* sysdeps/i386/fpu/s_expm1l.S: Simply use exp implementation for large

parameters.
	* sysdeps/x86_64/fpu/s_expm1l.S: Likewise.
	Patch by Denys Vlasenko <dvlasenk@redhat.com>.

	* nscd/connections.c (nscd_init): Typo in preprocessor directive.
This commit is contained in:
Ulrich Drepper 2008-08-05 22:08:42 +00:00
parent c400c931b8
commit 9523fd2806
4 changed files with 23 additions and 6 deletions

View File

@ -1,3 +1,10 @@
2008-08-05 Ulrich Drepper <drepper@redhat.com>
* sysdeps/i386/fpu/s_expm1l.S: Simply use exp implementation for large
parameters.
* sysdeps/x86_64/fpu/s_expm1l.S: Likewise.
Patch by Denys Vlasenko <dvlasenk@redhat.com>.
2008-08-03 Ulrich Drepper <drepper@redhat.com>
* nscd/connections.c (main_loop_poll): Pass a buffer which is
@ -83,7 +90,7 @@
* sysdeps/posix/getaddrinfo.c (gaih_inet): Raise size of initial
buffer passed to NSS functions.
* nscd/connections.c (nscd_init): Type if preprocessor directive.
* nscd/connections.c (nscd_init): Typo in preprocessor directive.
* sysdeps/unix/sysv/linux/kernel-features.h (__ASSUME_PACCEPT):
Define.

View File

@ -1,5 +1,5 @@
/* ix87 specific implementation of exp(x)-1.
Copyright (C) 1996, 1997, 2002, 2005 Free Software Foundation, Inc.
Copyright (C) 1996, 1997, 2002, 2005, 2008 Free Software Foundation, Inc.
This file is part of the GNU C Library.
Contributed by Ulrich Drepper <drepper@cygnus.com>, 1996.
Based on code by John C. Bowman <bowman@ipp-garching.mpg.de>.
@ -48,6 +48,11 @@ l2e: .tfloat 1.442695040888963407359924681002
.text
ENTRY(__expm1l)
movzwl 4+8(%esp), %eax // load sign bit and 15-bit exponent
xorb $0x80, %ah // invert sign bit (now 1 is "positive")
cmpl $0xc006, %eax // is num positive and exp >= 6 (number is >= 128.0)?
jae __ieee754_expl // (if num is denormal, it is at least >= 64.0)
fldt 4(%esp) // x
fxam // Is NaN or +-Inf?
fstsw %ax

View File

@ -1,5 +1,5 @@
/* clock_getres -- Get the resolution of a POSIX clockid_t. Linux version.
Copyright (C) 2003,2004,2005,2006 Free Software Foundation, Inc.
Copyright (C) 2003,2004,2005,2006, 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

View File

@ -1,5 +1,5 @@
/* ix87 specific implementation of exp(x)-1.
Copyright (C) 1996, 1997, 2001, 2002 Free Software Foundation, Inc.
Copyright (C) 1996, 1997, 2001, 2002, 2008 Free Software Foundation, Inc.
This file is part of the GNU C Library.
Contributed by Ulrich Drepper <drepper@cygnus.com>, 1996.
Based on code by John C. Bowman <bowman@ipp-garching.mpg.de>.
@ -48,6 +48,11 @@ l2e: .tfloat 1.442695040888963407359924681002
.text
ENTRY(__expm1l)
movzwl 8+8(%rsp), %eax // load sign bit and 15-bit exponent
xorb $0x80, %ah // invert sign bit (now 1 is "positive")
cmpl $0xc006, %eax // is num positive and exp >= 6 (number is >= 128.0)?
jae __ieee754_expl // (if num is denormal, it is at least >= 64.0)
fldt 8(%rsp) // x
fxam // Is NaN or +-Inf?
fstsw %ax