diff options
| author | H.J. Lu <hjl.tools@gmail.com> | 2023-08-11 08:04:08 -0700 |
|---|---|---|
| committer | H.J. Lu <hjl.tools@gmail.com> | 2023-08-14 08:14:19 -0700 |
| commit | 1b214630ce6f7e0099b8b6f87246246739b079cf (patch) | |
| tree | bed526aa9ce38f6c09e144cf27fc24128b0b12c0 /sysdeps/x86_64/fpu/multiarch/Makefile | |
| parent | dd2794d6680d322754464448a25916ee199ce678 (diff) | |
| download | glibc-1b214630ce6f7e0099b8b6f87246246739b079cf.tar.xz glibc-1b214630ce6f7e0099b8b6f87246246739b079cf.zip | |
x86_64: Add expm1 with FMA
On Skylake, it improves expm1 bench performance by:
Before After Improvement
max 70.204 68.054 3%
min 20.709 16.2 22%
mean 22.1221 16.7367 24%
NB: Add
extern long double __expm1l (long double);
extern long double __expm1f128 (long double);
for __typeof (__expm1l) and __typeof (__expm1f128) when __expm1 is
defined since __expm1 may be expanded in their declarations which
causes the build failure.
Diffstat (limited to 'sysdeps/x86_64/fpu/multiarch/Makefile')
| -rw-r--r-- | sysdeps/x86_64/fpu/multiarch/Makefile | 2 |
1 files changed, 2 insertions, 0 deletions
diff --git a/sysdeps/x86_64/fpu/multiarch/Makefile b/sysdeps/x86_64/fpu/multiarch/Makefile index f773255721..add339a876 100644 --- a/sysdeps/x86_64/fpu/multiarch/Makefile +++ b/sysdeps/x86_64/fpu/multiarch/Makefile @@ -37,6 +37,7 @@ libm-sysdep_routines += \ e_log2-fma \ e_pow-fma \ s_atan-fma \ + s_expm1-fma \ s_sin-fma \ s_sincos-fma \ s_tan-fma \ @@ -49,6 +50,7 @@ CFLAGS-e_log-fma.c = -mfma -mavx2 CFLAGS-e_log2-fma.c = -mfma -mavx2 CFLAGS-e_pow-fma.c = -mfma -mavx2 CFLAGS-s_atan-fma.c = -mfma -mavx2 +CFLAGS-s_expm1-fma.c = -mfma -mavx2 CFLAGS-s_sin-fma.c = -mfma -mavx2 CFLAGS-s_tan-fma.c = -mfma -mavx2 CFLAGS-s_sincos-fma.c = -mfma -mavx2 |