diff options
16 files changed, 817 insertions, 197 deletions
diff --git a/sysdeps/powerpc/powerpc64/le/fpu/multiarch/Makefile b/sysdeps/powerpc/powerpc64/le/fpu/multiarch/Makefile index 8747b02127..767805b510 100644 --- a/sysdeps/powerpc/powerpc64/le/fpu/multiarch/Makefile +++ b/sysdeps/powerpc/powerpc64/le/fpu/multiarch/Makefile @@ -1,10 +1,202 @@ ifeq ($(subdir),math) -libm-sysdep_routines += s_fmaf128-ppc64 s_fmaf128-power9 \ - w_sqrtf128-power9 w_sqrtf128-ppc64le -CFLAGS-s_fmaf128-ppc64.c += $(type-float128-CFLAGS) $(no-gnu-attribute-CFLAGS) -CFLAGS-s_fmaf128-power9.c += $(type-float128-CFLAGS) -mcpu=power9 $(no-gnu-attribute-CFLAGS) +# +# Only enable ifunc _Float128 support if the baseline cpu support +# is older than power9. +ifneq (yes,$(libc-submachine-power9)) +do_f128_multiarch = yes +endif + +# +# This is an ugly, but contained, mechanism to provide hardware optimized +# _Float128 and ldouble == ieee128 optimized routines for P9 and beyond +# hardware. At a very high level, we rely on ASM renames, and rarely +# macro renames to build two sets of _Float128 ABI, one with _power8 (the +# baseline powerpc64le cpu) and _power9 (the first powerpc64le cpu to introduce +# hardware support for _Float128). +# +# At a high level, we compile 3 files for each object file. +# 1. The baseline soft-float128, unsuffixed objects $(object).$(sfx) +# The symbols contained in these files is suffixed by _power8. +# 2. The hard-float128, power9, suffixed objects $(object)-power9.$(sfx). +# The symbols contained in these files is suffixed by _power9. +# 3. The IFUNC wrapper object to export ABI, $(object)-ifunc.$(sfx) +# This glues the above together and implements the ABI. +# +# 2 & 3 are automatically generated by Makefile rule. Placing the exported +# ABI into a separate file allows reuse of existing aliasing macros +# with minimal hassle. +# +# +# If the float128 ABI is expanded, and a new ifunc wrappers are desired, +# the following lists how to map new symbols from the shared headers into +# their local overrides here: +# +# float128_private.h +# +# is used to rename the ldouble == ieee128 object files. This takes +# it a step further and redirects symbols to a local name. This supports +# nearly all files in sysdeps/ieee754/float128, but not all _Float128 +# objects. However, this is only meant to be used internally to support +# compilation of ldbl-128 into float128. +# +# math-type-macros-float128.h +# +# renames symbols which are generated via shared templated in math/. +# +# math_private.h +# +# provides internal declarations for common macros and functions which +# are called from within libm. Note, float128_private.h duplicates +# some of these declarations as these headers are generally not included +# in the same translation unit. +# +# The above is supported by several header files as described below: +# +# float128-ifunc.h +# +# provides support for generating the IFUNC objects in part 3 above. +# This header is only included with wrapper functions. +# +# float128-ifunc-macros.h +# +# disables all first-order float128 aliasing macros used in libm, +# and libm wrappers around libc-symbols.h. +# +# float128-ifunc-redirect-macros.h +# +# provides macros which implement the appending of the suffix to +# symbols what have been selected. +# +# float128-ifunc-redirects.h +# +# provides ASM redirects for symbols which are redirected in the +# private copy of math.h used by glibc, but not declared by math_private.h +# +# float128-ifunc-redirects-mp.h +# +# provides ASM redirects which are used by math_private.h (the -mp suffix) +# and the interposer float128_private.h discussed late. +# +# Notably, this enforces a slightly different mechanism for machine specific +# overrides. Optimizations for all targets must all be reachable from the same +# file. See the history to fmaf128 or sqrtf128 to understand how this looks +# in practice. +# +ifeq ($(do_f128_multiarch),yes) + +gen-libm-f128-ifunc-routines = \ + e_acosf128 e_acoshf128 e_asinf128 e_atan2f128 e_atanhf128 e_coshf128 \ + e_expf128 e_fmodf128 e_hypotf128 e_j0f128 e_j1f128 e_jnf128 \ + e_lgammaf128_r e_logf128 e_log10f128 e_powf128 e_remainderf128 \ + e_sinhf128 e_sqrtf128 e_gammaf128_r e_ilogbf128 k_tanf128 s_asinhf128 \ + s_atanf128 s_cbrtf128 s_ceilf128 s_cosf128 s_erff128 s_expm1f128 \ + s_fabsf128 s_floorf128 s_log1pf128 s_logbf128 \ + s_rintf128 s_scalblnf128 s_sinf128 s_tanf128 \ + s_tanhf128 s_truncf128 s_remquof128 e_log2f128 \ + s_roundf128 s_nearbyintf128 s_sincosf128 s_fmaf128 s_lrintf128 \ + s_llrintf128 s_lroundf128 s_llroundf128 e_exp10f128 \ + m_modff128 m_scalbnf128 m_frexpf128 m_ldexpf128 x2y2m1f128 \ + gamma_productf128 lgamma_negf128 lgamma_productf128 s_roundevenf128 \ + cargf128 conjf128 cimagf128 crealf128 cabsf128 e_scalbf128 s_cacosf128 \ + s_cacoshf128 s_ccosf128 s_ccoshf128 s_casinf128 s_csinf128 \ + s_casinhf128 k_casinhf128 s_csinhf128 k_casinhf128 s_csinhf128 \ + s_catanhf128 s_catanf128 s_ctanf128 s_ctanhf128 s_cexpf128 s_clogf128 \ + s_cprojf128 s_csqrtf128 s_cpowf128 s_clog10f128 s_fdimf128 \ + s_fmaxf128 s_fminf128 w_ilogbf128 w_llogbf128 \ + w_log1pf128 w_scalblnf128 w_acosf128 \ + w_acoshf128 w_asinf128 w_atan2f128 w_atanhf128 w_coshf128 w_exp10f128 \ + w_exp2f128 w_fmodf128 w_hypotf128 w_j0f128 w_j1f128 w_jnf128 \ + w_logf128 w_log10f128 w_log2f128 w_powf128 w_remainderf128 \ + w_scalbf128 w_sinhf128 w_sqrtf128 w_tgammaf128 w_lgammaf128 \ + w_lgammaf128_r w_expf128 e_exp2f128 \ + k_sinf128 k_cosf128 k_sincosf128 e_rem_pio2f128 + + +f128-march-routines-p9 = $(addsuffix -power9,$(gen-libm-f128-ifunc-routines)) +f128-march-routines-ifunc = $(addsuffix -ifunc,$(gen-libm-f128-ifunc-routines)) +f128-march-routines = $(f128-march-routines-p9) $(f128-march-routines-ifunc) +f128-march-cpus = power9 + +libm-routines += $(f128-march-routines) +generated += $(f128-march-routines) + +CFLAGS-float128-ifunc.c += $(type-float128-CFLAGS) $(no-gnu-attribute-CFLAGS) + +# Copy special CFLAGS for some functions +CFLAGS-m_modff128-power9.c += -fsignaling-nans + +# Generate ifunc wrapper files and target specific wrappers around +# each routine above. Note, m_%.c files are fixed up to include +# s_%.c files. This is an artifact of the makefile rules which allow +# some files to be compiled for libc and libm. +$(objpfx)gen-float128-ifuncs.stmp: Makefile + $(make-target-directory) + for gcall in $(gen-libm-f128-ifunc-routines); do \ + ifile="$${gcall}"; \ + if [ $${gcall##m_} != $${gcall} ]; then \ + ifile="s_$${gcall##m_}"; \ + fi; \ + for cpu in $(f128-march-cpus); do \ + file=$(objpfx)$${gcall}-$${cpu}.c; \ + { \ + echo "#include <$${ifile}.c>"; \ + } > $${file}; \ + done; \ + name="$${gcall##?_}"; \ + pfx="$${gcall%%_*}"; \ + R=""; \ + r=""; \ + if [ $${gcall##m_} != $${gcall} ]; then \ + pfx="s"; \ + fi; \ + if [ $${#pfx} != 1 ]; then \ + pfx=""; \ + else \ + pfx="_$${pfx}"; \ + fi; \ + if [ $${name%%_r} != $${name} ]; then \ + R="_R"; \ + r="_r"; \ + name="$${name%%_r}"; \ + fi; \ + name="$${name%%f128}"; \ + decl="DECL_ALIAS$${pfx}_$${name}$${r}"; \ + compat="GEN_COMPAT$${pfx}_$${name}$${r}"; \ + declc="DECL_ALIAS$${R}$${pfx}"; \ + { \ + echo "#include <float128-ifunc.h>"; \ + echo "#ifndef $${decl}"; \ + echo "# define $${decl}(f) $${declc} (f)"; \ + echo "#endif"; \ + echo "#ifndef $${compat}"; \ + echo "# define $${compat}(f)"; \ + echo "#endif"; \ + echo "$${decl} ($${name});"; \ + echo "$${compat} ($${name});"; \ + } > $(objpfx)$${gcall}-ifunc.c; \ + done; \ + echo > $(@) + +$(foreach f,$(f128-march-routines),$(objpfx)$(f).c): $(objpfx)gen-float128-ifuncs.stmp + +enable-f128-ifunc-CFLAGS = -D_F128_ENABLE_IFUNC $(no-gnu-attributes-CFLAGS) $(type-float128-CFLAGS) + +# Enable IFUNC on baseline (power8) implementations +include $(o-iterator) +define o-iterator-doit +$(foreach f,$(gen-libm-f128-ifunc-routines),$(objpfx)$(f)$(o)): sysdep-CFLAGS += -D_F128_ENABLE_IFUNC +endef +object-suffixes-left := $(all-object-suffixes) +include $(o-iterator) + +# Likewise, but for power9. +include $(o-iterator) +define o-iterator-doit +$(foreach f,$(f128-march-routines-p9),$(objpfx)$(f)$(o)): sysdep-CFLAGS += $$(enable-f128-ifunc-CFLAGS) -mcpu=power9 +endef +object-suffixes-left := $(all-object-suffixes) +include $(o-iterator) -CFLAGS-w_sqrtf128-ppc64le.c += $(type-float128-CFLAGS) $(no-gnu-attribute-CFLAGS) -CFLAGS-w_sqrtf128-power9.c += $(type-float128-CFLAGS) -mcpu=power9 $(no-gnu-attribute-CFLAGS) +endif # do_f128_multiarch endif diff --git a/sysdeps/powerpc/powerpc64/le/fpu/multiarch/float128-ifunc-macros.h b/sysdeps/powerpc/powerpc64/le/fpu/multiarch/float128-ifunc-macros.h new file mode 100644 index 0000000000..15e7c96fd3 --- /dev/null +++ b/sysdeps/powerpc/powerpc64/le/fpu/multiarch/float128-ifunc-macros.h @@ -0,0 +1,56 @@ +/* _Float128 aliasing macro support for ifunc generation on PPC. + Copyright (C) 2020 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, see + <https://www.gnu.org/licenses/>. */ + +#ifndef _FLOAT128_IFUNC_MACROS_PPC64LE +#define _FLOAT128_IFUNC_MACROS_PPC64LE 1 + +/* Bring in the various alias-providing headers, and disable + _Float128 related macros. This prevents exporting any ABI + from _Float128 implementation objects. */ +#include <libm-alias-float128.h> +#include <libm-alias-finite.h> + +#undef libm_alias_float128_r +#undef libm_alias_finite +#undef libm_alias_exclusive_ldouble +#undef libm_alias_float128_other_r_ldbl +#undef declare_mgen_finite_alias +#undef declare_mgen_alias +#undef declare_mgen_alias_r + +#define libm_alias_finite(from, to) +#define libm_alias_float128_r(from, to, r) +#define libm_alias_exclusive_ldouble(from, to) +#define libm_alias_float128_other_r_ldbl(from, to, r) +#define declare_mgen_finite_alias(from, to) +#define declare_mgen_alias(from, to) +#define declare_mgen_alias_r(from, to) + +/* Likewise, disable hidden symbol support. This is not needed + for the implementation objects as the redirects already give + us this support. This also means any non-_Float128 headers + which provide hidden_def's should be included prior to this + header (e.g fenv.h). */ +#undef libm_hidden_def +#define libm_hidden_def(func) +#undef libm_hidden_proto +#define libm_hidden_proto(f) + +#include <float128-ifunc-redirect-macros.h> + +#endif /* _FLOAT128_IFUNC_MACROS_PPC64LE */ diff --git a/sysdeps/powerpc/powerpc64/le/fpu/multiarch/float128-ifunc-redirect-macros.h b/sysdeps/powerpc/powerpc64/le/fpu/multiarch/float128-ifunc-redirect-macros.h new file mode 100644 index 0000000000..baf5633aae --- /dev/null +++ b/sysdeps/powerpc/powerpc64/le/fpu/multiarch/float128-ifunc-redirect-macros.h @@ -0,0 +1,53 @@ +/* _Float128 aliasing macro support for ifunc generation on PPC. + Copyright (C) 2020 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, see + <https://www.gnu.org/licenses/>. */ + +#ifndef _FLOAT128_IFUNC_REDIRECT_MACROS_PPC64LE +#define _FLOAT128_IFUNC_REDIRECT_MACROS_PPC64LE 1 + +/* Define the redirection macros used throughout most of the IFUNC headers. + The variant is inferred via compiler options. + + F128_REDIR_PFX_R(function, destination_prefix, reentrant_suffix) + Redirect function, optionally suffixed by reentrant_suffix, to a function + named destination_prefix ## function ## variant ## reentrant_suffix. + + F128_SFX_APPEND(sym) + Append the the multiarch variant specific suffix to the sym. sym is not + expanded. This is sym ## variant. + + F128_REDIR_R(func, reentrant_suffix) + Redirect func to a function named function ## variant ## reentrant_suffix + + F128_REDIR(function) + Convience wrapper for F128_REDIR_R where function does not require + a suffix argument. + +*/ +#ifndef _ARCH_PWR9 +#define F128_REDIR_PFX_R(func, pfx, r) \ + extern __typeof(func ## r) func ## r __asm( #pfx #func "_power8" #r ); +#define F128_SFX_APPEND(x) x ## _power8 +#else +#define F128_REDIR_PFX_R(func, pfx, r) \ + extern __typeof(func ## r) func ## r __asm( #pfx #func "_power9" #r ); +#define F128_SFX_APPEND(x) x ## _power9 +#endif +#define F128_REDIR_R(func, r) F128_REDIR_PFX_R (func, , r) +#define F128_REDIR(func) F128_REDIR_R (func, ) + +#endif /*_FLOAT128_IFUNC_REDIRECT_MACROS_PPC64LE */ diff --git a/sysdeps/powerpc/powerpc64/le/fpu/multiarch/float128-ifunc-redirects-mp.h b/sysdeps/powerpc/powerpc64/le/fpu/multiarch/float128-ifunc-redirects-mp.h new file mode 100644 index 0000000000..3c8b6f1291 --- /dev/null +++ b/sysdeps/powerpc/powerpc64/le/fpu/multiarch/float128-ifunc-redirects-mp.h @@ -0,0 +1,64 @@ +/* _Float128 multiarch redirects shared with math_private.h + Copyright (C) 2020 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, see + <https://www.gnu.org/licenses/>. */ + +#ifndef _FLOAT128_IFUNC_REDIRECTS_MP_H +#define _FLOAT128_IFUNC_REDIRECTS_MP_H 1 + +#include <float128-ifunc-redirect-macros.h> + +F128_REDIR (__ieee754_acosf128) +F128_REDIR (__ieee754_acoshf128) +F128_REDIR (__ieee754_asinf128) +F128_REDIR (__ieee754_atan2f128) +F128_REDIR (__ieee754_atanhf128) +F128_REDIR (__ieee754_coshf128) +F128_REDIR (__ieee754_expf128) +F128_REDIR (__ieee754_exp10f128) +F128_REDIR (__ieee754_exp2f128) +F128_REDIR (__ieee754_fmodf128) +F128_REDIR (__ieee754_gammaf128) +F128_REDIR_R (__ieee754_gammaf128, _r) +F128_REDIR (__ieee754_hypotf128) +F128_REDIR (__ieee754_j0f128) +F128_REDIR (__ieee754_j1f128) +F128_REDIR (__ieee754_jnf128) +F128_REDIR (__ieee754_lgammaf128) +F128_REDIR_R (__ieee754_lgammaf128, _r) +F128_REDIR (__ieee754_logf128) +F128_REDIR (__ieee754_log10f128) +F128_REDIR (__ieee754_log2f128) +F128_REDIR (__ieee754_powf128) +F128_REDIR (__ieee754_remainderf128) +F128_REDIR (__ieee754_sinhf128) +F128_REDIR (__ieee754_sqrtf128) +F128_REDIR (__ieee754_y0f128) +F128_REDIR (__ieee754_y1f128) +F128_REDIR (__ieee754_ynf128) +F128_REDIR (__ieee754_scalbf128) +F128_REDIR (__ieee754_ilogbf128) +F128_REDIR (__ieee754_rem_pio2f128) +F128_REDIR (__kernel_sinf128) +F128_REDIR (__kernel_cosf128) +F128_REDIR (__kernel_tanf128) +F128_REDIR (__kernel_sincosf128) +F128_REDIR (__kernel_rem_pio2f128) +F128_REDIR (__x2y2m1f128) +F128_REDIR (__gamma_productf128) +F128_REDIR (__lgamma_negf128) + +#endif /*_FLOAT128_IFUNC_REDIRECTS_MP_H */ diff --git a/sysdeps/powerpc/powerpc64/le/fpu/multiarch/float128-ifunc-redirects.h b/sysdeps/powerpc/powerpc64/le/fpu/multiarch/float128-ifunc-redirects.h new file mode 100644 index 0000000000..a0f035965c --- /dev/null +++ b/sysdeps/powerpc/powerpc64/le/fpu/multiarch/float128-ifunc-redirects.h @@ -0,0 +1,43 @@ +/* _Float128 redirects for ppc64le multiarch env. + Copyright (C) 2020 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, see + <https://www.gnu.org/licenses/>. */ + +#ifndef _FLOAT128_IFUNC_REDIRECTS +#define _FLOAT128_IFUNC_REDIRECTS 1 + +#include <float128-ifunc-macros.h> + +F128_REDIR_PFX_R (sqrtf128, __,); +F128_REDIR_PFX_R (rintf128, __,); +F128_REDIR_PFX_R (ceilf128, __,); +F128_REDIR_PFX_R (floorf128, __,); +F128_REDIR_PFX_R (truncf128, __,); +F128_REDIR_PFX_R (roundf128, __,); +F128_REDIR_PFX_R (fabsf128, __,); + +extern __typeof (ldexpf128) F128_SFX_APPEND (__ldexpf128); + +#define __ldexpf128 F128_SFX_APPEND (__ldexpf128) + +/* libm_hidden_proto is disabled by the time we reach here. + Ensure some internally called functions are still called + without going through the PLT. Note, this code is only + included when building libm. */ +hidden_proto (__fpclassifyf128) +hidden_proto (__issignalingf128) + +#endif /* _FLOAT128_IFUNC_REDIRECTS */ diff --git a/sysdeps/powerpc/powerpc64/le/fpu/multiarch/float128-ifunc.h b/sysdeps/powerpc/powerpc64/le/fpu/multiarch/float128-ifunc.h new file mode 100644 index 0000000000..73099e0341 --- /dev/null +++ b/sysdeps/powerpc/powerpc64/le/fpu/multiarch/float128-ifunc.h @@ -0,0 +1,166 @@ +/* _Float128 ifunc ABI/ifunc generation macros. + Copyright (C) 2020 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, see + <https://www.gnu.org/licenses/>. */ + +#ifndef _FLOAT128_IFUNC_H +#define _FLOAT128_IFUNC_H 1 + +/* Disable hidden prototypes. These perform function renames which + prevent the ifunc functions from working. */ +#undef hidden_proto +#define hidden_proto(x) +#define NO_MATH_REDIRECT 1 + +/* Include the real math.h to avoid optimizations caused by include/math.h + (e.x fabsf128 prototype is masked by an inline definition).*/ +#include <math/math.h> +#include <math_private.h> +#include <complex.h> +#include <first-versions.h> +#include <shlib-compat.h> +#include "init-arch.h" + +#include <libm-alias-float128.h> +#include <libm-alias-finite.h> + +/* _F128_IFUNC2(func, from, r) + Generate an ifunc symbol func ## r from the symbols + from ## {power8, power9} ## r + + We use the PPC hwcap bit HAS_IEEE128 to select between the two with + the assumption all P9 features are available on such targets. */ +#define _F128_IFUNC2(func, from, r) \ + libc_ifunc (func ## r, (hwcap2 & PPC_FEATURE2_HAS_IEEE128) \ + ? from ## _power9 ## r : from ## _power8 ## r) + +/* _F128_IFUNC(func, r) + Similar to above, except the exported symbol name trivially remaps from |