mirror of
https://sourceware.org/git/glibc.git
synced 2024-11-27 03:33:33 +08:00
* sysdeps/ieee754/ldbl-128ibm/s_nearbyintl.c (__nearbyintl): Use
__nextafter instead of nextafter to avoid local PLT.
This commit is contained in:
parent
c3a0ead420
commit
3b6d574e01
@ -1,5 +1,7 @@
|
||||
2008-04-11 Ulrich Drepper <drepper@redhat.com>
|
||||
|
||||
* sysdeps/ieee754/ldbl-128ibm/s_nearbyintl.c (__nearbyintl): Use
|
||||
__nextafter instead of nextafter to avoid local PLT.
|
||||
* sysdeps/powerpc/fpu/e_sqrt.c: Avoid call to fetestexcept.
|
||||
* sysdeps/powerpc/fpu/e_sqrtf.c: Likewise.
|
||||
|
||||
|
@ -1,6 +1,6 @@
|
||||
/* Round to int long double floating-point values without raising inexact.
|
||||
IBM extended format long double version.
|
||||
Copyright (C) 2006 Free Software Foundation, Inc.
|
||||
Copyright (C) 2006, 2008 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
|
||||
@ -84,8 +84,8 @@ __nearbyintl (x)
|
||||
{
|
||||
/* Else the high double is pre rounded and we need to
|
||||
adjust for that. */
|
||||
|
||||
tau = nextafter (u.dd[0], 0.0);
|
||||
|
||||
tau = __nextafter (u.dd[0], 0.0);
|
||||
tau = (u.dd[0] - tau) * 2.0;
|
||||
high = u.dd[0] - tau;
|
||||
low = u.dd[1] + tau;
|
||||
@ -106,7 +106,7 @@ __nearbyintl (x)
|
||||
{
|
||||
/* Else the high double is pre rounded and we need to
|
||||
adjust for that. */
|
||||
tau = nextafter (u.dd[0], 0.0);
|
||||
tau = __nextafter (u.dd[0], 0.0);
|
||||
tau = (u.dd[0] - tau) * 2.0;
|
||||
high = u.dd[0] - tau;
|
||||
low = u.dd[1] + tau;
|
||||
|
Loading…
Reference in New Issue
Block a user