glibc/math/Makefile
Ulrich Drepper 3e5f555752 Update.
1997-03-25 02:15  Ulrich Drepper  <drepper@cygnus.com>

	* gmon/gmon.c: Optimize a bit by using more sets of records to
	write in a single writev call.

	* math/math.h: Add definitions of macros __MATHCALLX and __MATHDECLX.
	* math/mathcalls.h: Use __MATHCALLX for fabs, infnan, copysign,
	nan, isnan, nextafter, trunc, __fpclassify and signbit to mark as
	`const'.
	Use __MATHDECLX for isinf and finite.

	* sysdeps/generic/setenv.c [_LIBC]: Define __clearenv and make
	clearenv a weak alias.

	Implement complex exponential function.
	* sysdeps/libm-ieee754/s_cexp.c: New file.
	* sysdeps/libm-ieee754/s_cexpf.c: New file.
	* sysdeps/libm-ieee754/s_cexpl.c: New file.

	* sysdeps/libm-ieee754/s_nan.c: Define function as __nan and make
	nan a weak alias.
	* sysdeps/libm-ieee754/s_nanf.c: Likewise for nanf.
	* sysdeps/libm-ieee754/s_nanl.c: Likewise for nanl.

	* sysdeps/unix/sysv/linux/iovec.h: Don't use kernel header because
	of type clashes.  Add all definitions here.

	* sysdeps/unix/sysv/linux/shmat.c: Likewise.  Correct types according
	to XPG4.2.
	* sysdeps/unix/sysv/linux/shmdt.c: Likewise.
	* sysdeps/unix/sysv/linux/msgrcv.c: Likewise.

	* sysvipc/sys/shm.h (shmat, shmdt): Correct types.

	* sysdeps/unix/sysv/linux/sys/kd.h: Define _LINUX_TYPES_H to avoid
	use of kernel types.

1997-03-25 00:00  Ulrich Drepper  <drepper@cygnus.com>

	* sysdeps/unix/sysv/linux/netinet/ip.h (struct ip_timestamp):
	Correct typos.  Reported by a sun <asun@zoology.washington.edu>.

1997-03-20 21:58  Andreas Schwab  <schwab@issan.informatik.uni-dortmund.de>

	* sysdeps/libm-ieee754/s_asinhl.c: Fix sign of result.

1997-03-20 16:20  Andreas Schwab  <schwab@issan.informatik.uni-dortmund.de>

	* sysdeps/libm-ieee754/e_atan2l.c: Fix typo.

1997-03-20 14:23  Andreas Schwab  <schwab@issan.informatik.uni-dortmund.de>

	* sysdeps/m68k/fpu/__math.h (__logb, logb): Don't define.
	* sysdeps/m68k/fpu/s_logb.c: Removed.
	* sysdeps/m68k/fpu/s_logbf.c: Removed.
	* sysdeps/m68k/fpu/s_logbl.c: Removed.

	* sysdeps/m68k/fpu/__math.h (__ieee754_atan2, __ieee754_pow):
	Don't define here.
	* sysdeps/m68k/fpu/e_atan2.c: Rewritten.
	* sysdeps/m68k/fpu/e_pow.c: Likewise.
	* sysdeps/m68k/fpu/e_atan2f.c: Based on e_atan2.c.
	* sysdeps/m68k/fpu/e_atan2l.c: Likewise.
	* sysdeps/m68k/fpu/e_powf.c: Based on e_pow.c.
	* sysdeps/m68k/fpu/e_powl.c: Likewise.

	* sysdeps/m68k/fpu/s_log2.c: New file.
	* sysdeps/m68k/fpu/s_log2f.c: New file.
	* sysdeps/m68k/fpu/s_log2l.c: New file.
	* sysdeps/m68k/fpu/s_exp2.c: New file.
	* sysdeps/m68k/fpu/s_exp2f.c: New file.
	* sysdeps/m68k/fpu/s_exp2l.c: New file.

1997-03-20 14:46  Andreas Schwab  <schwab@issan.informatik.uni-dortmund.de>

	* Makeconfig (all): Make sure this is always the default goal.

1997-03-20 11:09  Andreas Schwab  <schwab@issan.informatik.uni-dortmund.de>

	* math/libm-test.c: Increase epsilon in many places.  Fix many
	typos.  Allow optinal argument also for short option.

1997-03-20 11:09  Andreas Schwab  <schwab@issan.informatik.uni-dortmund.de>

	* math/Makefile ($(objpfx)$(tests)): Link against libm.

1997-03-24 23:14  Ulrich Drepper  <drepper@cygnus.com>

	* sysdeps/libm-ieee754/s_fpclassifyl.c: Correct recognition of
	denormalized numbers.

	* sysdeps/i386/huge_val.h: Remove references to byte order macros.
	Don't include <endian.h>.

1997-03-19 15:18  Andreas Schwab  <schwab@issan.informatik.uni-dortmund.de>
	* sysdeps/m68k/huge_val.h: Remove references to byte order macros.
	Don't include <endian.h>.

1997-03-24 23:09  Ulrich Drepper  <drepper@cygnus.com>

	* inet/rcmd.c (iruserok): Use access instead of euidaccess.

1997-03-15 18:08  Andreas Schwab  <schwab@issan.informatik.uni-dortmund.de>

	* math/cmathcalls.h: Remove whitespace before function name in use
	of __MATHDECL macro, to make it compilable with a traditional
	preprocessor.

1997-03-24 15:31  Ulrich Drepper  <drepper@cygnus.com>

	* configure.in: Use AC_PROG_CC instead of AC_PROG_TOOL to find
	compiler.

1997-03-24 02:34  Ulrich Drepper  <drepper@cygnus.com>

	* sysdeps/libm-ieee754/s_finitel.c (__finitel): Shift return value
	by 31 positions to get 0/1 result.
	Patch by Joe Keane <jgk@jgk.org>.

1997-03-23 12:15  H.J. Lu  <hjl@gnu.ai.mit.edu>

	* posix/getopt.c (__getopt_nonoption_flags): Make it extern
	to prevent from ld linking in getopt.o even if there is another
	incompatible one.

	* posix/getopt_init.c (__getopt_nonoption_flags): Remove extern.

1997-03-23 23:30   Thorsten Kukuk  <kukuk@vt.uni-paderborn.de>

	* nis/nis_call.c (__do_niscall): Print message if cold start file
	does not exist.
	* nis/nis_file.c: Don't print error messages.
	* nis/nis_local_name.c (nis_local_host): Fix pointer errors.
	* nis/rpcsvc/ypclnt.h: Fix copyright and prototypes.
	* nis/rpcsvc/nis.h: Likewise.
	* nis/ypclnt.c: Fix prototypes.

1997-03-24 01:36  Ulrich Drepper  <drepper@cygnus.com>

	* sysdeps/libm-i387/e_scalb.S: Pop additional result before
	returning.
	* sysdeps/libm-i387/e_scalbf.S: Likewise.
	* sysdeps/libm-i387/e_scalbl.S: Likewise.
	Reported by Andreas Jaeger <aj@arthur.pfalz.de>.

	* elf/rtld.c (dl_main): Set main_map->l_opencount to 1 also if
	dynamic linker was called implicitly by the kernel.
	Reported by Eric Youngdale <eric@andante.jic.com>.

	* math/Makefile (CFLAGS-test-float.c, CFLAGS-test-double.c,
	CFLAGS-test-longdouble.c): New variables.  Set to -fno-inline to
	prevent clever optimizations which corrupt the tests.

1997-03-23 21:33  Andreas Jaeger  <aj@arthur.pfalz.de>

	* math/libm-test.c (scalb_test): New function.
	(sqrt_test): New function.
	(scalbn_test): New function.
	(ilogb_test): New function.
	(main): Added calls for new test functions.
	(ldexp_test): Add another test for ldexp(x,0).

1997-03-23 12:35  Andreas Jaeger  <aj@arthur.pfalz.de>

	* math/libm-test.c (remquo_test): Correct messages so that they
	match the tests.
	(copysign_test): Likewise.

1997-03-23 16:28  Ulrich Drepper  <drepper@cygnus.com>

	* stdio-common/Makefile (routines): Add printf_size.

	* stdio-common/printf.h: Add declarations for printf_size and
	printf_size_info.

	* stdio-common/printf-parse.h (__printf_arginfo_table): Correct
	declaration.
	(parse_one_spec): Test whether __printf_function_table is not NULL
	before using registered handlers and don't test
	__printf_arginfo_table.  Update nargs also when handler is
	registered.

	* stdio-common/vfprintf.c (vfprintf): Count number of specifiers
	processed in fast loop.

	* stdio-common/printf_size: New file.

1997-03-22 04:53  Ulgmon_out_readrich Drepper  <drepper@cygnus.com>
1997-03-25 01:45:04 +00:00

128 lines
4.6 KiB
Makefile

# Copyright (C) 1996, 1997 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 Library General Public License as
# published by the Free Software Foundation; either version 2 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
# Library General Public License for more details.
# You should have received a copy of the GNU Library General Public
# License along with the GNU C Library; see the file COPYING.LIB. If not,
# write to the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
# Boston, MA 02111-1307, USA.
# Makefile for the math library.
subdir := math
# Installed header files.
headers := math.h mathcalls.h __math.h huge_val.h nan.h \
fpu_control.h complex.h cmathcalls.h
# Internal header files.
distribute := math_private.h machine/asm.h machine/endian.h
# FPU support code.
aux := fpu_control setfpucw
# Build the -lm library.
extra-libs := libm
extra-libs-others = $(extra-libs)
libm-support = k_standard s_lib_version s_matherr s_signgam
libm-calls = e_acos e_acosh e_asin e_atan2 e_atanh e_cosh e_exp e_fmod \
e_hypot e_j0 e_j1 e_jn e_lgamma_r e_log e_log10 e_pow \
e_rem_pio2 e_remainder e_scalb e_sinh e_sqrt k_cos \
k_rem_pio2 k_sin k_tan s_asinh s_atan s_cbrt \
s_ceil s_copysign s_cos s_erf s_expm1 s_fabs s_finite \
s_floor s_frexp s_ilogb s_ldexp s_log1p s_logb \
s_modf s_nextafter s_rint s_scalbn s_significand \
s_sin s_tan s_tanh w_acos w_acosh w_asin \
w_atan2 w_atanh w_cosh w_drem w_exp w_fmod w_gamma \
w_gamma_r w_hypot w_j0 w_j1 w_jn w_lgamma w_lgamma_r \
w_log w_log10 w_pow w_remainder w_scalb w_sinh w_sqrt \
s_signbit s_fpclassify s_fmax s_fmin s_fdim s_nan s_trunc \
s_remquo \
conj cimag creal cabs s_cexp
libm-routines = $(libm-support) $(libm-calls) \
$(patsubst %_rf,%f_r,$(libm-calls:=f)) \
$(long-m-$(long-double-fcts))
long-m-routines = $(patsubst %_rl,%l_r,$(libm-calls:=l))
long-m-yes = $(long-m-routines)
distribute += $(long-m-yes:=.c)
# These functions are in libc instead of libm because __printf_fp
# calls them, so any program using printf will need them linked in,
# and we don't want to have to link every program with -lm.
calls = s_isinf s_isnan s_finite s_copysign s_modf s_scalbn s_frexp s_ldexp
routines = $(calls) $(calls:=f) $(long-c-$(long-double-fcts))
long-c-yes = $(calls:=l)
distribute += $(long-c-yes:=.c)
# Rules for the test suite.
tests = test-float test-double $(test-longdouble-$(long-double-fcts))
# We do the `long double' tests only if this data type is available and
# distrinct from `double'.
#
# XXX This test is disabled for now since the functions are too buggy.
#test-longdouble-yes = test-longdouble
CFLAGS-test-float.c = -fno-inline
CFLAGS-test-double.c = -fno-inline
CFLAGS-test-longdouble.c = -fno-inline
LDLIBS-test-float = libm
LDLIBS-test-double = libm
LDLIBS-test-longdouble = libm
distribute += libm-test.c
# The -lieee module sets the _LIB_VERSION_ switch to IEEE mode
# for error handling in the -lm functions.
install-lib += libieee.a
non-lib.a += libieee.a
extra-objs += libieee.a ieee-math.o
distribute += ieee-math.c
include ../Rules
# This file defines the default _LIB_VERSION variable that controls
# the error return conventions for the math functions.
CPPFLAGS-s_lib_version.c := -D_POSIX_MODE
# We don't want the fdlibm code to use the inline math functions,
# only the fdlibm code.
CPPFLAGS += -D__NO_MATH_INLINES
ifneq ($(long-double-fcts),yes)
# The `double' and `long double' types are the same on this machine.
# We won't compile the `long double' code at all. Tell the `double' code
# to define aliases for the `FUNCl' names. To avoid type conflicts in
# defining those aliases, tell <math.h> to declare the `FUNCl' names with
# `double' instead of `long double'.
CPPFLAGS += -DNO_LONG_DOUBLE -D_Mlong_double_=double
endif
# The fdlibm code generates a lot of these warnings but is otherwise clean.
override CFLAGS += -Wno-uninitialized -Wno-write-strings
# The -lieee library is actually an object file.
# The module just defines the _LIB_VERSION_ variable.
# It's not a library to make sure it is linked in instead of s_lib_version.o.
$(objpfx)libieee.a: $(objpfx)ieee-math.o
rm -f $@
ln $< $@
ifeq ($(build-shared),yes)
$(addprefix $(objpfx),$(tests)): $(objpfx)libm.so$(libm.so-version)
else
$(addprefix $(objpfx),$(tests)): $(objpfx)libm.a
endif