Adhemerval Zanella 2eb1cd2f47 math: Remove powerpc e_hypot ... The generic implementation is shows only slight worse performance: POWER10 reciprocal-throughput latency master 8.28478 13.7253 new hypot 7.21945 13.1933 POWER9 reciprocal-throughput latency master 13.4024 14.0967 new hypot 14.8479 15.8061 POWER8 reciprocal-throughput latency master 15.5767 16.8885 new hypot 16.5371 18.4057 One way to improve might to make gcc generate xsmaxdp/xsmindp for fmax/fmin (it onl does for -ffast-math, clang does for default options). Checked on powerpc64-linux-gnu (power8) and powerpc64le-linux-gnu (power9).		2021-12-13 09:08:07 -03:00
..
Makefile	…
e_sqrt.c	…
e_sqrtf.c	…
fclrexcpt.c	…
fe_mask.c	…
fe_nomask.c	…
fedisblxcpt.c	…
feenablxcpt.c	…
fegetenv.c	…
fegetexcept.c	…
fegetmode.c	…
fegetround.c	…
feholdexcpt.c	…
fenv_const.c	…
fenv_libc.h	[powerpc] Tighten contraints for asm constant parameters	2021-11-03 09:17:28 -05:00
fenv_private.h	…
fesetenv.c	…
fesetexcept.c	…
fesetmode.c	…
fesetround.c	…
feupdateenv.c	…
fgetexcptflg.c	…
fix-fp-int-compare-invalid.h	…
fraiseexcpt.c	…
fsetexcptflg.c	…
ftestexcept.c	…
get-rounding-mode.h	…
libm-test-ulps	powerpc: update libm test ulps	2021-10-06 10:50:33 -03:00
libm-test-ulps-name	…
math-barriers.h	…
math-tests-trap-force.h	…
math-use-builtins-fma.h	…
math-use-builtins-sqrt.h	…
math_ldbl.h	…
math_private.h	…
round_to_integer.h	…
s_ceil.c	…
s_ceilf.c	…
s_copysign.c	…
s_copysignf.c	…
s_floor.c	…
s_floorf.c	…
s_logb.c	…
s_logbf.c	…
s_logbl.c	…
s_lrintf.c	…
s_modf.c	…
s_modff.c	…
s_nearbyint.c	…
s_nearbyintf.c	…
s_rint.c	…
s_rintf.c	…
s_round.c	…
s_roundf.c	…
s_trunc.c	…
s_truncf.c	…
t_sqrt.c	…
tst-setcontext-fpscr.c	…