+      result |= STRTOD_TEST_FOREACH (test_strto, loc, s);
       s[4] = L_('a') + i - 10;
-      TEST (loc, s, f, float);
-      TEST (loc, s, d, double);
-      TEST (loc, s, ld, long double);
+      result |= STRTOD_TEST_FOREACH (test_strto, loc, s);
     }
   return result;
 }
diff --git a/stdlib/tst-strtod-round-skeleton.c b/stdlib/tst-strtod-round-skeleton.c
index 73603e6..1a9f663 100644
--- a/stdlib/tst-strtod-round-skeleton.c
+++ b/stdlib/tst-strtod-round-skeleton.c
@@ -93,6 +93,7 @@
 /* Add type specific choosing macros below.  */
 #define CHOOSE_f(f,...) f
 #define CHOOSE_d(f,d,...) d
+#define CHOOSE_f128(f,d,ld64i,ld64m,ld106,ld113,...) ld113
 /* long double is special, and handled above.  */
 
 /* Selector for expected result field of a given type.  */
diff --git a/stdlib/tst-strtod.h b/stdlib/tst-strtod.h
index bf5f901..e23ac3f 100644
--- a/stdlib/tst-strtod.h
+++ b/stdlib/tst-strtod.h
@@ -21,11 +21,24 @@
 
 #define FSTRLENMAX 128
 
+#include <bits/floatn.h>
+
+#define F128 __f128 ()
+
+/* Test strfromf128 and strtof128 on all platforms that provide them,
+   whether or not the type _Float128 is ABI-distinct from long double.  */
+#if __HAVE_FLOAT128
+# define IF_FLOAT128(x) x
+#else
+# define IF_FLOAT128(x)
+#endif
+
 /* Splat n variants of the same test for the various strtod functions.  */
-#define GEN_TEST_STRTOD_FOREACH(mfunc, ...)			 \
-    mfunc (  f,       float, strfromf, f, f, ##__VA_ARGS__)	 \
-    mfunc (  d,      double, strfromd,  ,  , ##__VA_ARGS__)	 \
-    mfunc ( ld, long double, strfroml, L, l, ##__VA_ARGS__)
+#define GEN_TEST_STRTOD_FOREACH(mfunc, ...)				      \
+  mfunc (  f,       float, strfromf, f, f, ##__VA_ARGS__)		      \
+  mfunc (  d,      double, strfromd,  ,  , ##__VA_ARGS__)		      \
+  mfunc ( ld, long double, strfroml, L, l, ##__VA_ARGS__)		      \
+  IF_FLOAT128 (mfunc (f128, _Float128, strfromf128, F128, f128, ##__VA_ARGS__))
 /* The arguments to the generated macros are:
    FSUF - Function suffix
    FTYPE - float type
@@ -36,13 +49,14 @@
 
 
 
-#define STRTOD_TEST_FOREACH(mfunc, ...)	  \
-({					  \
-   int result = 0;			  \
-   result |= mfunc ## f  (__VA_ARGS__);   \
-   result |= mfunc ## d  (__VA_ARGS__);   \
-   result |= mfunc ## ld (__VA_ARGS__);   \
-   result;				  \
+#define STRTOD_TEST_FOREACH(mfunc, ...)			\
+({							\
+   int result = 0;					\
+   result |= mfunc ## f  (__VA_ARGS__);			\
+   result |= mfunc ## d  (__VA_ARGS__);			\
+   result |= mfunc ## ld (__VA_ARGS__);			\
+   IF_FLOAT128 (result |= mfunc ## f128 (__VA_ARGS__));	\
+   result;						\
 })
 
 
diff --git a/sysdeps/ieee754/float128/Makefile b/sysdeps/ieee754/float128/Makefile
index c07586c..166e630 100644
--- a/sysdeps/ieee754/float128/Makefile
+++ b/sysdeps/ieee754/float128/Makefile
@@ -1,3 +1,8 @@
 ifeq ($(subdir),stdlib)
 routines += float1282mpn strfromf128
+routines += strtof128 strtof128_l strtof128_nan mpn2float128
+endif
+
+ifeq ($(subdir),wcsmbs)
+routines += wcstof128_l wcstof128 wcstof128_nan
 endif
diff --git a/sysdeps/ieee754/float128/Versions b/sysdeps/ieee754/float128/Versions
index caf2064..ef9df2d 100644
--- a/sysdeps/ieee754/float128/Versions
+++ b/sysdeps/ieee754/float128/Versions
@@ -4,7 +4,17 @@
 %endif
 libc {
   FLOAT128_VERSION {
+    __strtof128_internal;
+    __wcstof128_internal;
     strfromf128;
+    strtof128;
+    strtof128_l;
+    wcstof128;
+    wcstof128_l;
+  }
+  GLIBC_PRIVATE {
+    # For __nanf128.
+    __strtof128_nan;
   }
 }
 libm {
diff --git a/sysdeps/ieee754/float128/mpn2float128.c b/sysdeps/ieee754/float128/mpn2float128.c
new file mode 100644
index 0000000..12f703a
--- /dev/null
+++ b/sysdeps/ieee754/float128/mpn2float128.c
@@ -0,0 +1,54 @@
+/* Convert a number in multi-precision format to _Float128.
+   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 "gmp.h"
+#include "gmp-impl.h"
+#include <float.h>
+#include <math.h>
+
+#include "ieee754_float128.h"
+
+/* Convert a multi-precision integer of the needed number of bits (113 for
+   binary128) and an integral power of two to a `_Float128' in IEEE854
+   quad-precision format.  */
+
+_Float128
+__mpn_construct_float128 (mp_srcptr frac_ptr, int expt, int sign)
+{
+  union ieee854_float128 u;
+
+  u.ieee.negative = sign;
+  u.ieee.exponent = expt + IEEE854_FLOAT128_BIAS;
+#if BITS_PER_MP_LIMB == 32
+  u.ieee.mantissa3 = frac_ptr[0];
+  u.ieee.mantissa2 = frac_ptr[1];
+  u.ieee.mantissa1 = frac_ptr[2];
+  u.ieee.mantissa0 = frac_ptr[3] & (((mp_limb_t) 1
+				     << (FLT128_MANT_DIG - 96)) - 1);
+#elif BITS_PER_MP_LIMB == 64
+  u.ieee.mantissa3 = frac_ptr[0] & (((mp_limb_t) 1 << 32) - 1);
+  u.ieee.mantissa2 = frac_ptr[0] >> 32;
+  u.ieee.mantissa1 = frac_ptr[1] & (((mp_limb_t) 1 << 32) - 1);
+  u.ieee.mantissa0 = (frac_ptr[1] >> 32) & (((mp_limb_t) 1
+					     << (FLT128_MANT_DIG - 96)) - 1);
+#else
+  #error "mp_limb size " BITS_PER_MP_LIMB "not accounted for"
+#endif
+
+  return u.d;
+}
diff --git a/sysdeps/ieee754/float128/strtod_nan_float128.h b/sysdeps/ieee754/float128/strtod_nan_float128.h
new file mode 100644
index 0000000..79329fb
--- /dev/null
+++ b/sysdeps/ieee754/float128/strtod_nan_float128.h
@@ -0,0 +1,33 @@
+/* Convert string for NaN payload to corresponding NaN.  For _Float128.
+   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/>.  */
+
+#define FLOAT		_Float128
+#define SET_MANTISSA(flt, mant)				\
+  do							\
+    {							\
+      union ieee854_float128 u;				\
+      u.d = (flt);					\
+      u.ieee_nan.mantissa0 = 0;				\
+      u.ieee_nan.mantissa1 = 0;				\
+      u.ieee_nan.mantissa2 = (mant) >> 32;		\
+      u.ieee_nan.mantissa3 = (mant);			\
+      if ((u.ieee.mantissa0 | u.ieee.mantissa1		\
+	   | u.ieee.mantissa2 | u.ieee.mantissa3) != 0)	\
+	(flt) = u.d;					\
+    }							\
+  while (0)
diff --git a/sysdeps/ieee754/float128/strtof128.c b/sysdeps/ieee754/float128/strtof128.c
new file mode 100644
index 0000000..63cc80d
--- /dev/null
+++ b/sysdeps/ieee754/float128/strtof128.c
@@ -0,0 +1,34 @@
+/* strtof128 wrapper of strtof128_l.
+   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/>.  */
+
+/* The actual implementation for all floating point sizes is in strtod.c.
+   These macros tell it to produce the `_Float128' version, `strtof128'.  */
+
+#define FLOAT		_Float128
+#define FLT		FLT128
+#ifdef USE_WIDE_CHAR
+# define STRTOF		wcstof128
+# define STRTOF_L	__wcstof128_l
+#else
+# define STRTOF		strtof128
+# define STRTOF_L	__strtof128_l
+#endif
+
+#include <float128_private.h>
+
+#include <stdlib/strtod.c>
diff --git a/sysdeps/ieee754/float128/strtof128_l.c b/sysdeps/ieee754/float128/strtof128_l.c
new file mode 100644
index 0000000..0f768a3
--- /dev/null
+++ b/sysdeps/ieee754/float128/strtof128_l.c
@@ -0,0 +1,43 @@
+/* Convert string representing a number to a _Float128 value, with locale.
+   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 <xlocale.h>
+
+/* Bring in potential typedef for _Float128 early for declaration below.  */
+#include <bits/floatn.h>
+
+extern _Float128 ____strtof128_l_internal (const char *, char **,
+					   int, __locale_t);
+
+#define	FLOAT		_Float128
+#define	FLT		FLT128
+#ifdef USE_WIDE_CHAR
+# define STRTOF		wcstof128_l
+# define __STRTOF	__wcstof128_l
+# define STRTOF_NAN	__wcstof128_nan
+#else
+# define STRTOF		strtof128_l
+# define __STRTOF	__strtof128_l
+# define STRTOF_NAN	__strtof128_nan
+#endif
+#define	MPN2FLOAT	__mpn_construct_float128
+#define	FLOAT_HUGE_VAL	__builtin_huge_valf128 ()
+
+#include <float128_private.h>
+
+#include <stdlib/strtod_l.c>
diff --git a/sysdeps/ieee754/float128/strtof128_nan.c b/sysdeps/ieee754/float128/strtof128_nan.c
new file mode 100644
index 0000000..b9d79e3
--- /dev/null
+++ b/sysdeps/ieee754/float128/strtof128_nan.c
@@ -0,0 +1,28 @@
+/* Convert string for NaN payload to corresponding NaN.  Narrow
+   strings, __float128.
+   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 <math.h>
+
+#include <float128_private.h>
+
+#include <strtod_nan_narrow.h>
+#include <strtod_nan_float128.h>
+
+#define STRTOD_NAN __strtof128_nan
+#include <stdlib/strtod_nan_main.c>
diff --git a/sysdeps/ieee754/float128/wcstof128.c b/sysdeps/ieee754/float128/wcstof128.c
new file mode 100644
index 0000000..c5a17ba
--- /dev/null
+++ b/sysdeps/ieee754/float128/wcstof128.c
@@ -0,0 +1,30 @@
+/* wcstof128 wrapper of wcstof128_l.
+   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 <stddef.h>
+#include <xlocale.h>
+
+#define	USE_WIDE_CHAR	1
+
+/* Bring in _Float128 typedef if needed.  */
+#include <bits/floatn.h>
+
+extern _Float128 ____wcstof128_l_internal (const wchar_t *, wchar_t **, int,
+					   __locale_t);
+
+#include "strtof128.c"
diff --git a/sysdeps/ieee754/float128/wcstof128_l.c b/sysdeps/ieee754/float128/wcstof128_l.c
new file mode 100644
index 0000000..670b6af
--- /dev/null
+++ b/sysdeps/ieee754/float128/wcstof128_l.c
@@ -0,0 +1,30 @@
+/* Convert string representing a number to integer value, using given locale.
+   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 <stddef.h>
+#include <xlocale.h>
+
+
+#define	USE_WIDE_CHAR	1
+
+#include <bits/floatn.h>
+
+extern _Float128 ____wcstof128_l_internal (const wchar_t *, wchar_t **, int,
+					   __locale_t);
+
+#include "strtof128_l.c"
diff --git a/sysdeps/ieee754/float128/wcstof128_nan.c b/sysdeps/ieee754/float128/wcstof128_nan.c
new file mode 100644
index 0000000..abad881
--- /dev/null
+++ b/sysdeps/ieee754/float128/wcstof128_nan.c
@@ -0,0 +1,24 @@
+/* Convert string for NaN payload to corresponding NaN.  Wide strings,
+   _Float128.
+   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 <stdlib/strtod_nan_wide.h>
+#include <float128_private.h>
+#include <strtod_nan_float128.h>
+#define STRTOD_NAN __wcstof128_nan
+#include <stdlib/strtod_nan_main.c>
diff --git a/wcsmbs/Makefile b/wcsmbs/Makefile
index d6b214b..84cf986 100644
--- a/wcsmbs/Makefile
+++ b/wcsmbs/Makefile
@@ -81,6 +81,7 @@ CFLAGS-wcstoll.c = $(strtox-CFLAGS)
 CFLAGS-wcstoull.c = $(strtox-CFLAGS)
 CFLAGS-wcstod.c = $(strtox-CFLAGS)
 CFLAGS-wcstold.c = $(strtox-CFLAGS)
+CFLAGS-wcstof128.c = $(strtox-CFLAGS)
 CFLAGS-wcstof.c = $(strtox-CFLAGS)
 CFLAGS-wcstol_l.c = $(strtox-CFLAGS)
 CFLAGS-wcstoul_l.c = $(strtox-CFLAGS)
@@ -88,6 +89,7 @@ CFLAGS-wcstoll_l.c = $(strtox-CFLAGS)
 CFLAGS-wcstoull_l.c = $(strtox-CFLAGS)
 CFLAGS-wcstod_l.c = $(strtox-CFLAGS)
 CFLAGS-wcstold_l.c = $(strtox-CFLAGS)
+CFLAGS-wcstof128_l.c = $(strtox-CFLAGS)
 CFLAGS-wcstof_l.c = $(strtox-CFLAGS)
 CPPFLAGS-tst-wchar-h.c = -D_FORTIFY_SOURCE=2
 
diff --git a/wcsmbs/wchar.h b/wcsmbs/wchar.h
index 79be9fc..5be993d 100644
--- a/wcsmbs/wchar.h
+++ b/wcsmbs/wchar.h
@@ -28,6 +28,9 @@
 # include <bits/libc-header-start.h>
 #endif
 
+/* Gather machine dependent type support.  */
+#include <bits/floatn.h>
+
 #ifdef _WCHAR_H
 /* Get FILE definition.  */
 # define __need___FILE
@@ -426,6 +429,12 @@ extern long double wcstold (const wchar_t *__restrict __nptr,
 			    wchar_t **__restrict __endptr) __THROW;
 #endif /* C99 */
 
+/* Likewise for `_Float128' when support is enabled.  */
+#if __HAVE_FLOAT128 && defined __USE_GNU
+extern _Float128 wcstof128 (const wchar_t *__restrict __nptr,
+			    wchar_t **__restrict __endptr) __THROW;
+#endif
+
 
 /* Convert initial portion of wide string NPTR to `long int'
    representation.  */
@@ -518,6 +527,12 @@ extern float wcstof_l (const wchar_t *__restrict __nptr,
 extern long double wcstold_l (const wchar_t *__restrict __nptr,
 			      wchar_t **__restrict __endptr,
 			      __locale_t __loc) __THROW;
+
+# if __HAVE_FLOAT128
+extern _Float128 wcstof128_l (const wchar_t *__restrict __nptr,
+			      wchar_t **__restrict __endptr,
+			      __locale_t __loc) __THROW;
+# endif
 #endif	/* use GNU */
 
 
-- 
2.4.11

[PATCH v4 2/2] float128: Add strfromf128, strtof128, and wcstof128 to the manual

[Gabriel F. T. Gomes]

Changes since v3:

  - Use strtofNx instead of strtofN in the @deftypefunx item.
  - Added "@comment" to each new function definition.
  - Added "@comment GNU" to wcstofN[x] definitions.
  - Replaces use of @ref with @pxref.
  - Reworded the safety comments to make clear that both strfromfN and
    strfromfNx depend on both __printf_fp and __printf_fphex.
  - Fixed grammatical errors and applied language suggestions.

Changes since v2:

  - Fix typos and grammatical errors.
  - Move @var{N} into the @code{} blocks
  - Remove @var{N} from the function names and return types in the
    @deftypefun and @deftypefunx blocks, because @var{N} causes errors in
    'make html' and 'make info' (make 'pdf' was working fine).

Changes since v1:

  - Added _FloatN and _FloatNx to the section that discusses C
    floating-point types in the section Mathematics.
  - Reworded the descriptions of strfromf128 etc. into descriptions of
    strfromfN etc., to ease the changes needed in the manual when more
    types from TS 18661-3 get added to glibc.

-- 8< --
2016-11-14  Gabriel F. T. Gomes  <gftg@linux.vnet.ibm.com>

	* manual/math.texi (Mathematics): Add description for _FloatN and
	_FloatNx types and mentions that they are not support in glibc for
	any architecture, so far.
	* manual/arith.texi (wcstof, wcstold): Replace the mention to
	stdlib.h with wchar.h.
	(Parsing of Floats): Add descriptions for strtofN and wcstofN.
	(Printing of Floats): Add description for strfromfN.
---
 manual/arith.texi | 63 ++++++++++++++++++++++++++++++++++++++++++++++++-------
 manual/math.texi  | 11 ++++++++++
 2 files changed, 67 insertions(+), 7 deletions(-)

diff --git a/manual/arith.texi b/manual/arith.texi
index dec12a0..5c1dcdc 100644
--- a/manual/arith.texi
+++ b/manual/arith.texi
@@ -2980,6 +2980,7 @@ examining @var{errno} and @var{tailptr}.
 @comment ISO
 @deftypefunx {long double} strtold (const char *@var{string}, char **@var{tailptr})
 @safety{@prelim{}@mtsafe{@mtslocale{}}@assafe{}@acsafe{}}
+@comment See safety comments for strtod.
 These functions are analogous to @code{strtod}, but return @code{float}
 and @code{long double} values respectively.  They report errors in the
 same way as @code{strtod}.  @code{strtof} can be substantially faster
@@ -2990,23 +2991,52 @@ double} is a separate type).
 These functions have been GNU extensions and are new to @w{ISO C99}.
 @end deftypefun
 
+@comment stdlib.h
+@comment ISO/IEC TS 18661-3
+@deftypefun _FloatN strtofN (const char *@var{string}, char **@var{tailptr})
+@comment stdlib.h
+@comment ISO/IEC TS 18661-3
+@deftypefunx _FloatNx strtofNx (const char *@var{string}, char **@var{tailptr})
+@safety{@prelim{}@mtsafe{@mtslocale{}}@assafe{}@acsafe{}}
+@comment See safety comments for strtod.
+These functions are like @code{strtod}, except for the return type.
+
+They were introduced in @w{ISO/IEC TS 18661-3} and are available on machines
+that support the related types; @pxref{Mathematics}.
+@end deftypefun
+
 @comment wchar.h
 @comment ISO
 @deftypefun double wcstod (const wchar_t *restrict @var{string}, wchar_t **restrict @var{tailptr})
-@comment stdlib.h
+@comment wchar.h
 @comment ISO
 @deftypefunx float wcstof (const wchar_t *@var{string}, wchar_t **@var{tailptr})
-@comment stdlib.h
+@comment wchar.h
 @comment ISO
 @deftypefunx {long double} wcstold (const wchar_t *@var{string}, wchar_t **@var{tailptr})
+@comment wchar.h
+@comment GNU
+@deftypefunx _FloatN wcstofN (const wchar_t *@var{string}, wchar_t **@var{tailptr})
+@comment wchar.h
+@comment GNU
+@deftypefunx _FloatNx wcstofNx (const wchar_t *@var{string}, wchar_t **@var{tailptr})
 @safety{@prelim{}@mtsafe{@mtslocale{}}@assafe{}@acsafe{}}
-The @code{wcstod}, @code{wcstof}, and @code{wcstol} functions are
-equivalent in nearly all aspect to the @code{strtod}, @code{strtof}, and
-@code{strtold} functions but it handles wide character string.
+@comment See safety comments for strtod.
+The @code{wcstod}, @code{wcstof}, @code{wcstol}, @code{wcstof@var{N}},
+and @code{wcstof@var{N}x} functions are equivalent in nearly all aspects
+to the @code{strtod}, @code{strtof}, @code{strtold},
+@code{strtof@var{N}}, and @code{strtof@var{N}x} functions, but they
+handle wide character strings.
 
 The @code{wcstod} function was introduced in @w{Amendment 1} of @w{ISO
 C90}.  The @code{wcstof} and @code{wcstold} functions were introduced in
 @w{ISO C99}.
+
+The @code{wcstof@var{N}} and @code{wcstof@var{N}x} functions are not in
+any standard, but are added to provide completeness for the
+non-deprecated interface of wide character string to floating-point
+conversion functions.  They are only available on machines that support
+the related types; @pxref{Mathematics}.
 @end deftypefun
 
 @comment stdlib.h
@@ -3033,11 +3063,15 @@ The @samp{strfrom} functions are declared in @file{stdlib.h}.
 @comment stdlib.h
 @comment ISO/IEC TS 18661-1
 @deftypefun int strfromd (char *restrict @var{string}, size_t @var{size}, const char *restrict @var{format}, double @var{value})
+@comment stdlib.h
+@comment ISO/IEC TS 18661-1
 @deftypefunx int strfromf (char *restrict @var{string}, size_t @var{size}, const char *restrict @var{format}, float @var{value})
+@comment stdlib.h
+@comment ISO/IEC TS 18661-1
 @deftypefunx int strfroml (char *restrict @var{string}, size_t @var{size}, const char *restrict @var{format}, long double @var{value})
 @safety{@prelim{}@mtsafe{@mtslocale{}}@asunsafe{@ascuheap{}}@acunsafe{@acsmem{}}}
-@comment these functions depend on __printf_fp and __printf_fphex, which are
-@comment AS-unsafe (ascuheap) and AC-unsafe (acsmem).
+@comment All these functions depend on both __printf_fp and __printf_fphex,
+@comment which are both AS-unsafe (ascuheap) and AC-unsafe (acsmem).
 The functions @code{strfromd} (``string-from-double''), @code{strfromf}
 (``string-from-float''), and @code{strfroml} (``string-from-long-double'')
 convert the floating-point number @var{value} to a string of characters and
@@ -3064,6 +3098,21 @@ has been completely written if and only if the returned value is less than
 These functions were introduced by ISO/IEC TS 18661-1.
 @end deftypefun
 
+@comment stdlib.h
+@comment ISO/IEC TS 18661-3
+@deftypefun int strfromfN (char *restrict @var{string}, size_t @var{size}, const char *restrict @var{format}, _Float@var{N} @var{value})
+@comment stdlib.h
+@comment ISO/IEC TS 18661-3
+@deftypefunx int strfromfNx (char *restrict @var{string}, size_t @var{size}, const char *restrict @var{format}, _Float@var{N}x @var{value})
+@safety{@prelim{}@mtsafe{@mtslocale{}}@asunsafe{@ascuheap{}}@acunsafe{@acsmem{}}}
+@comment See safety comments for strfromd.
+These functions are like @code{strfromd}, except for the type of
+@code{value}.
+
+They were introduced in @w{ISO/IEC TS 18661-3} and are available on machines
+that support the related types; @pxref{Mathematics}.
+@end deftypefun
+
 @node System V Number Conversion
 @section Old-fashioned System V number-to-string functions
 
diff --git a/manual/math.texi b/manual/math.texi
index 69a0ace..53c2bc1 100644
--- a/manual/math.texi
+++ b/manual/math.texi
@@ -53,6 +53,17 @@ usually wise to pick the narrowest type that can accommodate your data.
 Not all machines have a distinct @code{long double} type; it may be the
 same as @code{double}.
 
+On some machines, @theglibc{} also provides @code{_Float@var{N}} and
+@code{_Float@var{N}x} types.  These types are defined in @w{ISO/IEC TS
+18661-3}, which extends @w{ISO C} and defines floating-point types that
+are not machine-dependent.  When such a type, such as @code{_Float128},
+is supported by @theglibc{}, extra variants for most of the mathematical
+functions provided for @code{double}, @code{float}, and @code{long
+double} are also provided for the supported type.
+
+Currently, support for @code{_Float@var{N}} or @code{_Float@var{N}x}
+types is not provided for any machine.
+
 @menu
 * Mathematical Constants::      Precise numeric values for often-used
                                  constants.
-- 
2.4.11

[PATCH v4 0/2] Add string functions and libm tests for _Float128

[Gabriel F. T. Gomes]

I added a "Changes since v3" to each patch in this series.

Gabriel F. T. Gomes (1):
  float128: Add strfromf128, strtof128, and wcstof128 to the manual

Paul E. Murphy (1):
  float128: Add strtof128, wcstof128, and related functions.

 include/gmp.h                                  |  4 ++
 include/stdlib.h                               | 29 ++++++++++++++
 include/wchar.h                                | 13 +++++++
 manual/arith.texi                              | 53 ++++++++++++++++++++++---
 manual/math.texi                               | 11 ++++++
 stdlib/Makefile                                |  1 +
 stdlib/stdlib.h                                | 14 +++++++
 stdlib/tst-strtod-nan-locale-main.c            | 36 ++++++++---------
 stdlib/tst-strtod-round-skeleton.c             |  1 +
 stdlib/tst-strtod.h                            | 36 +++++++++++------
 sysdeps/ieee754/float128/Makefile              |  5 +++
 sysdeps/ieee754/float128/Versions              | 10 +++++
 sysdeps/ieee754/float128/mpn2float128.c        | 54 ++++++++++++++++++++++++++
 sysdeps/ieee754/float128/strtod_nan_float128.h | 33 ++++++++++++++++
 sysdeps/ieee754/float128/strtof128.c           | 34 ++++++++++++++++
 sysdeps/ieee754/float128/strtof128_l.c         | 43 ++++++++++++++++++++
 sysdeps/ieee754/float128/strtof128_nan.c       | 28 +++++++++++++
 sysdeps/ieee754/float128/wcstof128.c           | 30 ++++++++++++++
 sysdeps/ieee754/float128/wcstof128_l.c         | 30 ++++++++++++++
 sysdeps/ieee754/float128/wcstof128_nan.c       | 24 ++++++++++++
 wcsmbs/Makefile                                |  2 +
 wcsmbs/wchar.h                                 | 15 +++++++
 22 files changed, 472 insertions(+), 34 deletions(-)
 create mode 100644 sysdeps/ieee754/float128/mpn2float128.c
 create mode 100644 sysdeps/ieee754/float128/strtod_nan_float128.h
 create mode 100644 sysdeps/ieee754/float128/strtof128.c
 create mode 100644 sysdeps/ieee754/float128/strtof128_l.c
 create mode 100644 sysdeps/ieee754/float128/strtof128_nan.c
 create mode 100644 sysdeps/ieee754/float128/wcstof128.c
 create mode 100644 sysdeps/ieee754/float128/wcstof128_l.c
 create mode 100644 sysdeps/ieee754/float128/wcstof128_nan.c

-- 
2.4.11

Re: [PATCH v4 0/2] Add string functions and libm tests for _Float128

[Joseph Myers]

These patches are OK.

-- 
Joseph S. Myers
joseph@codesourcery.com

Re: [PATCH v4 2/2] float128: Add strfromf128, strtof128, and wcstof128 to the manual

[Rical Jasan]

LGTM, thank you.

Rical

Re: [PATCH v4 0/2] Add string functions and libm tests for _Float128

[Gabriel F. T. Gomes]

On Fri, 9 Jun 2017 20:27:54 +0000
Joseph Myers <joseph@codesourcery.com> wrote:

> These patches are OK.

Thanks.  I have commit these and the remaining patches from
https://sourceware.org/ml/libc-alpha/2017-06/msg00131.html