S390: Use load-fp-integer instruction for roundeven functions.

If compiled with z196 zarch support, the load-fp-integer instruction
is used to implement roundeven, roundevenf, roundevenl.
Otherwise the common-code implementation is used.

sysdeps/s390/fpu/s_roundeven.c

@@ -0,0 +1,39 @@
+/* roundeven() - S390 version.
+   Copyright (C) 2019 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
+   <https://www.gnu.org/licenses/>.  */
+
+#ifdef HAVE_S390_MIN_Z196_ZARCH_ASM_SUPPORT
+# include <math.h>
+# include <libm-alias-double.h>
+
+double
+__roundeven (double x)
+{
+  double y;
+  /* The z196 zarch "load fp integer" (fidbra) instruction is rounding
+     x to the nearest integer with "ties to even" rounding mode
+     (M3-field: 4) where inexact exceptions are suppressed (M4-field: 4).  */
+  __asm__ ("fidbra %0,4,%1,4" : "=f" (y) : "f" (x));
+  return y;
+}
+hidden_def (__roundeven)
+libm_alias_double (__roundeven, roundeven)
+
+#else
+# include <sysdeps/ieee754/dbl-64/s_roundeven.c>
+#endif

sysdeps/s390/fpu/s_roundevenf.c

@@ -0,0 +1,38 @@
+/* roundevenf() - S390 version.
+   Copyright (C) 2019 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
+   <https://www.gnu.org/licenses/>.  */
+
+#ifdef HAVE_S390_MIN_Z196_ZARCH_ASM_SUPPORT
+# include <math.h>
+# include <libm-alias-float.h>
+
+float
+__roundevenf (float x)
+{
+  float y;
+  /* The z196 zarch "load fp integer" (fiebra) instruction is rounding
+     x to the nearest integer with "ties to even" rounding mode
+     (M3-field: 4) where inexact exceptions are suppressed (M4-field: 4).  */
+  __asm__ ("fiebra %0,4,%1,4" : "=f" (y) : "f" (x));
+  return y;
+}
+libm_alias_float (__roundeven, roundeven)
+
+#else
+# include <sysdeps/ieee754/flt-32/s_roundevenf.c>
+#endif

sysdeps/s390/fpu/s_roundevenl.c

@@ -0,0 +1,39 @@
+/* roundevenl() - S390 version.
+   Copyright (C) 2019 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
+   <https://www.gnu.org/licenses/>.  */
+
+#ifdef HAVE_S390_MIN_Z196_ZARCH_ASM_SUPPORT
+# include <math.h>
+# include <math_private.h>
+# include <libm-alias-ldouble.h>
+
+_Float128
+__roundevenl (_Float128 x)
+{
+  _Float128 y;
+  /* The z196 zarch "load fp integer" (fixbra) instruction is rounding
+     x to the nearest integer with "ties to even" rounding mode
+     (M3-field: 4) where inexact exceptions are suppressed (M4-field: 4).  */
+  __asm__ ("fixbra %0,4,%1,4" : "=f" (y) : "f" (x));
+  return y;
+}
+libm_alias_ldouble (__roundeven, roundeven)
+
+#else
+# include <sysdeps/ieee754/ldbl-128/s_roundevenl.c>
+#endif