AuroraMiddleware/glibc
1
0
Fork 0
mirror of https://sourceware.org/git/glibc.git synced 2024-12-31 23:11:09 +00:00
Code Issues Packages Projects Releases Wiki Activity
d8742dd82f
glibc/sysdeps/ieee754/ldbl-opt/Makefile
Joseph Myers d8742dd82f Add narrowing add functions. 
This patch adds the narrowing add functions from TS 18661-1 to glibc's
libm: fadd, faddl, daddl, f32addf64, f32addf32x, f32xaddf64 for all
configurations; f32addf64x, f32addf128, f64addf64x, f64addf128,
f32xaddf64x, f32xaddf128, f64xaddf128 for configurations with
_Float64x and _Float128; __nldbl_daddl for ldbl-opt.  As discussed for
the build infrastructure patch, tgmath.h support is deliberately
deferred, and FP_FAST_* macros are not applicable without optimized
function implementations.

Function implementations are added for all relevant pairs of formats
(including certain cases of a format and itself where more than one
type has that format).  The main implementations use round-to-odd, or
a trivial computation in the case where both formats are the same or
where the wider format is IBM long double (in which case we don't
attempt to be correctly rounding).  The sysdeps/ieee754/soft-fp
implementations use soft-fp, and are used automatically for
configurations without exceptions and rounding modes by virtue of
existing Implies files.  As previously discussed, optimized versions
for particular architectures are possible, but not included.

i386 gets a special version of f32xaddf64 to avoid problems with
double rounding (similar to the existing fdim version), since this
function must round just once without an intermediate rounding to long
double.  (No such special version is needed for any other function,
because the nontrivial functions use round-to-odd, which does the
intermediate computation with the rounding mode set to round-to-zero,
and double rounding is OK except in round-to-nearest mode, so is OK
for that intermediate round-to-zero computation.)  mul and div will
need slightly different special versions for i386 (using round-to-odd
on long double instead of precision control) because of the
possibility of inexact intermediate results in the subnormal range for
double.

To reduce duplication among the different function implementations,
math-narrow.h gets macros CHECK_NARROW_ADD, NARROW_ADD_ROUND_TO_ODD
and NARROW_ADD_TRIVIAL.

In the trivial cases and for any architecture-specific optimized
implementations, the overhead of the errno setting might be
significant, but I think that's best handled through compiler built-in
functions rather than providing separate no-errno versions in glibc
(and likewise there are no __*_finite entry points for these function
provided, __*_finite effectively being no-errno versions at present in
most cases).

Tested for x86_64 and x86, with both GCC 6 and GCC 7.  Tested for
mips64 (all three ABIs, both hard and soft float) and powerpc with GCC
7.  Tested with build-many-glibcs.py with both GCC 6 and GCC 7.

	* math/Makefile (libm-narrow-fns): Add add.
	(libm-test-funcs-narrow): Likewise.
	* math/Versions (GLIBC_2.28): Add narrowing add functions.
	* math/bits/mathcalls-narrow.h (add): Use __MATHCALL_NARROW .
	* math/gen-auto-libm-tests.c (test_functions): Add add.
	* math/math-narrow.h (CHECK_NARROW_ADD): New macro.
	(NARROW_ADD_ROUND_TO_ODD): Likewise.
	(NARROW_ADD_TRIVIAL): Likewise.
	* sysdeps/ieee754/float128/float128_private.h (__faddl): New
	macro.
	(__daddl): Likewise.
	* sysdeps/ieee754/ldbl-opt/Makefile (libnldbl-calls): Add fadd and
	dadd.
	(CFLAGS-nldbl-dadd.c): New variable.
	(CFLAGS-nldbl-fadd.c): Likewise.
	* sysdeps/ieee754/ldbl-opt/Versions (GLIBC_2.28): Add
	__nldbl_daddl.
	* sysdeps/ieee754/ldbl-opt/nldbl-compat.h (__nldbl_daddl): New
	prototype.
	* manual/arith.texi (Misc FP Arithmetic): Document fadd, faddl,
	daddl, fMaddfN, fMaddfNx, fMxaddfN and fMxaddfNx.
	* math/auto-libm-test-in: Add tests of add.
	* math/auto-libm-test-out-narrow-add: New generated file.
	* math/libm-test-narrow-add.inc: New file.
	* sysdeps/i386/fpu/s_f32xaddf64.c: Likewise.
	* sysdeps/ieee754/dbl-64/s_f32xaddf64.c: Likewise.
	* sysdeps/ieee754/dbl-64/s_fadd.c: Likewise.
	* sysdeps/ieee754/float128/s_f32addf128.c: Likewise.
	* sysdeps/ieee754/float128/s_f64addf128.c: Likewise.
	* sysdeps/ieee754/float128/s_f64xaddf128.c: Likewise.
	* sysdeps/ieee754/ldbl-128/s_daddl.c: Likewise.
	* sysdeps/ieee754/ldbl-128/s_f64xaddf128.c: Likewise.
	* sysdeps/ieee754/ldbl-128/s_faddl.c: Likewise.
	* sysdeps/ieee754/ldbl-128ibm/s_daddl.c: Likewise.
	* sysdeps/ieee754/ldbl-128ibm/s_faddl.c: Likewise.
	* sysdeps/ieee754/ldbl-96/s_daddl.c: Likewise.
	* sysdeps/ieee754/ldbl-96/s_faddl.c: Likewise.
	* sysdeps/ieee754/ldbl-opt/nldbl-dadd.c: Likewise.
	* sysdeps/ieee754/ldbl-opt/nldbl-fadd.c: Likewise.
	* sysdeps/ieee754/soft-fp/s_daddl.c: Likewise.
	* sysdeps/ieee754/soft-fp/s_fadd.c: Likewise.
	* sysdeps/ieee754/soft-fp/s_faddl.c: Likewise.
	* sysdeps/powerpc/fpu/libm-test-ulps: Update.
	* sysdeps/mach/hurd/i386/libm.abilist: Likewise.
	* sysdeps/unix/sysv/linux/aarch64/libm.abilist: Likewise.
	* sysdeps/unix/sysv/linux/alpha/libm.abilist: Likewise.
	* sysdeps/unix/sysv/linux/arm/libm.abilist: Likewise.
	* sysdeps/unix/sysv/linux/hppa/libm.abilist: Likewise.
	* sysdeps/unix/sysv/linux/i386/libm.abilist: Likewise.
	* sysdeps/unix/sysv/linux/ia64/libm.abilist: Likewise.
	* sysdeps/unix/sysv/linux/m68k/coldfire/libm.abilist: Likewise.
	* sysdeps/unix/sysv/linux/m68k/m680x0/libm.abilist: Likewise.
	* sysdeps/unix/sysv/linux/microblaze/libm.abilist: Likewise.
	* sysdeps/unix/sysv/linux/mips/mips32/libm.abilist: Likewise.
	* sysdeps/unix/sysv/linux/mips/mips64/libm.abilist: Likewise.
	* sysdeps/unix/sysv/linux/nios2/libm.abilist: Likewise.
	* sysdeps/unix/sysv/linux/powerpc/powerpc32/fpu/libm.abilist: Likewise.
	* sysdeps/unix/sysv/linux/powerpc/powerpc32/nofpu/libm.abilist: Likewise.
	* sysdeps/unix/sysv/linux/powerpc/powerpc64/libm-le.abilist: Likewise.
	* sysdeps/unix/sysv/linux/powerpc/powerpc64/libm.abilist: Likewise.
	* sysdeps/unix/sysv/linux/riscv/rv64/libm.abilist: Likewise.
	* sysdeps/unix/sysv/linux/s390/s390-32/libm.abilist: Likewise.
	* sysdeps/unix/sysv/linux/s390/s390-64/libm.abilist: Likewise.
	* sysdeps/unix/sysv/linux/sh/libm.abilist: Likewise.
	* sysdeps/unix/sysv/linux/sparc/sparc32/libm.abilist: Likewise.
	* sysdeps/unix/sysv/linux/sparc/sparc64/libm.abilist: Likewise.
	* sysdeps/unix/sysv/linux/tile/tilegx32/libm.abilist: Likewise.
	* sysdeps/unix/sysv/linux/tile/tilegx64/libm.abilist: Likewise.
	* sysdeps/unix/sysv/linux/x86_64/64/libm.abilist: Likewise.
	* sysdeps/unix/sysv/linux/x86_64/x32/libm.abilist: Likewise.
2018-02-10 02:08:43 +00:00

172 lines
7.4 KiB
Makefile
Raw Blame History

# The`long double' type is a distinct type we support if
# -mlong-double-128 option is used (or when it becomes a default
# when -mlong-double-64 is not used).
long-double-fcts = yes
ifeq (,$(filter -mlong-double-128,$(sysdep-CFLAGS)))
sysdep-CFLAGS += -mlong-double-128
endif
ifeq ($(subdir),math)
libm-routines += s_nexttowardfd
routines += math_ldbl_opt nldbl-compat
extra-libs += libnldbl
libnldbl-calls = asprintf dprintf fprintf fscanf fwprintf fwscanf iovfscanf \
obstack_printf obstack_vprintf printf scanf snprintf \
sprintf sscanf swprintf swscanf vasprintf vdprintf vfprintf \
vfscanf vfwprintf vfwscanf vprintf vscanf vsnprintf \
vsprintf vsscanf vswprintf vswscanf vwprintf vwscanf \
wprintf wscanf printf_fp printf_size \
fprintf_chk fwprintf_chk printf_chk snprintf_chk sprintf_chk \
swprintf_chk vfprintf_chk vfwprintf_chk vprintf_chk \
vsnprintf_chk vsprintf_chk vswprintf_chk vwprintf_chk \
wprintf_chk asprintf_chk vasprintf_chk dprintf_chk \
vdprintf_chk obstack_printf_chk obstack_vprintf_chk \
syslog syslog_chk vsyslog vsyslog_chk \
strfmon strfmon_l \
strfroml \
strtold strtold_l strtoldint wcstold wcstold_l wcstoldint \
qecvt qfcvt qgcvt qecvt_r qfcvt_r \
isinf isnan finite signbit scalb log2 lgamma_r ceil \
significand acos asin atan atan2 cos sin tan cosh sinh \
tanh acosh asinh atanh exp log log10 exp10 expm1 \
log1p logb exp2 sqrt cbrt fabs floor j0 j1 y0 y1 erf erfc \
lgamma tgamma gamma rint nearbyint round trunc \
copysign fdim fmax fmin nextafter pow hypot fmod \
remainder ldexp scalbn frexp modf scalbln fma nan sincos \
jn yn ilogb remquo lrint lround llrint llround nexttowardf \
nexttoward conj cacos cacosh casin catan catanh ccos ccosh \
casinh cexp clog cproj csin csinh csqrt ctan ctanh cpow \
cabs carg cimag creal clog10 \
isoc99_scanf isoc99_fscanf isoc99_sscanf \
isoc99_vscanf isoc99_vfscanf isoc99_vsscanf \
isoc99_wscanf isoc99_fwscanf isoc99_swscanf \
isoc99_vwscanf isoc99_vfwscanf isoc99_vswscanf \
nextup nextdown totalorder totalordermag getpayload \
canonicalize setpayload setpayloadsig llogb fmaxmag fminmag \
roundeven fromfp ufromfp fromfpx ufromfpx fadd dadd
libnldbl-routines = $(libnldbl-calls:%=nldbl-%)
libnldbl-inhibit-o = $(object-suffixes)
libnldbl-static-only-routines = $(libnldbl-routines)
extra-objs += $(addsuffix .oS, $(libnldbl-routines))
CFLAGS-nldbl-acos.c = -fno-builtin-acosl
CFLAGS-nldbl-acosh.c = -fno-builtin-acoshl
CFLAGS-nldbl-asin.c = -fno-builtin-asinl
CFLAGS-nldbl-asinh.c = -fno-builtin-asinhl
CFLAGS-nldbl-atan.c = -fno-builtin-atanl
CFLAGS-nldbl-atan2.c = -fno-builtin-atan2l
CFLAGS-nldbl-atanh.c = -fno-builtin-atanhl
CFLAGS-nldbl-cabs.c = -fno-builtin-cabsl
CFLAGS-nldbl-cacos.c = -fno-builtin-cacosl
CFLAGS-nldbl-cacosh.c = -fno-builtin-cacoshl
CFLAGS-nldbl-canonicalize.c = -fno-builtin-canonicalizel
CFLAGS-nldbl-carg.c = -fno-builtin-cargl
CFLAGS-nldbl-casin.c = -fno-builtin-casinl
CFLAGS-nldbl-casinh.c = -fno-builtin-casinhl
CFLAGS-nldbl-catan.c = -fno-builtin-catanl
CFLAGS-nldbl-catanh.c = -fno-builtin-catanhl
CFLAGS-nldbl-cbrt.c = -fno-builtin-cbrtl
CFLAGS-nldbl-ccos.c = -fno-builtin-ccosl
CFLAGS-nldbl-ccosh.c = -fno-builtin-ccoshl
CFLAGS-nldbl-ceil.c = -fno-builtin-ceill
CFLAGS-nldbl-cexp.c = -fno-builtin-cexpl
CFLAGS-nldbl-cimag.c = -fno-builtin-cimagl
CFLAGS-nldbl-clog.c = -fno-builtin-clogl
CFLAGS-nldbl-clog10.c = -fno-builtin-clog10l
CFLAGS-nldbl-conj.c = -fno-builtin-conjl
CFLAGS-nldbl-copysign.c = -fno-builtin-copysignl
CFLAGS-nldbl-cos.c = -fno-builtin-cosl
CFLAGS-nldbl-cosh.c = -fno-builtin-coshl
CFLAGS-nldbl-cpow.c = -fno-builtin-cpowl
CFLAGS-nldbl-cproj.c = -fno-builtin-cprojl
CFLAGS-nldbl-creal.c = -fno-builtin-creall
CFLAGS-nldbl-csin.c = -fno-builtin-csinl
CFLAGS-nldbl-csinh.c = -fno-builtin-csinhl
CFLAGS-nldbl-csqrt.c = -fno-builtin-csqrtl
CFLAGS-nldbl-ctan.c = -fno-builtin-ctanl
CFLAGS-nldbl-ctanh.c = -fno-builtin-ctanhl
CFLAGS-nldbl-dadd.c = -fno-builtin-daddl
CFLAGS-nldbl-erf.c = -fno-builtin-erfl
CFLAGS-nldbl-erfc.c = -fno-builtin-erfcl
CFLAGS-nldbl-exp.c = -fno-builtin-expl
CFLAGS-nldbl-exp10.c = -fno-builtin-exp10l
CFLAGS-nldbl-exp2.c = -fno-builtin-exp2l
CFLAGS-nldbl-expm1.c = -fno-builtin-expm1l
CFLAGS-nldbl-fabs.c = -fno-builtin-fabsl
CFLAGS-nldbl-fadd.c = -fno-builtin-faddl
CFLAGS-nldbl-fdim.c = -fno-builtin-fdiml
CFLAGS-nldbl-finite.c = -fno-builtin-finitel
CFLAGS-nldbl-floor.c = -fno-builtin-floorl
CFLAGS-nldbl-fma.c = -fno-builtin-fmal
CFLAGS-nldbl-fmax.c = -fno-builtin-fmaxl
CFLAGS-nldbl-fmaxmag.c = -fno-builtin-fmaxmagl
CFLAGS-nldbl-fmin.c = -fno-builtin-fminl
CFLAGS-nldbl-fminmag.c = -fno-builtin-fminmagl
CFLAGS-nldbl-fmod.c = -fno-builtin-fmodl
CFLAGS-nldbl-frexp.c = -fno-builtin-frexpl
CFLAGS-nldbl-fromfp.c = -fno-builtin-fromfpl
CFLAGS-nldbl-fromfpx.c = -fno-builtin-fromfpxl
CFLAGS-nldbl-gamma.c = -fno-builtin-gammal
CFLAGS-nldbl-getpayload.c = -fno-builtin-getpayloadl
CFLAGS-nldbl-hypot.c = -fno-builtin-hypotl
CFLAGS-nldbl-ilogb.c = -fno-builtin-ilogbl
CFLAGS-nldbl-isinf.c = -fno-builtin-isinfl
CFLAGS-nldbl-isnan.c = -fno-builtin-isnanl
CFLAGS-nldbl-j0.c = -fno-builtin-j0l
CFLAGS-nldbl-j1.c = -fno-builtin-j1l
CFLAGS-nldbl-jn.c = -fno-builtin-jnl
CFLAGS-nldbl-ldexp.c = -fno-builtin-ldexpl
CFLAGS-nldbl-lgamma.c = -fno-builtin-lgammal
CFLAGS-nldbl-lgamma_r.c = -fno-builtin-lgammal_r
CFLAGS-nldbl-llogb.c = -fno-builtin-llogbl
CFLAGS-nldbl-llrint.c = -fno-builtin-llrintl
CFLAGS-nldbl-llround.c = -fno-builtin-llroundl
CFLAGS-nldbl-log.c = -fno-builtin-logl
CFLAGS-nldbl-log10.c = -fno-builtin-log10l
CFLAGS-nldbl-log1p.c = -fno-builtin-log1pl
CFLAGS-nldbl-log2.c = -fno-builtin-log2l
CFLAGS-nldbl-logb.c = -fno-builtin-logbl
CFLAGS-nldbl-lrint.c = -fno-builtin-lrintl
CFLAGS-nldbl-lround.c = -fno-builtin-lroundl
CFLAGS-nldbl-modf.c = -fno-builtin-modfl
CFLAGS-nldbl-nan.c = -fno-builtin-nanl
CFLAGS-nldbl-nearbyint.c = -fno-builtin-nearbyintl
CFLAGS-nldbl-nextafter.c = -fno-builtin-nextafterl
CFLAGS-nldbl-nextdown.c = -fno-builtin-nextdownl
CFLAGS-nldbl-nexttoward.c = -fno-builtin-nexttoward -fno-builtin-nexttowardl
CFLAGS-nldbl-nexttowardf.c = -fno-builtin-nexttowardf
CFLAGS-nldbl-nextup.c = -fno-builtin-nextupl
CFLAGS-nldbl-pow.c = -fno-builtin-powl
CFLAGS-nldbl-remainder.c = -fno-builtin-remainderl -fno-builtin-dreml
CFLAGS-nldbl-remquo.c = -fno-builtin-remquol
CFLAGS-nldbl-rint.c = -fno-builtin-rintl
CFLAGS-nldbl-round.c = -fno-builtin-roundl
CFLAGS-nldbl-roundeven.c = -fno-builtin-roundevenl
CFLAGS-nldbl-scalb.c = -fno-builtin-scalbl
CFLAGS-nldbl-scalbln.c = -fno-builtin-scalblnl
CFLAGS-nldbl-scalbn.c = -fno-builtin-scalbnl
CFLAGS-nldbl-setpayload.c = -fno-builtin-setpayloadl
CFLAGS-nldbl-setpayloadsig.c = -fno-builtin-setpayloadsigl
CFLAGS-nldbl-significand.c = -fno-builtin-significandl
CFLAGS-nldbl-sin.c = -fno-builtin-sinl
CFLAGS-nldbl-sincos.c = -fno-builtin-sincosl
CFLAGS-nldbl-sinh.c = -fno-builtin-sinhl
CFLAGS-nldbl-sqrt.c = -fno-builtin-sqrtl
CFLAGS-nldbl-tan.c = -fno-builtin-tanl
CFLAGS-nldbl-tanh.c = -fno-builtin-tanhl
CFLAGS-nldbl-tgamma.c = -fno-builtin-tgammal
CFLAGS-nldbl-totalorder.c = -fno-builtin-totalorderl
CFLAGS-nldbl-totalordermag.c = -fno-builtin-totalordermagl
CFLAGS-nldbl-trunc.c = -fno-builtin-truncl
CFLAGS-nldbl-ufromfp.c = -fno-builtin-ufromfpl
CFLAGS-nldbl-ufromfpx.c = -fno-builtin-ufromfpxl
CFLAGS-nldbl-y0.c = -fno-builtin-y0l
CFLAGS-nldbl-y1.c = -fno-builtin-y1l
CFLAGS-nldbl-yn.c = -fno-builtin-ynl
tests += test-narrow-macros-ldbl-64
CFLAGS-test-narrow-macros-ldbl-64.c += -mlong-double-64
endif
Reference in New Issue View Git Blame Copy Permalink