diff options
author | Adhemerval Zanella <adhemerval.zanella@linaro.org> | 2020-03-16 15:58:45 -0300 |
---|---|---|
committer | Adhemerval Zanella <adhemerval.zanella@linaro.org> | 2020-03-19 11:45:44 -0300 |
commit | 4bad2e014e5e2a835cca6139b5e5a0f42ed0ea05 (patch) | |
tree | 78ee7dcfdbdf9458904e34a0fbc31f0c40f900ed /sysdeps/m68k/m680x0/fpu/s_rint_template.c | |
parent | 649fdf039b1572fca1880a6a643c857bde505456 (diff) | |
download | glibc-4bad2e014e5e2a835cca6139b5e5a0f42ed0ea05.tar.gz glibc-4bad2e014e5e2a835cca6139b5e5a0f42ed0ea05.tar.xz glibc-4bad2e014e5e2a835cca6139b5e5a0f42ed0ea05.zip |
m68k: Remove mathinline.h
This is similar to x86 (da75c1b180f9355a) and powerpc (32ea72999693b98e) mathinline.h removal. The required macros to build the fpu routines are moved to mathimpl.h, while the inline optimization macros for atan, tanh, rint, log1p, significand, trunc, floor, ceil, isinf, finite, scalbn, isnan, scalbln, nearbyint, lrint, and sincos are removed. The gcc bug https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94204 was created to track builtin support. Checked with a build against m68k-linux-gnu, resulting binaries are similar with and without the patch.
Diffstat (limited to 'sysdeps/m68k/m680x0/fpu/s_rint_template.c')
-rw-r--r-- | sysdeps/m68k/m680x0/fpu/s_rint_template.c | 1 |
1 files changed, 1 insertions, 0 deletions
diff --git a/sysdeps/m68k/m680x0/fpu/s_rint_template.c b/sysdeps/m68k/m680x0/fpu/s_rint_template.c index a5fff65f8d..eb38400118 100644 --- a/sysdeps/m68k/m680x0/fpu/s_rint_template.c +++ b/sysdeps/m68k/m680x0/fpu/s_rint_template.c @@ -17,6 +17,7 @@ <https://www.gnu.org/licenses/>. */ #include <math.h> +#include "mathimpl.h" FLOAT M_DECL_FUNC (__rint) (FLOAT x) |