Fix lrint, llrint, lround, llround missing exceptions for MIPS (bug 16399).

For 32-bit MIPS and some other systems, various of the lrint, llrint,
lround, llround functions can be missing exceptions on overflow
because casts do not (in current GCC) result in the proper
exceptions.  In the MIPS case there are two problems here: MIPS I code
generation uses an assembler macro that doesn't raise exceptions,
while the libgcc conversions of floating-point values to long long
also do not raise "invalid" on all overflow cases (and can raise
spurious "inexact").

This patch adds support in the generic code (only the functions for
which this problem has actually been seen) for forcing the "invalid"
exception in the problem cases, and enables that support for the
affected MIPS cases.

Tested for MIPS; also tested for x86_64 and x86 that installed
stripped shared libraries are unchanged by this patch.

	[BZ #16399]
	* sysdeps/generic/fix-fp-int-convert-overflow.h: New file.
	* sysdeps/ieee754/dbl-64/s_llrint.c: Include <fenv.h>, <limits.h>
	and <fix-fp-int-convert-overflow.h>.
	(__llrint) [FE_INVALID]: Force FE_INVALID exception as needed if
	FIX_DBL_LLONG_CONVERT_OVERFLOW.
	* sysdeps/ieee754/dbl-64/s_llround.c: Include <fenv.h>, <limits.h>
	and <fix-fp-int-convert-overflow.h>.
	(__llround) [FE_INVALID]: Force FE_INVALID exception as needed if
	FIX_DBL_LLONG_CONVERT_OVERFLOW.
	* sysdeps/ieee754/dbl-64/s_lrint.c: Include
	<fix-fp-int-convert-overflow.h>.
	(__lrint) [FE_INVALID]: Force FE_INVALID exception as needed if
	FIX_DBL_LLONG_CONVERT_OVERFLOW.
	* sysdeps/ieee754/dbl-64/s_lround.c: Include
	<fix-fp-int-convert-overflow.h>.
	(__lround) [FE_INVALID]: Force FE_INVALID exception as needed if
	FIX_DBL_LLONG_CONVERT_OVERFLOW.
	* sysdeps/ieee754/flt-32/s_llrintf.c: Include <fenv.h>, <limits.h>
	and <fix-fp-int-convert-overflow.h>.
	(__llrintf) [FE_INVALID]: Force FE_INVALID exception as needed if
	FIX_DBL_LLONG_CONVERT_OVERFLOW.
	* sysdeps/ieee754/flt-32/s_llroundf.c: Include <fenv.h>,
	<limits.h> and <fix-fp-int-convert-overflow.h>.
	(__llroundf) [FE_INVALID]: Force FE_INVALID exception as needed if
	FIX_DBL_LLONG_CONVERT_OVERFLOW.
	* sysdeps/ieee754/flt-32/s_lrintf.c: Include <fenv.h>, <limits.h>
	and <fix-fp-int-convert-overflow.h>.
	(__lrintf) [FE_INVALID]: Force FE_INVALID exception as needed if
	FIX_DBL_LLONG_CONVERT_OVERFLOW.
	* sysdeps/ieee754/flt-32/s_lroundf.c: Include <fenv.h>, <limits.h>
	and <fix-fp-int-convert-overflow.h>.
	(__lroundf) [FE_INVALID]: Force FE_INVALID exception as needed if
	FIX_DBL_LLONG_CONVERT_OVERFLOW.
	* sysdeps/mips/mips32/fpu/fix-fp-int-convert-overflow.h: New file.

12 files changed, 212 insertions, 24 deletions

diff --git a/ChangeLog b/ChangeLog
index c47fd6d899..1e9a477895 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,41 @@
+2015-10-09  Joseph Myers  <joseph@codesourcery.com>
+
+	[BZ #16399]
+	* sysdeps/generic/fix-fp-int-convert-overflow.h: New file.
+	* sysdeps/ieee754/dbl-64/s_llrint.c: Include <fenv.h>, <limits.h>
+	and <fix-fp-int-convert-overflow.h>.
+	(__llrint) [FE_INVALID]: Force FE_INVALID exception as needed if
+	FIX_DBL_LLONG_CONVERT_OVERFLOW.
+	* sysdeps/ieee754/dbl-64/s_llround.c: Include <fenv.h>, <limits.h>
+	and <fix-fp-int-convert-overflow.h>.
+	(__llround) [FE_INVALID]: Force FE_INVALID exception as needed if
+	FIX_DBL_LLONG_CONVERT_OVERFLOW.
+	* sysdeps/ieee754/dbl-64/s_lrint.c: Include
+	<fix-fp-int-convert-overflow.h>.
+	(__lrint) [FE_INVALID]: Force FE_INVALID exception as needed if
+	FIX_DBL_LLONG_CONVERT_OVERFLOW.
+	* sysdeps/ieee754/dbl-64/s_lround.c: Include
+	<fix-fp-int-convert-overflow.h>.
+	(__lround) [FE_INVALID]: Force FE_INVALID exception as needed if
+	FIX_DBL_LLONG_CONVERT_OVERFLOW.
+	* sysdeps/ieee754/flt-32/s_llrintf.c: Include <fenv.h>, <limits.h>
+	and <fix-fp-int-convert-overflow.h>.
+	(__llrintf) [FE_INVALID]: Force FE_INVALID exception as needed if
+	FIX_DBL_LLONG_CONVERT_OVERFLOW.
+	* sysdeps/ieee754/flt-32/s_llroundf.c: Include <fenv.h>,
+	<limits.h> and <fix-fp-int-convert-overflow.h>.
+	(__llroundf) [FE_INVALID]: Force FE_INVALID exception as needed if
+	FIX_DBL_LLONG_CONVERT_OVERFLOW.
+	* sysdeps/ieee754/flt-32/s_lrintf.c: Include <fenv.h>, <limits.h>
+	and <fix-fp-int-convert-overflow.h>.
+	(__lrintf) [FE_INVALID]: Force FE_INVALID exception as needed if
+	FIX_DBL_LLONG_CONVERT_OVERFLOW.
+	* sysdeps/ieee754/flt-32/s_lroundf.c: Include <fenv.h>, <limits.h>
+	and <fix-fp-int-convert-overflow.h>.
+	(__lroundf) [FE_INVALID]: Force FE_INVALID exception as needed if
+	FIX_DBL_LLONG_CONVERT_OVERFLOW.
+	* sysdeps/mips/mips32/fpu/fix-fp-int-convert-overflow.h: New file.
+
 2015-10-09  Carlos O'Donell  <carlos@redhat.com>
 
 	[BZ #18589]
diff --git a/NEWS b/NEWS
index 5c239f3988..32e4c64c07 100644
--- a/NEWS
+++ b/NEWS
@@ -10,16 +10,16 @@ Version 2.23
 * The following bugs are resolved with this release:
 
   887, 2542, 2543, 2558, 2898, 4404, 6803, 10432, 14341, 14912, 15367,
-  15384, 15786, 15918, 16141, 16296, 16347, 16415, 16517, 16519, 16520,
-  16521, 16620, 16734, 16973, 16985, 17118, 17243, 17244, 17250, 17441,
-  17787, 17886, 17887, 17905, 18084, 18086, 18240, 18265, 18370, 18421,
-  18480, 18525, 18595, 18589, 18610, 18618, 18647, 18661, 18674, 18675,
-  18681, 18724, 18757, 18778, 18781, 18787, 18789, 18790, 18795, 18796,
-  18803, 18820, 18823, 18824, 18825, 18857, 18863, 18870, 18872, 18873,
-  18875, 18887, 18921, 18951, 18952, 18956, 18961, 18966, 18967, 18969,
-  18970, 18977, 18980, 18981, 18985, 19003, 19012, 19016, 19018, 19032,
-  19046, 19049, 19050, 19059, 19071, 19076, 19077, 19078, 19079, 19085,
-  19086, 19088, 19094, 19095.
+  15384, 15786, 15918, 16141, 16296, 16347, 16399, 16415, 16517, 16519,
+  16520, 16521, 16620, 16734, 16973, 16985, 17118, 17243, 17244, 17250,
+  17441, 17787, 17886, 17887, 17905, 18084, 18086, 18240, 18265, 18370,
+  18421, 18480, 18525, 18595, 18589, 18610, 18618, 18647, 18661, 18674,
+  18675, 18681, 18724, 18757, 18778, 18781, 18787, 18789, 18790, 18795,
+  18796, 18803, 18820, 18823, 18824, 18825, 18857, 18863, 18870, 18872,
+  18873, 18875, 18887, 18921, 18951, 18952, 18956, 18961, 18966, 18967,
+  18969, 18970, 18977, 18980, 18981, 18985, 19003, 19012, 19016, 19018,
+  19032, 19046, 19049, 19050, 19059, 19071, 19076, 19077, 19078, 19079,
+  19085, 19086, 19088, 19094, 19095.
 
 * The obsolete header <regexp.h> has been removed.  Programs that require
   this header must be updated to use <regex.h> instead.
diff --git a/sysdeps/generic/fix-fp-int-convert-overflow.h b/sysdeps/generic/fix-fp-int-convert-overflow.h
new file mode 100644
index 0000000000..d29211b77f
--- /dev/null
+++ b/sysdeps/generic/fix-fp-int-convert-overflow.h
@@ -0,0 +1,31 @@
+/* Fix for conversion of floating point to integer overflow.  Generic version.
+   Copyright (C) 2015 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/>.  */
+
+#ifndef FIX_FP_INT_CONVERT_OVERFLOW_H
+#define FIX_FP_INT_CONVERT_OVERFLOW_H	1
+
+/* Define these macros to 1 to workaround conversions of out-of-range
+   floating-point numbers to integer types failing to raise the
+   "invalid" exception, or raising spurious "inexact" or other
+   exceptions.  */
+#define FIX_FLT_LONG_CONVERT_OVERFLOW 0
+#define FIX_FLT_LLONG_CONVERT_OVERFLOW 0
+#define FIX_DBL_LONG_CONVERT_OVERFLOW 0
+#define FIX_DBL_LLONG_CONVERT_OVERFLOW 0
+
+#endif /* fix-fp-int-convert-overflow.h */
diff --git a/sysdeps/ieee754/dbl-64/s_llrint.c b/sysdeps/ieee754/dbl-64/s_llrint.c
index 048331e9a7..1673527ba4 100644
--- a/sysdeps/ieee754/dbl-64/s_llrint.c
+++ b/sysdeps/ieee754/dbl-64/s_llrint.c
@@ -18,9 +18,12 @@
    License along with the GNU C Library; if not, see
    <http://www.gnu.org/licenses/>.  */
 
+#include <fenv.h>
+#include <limits.h>
 #include <math.h>
 
 #include <math_private.h>
+#include <fix-fp-int-convert-overflow.h>
 
 static const double two52[2] =
 {
@@ -77,8 +80,16 @@ __llrint (double x)
     }
   else
     {
-      /* The number is too large.  It is left implementation defined
-	 what happens.  */
+#ifdef FE_INVALID
+      /* The number is too large.  Unless it rounds to LLONG_MIN,
+	 FE_INVALID must be raised and the return value is
+	 unspecified.  */
+      if (FIX_DBL_LLONG_CONVERT_OVERFLOW && x != (double) LLONG_MIN)
+	{
+	  feraiseexcept (FE_INVALID);
+	  return sx == 0 ? LLONG_MAX : LLONG_MIN;
+	}
+#endif
       return (long long int) x;
     }
 
diff --git a/sysdeps/ieee754/dbl-64/s_llround.c b/sysdeps/ieee754/dbl-64/s_llround.c
index c8b3c190b5..77ad3c3e89 100644
--- a/sysdeps/ieee754/dbl-64/s_llround.c
+++ b/sysdeps/ieee754/dbl-64/s_llround.c
@@ -17,9 +17,12 @@
    License along with the GNU C Library; if not, see
    <http://www.gnu.org/licenses/>.  */
 
+#include <fenv.h>
+#include <limits.h>
 #include <math.h>
 
 #include <math_private.h>
+#include <fix-fp-int-convert-overflow.h>
 
 
 long long int
@@ -65,8 +68,16 @@ __llround (double x)
     }
   else
     {
-      /* The number is too large.  It is left implementation defined
-         what happens.  */
+#ifdef FE_INVALID
+      /* The number is too large.  Unless it rounds to LLONG_MIN,
+	 FE_INVALID must be raised and the return value is
+	 unspecified.  */
+      if (FIX_DBL_LLONG_CONVERT_OVERFLOW && x != (double) LLONG_MIN)
+	{
+	  feraiseexcept (FE_INVALID);
+	  return sign == 1 ? LLONG_MAX : LLONG_MIN;
+	}
+#endif
       return (long long int) x;
     }
 
diff --git a/sysdeps/ieee754/dbl-64/s_lrint.c b/sysdeps/ieee754/dbl-64/s_lrint.c
index d004594bc2..3fdb05ecf7 100644
--- a/sysdeps/ieee754/dbl-64/s_lrint.c
+++ b/sysdeps/ieee754/dbl-64/s_lrint.c
@@ -23,6 +23,7 @@
 #include <math.h>
 
 #include <math_private.h>
+#include <fix-fp-int-convert-overflow.h>
 
 static const double two52[2] =
 {
@@ -107,6 +108,11 @@ __lrint (double x)
 	  feraiseexcept (t == LONG_MIN ? FE_INEXACT : FE_INVALID);
 	  return LONG_MIN;
 	}
+      else if (FIX_DBL_LONG_CONVERT_OVERFLOW && x != (double) LONG_MIN)
+	{
+	  feraiseexcept (FE_INVALID);
+	  return sx == 0 ? LONG_MAX : LONG_MIN;
+	}
 #endif
       return (long int) x;
     }
diff --git a/sysdeps/ieee754/dbl-64/s_lround.c b/sysdeps/ieee754/dbl-64/s_lround.c
index 91b17b0de9..699eafb010 100644
--- a/sysdeps/ieee754/dbl-64/s_lround.c
+++ b/sysdeps/ieee754/dbl-64/s_lround.c
@@ -22,6 +22,7 @@
 #include <math.h>
 
 #include <math_private.h>
+#include <fix-fp-int-convert-overflow.h>
 
 
 long int
@@ -80,8 +81,18 @@ __lround (double x)
 	 FE_INVALID must be raised and the return value is
 	 unspecified.  */
 #ifdef FE_INVALID
-      if (sizeof (long int) == 4
-	  && x <= (double) LONG_MIN - 0.5)
+      if (FIX_DBL_LONG_CONVERT_OVERFLOW
+	  && !(sign == -1
+	       && (sizeof (long int) == 4
+		   ? x > (double) LONG_MIN - 0.5
+		   : x >= (double) LONG_MIN)))
+	{
+	  feraiseexcept (FE_INVALID);
+	  return sign == 1 ? LONG_MAX : LONG_MIN;
+	}
+      else if (!FIX_DBL_LONG_CONVERT_OVERFLOW
+	       && sizeof (long int) == 4
+	       && x <= (double) LONG_MIN - 0.5)
 	{
 	  /* If truncation produces LONG_MIN, the cast will not raise
 	     the exception, but may raise "inexact".  */
diff --git a/sysdeps/ieee754/flt-32/s_llrintf.c b/sysdeps/ieee754/flt-32/s_llrintf.c
index d7588d04cf..debac8d3cc 100644
--- a/sysdeps/ieee754/flt-32/s_llrintf.c
+++ b/sysdeps/ieee754/flt-32/s_llrintf.c
@@ -18,9 +18,12 @@
    License along with the GNU C Library; if not, see
    <http://www.gnu.org/licenses/>.  */
 
+#include <fenv.h>
+#include <limits.h>
 #include <math.h>
 
 #include <math_private.h>
+#include <fix-fp-int-convert-overflow.h>
 
 static const float two23[2] =
 {
@@ -64,8 +67,16 @@ __llrintf (float x)
     }
   else
     {
-      /* The number is too large.  It is left implementation defined
-	 what happens.  */
+#ifdef FE_INVALID
+      /* The number is too large.  Unless it rounds to LLONG_MIN,
+	 FE_INVALID must be raised and the return value is
+	 unspecified.  */
+      if (FIX_FLT_LLONG_CONVERT_OVERFLOW && x != (float) LLONG_MIN)
+	{
+	  feraiseexcept (FE_INVALID);
+	  return sx == 0 ? LLONG_MAX : LLONG_MIN;
+	}
+#endif
       return (long long int) x;
     }
 
diff --git a/sysdeps/ieee754/flt-32/s_llroundf.c b/sysdeps/ieee754/flt-32/s_llroundf.c
index f695c63710..0404231dfb 100644
--- a/sysdeps/ieee754/flt-32/s_llroundf.c
+++ b/sysdeps/ieee754/flt-32/s_llroundf.c
@@ -17,9 +17,12 @@
    License along with the GNU C Library; if not, see
    <http://www.gnu.org/licenses/>.  */
 
+#include <fenv.h>
+#include <limits.h>
 #include <math.h>
 
 #include <math_private.h>
+#include <fix-fp-int-convert-overflow.h>
 
 
 long long int
@@ -51,8 +54,16 @@ __llroundf (float x)
     }
   else
     {
-      /* The number is too large.  It is left implementation defined
-	 what happens.  */
+#ifdef FE_INVALID
+      /* The number is too large.  Unless it rounds to LLONG_MIN,
+	 FE_INVALID must be raised and the return value is
+	 unspecified.  */
+      if (FIX_FLT_LLONG_CONVERT_OVERFLOW && x != (float) LLONG_MIN)
+	{
+	  feraiseexcept (FE_INVALID);
+	  return sign == 1 ? LLONG_MAX : LLONG_MIN;
+	}
+#endif
       return (long long int) x;
     }
 
diff --git a/sysdeps/ieee754/flt-32/s_lrintf.c b/sysdeps/ieee754/flt-32/s_lrintf.c
index bc60fd2311..2eb242d828 100644
--- a/sysdeps/ieee754/flt-32/s_lrintf.c
+++ b/sysdeps/ieee754/flt-32/s_lrintf.c
@@ -18,9 +18,12 @@
    License along with the GNU C Library; if not, see
    <http://www.gnu.org/licenses/>.  */
 
+#include <fenv.h>
+#include <limits.h>
 #include <math.h>
 
 #include <math_private.h>
+#include <fix-fp-int-convert-overflow.h>
 
 static const float two23[2] =
 {
@@ -64,8 +67,16 @@ __lrintf (float x)
     }
   else
     {
-      /* The number is too large.  It is left implementation defined
-	 what happens.  */
+#ifdef FE_INVALID
+      /* The number is too large.  Unless it rounds to LONG_MIN,
+	 FE_INVALID must be raised and the return value is
+	 unspecified.  */
+      if (FIX_FLT_LONG_CONVERT_OVERFLOW && x != (float) LONG_MIN)
+	{
+	  feraiseexcept (FE_INVALID);
+	  return sx == 0 ? LONG_MAX : LONG_MIN;
+	}
+#endif
       return (long int) x;
     }
 
diff --git a/sysdeps/ieee754/flt-32/s_lroundf.c b/sysdeps/ieee754/flt-32/s_lroundf.c
index f20352a3c8..da3551c27d 100644
--- a/sysdeps/ieee754/flt-32/s_lroundf.c
+++ b/sysdeps/ieee754/flt-32/s_lroundf.c
@@ -17,9 +17,12 @@
    License along with the GNU C Library; if not, see
    <http://www.gnu.org/licenses/>.  */
 
+#include <fenv.h>
+#include <limits.h>
 #include <math.h>
 
 #include <math_private.h>
+#include <fix-fp-int-convert-overflow.h>
 
 
 long int
@@ -51,8 +54,16 @@ __lroundf (float x)
     }
   else
     {
-      /* The number is too large.  It is left implementation defined
-	 what happens.  */
+#ifdef FE_INVALID
+      /* The number is too large.  Unless it rounds to LONG_MIN,
+	 FE_INVALID must be raised and the return value is
+	 unspecified.  */
+      if (FIX_FLT_LONG_CONVERT_OVERFLOW && x != (float) LONG_MIN)
+	{
+	  feraiseexcept (FE_INVALID);
+	  return sign == 1 ? LONG_MAX : LONG_MIN;
+	}
+#endif
       return (long int) x;
     }
 
diff --git a/sysdeps/mips/mips32/fpu/fix-fp-int-convert-overflow.h b/sysdeps/mips/mips32/fpu/fix-fp-int-convert-overflow.h
new file mode 100644
index 0000000000..ee14f7e1b8
--- /dev/null
+++ b/sysdeps/mips/mips32/fpu/fix-fp-int-convert-overflow.h
@@ -0,0 +1,36 @@
+/* Fix for conversion of floating point to integer overflow.  MIPS version.
+   Copyright (C) 2015 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/>.  */
+
+#ifndef FIX_FP_INT_CONVERT_OVERFLOW_H
+#define FIX_FP_INT_CONVERT_OVERFLOW_H	1
+
+/* As of GCC 5, the generic libgcc2.c conversions from floating point
+   to long long may not raise the correct exceptions on overflow (and
+   may raise spurious "inexact" exceptions even in non-overflow cases,
+   see <https://gcc.gnu.org/bugzilla/show_bug.cgi?id=59412>).  */
+#define FIX_FLT_LLONG_CONVERT_OVERFLOW 1
+#define FIX_DBL_LLONG_CONVERT_OVERFLOW 1
+
+/* As of GCC 5 and binutils 2.25, for MIPS I GCC generates calls to
+   assembler macros for conversions from floating point to integer
+   types, and those macros save and restore the whole of FSCR and so
+   lose exceptions.  */
+#define FIX_FLT_LONG_CONVERT_OVERFLOW (__mips == 1)
+#define FIX_DBL_LONG_CONVERT_OVERFLOW (__mips == 1)
+
+#endif /* fix-fp-int-convert-overflow.h */