]> git.ipfire.org Git - thirdparty/glibc.git/commit
Use GCC builtins for nearbyint functions if desired.
authorStefan Liebler <stli@linux.ibm.com>
Wed, 11 Dec 2019 14:09:18 +0000 (15:09 +0100)
committerStefan Liebler <stli@linux.ibm.com>
Wed, 11 Dec 2019 14:12:15 +0000 (15:12 +0100)
commitae3577f607b50bf3ce9b0877e43ad2508c9da61b
treeb284b10fc26c0454a71be353759f3d2f892b06e0
parent36e9acbd5cb2c330c0d53195db4a0ee31f2c3097
Use GCC builtins for nearbyint functions if desired.

This patch is using the corresponding GCC builtin for nearbyintf, nearbyint,
nearbintl and nearbyintf128 if the USE_FUNCTION_BUILTIN macros are defined to one
in math-use-builtins.h.

This is the case for s390 if build with at least --march=z196 --mzarch.
Otherwise the generic implementation is used.  The code of the generic
implementation is not changed.

Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
sysdeps/generic/math-use-builtins.h [new file with mode: 0644]
sysdeps/ieee754/dbl-64/s_nearbyint.c
sysdeps/ieee754/float128/float128_private.h
sysdeps/ieee754/flt-32/s_nearbyintf.c
sysdeps/ieee754/ldbl-128/s_nearbyintl.c
sysdeps/s390/fpu/math-use-builtins.h [new file with mode: 0644]