mirror/glibc
2
0
Fork 0
mirror of git://sourceware.org/git/glibc.git synced 2024-11-21 01:12:26 +08:00
Code Issues Packages Projects Releases Wiki Activity

Remove x86 mathinline.h sinh, cosh, tanh inlines.

Browse Source 
Continuing the removal of bits/mathinline.h inlines that would better
be done by the compiler, this patch removes x86 inlines for sinh, cosh
and tanh functions (inlines only previously present for fast-math,
non-SSE 32-bit x86).  I've filed
<https://gcc.gnu.org/bugzilla/show_bug.cgi?id=88556> for adding such
inlines as an optimization in GCC.

I believe the only remaining part of the x86 bits/mathinline.h that
does anything useful with current compilers after this patch is the
__LIBC_INTERNAL_MATH_INLINES inline of __ieee754_atan2l; I intend to
remove the whole header and move that inline to a sysdeps
math_private.h header in a subsequent patch.

Tested for x86_64 and x86.

	* sysdeps/x86/fpu/bits/mathinline.h (sinh): Remove inline
	definition.
	(cosh): Likewise.
	(tanh): Likewise.
This commit is contained in:
Joseph Myers 2018-12-19 21:48:03 +00:00
parent 1616d034b6
commit 515f463f52
2 changed files with 7 additions and 16 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
ChangeLog
View File

@ -1,3 +1,10 @@
2018-12-19 Joseph Myers <joseph@codesourcery.com>
* sysdeps/x86/fpu/bits/mathinline.h (sinh): Remove inline
definition.
(cosh): Likewise.
(tanh): Likewise.
2018-12-19 Tulio Magno Quites Machado Filho <tuliom@linux.ibm.com>
* elf/dl-sysdep.c (auxvars): Add AT_L1I_CACHESIZE,

16
sysdeps/x86/fpu/bits/mathinline.h
View File

@ -263,22 +263,6 @@ __inline_mathcode_ (long double, __sgn1l, __x, \
return __n.__xld)
# ifdef __FAST_MATH__
/* The argument range of the inline version of sinhl is slightly reduced. */
__inline_mathcodeNP (sinh, __x, \
register long double __exm1 = __expm1l (__fabsl (__x)); \
return 0.5 * (__exm1 / (__exm1 + 1.0) + __exm1) * __sgn1l (__x))
__inline_mathcodeNP (cosh, __x, \
register long double __ex = __expl (__x); \
return 0.5 * (__ex + 1.0 / __ex))
__inline_mathcodeNP (tanh, __x, \
register long double __exm1 = __expm1l (-__fabsl (__x + __x)); \
return __exm1 / (__exm1 + 2.0) * __sgn1l (-__x))
# endif
/* Optimized versions for some non-standardized functions. */
# ifdef __USE_ISOC99