about summary refs log tree commit diff
path: root/sysdeps/ieee754/dbl-64/s_fmaf.c
diff options
context:
space:
mode:
authorJakub Jelinek <jakub@redhat.com>2010-10-11 09:27:05 -0400
committerUlrich Drepper <drepper@gmail.com>2010-10-11 09:27:05 -0400
commit9ff8d36f27aaf390fd074976cd2e867ba41f2588 (patch)
tree694d6d7482b202d35b90d9e958d322f4c4a1b83b /sysdeps/ieee754/dbl-64/s_fmaf.c
parentc579b202990c48710e133b58b44ada32d0a7d458 (diff)
downloadglibc-9ff8d36f27aaf390fd074976cd2e867ba41f2588.tar.gz
glibc-9ff8d36f27aaf390fd074976cd2e867ba41f2588.tar.xz
glibc-9ff8d36f27aaf390fd074976cd2e867ba41f2588.zip
Correct implementation of fmaf.
Diffstat (limited to 'sysdeps/ieee754/dbl-64/s_fmaf.c')
-rw-r--r--sysdeps/ieee754/dbl-64/s_fmaf.c50
1 files changed, 50 insertions, 0 deletions
diff --git a/sysdeps/ieee754/dbl-64/s_fmaf.c b/sysdeps/ieee754/dbl-64/s_fmaf.c
new file mode 100644
index 0000000000..9721876601
--- /dev/null
+++ b/sysdeps/ieee754/dbl-64/s_fmaf.c
@@ -0,0 +1,50 @@
+/* Compute x * y + z as ternary operation.
+   Copyright (C) 2010 Free Software Foundation, Inc.
+   This file is part of the GNU C Library.
+   Contributed by Jakub Jelinek <jakub@redhat.com>, 2010.
+
+   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, write to the Free
+   Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
+   02111-1307 USA.  */
+
+#include <math.h>
+#include <fenv.h>
+#include <ieee754.h>
+
+/* This implementation relies on double being more than twice as
+   precise as float and uses rounding to odd in order to avoid problems
+   with double rounding.
+   See a paper by Boldo and Melquiond:
+   http://www.lri.fr/~melquion/doc/08-tc.pdf  */
+
+float
+__fmaf (float x, float y, float z)
+{
+  fenv_t env;
+  /* Multiplication is always exact.  */
+  double temp = (double) x * (double) y;
+  union ieee754_double u;
+  feholdexcept (&env);
+  fesetround (FE_TOWARDZERO);
+  /* Perform addition with round to odd.  */
+  u.d = temp + (double) z;
+  if ((u.ieee.mantissa1 & 1) == 0 && u.ieee.exponent != 0xff)
+    u.ieee.mantissa1 |= fetestexcept (FE_INEXACT) != 0;
+  feupdateenv (&env);
+  /* And finally truncation with round to nearest.  */
+  return (float) u.d;
+}
+#ifndef __fmaf
+weak_alias (__fmaf, fmaf)
+#endif