Update.
1997-06-19 19:38 Ulrich Drepper <drepper@cygnus.com>
* features.h: Define __STDC_IEC_559__ and _STDC_IEC_559_COMPLEX__.
* elf/dl-minimal.c (__dcgettext): Remove assertion.
* inet/rcmd.c: Correct a few typos. Reported by Eric Troan.
* manual/Makefile (distribute): Add dir.
* manual/dir: New file.
* math/Makefile (libm-support): Rename s_rinttol, s_rinttoll,
s_roundtol, and s_roundtoll to s_lrint, s_llrint, s_lround,
and s_llround respectively.
(libm-calls): Add e_gamma_r.
* math/libm-test.c (check_int_exc): New function.
(signbit_test): Remove test for sign of NaN.
(gamma_test): Clear exception after test of existence.
Correct tests and and epsilons.
(lgamma_test): Likewise.
(ilogb_test): Correct all tests.
(scalb_test): Rewrite.
(rinttol_test): Rename to lrint_test and correct tests.
(rinttoll_test): Likewise.
(roundtol_test): Likewise.
(roundtoll_test): Likewise.
(main): Call lrint/lround functions instead of rinttol/roundtol.
* math/math.h: Change prototypes for rinttol/roundtol.
* math/mathcalls.h: Rearrange prototypes according to ISO C9X draft.
* sysdeps/generic/mathbits.h: Define FP_ILOGB0 and FP_ILOGBNAN.
* sysdeps/i386/mathbits.h: Likewise.
* sysdeps/libm-i387/e_scalb.S: Handle special cases correctly.
* sysdeps/libm-i387/e_scalbf.S: Likewise.
* sysdeps/libm-i387/e_scalbl.S: Likewise.
* sysdeps/libm-i387/s_asinh.S: Handle -inf correctly.
* sysdeps/libm-i387/s_asinhf.S: Likewise.
* sysdeps/libm-i387/s_asinhl.S: Likewise.
* sysdeps/libm-i387/s_ilogb.S: Optimize.
* sysdeps/libm-i387/s_ilogbf.S: Likewise.
* sysdeps/libm-i387/s_ilogbl.S: Likewise.
* sysdeps/libm-i387/s_rinttol.S: Rename to...
* sysdeps/libm-i387/s_lrint.S: ...this.
* sysdeps/libm-i387/s_rinttoll.S: Rename to...
* sysdeps/libm-i387/s_llrint.S: ...this.
* sysdeps/libm-i387/s_remquo.S: Correctly set sign of remainder.
* sysdeps/libm-i387/s_remquof.S: Likewise.
* sysdeps/libm-i387/s_remquol.S: Likewise.
* sysdeps/libm-i387/e_gamma_r.c: New file. Implementation of gamma
function according to ISO C.
* sysdeps/libm-i387/e_gammaf_r.c: New file.
* sysdeps/libm-i387/e_gammal_r.c: New file.
* sysdeps/libm-i387/e_lgamma_r.c: Don't let optimize compile the
generation of exceptions away.
* sysdeps/libm-i387/e_lgammaf_r.c: Likewise.
* sysdeps/libm-i387/k_standard.c: Correct return value for infinity
points of gamma function when not SVID mode.
* sysdeps/libm-i387/s_rinttoll.c: Renamed to...
* sysdeps/libm-i387/s_llrint.c: ...this.
* sysdeps/libm-i387/s_rinttol.c: Renamed to...
* sysdeps/libm-i387/s_lrint.c: ...this.
* sysdeps/libm-i387/s_roundtoll.c: Renamed to...
* sysdeps/libm-i387/s_llround.c: ...this.
* sysdeps/libm-i387/s_roundtol.c: Renamed to...
* sysdeps/libm-i387/s_lround.c: ..this.
* sysdeps/libm-i387/s_scalbn.c: Change second parameter according to
ISO C.
* sysdeps/libm-i387/s_scalbnf.c: Likewise.
* sysdeps/libm-i387/s_scalbnl.c: Likewise.
* sysdeps/libm-i387/w_gamma.c: Call __ieee754_gamma_r if library
mode is not _SVID_.
* sysdeps/libm-i387/w_gammaf.c: Likewise.
* sysdeps/libm-i387/w_gammal.c: Likewise.
* sysdeps/m68k/fpu/__math.h: Rename __rinttol to __lrint and
rinttol to lrint.
* sysdeps/m68k/fpu/s_rinttol.c: Renamed to...
* sysdeps/m68k/fpu/s_lrint.c: ...this.
* sysdeps/m68k/fpu/s_rinttoll.c: Renamed to...
* sysdeps/m68k/fpu/s_llrint.c: ...this.
* md5-crypt/Makefile: Link md5test program with md5.o.
* stdio-common/temptest.c: Don't use __stdio_gen_tempname which
is not exported by the libc.so.
* stdio-common/vfscanf.c: Correct scanning of strings after last
change.
* sysdeps/unix/sysv/linux/i386/sysdep.S: Use .comm to define errno.
1997-06-19 07:37 H.J. Lu <hjl@gnu.ai.mit.edu>
* time/tzfile.c (__tzfile_read): Store getc () return in int.
1997-06-13 Miles Bader <miles@gnu.ai.mit.edu>
* argp/argp-parse.c (argp_version_parser): Include `(PROGRAM ERROR)' in
the no-version error text to indicate that something's fucked.
[!_] (N_): New macro.
(argp_default_options, argp_version_options): Wrap doc strings in N_().
1997-06-12 Miles Bader <miles@gnu.ai.mit.edu>
* argp/argp-parse.c (parser_parse_opt): Detect and report unhandled
options here.
(parser_parse_arg): Handle ARGP_KEY_ARGS here.
Adjust NEXT pointer back if we fail to parse anything.
(parser_parse_next): Simplify arg code. Leave state NEXT frobbing
to parser_parse_arg.
1997-06-11 Miles Bader <miles@gnu.ai.mit.edu>
* argp/argp.h (ARGP_KEY_ARGS, ARGP_KEY_FINI): New macros.
* argp/argp-parse.c (parser_finalize): Do another pass over the
parsers with ARGP_KEY_FINI.
1997-06-18 Miles Bader <miles@gnu.ai.mit.edu>
* string/Makefile (routines): Add argz-replace.
1997-06-16 00:16 Miles Bader <miles@gnu.ai.mit.edu>
* manual/string.texi (Argz Functions): Document argz_replace.
1997-06-12 Miles Bader <miles@gnu.ai.mit.edu>
* string/argz.h (__argz_replace, argz_replace): New declarations.
* string/argz-replace.c: New file.
1997-06-11 Miles Bader <miles@gnu.ai.mit.edu>
* manual/argp.texi (Argp Special Keys): Document ARGP_KEY_ARGS.
1997-06-16 23:08 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* Makeconfig (libc-map): Remove definition.
* Makerules (libc-map): Define it here, using the full name.
(load-map-file): Remove case for empty $(..).
($(common-objpfx)libc.so): Revert last change.
1997-06-17 22:18 Mark Kettenis <kettenis@phys.uva.nl>
* login/programs/utmpd.c (main): Improve signal handling.
* login/programs/request.c (do_setutent, do_updwtmp):
Allow arbitrary length filenames.
* login/programs/utmpd.h (setutent_request, updwtmp_request):
Get rid of fixed length file field.
* login/utmp_daemon.c (do_setutent, do_updwtmp):
Allow arbitrary length filenames.
* login/programs/request.c (do_pututline):
Don't fail if connection->position is -1 on entry.
1997-06-15 16:32 Mark Kettenis <kettenis@phys.uva.nl>
* login/utmp_file.c (updwtmp_file): Use the same method for
appending an entry as in pututline_file.
1997-06-11 18:59 Mark Kettenis <kettenis@phys.uva.nl>
* login/utmp_file.c (getutent_r_file):
Use read lock instead of write lock.
(getutline_r_file, internal_getut_r): Lock utmp file.
(updwtmp_file): Use fcntl to lock file instead of flock.
1997-06-18 00:11 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* stdio-common/vfscanf.c (inchar, ungetc): Don't count EOF as
character read in/put back.
* stdio-common/tstscanf.c: Add test case for this.
1997-06-17 22:17 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* libc.map: Add more symbols.
1997-06-18 12:01 Ulrich Drepper <drepper@cygnus.com>
* manual/Makefile (install): Make sure `dir' file exists if we use
install-info.
1997-06-17 19:32 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* manual/Makefile (info): Depend on dir-add.info.
1997-06-17 17:19 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* elf/ldd.bash.in: Fix spacing in message.
1997-06-17 14:28 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* elf/do-rel.h (elf_dynamic_do_rel): Always use version
information if available.
1997-06-17 11:34 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sunrpc/Makefile ($(rpcsvc:%.x=$(objpfx)rpcsvc/%.h)): Make
command non-empty to force make to recheck modification time.
($(rpcsvc:%.x=$(objpfx)x%.c)): Likewise.
1997-06-17 00:26 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/stub/e_acoshl.c: Set errno to ENOSYS.
* sysdeps/stub/e_acosl.c: Likewise.
* sysdeps/stub/e_asinl.c: Likewise.
* sysdeps/stub/e_atan2l.c: Likewise.
* sysdeps/stub/e_expl.c: Likewise.
* sysdeps/stub/e_fmodl.c: Likewise.
* sysdeps/stub/e_j0l.c: Likewise.
* sysdeps/stub/e_j1l.c: Likewise.
* sysdeps/stub/e_jnl.c: Likewise.
* sysdeps/stub/e_lgammal_r.c: Likewise.
* sysdeps/stub/e_log10l.c: Likewise.
* sysdeps/stub/e_logl.c: Likewise.
* sysdeps/stub/e_powl.c: Likewise.
* sysdeps/stub/e_rem_pio2l.c: Likewise.
* sysdeps/stub/e_sqrtl.c: Likewise.
* sysdeps/stub/k_cosl.c: Likewise.
* sysdeps/stub/k_rem_pio2l.c: Likewise.
* sysdeps/stub/k_sinl.c: Likewise.
* sysdeps/stub/k_tanl.c: Likewise.
* sysdeps/stub/s_atanl.c: Likewise.
* sysdeps/stub/s_erfl.c: Likewise.
* sysdeps/stub/s_exp2.c: Likewise.
* sysdeps/stub/s_exp2f.c: Likewise.
* sysdeps/stub/s_exp2l.c: Likewise.
* sysdeps/stub/s_expm1l.c: Likewise.
* sysdeps/stub/s_log1pl.c: Likewise.
* sysdeps/stub/s_log2l.c: Likewise.
1997-06-18 11:46 Ulrich Drepper <drepper@cygnus.com>
* sysdeps/libm-ieee754/w_gamma.c: If _LIB_VERSION is _SVID_ compute
result as before last change.
* sysdeps/libm-ieee754/w_gammaf.c: Likewise.
* sysdeps/libm-ieee754/w_gammal.c: Likewise.
1997-06-16 23:37 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/libm-ieee754/s_remquo.c: Fix off-by-one when computing
quotient.
* sysdeps/libm-ieee754/s_remquof.c: Likewise.
* sysdeps/libm-ieee754/s_remquol.c: Likewise.
* sysdeps/m68k/fpu/s_remquo.c: Remove FIXME and special case for
quotient.
* sysdeps/libm-ieee754/w_gamma.c: Add missing call to exp
function. Don't use global signgam.
* sysdeps/libm-ieee754/w_gammaf.c: Likewise.
* sysdeps/libm-ieee754/w_gammal.c: Likewise.
* math/Makefile (libm-calls): Remove w_gamma_r.
* sysdeps/libm-ieee754/w_gamma_r.c: Remove file.
* sysdeps/libm-ieee754/w_gammaf_r.c: Remove file.
* sysdeps/libm-ieee754/w_gammal_r.c: Remove file.
* math/libm-test.c (atanh_test): Declare x only if needed.
(signbit_test): Fix typo.
(gamma_test): Check whether function is implemented. Add
epsilons.
(lgamma_test): Likewise.
(fmod_test): Add epsilons.
(exp2_test): Use right function for existence test.
1997-06-07 09:20 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/powerpc/Dist: Add fenv_const.c.
* sysdeps/unix/sysv/linux/Dist: Add net/if_slip.h.
* sysdeps/unix/sysv/linux/powerpc/Dist: Add init-first.h and
syscall.h.
* sysdeps/unix/sysv/linux/sparc/Dist: Add init-first.h.
* string/Makefile (distribute): Add tst-svc.expect.
* nis/Makefile (distribute): Add nis_intern.h and Banner.
* elf/Makefile (distribute): Add dl-hash.h.
* Rules (subdir_echo-distinfo): Add headers from $(distribute).
* login/Makefile (others): Add utmpdump.
* sysdeps/unix/sysv/linux/Makefile (sysdep_headers): Add
net/if_slip.h.
* manual/Makefile (dir-add.texi): Also look in indirectly included
files.
1997-06-16 23:15 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/unix/sysv/linux/configure.in: Remove commands to
generate stdio_lim.h.
* sysdeps/unix/sysv/linux/mk-stdiolim.c: Remove.
* sysdeps/unix/sysv/linux/Makefile: Add rules to generate
stdio_lim.h here.
(common-generated): Add generated files.
(inhibit-stdio_lim): Define.
* sysdeps/posix/Makefile [$(inhibit-stdio_lim)=yes]: Disable rules
to generate stdio_lim.h.
* sysdeps/unix/sysv/linux/Makefile: Suppress inclusion of
dependecy files if no_deps is set.
($(objpfx)syscall-%.d): Add header file as target to dependency
generation.
1997-06-14 19:19 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/unix/sysv/linux/arm/Dist: New file.
1997-06-14 17:59 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* extra-lib.mk (others): Depend on versioned shared library, not
the unversioned one.
* Makerules (build-shlib): Don't make the version link here.
($(common-objpfx)libc.so$(libc.so-version)): New rule for libc
version link.
1997-06-16 03:07 Ulrich Drepper <drepper@cygnus.com>
* sysdeps/libm-ieee754/k_standard.c: Undo change of Tue Aug 6
01:13:56 1996.
* argp/argp-help.c (argp_args_usage): Supply correct argp to
filter_doc.
* argp/argp-help.c (hol_add_cluster): Initialize CL->depth.
* argp/argp-help.c (_help): Supply STATE to argp_args_usage.
* argp/argp.h (ARGP_KEY_HELP_ARGS_DOC): New macro.
* argp/argp-fmtstream.c: Add casts to prevent warnings.
* argp/argp.h (OPTION_NO_USAGE): New macro.
* argp/argp-help.c (usage_long_opt, usage_argful_short_opt,
* argp/argp-fmtstream.c (__argp_fmtstream_update): Account for case
* argp/argp-help.c <stddef.h>: New include.
* argp/argp.h (argp_state_help, __argp_state_help, argp_usage,
* argp/argp.h (argp_program_bug_address): Make const.
* argp/argp-parse.c (argp_default_parser): Set STATE->name for OPT_PROGNAME.
* argp/argp-help.c (__argp_error, __argp_failure, __argp_state_help):
* argp/argp-parse.c (parser_init): Set PARSER->state.flags.
1997-06-19 19:11:22 +00:00
|
|
|
/* Implementation of gamma function according to ISO C.
|
2018-01-01 00:32:25 +00:00
|
|
|
Copyright (C) 1997-2018 Free Software Foundation, Inc.
|
1997-04-08 23:42:08 +00:00
|
|
|
This file is part of the GNU C Library.
|
|
|
|
Contributed by Ulrich Drepper <drepper@cygnus.com>, 1997.
|
|
|
|
|
|
|
|
The GNU C Library is free software; you can redistribute it and/or
|
2001-07-06 04:58:11 +00:00
|
|
|
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.
|
1997-04-08 23:42:08 +00:00
|
|
|
|
|
|
|
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
|
2001-07-06 04:58:11 +00:00
|
|
|
Lesser General Public License for more details.
|
1997-04-08 23:42:08 +00:00
|
|
|
|
2001-07-06 04:58:11 +00:00
|
|
|
You should have received a copy of the GNU Lesser General Public
|
2012-02-09 23:18:22 +00:00
|
|
|
License along with the GNU C Library; if not, see
|
|
|
|
<http://www.gnu.org/licenses/>. */
|
1997-04-08 23:42:08 +00:00
|
|
|
|
Update.
1997-06-19 19:38 Ulrich Drepper <drepper@cygnus.com>
* features.h: Define __STDC_IEC_559__ and _STDC_IEC_559_COMPLEX__.
* elf/dl-minimal.c (__dcgettext): Remove assertion.
* inet/rcmd.c: Correct a few typos. Reported by Eric Troan.
* manual/Makefile (distribute): Add dir.
* manual/dir: New file.
* math/Makefile (libm-support): Rename s_rinttol, s_rinttoll,
s_roundtol, and s_roundtoll to s_lrint, s_llrint, s_lround,
and s_llround respectively.
(libm-calls): Add e_gamma_r.
* math/libm-test.c (check_int_exc): New function.
(signbit_test): Remove test for sign of NaN.
(gamma_test): Clear exception after test of existence.
Correct tests and and epsilons.
(lgamma_test): Likewise.
(ilogb_test): Correct all tests.
(scalb_test): Rewrite.
(rinttol_test): Rename to lrint_test and correct tests.
(rinttoll_test): Likewise.
(roundtol_test): Likewise.
(roundtoll_test): Likewise.
(main): Call lrint/lround functions instead of rinttol/roundtol.
* math/math.h: Change prototypes for rinttol/roundtol.
* math/mathcalls.h: Rearrange prototypes according to ISO C9X draft.
* sysdeps/generic/mathbits.h: Define FP_ILOGB0 and FP_ILOGBNAN.
* sysdeps/i386/mathbits.h: Likewise.
* sysdeps/libm-i387/e_scalb.S: Handle special cases correctly.
* sysdeps/libm-i387/e_scalbf.S: Likewise.
* sysdeps/libm-i387/e_scalbl.S: Likewise.
* sysdeps/libm-i387/s_asinh.S: Handle -inf correctly.
* sysdeps/libm-i387/s_asinhf.S: Likewise.
* sysdeps/libm-i387/s_asinhl.S: Likewise.
* sysdeps/libm-i387/s_ilogb.S: Optimize.
* sysdeps/libm-i387/s_ilogbf.S: Likewise.
* sysdeps/libm-i387/s_ilogbl.S: Likewise.
* sysdeps/libm-i387/s_rinttol.S: Rename to...
* sysdeps/libm-i387/s_lrint.S: ...this.
* sysdeps/libm-i387/s_rinttoll.S: Rename to...
* sysdeps/libm-i387/s_llrint.S: ...this.
* sysdeps/libm-i387/s_remquo.S: Correctly set sign of remainder.
* sysdeps/libm-i387/s_remquof.S: Likewise.
* sysdeps/libm-i387/s_remquol.S: Likewise.
* sysdeps/libm-i387/e_gamma_r.c: New file. Implementation of gamma
function according to ISO C.
* sysdeps/libm-i387/e_gammaf_r.c: New file.
* sysdeps/libm-i387/e_gammal_r.c: New file.
* sysdeps/libm-i387/e_lgamma_r.c: Don't let optimize compile the
generation of exceptions away.
* sysdeps/libm-i387/e_lgammaf_r.c: Likewise.
* sysdeps/libm-i387/k_standard.c: Correct return value for infinity
points of gamma function when not SVID mode.
* sysdeps/libm-i387/s_rinttoll.c: Renamed to...
* sysdeps/libm-i387/s_llrint.c: ...this.
* sysdeps/libm-i387/s_rinttol.c: Renamed to...
* sysdeps/libm-i387/s_lrint.c: ...this.
* sysdeps/libm-i387/s_roundtoll.c: Renamed to...
* sysdeps/libm-i387/s_llround.c: ...this.
* sysdeps/libm-i387/s_roundtol.c: Renamed to...
* sysdeps/libm-i387/s_lround.c: ..this.
* sysdeps/libm-i387/s_scalbn.c: Change second parameter according to
ISO C.
* sysdeps/libm-i387/s_scalbnf.c: Likewise.
* sysdeps/libm-i387/s_scalbnl.c: Likewise.
* sysdeps/libm-i387/w_gamma.c: Call __ieee754_gamma_r if library
mode is not _SVID_.
* sysdeps/libm-i387/w_gammaf.c: Likewise.
* sysdeps/libm-i387/w_gammal.c: Likewise.
* sysdeps/m68k/fpu/__math.h: Rename __rinttol to __lrint and
rinttol to lrint.
* sysdeps/m68k/fpu/s_rinttol.c: Renamed to...
* sysdeps/m68k/fpu/s_lrint.c: ...this.
* sysdeps/m68k/fpu/s_rinttoll.c: Renamed to...
* sysdeps/m68k/fpu/s_llrint.c: ...this.
* md5-crypt/Makefile: Link md5test program with md5.o.
* stdio-common/temptest.c: Don't use __stdio_gen_tempname which
is not exported by the libc.so.
* stdio-common/vfscanf.c: Correct scanning of strings after last
change.
* sysdeps/unix/sysv/linux/i386/sysdep.S: Use .comm to define errno.
1997-06-19 07:37 H.J. Lu <hjl@gnu.ai.mit.edu>
* time/tzfile.c (__tzfile_read): Store getc () return in int.
1997-06-13 Miles Bader <miles@gnu.ai.mit.edu>
* argp/argp-parse.c (argp_version_parser): Include `(PROGRAM ERROR)' in
the no-version error text to indicate that something's fucked.
[!_] (N_): New macro.
(argp_default_options, argp_version_options): Wrap doc strings in N_().
1997-06-12 Miles Bader <miles@gnu.ai.mit.edu>
* argp/argp-parse.c (parser_parse_opt): Detect and report unhandled
options here.
(parser_parse_arg): Handle ARGP_KEY_ARGS here.
Adjust NEXT pointer back if we fail to parse anything.
(parser_parse_next): Simplify arg code. Leave state NEXT frobbing
to parser_parse_arg.
1997-06-11 Miles Bader <miles@gnu.ai.mit.edu>
* argp/argp.h (ARGP_KEY_ARGS, ARGP_KEY_FINI): New macros.
* argp/argp-parse.c (parser_finalize): Do another pass over the
parsers with ARGP_KEY_FINI.
1997-06-18 Miles Bader <miles@gnu.ai.mit.edu>
* string/Makefile (routines): Add argz-replace.
1997-06-16 00:16 Miles Bader <miles@gnu.ai.mit.edu>
* manual/string.texi (Argz Functions): Document argz_replace.
1997-06-12 Miles Bader <miles@gnu.ai.mit.edu>
* string/argz.h (__argz_replace, argz_replace): New declarations.
* string/argz-replace.c: New file.
1997-06-11 Miles Bader <miles@gnu.ai.mit.edu>
* manual/argp.texi (Argp Special Keys): Document ARGP_KEY_ARGS.
1997-06-16 23:08 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* Makeconfig (libc-map): Remove definition.
* Makerules (libc-map): Define it here, using the full name.
(load-map-file): Remove case for empty $(..).
($(common-objpfx)libc.so): Revert last change.
1997-06-17 22:18 Mark Kettenis <kettenis@phys.uva.nl>
* login/programs/utmpd.c (main): Improve signal handling.
* login/programs/request.c (do_setutent, do_updwtmp):
Allow arbitrary length filenames.
* login/programs/utmpd.h (setutent_request, updwtmp_request):
Get rid of fixed length file field.
* login/utmp_daemon.c (do_setutent, do_updwtmp):
Allow arbitrary length filenames.
* login/programs/request.c (do_pututline):
Don't fail if connection->position is -1 on entry.
1997-06-15 16:32 Mark Kettenis <kettenis@phys.uva.nl>
* login/utmp_file.c (updwtmp_file): Use the same method for
appending an entry as in pututline_file.
1997-06-11 18:59 Mark Kettenis <kettenis@phys.uva.nl>
* login/utmp_file.c (getutent_r_file):
Use read lock instead of write lock.
(getutline_r_file, internal_getut_r): Lock utmp file.
(updwtmp_file): Use fcntl to lock file instead of flock.
1997-06-18 00:11 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* stdio-common/vfscanf.c (inchar, ungetc): Don't count EOF as
character read in/put back.
* stdio-common/tstscanf.c: Add test case for this.
1997-06-17 22:17 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* libc.map: Add more symbols.
1997-06-18 12:01 Ulrich Drepper <drepper@cygnus.com>
* manual/Makefile (install): Make sure `dir' file exists if we use
install-info.
1997-06-17 19:32 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* manual/Makefile (info): Depend on dir-add.info.
1997-06-17 17:19 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* elf/ldd.bash.in: Fix spacing in message.
1997-06-17 14:28 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* elf/do-rel.h (elf_dynamic_do_rel): Always use version
information if available.
1997-06-17 11:34 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sunrpc/Makefile ($(rpcsvc:%.x=$(objpfx)rpcsvc/%.h)): Make
command non-empty to force make to recheck modification time.
($(rpcsvc:%.x=$(objpfx)x%.c)): Likewise.
1997-06-17 00:26 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/stub/e_acoshl.c: Set errno to ENOSYS.
* sysdeps/stub/e_acosl.c: Likewise.
* sysdeps/stub/e_asinl.c: Likewise.
* sysdeps/stub/e_atan2l.c: Likewise.
* sysdeps/stub/e_expl.c: Likewise.
* sysdeps/stub/e_fmodl.c: Likewise.
* sysdeps/stub/e_j0l.c: Likewise.
* sysdeps/stub/e_j1l.c: Likewise.
* sysdeps/stub/e_jnl.c: Likewise.
* sysdeps/stub/e_lgammal_r.c: Likewise.
* sysdeps/stub/e_log10l.c: Likewise.
* sysdeps/stub/e_logl.c: Likewise.
* sysdeps/stub/e_powl.c: Likewise.
* sysdeps/stub/e_rem_pio2l.c: Likewise.
* sysdeps/stub/e_sqrtl.c: Likewise.
* sysdeps/stub/k_cosl.c: Likewise.
* sysdeps/stub/k_rem_pio2l.c: Likewise.
* sysdeps/stub/k_sinl.c: Likewise.
* sysdeps/stub/k_tanl.c: Likewise.
* sysdeps/stub/s_atanl.c: Likewise.
* sysdeps/stub/s_erfl.c: Likewise.
* sysdeps/stub/s_exp2.c: Likewise.
* sysdeps/stub/s_exp2f.c: Likewise.
* sysdeps/stub/s_exp2l.c: Likewise.
* sysdeps/stub/s_expm1l.c: Likewise.
* sysdeps/stub/s_log1pl.c: Likewise.
* sysdeps/stub/s_log2l.c: Likewise.
1997-06-18 11:46 Ulrich Drepper <drepper@cygnus.com>
* sysdeps/libm-ieee754/w_gamma.c: If _LIB_VERSION is _SVID_ compute
result as before last change.
* sysdeps/libm-ieee754/w_gammaf.c: Likewise.
* sysdeps/libm-ieee754/w_gammal.c: Likewise.
1997-06-16 23:37 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/libm-ieee754/s_remquo.c: Fix off-by-one when computing
quotient.
* sysdeps/libm-ieee754/s_remquof.c: Likewise.
* sysdeps/libm-ieee754/s_remquol.c: Likewise.
* sysdeps/m68k/fpu/s_remquo.c: Remove FIXME and special case for
quotient.
* sysdeps/libm-ieee754/w_gamma.c: Add missing call to exp
function. Don't use global signgam.
* sysdeps/libm-ieee754/w_gammaf.c: Likewise.
* sysdeps/libm-ieee754/w_gammal.c: Likewise.
* math/Makefile (libm-calls): Remove w_gamma_r.
* sysdeps/libm-ieee754/w_gamma_r.c: Remove file.
* sysdeps/libm-ieee754/w_gammaf_r.c: Remove file.
* sysdeps/libm-ieee754/w_gammal_r.c: Remove file.
* math/libm-test.c (atanh_test): Declare x only if needed.
(signbit_test): Fix typo.
(gamma_test): Check whether function is implemented. Add
epsilons.
(lgamma_test): Likewise.
(fmod_test): Add epsilons.
(exp2_test): Use right function for existence test.
1997-06-07 09:20 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/powerpc/Dist: Add fenv_const.c.
* sysdeps/unix/sysv/linux/Dist: Add net/if_slip.h.
* sysdeps/unix/sysv/linux/powerpc/Dist: Add init-first.h and
syscall.h.
* sysdeps/unix/sysv/linux/sparc/Dist: Add init-first.h.
* string/Makefile (distribute): Add tst-svc.expect.
* nis/Makefile (distribute): Add nis_intern.h and Banner.
* elf/Makefile (distribute): Add dl-hash.h.
* Rules (subdir_echo-distinfo): Add headers from $(distribute).
* login/Makefile (others): Add utmpdump.
* sysdeps/unix/sysv/linux/Makefile (sysdep_headers): Add
net/if_slip.h.
* manual/Makefile (dir-add.texi): Also look in indirectly included
files.
1997-06-16 23:15 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/unix/sysv/linux/configure.in: Remove commands to
generate stdio_lim.h.
* sysdeps/unix/sysv/linux/mk-stdiolim.c: Remove.
* sysdeps/unix/sysv/linux/Makefile: Add rules to generate
stdio_lim.h here.
(common-generated): Add generated files.
(inhibit-stdio_lim): Define.
* sysdeps/posix/Makefile [$(inhibit-stdio_lim)=yes]: Disable rules
to generate stdio_lim.h.
* sysdeps/unix/sysv/linux/Makefile: Suppress inclusion of
dependecy files if no_deps is set.
($(objpfx)syscall-%.d): Add header file as target to dependency
generation.
1997-06-14 19:19 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/unix/sysv/linux/arm/Dist: New file.
1997-06-14 17:59 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* extra-lib.mk (others): Depend on versioned shared library, not
the unversioned one.
* Makerules (build-shlib): Don't make the version link here.
($(common-objpfx)libc.so$(libc.so-version)): New rule for libc
version link.
1997-06-16 03:07 Ulrich Drepper <drepper@cygnus.com>
* sysdeps/libm-ieee754/k_standard.c: Undo change of Tue Aug 6
01:13:56 1996.
* argp/argp-help.c (argp_args_usage): Supply correct argp to
filter_doc.
* argp/argp-help.c (hol_add_cluster): Initialize CL->depth.
* argp/argp-help.c (_help): Supply STATE to argp_args_usage.
* argp/argp.h (ARGP_KEY_HELP_ARGS_DOC): New macro.
* argp/argp-fmtstream.c: Add casts to prevent warnings.
* argp/argp.h (OPTION_NO_USAGE): New macro.
* argp/argp-help.c (usage_long_opt, usage_argful_short_opt,
* argp/argp-fmtstream.c (__argp_fmtstream_update): Account for case
* argp/argp-help.c <stddef.h>: New include.
* argp/argp.h (argp_state_help, __argp_state_help, argp_usage,
* argp/argp.h (argp_program_bug_address): Make const.
* argp/argp-parse.c (argp_default_parser): Set STATE->name for OPT_PROGNAME.
* argp/argp-help.c (__argp_error, __argp_failure, __argp_state_help):
* argp/argp-parse.c (parser_init): Set PARSER->state.flags.
1997-06-19 19:11:22 +00:00
|
|
|
#include <math.h>
|
|
|
|
#include <math_private.h>
|
2013-05-08 11:58:18 +00:00
|
|
|
#include <float.h>
|
Update.
1997-06-19 19:38 Ulrich Drepper <drepper@cygnus.com>
* features.h: Define __STDC_IEC_559__ and _STDC_IEC_559_COMPLEX__.
* elf/dl-minimal.c (__dcgettext): Remove assertion.
* inet/rcmd.c: Correct a few typos. Reported by Eric Troan.
* manual/Makefile (distribute): Add dir.
* manual/dir: New file.
* math/Makefile (libm-support): Rename s_rinttol, s_rinttoll,
s_roundtol, and s_roundtoll to s_lrint, s_llrint, s_lround,
and s_llround respectively.
(libm-calls): Add e_gamma_r.
* math/libm-test.c (check_int_exc): New function.
(signbit_test): Remove test for sign of NaN.
(gamma_test): Clear exception after test of existence.
Correct tests and and epsilons.
(lgamma_test): Likewise.
(ilogb_test): Correct all tests.
(scalb_test): Rewrite.
(rinttol_test): Rename to lrint_test and correct tests.
(rinttoll_test): Likewise.
(roundtol_test): Likewise.
(roundtoll_test): Likewise.
(main): Call lrint/lround functions instead of rinttol/roundtol.
* math/math.h: Change prototypes for rinttol/roundtol.
* math/mathcalls.h: Rearrange prototypes according to ISO C9X draft.
* sysdeps/generic/mathbits.h: Define FP_ILOGB0 and FP_ILOGBNAN.
* sysdeps/i386/mathbits.h: Likewise.
* sysdeps/libm-i387/e_scalb.S: Handle special cases correctly.
* sysdeps/libm-i387/e_scalbf.S: Likewise.
* sysdeps/libm-i387/e_scalbl.S: Likewise.
* sysdeps/libm-i387/s_asinh.S: Handle -inf correctly.
* sysdeps/libm-i387/s_asinhf.S: Likewise.
* sysdeps/libm-i387/s_asinhl.S: Likewise.
* sysdeps/libm-i387/s_ilogb.S: Optimize.
* sysdeps/libm-i387/s_ilogbf.S: Likewise.
* sysdeps/libm-i387/s_ilogbl.S: Likewise.
* sysdeps/libm-i387/s_rinttol.S: Rename to...
* sysdeps/libm-i387/s_lrint.S: ...this.
* sysdeps/libm-i387/s_rinttoll.S: Rename to...
* sysdeps/libm-i387/s_llrint.S: ...this.
* sysdeps/libm-i387/s_remquo.S: Correctly set sign of remainder.
* sysdeps/libm-i387/s_remquof.S: Likewise.
* sysdeps/libm-i387/s_remquol.S: Likewise.
* sysdeps/libm-i387/e_gamma_r.c: New file. Implementation of gamma
function according to ISO C.
* sysdeps/libm-i387/e_gammaf_r.c: New file.
* sysdeps/libm-i387/e_gammal_r.c: New file.
* sysdeps/libm-i387/e_lgamma_r.c: Don't let optimize compile the
generation of exceptions away.
* sysdeps/libm-i387/e_lgammaf_r.c: Likewise.
* sysdeps/libm-i387/k_standard.c: Correct return value for infinity
points of gamma function when not SVID mode.
* sysdeps/libm-i387/s_rinttoll.c: Renamed to...
* sysdeps/libm-i387/s_llrint.c: ...this.
* sysdeps/libm-i387/s_rinttol.c: Renamed to...
* sysdeps/libm-i387/s_lrint.c: ...this.
* sysdeps/libm-i387/s_roundtoll.c: Renamed to...
* sysdeps/libm-i387/s_llround.c: ...this.
* sysdeps/libm-i387/s_roundtol.c: Renamed to...
* sysdeps/libm-i387/s_lround.c: ..this.
* sysdeps/libm-i387/s_scalbn.c: Change second parameter according to
ISO C.
* sysdeps/libm-i387/s_scalbnf.c: Likewise.
* sysdeps/libm-i387/s_scalbnl.c: Likewise.
* sysdeps/libm-i387/w_gamma.c: Call __ieee754_gamma_r if library
mode is not _SVID_.
* sysdeps/libm-i387/w_gammaf.c: Likewise.
* sysdeps/libm-i387/w_gammal.c: Likewise.
* sysdeps/m68k/fpu/__math.h: Rename __rinttol to __lrint and
rinttol to lrint.
* sysdeps/m68k/fpu/s_rinttol.c: Renamed to...
* sysdeps/m68k/fpu/s_lrint.c: ...this.
* sysdeps/m68k/fpu/s_rinttoll.c: Renamed to...
* sysdeps/m68k/fpu/s_llrint.c: ...this.
* md5-crypt/Makefile: Link md5test program with md5.o.
* stdio-common/temptest.c: Don't use __stdio_gen_tempname which
is not exported by the libc.so.
* stdio-common/vfscanf.c: Correct scanning of strings after last
change.
* sysdeps/unix/sysv/linux/i386/sysdep.S: Use .comm to define errno.
1997-06-19 07:37 H.J. Lu <hjl@gnu.ai.mit.edu>
* time/tzfile.c (__tzfile_read): Store getc () return in int.
1997-06-13 Miles Bader <miles@gnu.ai.mit.edu>
* argp/argp-parse.c (argp_version_parser): Include `(PROGRAM ERROR)' in
the no-version error text to indicate that something's fucked.
[!_] (N_): New macro.
(argp_default_options, argp_version_options): Wrap doc strings in N_().
1997-06-12 Miles Bader <miles@gnu.ai.mit.edu>
* argp/argp-parse.c (parser_parse_opt): Detect and report unhandled
options here.
(parser_parse_arg): Handle ARGP_KEY_ARGS here.
Adjust NEXT pointer back if we fail to parse anything.
(parser_parse_next): Simplify arg code. Leave state NEXT frobbing
to parser_parse_arg.
1997-06-11 Miles Bader <miles@gnu.ai.mit.edu>
* argp/argp.h (ARGP_KEY_ARGS, ARGP_KEY_FINI): New macros.
* argp/argp-parse.c (parser_finalize): Do another pass over the
parsers with ARGP_KEY_FINI.
1997-06-18 Miles Bader <miles@gnu.ai.mit.edu>
* string/Makefile (routines): Add argz-replace.
1997-06-16 00:16 Miles Bader <miles@gnu.ai.mit.edu>
* manual/string.texi (Argz Functions): Document argz_replace.
1997-06-12 Miles Bader <miles@gnu.ai.mit.edu>
* string/argz.h (__argz_replace, argz_replace): New declarations.
* string/argz-replace.c: New file.
1997-06-11 Miles Bader <miles@gnu.ai.mit.edu>
* manual/argp.texi (Argp Special Keys): Document ARGP_KEY_ARGS.
1997-06-16 23:08 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* Makeconfig (libc-map): Remove definition.
* Makerules (libc-map): Define it here, using the full name.
(load-map-file): Remove case for empty $(..).
($(common-objpfx)libc.so): Revert last change.
1997-06-17 22:18 Mark Kettenis <kettenis@phys.uva.nl>
* login/programs/utmpd.c (main): Improve signal handling.
* login/programs/request.c (do_setutent, do_updwtmp):
Allow arbitrary length filenames.
* login/programs/utmpd.h (setutent_request, updwtmp_request):
Get rid of fixed length file field.
* login/utmp_daemon.c (do_setutent, do_updwtmp):
Allow arbitrary length filenames.
* login/programs/request.c (do_pututline):
Don't fail if connection->position is -1 on entry.
1997-06-15 16:32 Mark Kettenis <kettenis@phys.uva.nl>
* login/utmp_file.c (updwtmp_file): Use the same method for
appending an entry as in pututline_file.
1997-06-11 18:59 Mark Kettenis <kettenis@phys.uva.nl>
* login/utmp_file.c (getutent_r_file):
Use read lock instead of write lock.
(getutline_r_file, internal_getut_r): Lock utmp file.
(updwtmp_file): Use fcntl to lock file instead of flock.
1997-06-18 00:11 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* stdio-common/vfscanf.c (inchar, ungetc): Don't count EOF as
character read in/put back.
* stdio-common/tstscanf.c: Add test case for this.
1997-06-17 22:17 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* libc.map: Add more symbols.
1997-06-18 12:01 Ulrich Drepper <drepper@cygnus.com>
* manual/Makefile (install): Make sure `dir' file exists if we use
install-info.
1997-06-17 19:32 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* manual/Makefile (info): Depend on dir-add.info.
1997-06-17 17:19 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* elf/ldd.bash.in: Fix spacing in message.
1997-06-17 14:28 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* elf/do-rel.h (elf_dynamic_do_rel): Always use version
information if available.
1997-06-17 11:34 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sunrpc/Makefile ($(rpcsvc:%.x=$(objpfx)rpcsvc/%.h)): Make
command non-empty to force make to recheck modification time.
($(rpcsvc:%.x=$(objpfx)x%.c)): Likewise.
1997-06-17 00:26 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/stub/e_acoshl.c: Set errno to ENOSYS.
* sysdeps/stub/e_acosl.c: Likewise.
* sysdeps/stub/e_asinl.c: Likewise.
* sysdeps/stub/e_atan2l.c: Likewise.
* sysdeps/stub/e_expl.c: Likewise.
* sysdeps/stub/e_fmodl.c: Likewise.
* sysdeps/stub/e_j0l.c: Likewise.
* sysdeps/stub/e_j1l.c: Likewise.
* sysdeps/stub/e_jnl.c: Likewise.
* sysdeps/stub/e_lgammal_r.c: Likewise.
* sysdeps/stub/e_log10l.c: Likewise.
* sysdeps/stub/e_logl.c: Likewise.
* sysdeps/stub/e_powl.c: Likewise.
* sysdeps/stub/e_rem_pio2l.c: Likewise.
* sysdeps/stub/e_sqrtl.c: Likewise.
* sysdeps/stub/k_cosl.c: Likewise.
* sysdeps/stub/k_rem_pio2l.c: Likewise.
* sysdeps/stub/k_sinl.c: Likewise.
* sysdeps/stub/k_tanl.c: Likewise.
* sysdeps/stub/s_atanl.c: Likewise.
* sysdeps/stub/s_erfl.c: Likewise.
* sysdeps/stub/s_exp2.c: Likewise.
* sysdeps/stub/s_exp2f.c: Likewise.
* sysdeps/stub/s_exp2l.c: Likewise.
* sysdeps/stub/s_expm1l.c: Likewise.
* sysdeps/stub/s_log1pl.c: Likewise.
* sysdeps/stub/s_log2l.c: Likewise.
1997-06-18 11:46 Ulrich Drepper <drepper@cygnus.com>
* sysdeps/libm-ieee754/w_gamma.c: If _LIB_VERSION is _SVID_ compute
result as before last change.
* sysdeps/libm-ieee754/w_gammaf.c: Likewise.
* sysdeps/libm-ieee754/w_gammal.c: Likewise.
1997-06-16 23:37 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/libm-ieee754/s_remquo.c: Fix off-by-one when computing
quotient.
* sysdeps/libm-ieee754/s_remquof.c: Likewise.
* sysdeps/libm-ieee754/s_remquol.c: Likewise.
* sysdeps/m68k/fpu/s_remquo.c: Remove FIXME and special case for
quotient.
* sysdeps/libm-ieee754/w_gamma.c: Add missing call to exp
function. Don't use global signgam.
* sysdeps/libm-ieee754/w_gammaf.c: Likewise.
* sysdeps/libm-ieee754/w_gammal.c: Likewise.
* math/Makefile (libm-calls): Remove w_gamma_r.
* sysdeps/libm-ieee754/w_gamma_r.c: Remove file.
* sysdeps/libm-ieee754/w_gammaf_r.c: Remove file.
* sysdeps/libm-ieee754/w_gammal_r.c: Remove file.
* math/libm-test.c (atanh_test): Declare x only if needed.
(signbit_test): Fix typo.
(gamma_test): Check whether function is implemented. Add
epsilons.
(lgamma_test): Likewise.
(fmod_test): Add epsilons.
(exp2_test): Use right function for existence test.
1997-06-07 09:20 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/powerpc/Dist: Add fenv_const.c.
* sysdeps/unix/sysv/linux/Dist: Add net/if_slip.h.
* sysdeps/unix/sysv/linux/powerpc/Dist: Add init-first.h and
syscall.h.
* sysdeps/unix/sysv/linux/sparc/Dist: Add init-first.h.
* string/Makefile (distribute): Add tst-svc.expect.
* nis/Makefile (distribute): Add nis_intern.h and Banner.
* elf/Makefile (distribute): Add dl-hash.h.
* Rules (subdir_echo-distinfo): Add headers from $(distribute).
* login/Makefile (others): Add utmpdump.
* sysdeps/unix/sysv/linux/Makefile (sysdep_headers): Add
net/if_slip.h.
* manual/Makefile (dir-add.texi): Also look in indirectly included
files.
1997-06-16 23:15 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/unix/sysv/linux/configure.in: Remove commands to
generate stdio_lim.h.
* sysdeps/unix/sysv/linux/mk-stdiolim.c: Remove.
* sysdeps/unix/sysv/linux/Makefile: Add rules to generate
stdio_lim.h here.
(common-generated): Add generated files.
(inhibit-stdio_lim): Define.
* sysdeps/posix/Makefile [$(inhibit-stdio_lim)=yes]: Disable rules
to generate stdio_lim.h.
* sysdeps/unix/sysv/linux/Makefile: Suppress inclusion of
dependecy files if no_deps is set.
($(objpfx)syscall-%.d): Add header file as target to dependency
generation.
1997-06-14 19:19 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/unix/sysv/linux/arm/Dist: New file.
1997-06-14 17:59 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* extra-lib.mk (others): Depend on versioned shared library, not
the unversioned one.
* Makerules (build-shlib): Don't make the version link here.
($(common-objpfx)libc.so$(libc.so-version)): New rule for libc
version link.
1997-06-16 03:07 Ulrich Drepper <drepper@cygnus.com>
* sysdeps/libm-ieee754/k_standard.c: Undo change of Tue Aug 6
01:13:56 1996.
* argp/argp-help.c (argp_args_usage): Supply correct argp to
filter_doc.
* argp/argp-help.c (hol_add_cluster): Initialize CL->depth.
* argp/argp-help.c (_help): Supply STATE to argp_args_usage.
* argp/argp.h (ARGP_KEY_HELP_ARGS_DOC): New macro.
* argp/argp-fmtstream.c: Add casts to prevent warnings.
* argp/argp.h (OPTION_NO_USAGE): New macro.
* argp/argp-help.c (usage_long_opt, usage_argful_short_opt,
* argp/argp-fmtstream.c (__argp_fmtstream_update): Account for case
* argp/argp-help.c <stddef.h>: New include.
* argp/argp.h (argp_state_help, __argp_state_help, argp_usage,
* argp/argp.h (argp_program_bug_address): Make const.
* argp/argp-parse.c (argp_default_parser): Set STATE->name for OPT_PROGNAME.
* argp/argp-help.c (__argp_error, __argp_failure, __argp_state_help):
* argp/argp-parse.c (parser_init): Set PARSER->state.flags.
1997-06-19 19:11:22 +00:00
|
|
|
|
2013-05-08 11:58:18 +00:00
|
|
|
/* Coefficients B_2k / 2k(2k-1) of x^-(2k-1) inside exp in Stirling's
|
|
|
|
approximation to gamma function. */
|
|
|
|
|
|
|
|
static const double gamma_coeff[] =
|
|
|
|
{
|
|
|
|
0x1.5555555555555p-4,
|
|
|
|
-0xb.60b60b60b60b8p-12,
|
|
|
|
0x3.4034034034034p-12,
|
|
|
|
-0x2.7027027027028p-12,
|
|
|
|
0x3.72a3c5631fe46p-12,
|
|
|
|
-0x7.daac36664f1f4p-12,
|
|
|
|
};
|
|
|
|
|
|
|
|
#define NCOEFF (sizeof (gamma_coeff) / sizeof (gamma_coeff[0]))
|
|
|
|
|
|
|
|
/* Return gamma (X), for positive X less than 184, in the form R *
|
|
|
|
2^(*EXP2_ADJ), where R is the return value and *EXP2_ADJ is set to
|
|
|
|
avoid overflow or underflow in intermediate calculations. */
|
|
|
|
|
|
|
|
static double
|
|
|
|
gamma_positive (double x, int *exp2_adj)
|
|
|
|
{
|
|
|
|
int local_signgam;
|
|
|
|
if (x < 0.5)
|
|
|
|
{
|
|
|
|
*exp2_adj = 0;
|
|
|
|
return __ieee754_exp (__ieee754_lgamma_r (x + 1, &local_signgam)) / x;
|
|
|
|
}
|
|
|
|
else if (x <= 1.5)
|
|
|
|
{
|
|
|
|
*exp2_adj = 0;
|
|
|
|
return __ieee754_exp (__ieee754_lgamma_r (x, &local_signgam));
|
|
|
|
}
|
|
|
|
else if (x < 6.5)
|
|
|
|
{
|
|
|
|
/* Adjust into the range for using exp (lgamma). */
|
|
|
|
*exp2_adj = 0;
|
|
|
|
double n = __ceil (x - 1.5);
|
|
|
|
double x_adj = x - n;
|
|
|
|
double eps;
|
|
|
|
double prod = __gamma_product (x_adj, 0, n, &eps);
|
|
|
|
return (__ieee754_exp (__ieee754_lgamma_r (x_adj, &local_signgam))
|
|
|
|
* prod * (1.0 + eps));
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
double eps = 0;
|
|
|
|
double x_eps = 0;
|
|
|
|
double x_adj = x;
|
|
|
|
double prod = 1;
|
|
|
|
if (x < 12.0)
|
|
|
|
{
|
|
|
|
/* Adjust into the range for applying Stirling's
|
|
|
|
approximation. */
|
|
|
|
double n = __ceil (12.0 - x);
|
2015-09-23 18:14:57 +00:00
|
|
|
x_adj = math_narrow_eval (x + n);
|
2013-05-08 11:58:18 +00:00
|
|
|
x_eps = (x - (x_adj - n));
|
|
|
|
prod = __gamma_product (x_adj - n, x_eps, n, &eps);
|
|
|
|
}
|
|
|
|
/* The result is now gamma (X_ADJ + X_EPS) / (PROD * (1 + EPS)).
|
|
|
|
Compute gamma (X_ADJ + X_EPS) using Stirling's approximation,
|
|
|
|
starting by computing pow (X_ADJ, X_ADJ) with a power of 2
|
|
|
|
factored out. */
|
|
|
|
double exp_adj = -eps;
|
|
|
|
double x_adj_int = __round (x_adj);
|
|
|
|
double x_adj_frac = x_adj - x_adj_int;
|
|
|
|
int x_adj_log2;
|
|
|
|
double x_adj_mant = __frexp (x_adj, &x_adj_log2);
|
|
|
|
if (x_adj_mant < M_SQRT1_2)
|
|
|
|
{
|
|
|
|
x_adj_log2--;
|
|
|
|
x_adj_mant *= 2.0;
|
|
|
|
}
|
|
|
|
*exp2_adj = x_adj_log2 * (int) x_adj_int;
|
|
|
|
double ret = (__ieee754_pow (x_adj_mant, x_adj)
|
|
|
|
* __ieee754_exp2 (x_adj_log2 * x_adj_frac)
|
|
|
|
* __ieee754_exp (-x_adj)
|
|
|
|
* __ieee754_sqrt (2 * M_PI / x_adj)
|
|
|
|
/ prod);
|
Improve tgamma accuracy (bug 18613).
In non-default rounding modes, tgamma can be slightly less accurate
than permitted by glibc's accuracy goals.
Part of the problem is error accumulation, addressed in this patch by
setting round-to-nearest for internal computations. However, there
was also a bug in the code dealing with computing pow (x + n, x + n)
where x + n is not exactly representable, providing another source of
error even in round-to-nearest mode; it was necessary to address both
bugs to get errors for all testcases within glibc's accuracy goals.
Given this second fix, accuracy in round-to-nearest mode is also
improved (hence regeneration of ulps for tgamma should be from scratch
- truncate libm-test-ulps or at least remove existing tgamma entries -
so that the expected ulps can be reduced).
Some additional complications also arose. Certain tgamma tests should
strictly, according to IEEE semantics, overflow or not depending on
the rounding mode; this is beyond the scope of glibc's accuracy goals
for any function without exactly-determined results, but
gen-auto-libm-tests doesn't handle being lax there as it does for
underflow. (libm-test.inc also doesn't handle being lax about whether
the result in cases very close to the overflow threshold is infinity
or a finite value close to overflow, but that doesn't cause problems
in this case though I've seen it cause problems with random test
generation for some functions.) Thus, spurious-overflow markings,
with a comment, are added to auto-libm-test-in (no bug in Bugzilla
because the issue is with the testsuite, not a user-visible bug in
glibc). And on x86, after the patch I saw ERANGE issues as previously
reported by Carlos (see my commentary in
<https://sourceware.org/ml/libc-alpha/2015-01/msg00485.html>), which
needed addressing by ensuring excess range and precision were
eliminated at various points if FLT_EVAL_METHOD != 0.
I also noticed and fixed a cosmetic issue where 1.0f was used in long
double functions and should have been 1.0L.
This completes the move of all functions to testing in all rounding
modes with ALL_RM_TEST, so gen-libm-have-vector-test.sh is updated to
remove the workaround for some functions not using ALL_RM_TEST.
Tested for x86_64, x86, mips64 and powerpc.
[BZ #18613]
* sysdeps/ieee754/dbl-64/e_gamma_r.c (gamma_positive): Take log of
X_ADJ not X when adjusting exponent.
(__ieee754_gamma_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed.
* sysdeps/ieee754/flt-32/e_gammaf_r.c (gammaf_positive): Take log
of X_ADJ not X when adjusting exponent.
(__ieee754_gammaf_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed.
* sysdeps/ieee754/ldbl-128/e_gammal_r.c (gammal_positive): Take
log of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* sysdeps/ieee754/ldbl-128ibm/e_gammal_r.c (gammal_positive): Take
log of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* sysdeps/ieee754/ldbl-96/e_gammal_r.c (gammal_positive): Take log
of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* math/libm-test.inc (tgamma_test_data): Remove one test. Moved
to auto-libm-test-in.
(tgamma_test): Use ALL_RM_TEST.
* math/auto-libm-test-in: Add one test of tgamma. Mark some other
tests of tgamma with spurious-overflow.
* math/auto-libm-test-out: Regenerated.
* math/gen-libm-have-vector-test.sh: Do not check for START.
* sysdeps/i386/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Likewise.
2015-06-29 23:29:35 +00:00
|
|
|
exp_adj += x_eps * __ieee754_log (x_adj);
|
2013-05-08 11:58:18 +00:00
|
|
|
double bsum = gamma_coeff[NCOEFF - 1];
|
|
|
|
double x_adj2 = x_adj * x_adj;
|
|
|
|
for (size_t i = 1; i <= NCOEFF - 1; i++)
|
|
|
|
bsum = bsum / x_adj2 + gamma_coeff[NCOEFF - 1 - i];
|
|
|
|
exp_adj += bsum / x_adj;
|
|
|
|
return ret + ret * __expm1 (exp_adj);
|
|
|
|
}
|
|
|
|
}
|
Update.
1997-06-19 19:38 Ulrich Drepper <drepper@cygnus.com>
* features.h: Define __STDC_IEC_559__ and _STDC_IEC_559_COMPLEX__.
* elf/dl-minimal.c (__dcgettext): Remove assertion.
* inet/rcmd.c: Correct a few typos. Reported by Eric Troan.
* manual/Makefile (distribute): Add dir.
* manual/dir: New file.
* math/Makefile (libm-support): Rename s_rinttol, s_rinttoll,
s_roundtol, and s_roundtoll to s_lrint, s_llrint, s_lround,
and s_llround respectively.
(libm-calls): Add e_gamma_r.
* math/libm-test.c (check_int_exc): New function.
(signbit_test): Remove test for sign of NaN.
(gamma_test): Clear exception after test of existence.
Correct tests and and epsilons.
(lgamma_test): Likewise.
(ilogb_test): Correct all tests.
(scalb_test): Rewrite.
(rinttol_test): Rename to lrint_test and correct tests.
(rinttoll_test): Likewise.
(roundtol_test): Likewise.
(roundtoll_test): Likewise.
(main): Call lrint/lround functions instead of rinttol/roundtol.
* math/math.h: Change prototypes for rinttol/roundtol.
* math/mathcalls.h: Rearrange prototypes according to ISO C9X draft.
* sysdeps/generic/mathbits.h: Define FP_ILOGB0 and FP_ILOGBNAN.
* sysdeps/i386/mathbits.h: Likewise.
* sysdeps/libm-i387/e_scalb.S: Handle special cases correctly.
* sysdeps/libm-i387/e_scalbf.S: Likewise.
* sysdeps/libm-i387/e_scalbl.S: Likewise.
* sysdeps/libm-i387/s_asinh.S: Handle -inf correctly.
* sysdeps/libm-i387/s_asinhf.S: Likewise.
* sysdeps/libm-i387/s_asinhl.S: Likewise.
* sysdeps/libm-i387/s_ilogb.S: Optimize.
* sysdeps/libm-i387/s_ilogbf.S: Likewise.
* sysdeps/libm-i387/s_ilogbl.S: Likewise.
* sysdeps/libm-i387/s_rinttol.S: Rename to...
* sysdeps/libm-i387/s_lrint.S: ...this.
* sysdeps/libm-i387/s_rinttoll.S: Rename to...
* sysdeps/libm-i387/s_llrint.S: ...this.
* sysdeps/libm-i387/s_remquo.S: Correctly set sign of remainder.
* sysdeps/libm-i387/s_remquof.S: Likewise.
* sysdeps/libm-i387/s_remquol.S: Likewise.
* sysdeps/libm-i387/e_gamma_r.c: New file. Implementation of gamma
function according to ISO C.
* sysdeps/libm-i387/e_gammaf_r.c: New file.
* sysdeps/libm-i387/e_gammal_r.c: New file.
* sysdeps/libm-i387/e_lgamma_r.c: Don't let optimize compile the
generation of exceptions away.
* sysdeps/libm-i387/e_lgammaf_r.c: Likewise.
* sysdeps/libm-i387/k_standard.c: Correct return value for infinity
points of gamma function when not SVID mode.
* sysdeps/libm-i387/s_rinttoll.c: Renamed to...
* sysdeps/libm-i387/s_llrint.c: ...this.
* sysdeps/libm-i387/s_rinttol.c: Renamed to...
* sysdeps/libm-i387/s_lrint.c: ...this.
* sysdeps/libm-i387/s_roundtoll.c: Renamed to...
* sysdeps/libm-i387/s_llround.c: ...this.
* sysdeps/libm-i387/s_roundtol.c: Renamed to...
* sysdeps/libm-i387/s_lround.c: ..this.
* sysdeps/libm-i387/s_scalbn.c: Change second parameter according to
ISO C.
* sysdeps/libm-i387/s_scalbnf.c: Likewise.
* sysdeps/libm-i387/s_scalbnl.c: Likewise.
* sysdeps/libm-i387/w_gamma.c: Call __ieee754_gamma_r if library
mode is not _SVID_.
* sysdeps/libm-i387/w_gammaf.c: Likewise.
* sysdeps/libm-i387/w_gammal.c: Likewise.
* sysdeps/m68k/fpu/__math.h: Rename __rinttol to __lrint and
rinttol to lrint.
* sysdeps/m68k/fpu/s_rinttol.c: Renamed to...
* sysdeps/m68k/fpu/s_lrint.c: ...this.
* sysdeps/m68k/fpu/s_rinttoll.c: Renamed to...
* sysdeps/m68k/fpu/s_llrint.c: ...this.
* md5-crypt/Makefile: Link md5test program with md5.o.
* stdio-common/temptest.c: Don't use __stdio_gen_tempname which
is not exported by the libc.so.
* stdio-common/vfscanf.c: Correct scanning of strings after last
change.
* sysdeps/unix/sysv/linux/i386/sysdep.S: Use .comm to define errno.
1997-06-19 07:37 H.J. Lu <hjl@gnu.ai.mit.edu>
* time/tzfile.c (__tzfile_read): Store getc () return in int.
1997-06-13 Miles Bader <miles@gnu.ai.mit.edu>
* argp/argp-parse.c (argp_version_parser): Include `(PROGRAM ERROR)' in
the no-version error text to indicate that something's fucked.
[!_] (N_): New macro.
(argp_default_options, argp_version_options): Wrap doc strings in N_().
1997-06-12 Miles Bader <miles@gnu.ai.mit.edu>
* argp/argp-parse.c (parser_parse_opt): Detect and report unhandled
options here.
(parser_parse_arg): Handle ARGP_KEY_ARGS here.
Adjust NEXT pointer back if we fail to parse anything.
(parser_parse_next): Simplify arg code. Leave state NEXT frobbing
to parser_parse_arg.
1997-06-11 Miles Bader <miles@gnu.ai.mit.edu>
* argp/argp.h (ARGP_KEY_ARGS, ARGP_KEY_FINI): New macros.
* argp/argp-parse.c (parser_finalize): Do another pass over the
parsers with ARGP_KEY_FINI.
1997-06-18 Miles Bader <miles@gnu.ai.mit.edu>
* string/Makefile (routines): Add argz-replace.
1997-06-16 00:16 Miles Bader <miles@gnu.ai.mit.edu>
* manual/string.texi (Argz Functions): Document argz_replace.
1997-06-12 Miles Bader <miles@gnu.ai.mit.edu>
* string/argz.h (__argz_replace, argz_replace): New declarations.
* string/argz-replace.c: New file.
1997-06-11 Miles Bader <miles@gnu.ai.mit.edu>
* manual/argp.texi (Argp Special Keys): Document ARGP_KEY_ARGS.
1997-06-16 23:08 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* Makeconfig (libc-map): Remove definition.
* Makerules (libc-map): Define it here, using the full name.
(load-map-file): Remove case for empty $(..).
($(common-objpfx)libc.so): Revert last change.
1997-06-17 22:18 Mark Kettenis <kettenis@phys.uva.nl>
* login/programs/utmpd.c (main): Improve signal handling.
* login/programs/request.c (do_setutent, do_updwtmp):
Allow arbitrary length filenames.
* login/programs/utmpd.h (setutent_request, updwtmp_request):
Get rid of fixed length file field.
* login/utmp_daemon.c (do_setutent, do_updwtmp):
Allow arbitrary length filenames.
* login/programs/request.c (do_pututline):
Don't fail if connection->position is -1 on entry.
1997-06-15 16:32 Mark Kettenis <kettenis@phys.uva.nl>
* login/utmp_file.c (updwtmp_file): Use the same method for
appending an entry as in pututline_file.
1997-06-11 18:59 Mark Kettenis <kettenis@phys.uva.nl>
* login/utmp_file.c (getutent_r_file):
Use read lock instead of write lock.
(getutline_r_file, internal_getut_r): Lock utmp file.
(updwtmp_file): Use fcntl to lock file instead of flock.
1997-06-18 00:11 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* stdio-common/vfscanf.c (inchar, ungetc): Don't count EOF as
character read in/put back.
* stdio-common/tstscanf.c: Add test case for this.
1997-06-17 22:17 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* libc.map: Add more symbols.
1997-06-18 12:01 Ulrich Drepper <drepper@cygnus.com>
* manual/Makefile (install): Make sure `dir' file exists if we use
install-info.
1997-06-17 19:32 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* manual/Makefile (info): Depend on dir-add.info.
1997-06-17 17:19 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* elf/ldd.bash.in: Fix spacing in message.
1997-06-17 14:28 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* elf/do-rel.h (elf_dynamic_do_rel): Always use version
information if available.
1997-06-17 11:34 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sunrpc/Makefile ($(rpcsvc:%.x=$(objpfx)rpcsvc/%.h)): Make
command non-empty to force make to recheck modification time.
($(rpcsvc:%.x=$(objpfx)x%.c)): Likewise.
1997-06-17 00:26 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/stub/e_acoshl.c: Set errno to ENOSYS.
* sysdeps/stub/e_acosl.c: Likewise.
* sysdeps/stub/e_asinl.c: Likewise.
* sysdeps/stub/e_atan2l.c: Likewise.
* sysdeps/stub/e_expl.c: Likewise.
* sysdeps/stub/e_fmodl.c: Likewise.
* sysdeps/stub/e_j0l.c: Likewise.
* sysdeps/stub/e_j1l.c: Likewise.
* sysdeps/stub/e_jnl.c: Likewise.
* sysdeps/stub/e_lgammal_r.c: Likewise.
* sysdeps/stub/e_log10l.c: Likewise.
* sysdeps/stub/e_logl.c: Likewise.
* sysdeps/stub/e_powl.c: Likewise.
* sysdeps/stub/e_rem_pio2l.c: Likewise.
* sysdeps/stub/e_sqrtl.c: Likewise.
* sysdeps/stub/k_cosl.c: Likewise.
* sysdeps/stub/k_rem_pio2l.c: Likewise.
* sysdeps/stub/k_sinl.c: Likewise.
* sysdeps/stub/k_tanl.c: Likewise.
* sysdeps/stub/s_atanl.c: Likewise.
* sysdeps/stub/s_erfl.c: Likewise.
* sysdeps/stub/s_exp2.c: Likewise.
* sysdeps/stub/s_exp2f.c: Likewise.
* sysdeps/stub/s_exp2l.c: Likewise.
* sysdeps/stub/s_expm1l.c: Likewise.
* sysdeps/stub/s_log1pl.c: Likewise.
* sysdeps/stub/s_log2l.c: Likewise.
1997-06-18 11:46 Ulrich Drepper <drepper@cygnus.com>
* sysdeps/libm-ieee754/w_gamma.c: If _LIB_VERSION is _SVID_ compute
result as before last change.
* sysdeps/libm-ieee754/w_gammaf.c: Likewise.
* sysdeps/libm-ieee754/w_gammal.c: Likewise.
1997-06-16 23:37 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/libm-ieee754/s_remquo.c: Fix off-by-one when computing
quotient.
* sysdeps/libm-ieee754/s_remquof.c: Likewise.
* sysdeps/libm-ieee754/s_remquol.c: Likewise.
* sysdeps/m68k/fpu/s_remquo.c: Remove FIXME and special case for
quotient.
* sysdeps/libm-ieee754/w_gamma.c: Add missing call to exp
function. Don't use global signgam.
* sysdeps/libm-ieee754/w_gammaf.c: Likewise.
* sysdeps/libm-ieee754/w_gammal.c: Likewise.
* math/Makefile (libm-calls): Remove w_gamma_r.
* sysdeps/libm-ieee754/w_gamma_r.c: Remove file.
* sysdeps/libm-ieee754/w_gammaf_r.c: Remove file.
* sysdeps/libm-ieee754/w_gammal_r.c: Remove file.
* math/libm-test.c (atanh_test): Declare x only if needed.
(signbit_test): Fix typo.
(gamma_test): Check whether function is implemented. Add
epsilons.
(lgamma_test): Likewise.
(fmod_test): Add epsilons.
(exp2_test): Use right function for existence test.
1997-06-07 09:20 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/powerpc/Dist: Add fenv_const.c.
* sysdeps/unix/sysv/linux/Dist: Add net/if_slip.h.
* sysdeps/unix/sysv/linux/powerpc/Dist: Add init-first.h and
syscall.h.
* sysdeps/unix/sysv/linux/sparc/Dist: Add init-first.h.
* string/Makefile (distribute): Add tst-svc.expect.
* nis/Makefile (distribute): Add nis_intern.h and Banner.
* elf/Makefile (distribute): Add dl-hash.h.
* Rules (subdir_echo-distinfo): Add headers from $(distribute).
* login/Makefile (others): Add utmpdump.
* sysdeps/unix/sysv/linux/Makefile (sysdep_headers): Add
net/if_slip.h.
* manual/Makefile (dir-add.texi): Also look in indirectly included
files.
1997-06-16 23:15 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/unix/sysv/linux/configure.in: Remove commands to
generate stdio_lim.h.
* sysdeps/unix/sysv/linux/mk-stdiolim.c: Remove.
* sysdeps/unix/sysv/linux/Makefile: Add rules to generate
stdio_lim.h here.
(common-generated): Add generated files.
(inhibit-stdio_lim): Define.
* sysdeps/posix/Makefile [$(inhibit-stdio_lim)=yes]: Disable rules
to generate stdio_lim.h.
* sysdeps/unix/sysv/linux/Makefile: Suppress inclusion of
dependecy files if no_deps is set.
($(objpfx)syscall-%.d): Add header file as target to dependency
generation.
1997-06-14 19:19 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/unix/sysv/linux/arm/Dist: New file.
1997-06-14 17:59 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* extra-lib.mk (others): Depend on versioned shared library, not
the unversioned one.
* Makerules (build-shlib): Don't make the version link here.
($(common-objpfx)libc.so$(libc.so-version)): New rule for libc
version link.
1997-06-16 03:07 Ulrich Drepper <drepper@cygnus.com>
* sysdeps/libm-ieee754/k_standard.c: Undo change of Tue Aug 6
01:13:56 1996.
* argp/argp-help.c (argp_args_usage): Supply correct argp to
filter_doc.
* argp/argp-help.c (hol_add_cluster): Initialize CL->depth.
* argp/argp-help.c (_help): Supply STATE to argp_args_usage.
* argp/argp.h (ARGP_KEY_HELP_ARGS_DOC): New macro.
* argp/argp-fmtstream.c: Add casts to prevent warnings.
* argp/argp.h (OPTION_NO_USAGE): New macro.
* argp/argp-help.c (usage_long_opt, usage_argful_short_opt,
* argp/argp-fmtstream.c (__argp_fmtstream_update): Account for case
* argp/argp-help.c <stddef.h>: New include.
* argp/argp.h (argp_state_help, __argp_state_help, argp_usage,
* argp/argp.h (argp_program_bug_address): Make const.
* argp/argp-parse.c (argp_default_parser): Set STATE->name for OPT_PROGNAME.
* argp/argp-help.c (__argp_error, __argp_failure, __argp_state_help):
* argp/argp-parse.c (parser_init): Set PARSER->state.flags.
1997-06-19 19:11:22 +00:00
|
|
|
|
|
|
|
double
|
|
|
|
__ieee754_gamma_r (double x, int *signgamp)
|
|
|
|
{
|
|
|
|
int32_t hx;
|
2017-08-03 19:55:04 +00:00
|
|
|
uint32_t lx;
|
Improve tgamma accuracy (bug 18613).
In non-default rounding modes, tgamma can be slightly less accurate
than permitted by glibc's accuracy goals.
Part of the problem is error accumulation, addressed in this patch by
setting round-to-nearest for internal computations. However, there
was also a bug in the code dealing with computing pow (x + n, x + n)
where x + n is not exactly representable, providing another source of
error even in round-to-nearest mode; it was necessary to address both
bugs to get errors for all testcases within glibc's accuracy goals.
Given this second fix, accuracy in round-to-nearest mode is also
improved (hence regeneration of ulps for tgamma should be from scratch
- truncate libm-test-ulps or at least remove existing tgamma entries -
so that the expected ulps can be reduced).
Some additional complications also arose. Certain tgamma tests should
strictly, according to IEEE semantics, overflow or not depending on
the rounding mode; this is beyond the scope of glibc's accuracy goals
for any function without exactly-determined results, but
gen-auto-libm-tests doesn't handle being lax there as it does for
underflow. (libm-test.inc also doesn't handle being lax about whether
the result in cases very close to the overflow threshold is infinity
or a finite value close to overflow, but that doesn't cause problems
in this case though I've seen it cause problems with random test
generation for some functions.) Thus, spurious-overflow markings,
with a comment, are added to auto-libm-test-in (no bug in Bugzilla
because the issue is with the testsuite, not a user-visible bug in
glibc). And on x86, after the patch I saw ERANGE issues as previously
reported by Carlos (see my commentary in
<https://sourceware.org/ml/libc-alpha/2015-01/msg00485.html>), which
needed addressing by ensuring excess range and precision were
eliminated at various points if FLT_EVAL_METHOD != 0.
I also noticed and fixed a cosmetic issue where 1.0f was used in long
double functions and should have been 1.0L.
This completes the move of all functions to testing in all rounding
modes with ALL_RM_TEST, so gen-libm-have-vector-test.sh is updated to
remove the workaround for some functions not using ALL_RM_TEST.
Tested for x86_64, x86, mips64 and powerpc.
[BZ #18613]
* sysdeps/ieee754/dbl-64/e_gamma_r.c (gamma_positive): Take log of
X_ADJ not X when adjusting exponent.
(__ieee754_gamma_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed.
* sysdeps/ieee754/flt-32/e_gammaf_r.c (gammaf_positive): Take log
of X_ADJ not X when adjusting exponent.
(__ieee754_gammaf_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed.
* sysdeps/ieee754/ldbl-128/e_gammal_r.c (gammal_positive): Take
log of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* sysdeps/ieee754/ldbl-128ibm/e_gammal_r.c (gammal_positive): Take
log of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* sysdeps/ieee754/ldbl-96/e_gammal_r.c (gammal_positive): Take log
of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* math/libm-test.inc (tgamma_test_data): Remove one test. Moved
to auto-libm-test-in.
(tgamma_test): Use ALL_RM_TEST.
* math/auto-libm-test-in: Add one test of tgamma. Mark some other
tests of tgamma with spurious-overflow.
* math/auto-libm-test-out: Regenerated.
* math/gen-libm-have-vector-test.sh: Do not check for START.
* sysdeps/i386/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Likewise.
2015-06-29 23:29:35 +00:00
|
|
|
double ret;
|
Update.
1997-06-19 19:38 Ulrich Drepper <drepper@cygnus.com>
* features.h: Define __STDC_IEC_559__ and _STDC_IEC_559_COMPLEX__.
* elf/dl-minimal.c (__dcgettext): Remove assertion.
* inet/rcmd.c: Correct a few typos. Reported by Eric Troan.
* manual/Makefile (distribute): Add dir.
* manual/dir: New file.
* math/Makefile (libm-support): Rename s_rinttol, s_rinttoll,
s_roundtol, and s_roundtoll to s_lrint, s_llrint, s_lround,
and s_llround respectively.
(libm-calls): Add e_gamma_r.
* math/libm-test.c (check_int_exc): New function.
(signbit_test): Remove test for sign of NaN.
(gamma_test): Clear exception after test of existence.
Correct tests and and epsilons.
(lgamma_test): Likewise.
(ilogb_test): Correct all tests.
(scalb_test): Rewrite.
(rinttol_test): Rename to lrint_test and correct tests.
(rinttoll_test): Likewise.
(roundtol_test): Likewise.
(roundtoll_test): Likewise.
(main): Call lrint/lround functions instead of rinttol/roundtol.
* math/math.h: Change prototypes for rinttol/roundtol.
* math/mathcalls.h: Rearrange prototypes according to ISO C9X draft.
* sysdeps/generic/mathbits.h: Define FP_ILOGB0 and FP_ILOGBNAN.
* sysdeps/i386/mathbits.h: Likewise.
* sysdeps/libm-i387/e_scalb.S: Handle special cases correctly.
* sysdeps/libm-i387/e_scalbf.S: Likewise.
* sysdeps/libm-i387/e_scalbl.S: Likewise.
* sysdeps/libm-i387/s_asinh.S: Handle -inf correctly.
* sysdeps/libm-i387/s_asinhf.S: Likewise.
* sysdeps/libm-i387/s_asinhl.S: Likewise.
* sysdeps/libm-i387/s_ilogb.S: Optimize.
* sysdeps/libm-i387/s_ilogbf.S: Likewise.
* sysdeps/libm-i387/s_ilogbl.S: Likewise.
* sysdeps/libm-i387/s_rinttol.S: Rename to...
* sysdeps/libm-i387/s_lrint.S: ...this.
* sysdeps/libm-i387/s_rinttoll.S: Rename to...
* sysdeps/libm-i387/s_llrint.S: ...this.
* sysdeps/libm-i387/s_remquo.S: Correctly set sign of remainder.
* sysdeps/libm-i387/s_remquof.S: Likewise.
* sysdeps/libm-i387/s_remquol.S: Likewise.
* sysdeps/libm-i387/e_gamma_r.c: New file. Implementation of gamma
function according to ISO C.
* sysdeps/libm-i387/e_gammaf_r.c: New file.
* sysdeps/libm-i387/e_gammal_r.c: New file.
* sysdeps/libm-i387/e_lgamma_r.c: Don't let optimize compile the
generation of exceptions away.
* sysdeps/libm-i387/e_lgammaf_r.c: Likewise.
* sysdeps/libm-i387/k_standard.c: Correct return value for infinity
points of gamma function when not SVID mode.
* sysdeps/libm-i387/s_rinttoll.c: Renamed to...
* sysdeps/libm-i387/s_llrint.c: ...this.
* sysdeps/libm-i387/s_rinttol.c: Renamed to...
* sysdeps/libm-i387/s_lrint.c: ...this.
* sysdeps/libm-i387/s_roundtoll.c: Renamed to...
* sysdeps/libm-i387/s_llround.c: ...this.
* sysdeps/libm-i387/s_roundtol.c: Renamed to...
* sysdeps/libm-i387/s_lround.c: ..this.
* sysdeps/libm-i387/s_scalbn.c: Change second parameter according to
ISO C.
* sysdeps/libm-i387/s_scalbnf.c: Likewise.
* sysdeps/libm-i387/s_scalbnl.c: Likewise.
* sysdeps/libm-i387/w_gamma.c: Call __ieee754_gamma_r if library
mode is not _SVID_.
* sysdeps/libm-i387/w_gammaf.c: Likewise.
* sysdeps/libm-i387/w_gammal.c: Likewise.
* sysdeps/m68k/fpu/__math.h: Rename __rinttol to __lrint and
rinttol to lrint.
* sysdeps/m68k/fpu/s_rinttol.c: Renamed to...
* sysdeps/m68k/fpu/s_lrint.c: ...this.
* sysdeps/m68k/fpu/s_rinttoll.c: Renamed to...
* sysdeps/m68k/fpu/s_llrint.c: ...this.
* md5-crypt/Makefile: Link md5test program with md5.o.
* stdio-common/temptest.c: Don't use __stdio_gen_tempname which
is not exported by the libc.so.
* stdio-common/vfscanf.c: Correct scanning of strings after last
change.
* sysdeps/unix/sysv/linux/i386/sysdep.S: Use .comm to define errno.
1997-06-19 07:37 H.J. Lu <hjl@gnu.ai.mit.edu>
* time/tzfile.c (__tzfile_read): Store getc () return in int.
1997-06-13 Miles Bader <miles@gnu.ai.mit.edu>
* argp/argp-parse.c (argp_version_parser): Include `(PROGRAM ERROR)' in
the no-version error text to indicate that something's fucked.
[!_] (N_): New macro.
(argp_default_options, argp_version_options): Wrap doc strings in N_().
1997-06-12 Miles Bader <miles@gnu.ai.mit.edu>
* argp/argp-parse.c (parser_parse_opt): Detect and report unhandled
options here.
(parser_parse_arg): Handle ARGP_KEY_ARGS here.
Adjust NEXT pointer back if we fail to parse anything.
(parser_parse_next): Simplify arg code. Leave state NEXT frobbing
to parser_parse_arg.
1997-06-11 Miles Bader <miles@gnu.ai.mit.edu>
* argp/argp.h (ARGP_KEY_ARGS, ARGP_KEY_FINI): New macros.
* argp/argp-parse.c (parser_finalize): Do another pass over the
parsers with ARGP_KEY_FINI.
1997-06-18 Miles Bader <miles@gnu.ai.mit.edu>
* string/Makefile (routines): Add argz-replace.
1997-06-16 00:16 Miles Bader <miles@gnu.ai.mit.edu>
* manual/string.texi (Argz Functions): Document argz_replace.
1997-06-12 Miles Bader <miles@gnu.ai.mit.edu>
* string/argz.h (__argz_replace, argz_replace): New declarations.
* string/argz-replace.c: New file.
1997-06-11 Miles Bader <miles@gnu.ai.mit.edu>
* manual/argp.texi (Argp Special Keys): Document ARGP_KEY_ARGS.
1997-06-16 23:08 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* Makeconfig (libc-map): Remove definition.
* Makerules (libc-map): Define it here, using the full name.
(load-map-file): Remove case for empty $(..).
($(common-objpfx)libc.so): Revert last change.
1997-06-17 22:18 Mark Kettenis <kettenis@phys.uva.nl>
* login/programs/utmpd.c (main): Improve signal handling.
* login/programs/request.c (do_setutent, do_updwtmp):
Allow arbitrary length filenames.
* login/programs/utmpd.h (setutent_request, updwtmp_request):
Get rid of fixed length file field.
* login/utmp_daemon.c (do_setutent, do_updwtmp):
Allow arbitrary length filenames.
* login/programs/request.c (do_pututline):
Don't fail if connection->position is -1 on entry.
1997-06-15 16:32 Mark Kettenis <kettenis@phys.uva.nl>
* login/utmp_file.c (updwtmp_file): Use the same method for
appending an entry as in pututline_file.
1997-06-11 18:59 Mark Kettenis <kettenis@phys.uva.nl>
* login/utmp_file.c (getutent_r_file):
Use read lock instead of write lock.
(getutline_r_file, internal_getut_r): Lock utmp file.
(updwtmp_file): Use fcntl to lock file instead of flock.
1997-06-18 00:11 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* stdio-common/vfscanf.c (inchar, ungetc): Don't count EOF as
character read in/put back.
* stdio-common/tstscanf.c: Add test case for this.
1997-06-17 22:17 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* libc.map: Add more symbols.
1997-06-18 12:01 Ulrich Drepper <drepper@cygnus.com>
* manual/Makefile (install): Make sure `dir' file exists if we use
install-info.
1997-06-17 19:32 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* manual/Makefile (info): Depend on dir-add.info.
1997-06-17 17:19 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* elf/ldd.bash.in: Fix spacing in message.
1997-06-17 14:28 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* elf/do-rel.h (elf_dynamic_do_rel): Always use version
information if available.
1997-06-17 11:34 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sunrpc/Makefile ($(rpcsvc:%.x=$(objpfx)rpcsvc/%.h)): Make
command non-empty to force make to recheck modification time.
($(rpcsvc:%.x=$(objpfx)x%.c)): Likewise.
1997-06-17 00:26 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/stub/e_acoshl.c: Set errno to ENOSYS.
* sysdeps/stub/e_acosl.c: Likewise.
* sysdeps/stub/e_asinl.c: Likewise.
* sysdeps/stub/e_atan2l.c: Likewise.
* sysdeps/stub/e_expl.c: Likewise.
* sysdeps/stub/e_fmodl.c: Likewise.
* sysdeps/stub/e_j0l.c: Likewise.
* sysdeps/stub/e_j1l.c: Likewise.
* sysdeps/stub/e_jnl.c: Likewise.
* sysdeps/stub/e_lgammal_r.c: Likewise.
* sysdeps/stub/e_log10l.c: Likewise.
* sysdeps/stub/e_logl.c: Likewise.
* sysdeps/stub/e_powl.c: Likewise.
* sysdeps/stub/e_rem_pio2l.c: Likewise.
* sysdeps/stub/e_sqrtl.c: Likewise.
* sysdeps/stub/k_cosl.c: Likewise.
* sysdeps/stub/k_rem_pio2l.c: Likewise.
* sysdeps/stub/k_sinl.c: Likewise.
* sysdeps/stub/k_tanl.c: Likewise.
* sysdeps/stub/s_atanl.c: Likewise.
* sysdeps/stub/s_erfl.c: Likewise.
* sysdeps/stub/s_exp2.c: Likewise.
* sysdeps/stub/s_exp2f.c: Likewise.
* sysdeps/stub/s_exp2l.c: Likewise.
* sysdeps/stub/s_expm1l.c: Likewise.
* sysdeps/stub/s_log1pl.c: Likewise.
* sysdeps/stub/s_log2l.c: Likewise.
1997-06-18 11:46 Ulrich Drepper <drepper@cygnus.com>
* sysdeps/libm-ieee754/w_gamma.c: If _LIB_VERSION is _SVID_ compute
result as before last change.
* sysdeps/libm-ieee754/w_gammaf.c: Likewise.
* sysdeps/libm-ieee754/w_gammal.c: Likewise.
1997-06-16 23:37 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/libm-ieee754/s_remquo.c: Fix off-by-one when computing
quotient.
* sysdeps/libm-ieee754/s_remquof.c: Likewise.
* sysdeps/libm-ieee754/s_remquol.c: Likewise.
* sysdeps/m68k/fpu/s_remquo.c: Remove FIXME and special case for
quotient.
* sysdeps/libm-ieee754/w_gamma.c: Add missing call to exp
function. Don't use global signgam.
* sysdeps/libm-ieee754/w_gammaf.c: Likewise.
* sysdeps/libm-ieee754/w_gammal.c: Likewise.
* math/Makefile (libm-calls): Remove w_gamma_r.
* sysdeps/libm-ieee754/w_gamma_r.c: Remove file.
* sysdeps/libm-ieee754/w_gammaf_r.c: Remove file.
* sysdeps/libm-ieee754/w_gammal_r.c: Remove file.
* math/libm-test.c (atanh_test): Declare x only if needed.
(signbit_test): Fix typo.
(gamma_test): Check whether function is implemented. Add
epsilons.
(lgamma_test): Likewise.
(fmod_test): Add epsilons.
(exp2_test): Use right function for existence test.
1997-06-07 09:20 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/powerpc/Dist: Add fenv_const.c.
* sysdeps/unix/sysv/linux/Dist: Add net/if_slip.h.
* sysdeps/unix/sysv/linux/powerpc/Dist: Add init-first.h and
syscall.h.
* sysdeps/unix/sysv/linux/sparc/Dist: Add init-first.h.
* string/Makefile (distribute): Add tst-svc.expect.
* nis/Makefile (distribute): Add nis_intern.h and Banner.
* elf/Makefile (distribute): Add dl-hash.h.
* Rules (subdir_echo-distinfo): Add headers from $(distribute).
* login/Makefile (others): Add utmpdump.
* sysdeps/unix/sysv/linux/Makefile (sysdep_headers): Add
net/if_slip.h.
* manual/Makefile (dir-add.texi): Also look in indirectly included
files.
1997-06-16 23:15 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/unix/sysv/linux/configure.in: Remove commands to
generate stdio_lim.h.
* sysdeps/unix/sysv/linux/mk-stdiolim.c: Remove.
* sysdeps/unix/sysv/linux/Makefile: Add rules to generate
stdio_lim.h here.
(common-generated): Add generated files.
(inhibit-stdio_lim): Define.
* sysdeps/posix/Makefile [$(inhibit-stdio_lim)=yes]: Disable rules
to generate stdio_lim.h.
* sysdeps/unix/sysv/linux/Makefile: Suppress inclusion of
dependecy files if no_deps is set.
($(objpfx)syscall-%.d): Add header file as target to dependency
generation.
1997-06-14 19:19 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/unix/sysv/linux/arm/Dist: New file.
1997-06-14 17:59 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* extra-lib.mk (others): Depend on versioned shared library, not
the unversioned one.
* Makerules (build-shlib): Don't make the version link here.
($(common-objpfx)libc.so$(libc.so-version)): New rule for libc
version link.
1997-06-16 03:07 Ulrich Drepper <drepper@cygnus.com>
* sysdeps/libm-ieee754/k_standard.c: Undo change of Tue Aug 6
01:13:56 1996.
* argp/argp-help.c (argp_args_usage): Supply correct argp to
filter_doc.
* argp/argp-help.c (hol_add_cluster): Initialize CL->depth.
* argp/argp-help.c (_help): Supply STATE to argp_args_usage.
* argp/argp.h (ARGP_KEY_HELP_ARGS_DOC): New macro.
* argp/argp-fmtstream.c: Add casts to prevent warnings.
* argp/argp.h (OPTION_NO_USAGE): New macro.
* argp/argp-help.c (usage_long_opt, usage_argful_short_opt,
* argp/argp-fmtstream.c (__argp_fmtstream_update): Account for case
* argp/argp-help.c <stddef.h>: New include.
* argp/argp.h (argp_state_help, __argp_state_help, argp_usage,
* argp/argp.h (argp_program_bug_address): Make const.
* argp/argp-parse.c (argp_default_parser): Set STATE->name for OPT_PROGNAME.
* argp/argp-help.c (__argp_error, __argp_failure, __argp_state_help):
* argp/argp-parse.c (parser_init): Set PARSER->state.flags.
1997-06-19 19:11:22 +00:00
|
|
|
|
|
|
|
EXTRACT_WORDS (hx, lx, x);
|
|
|
|
|
2014-02-10 13:45:42 +00:00
|
|
|
if (__glibc_unlikely (((hx & 0x7fffffff) | lx) == 0))
|
1999-06-28 15:52:36 +00:00
|
|
|
{
|
2004-11-21 19:58:34 +00:00
|
|
|
/* Return value for x == 0 is Inf with divide by zero exception. */
|
1999-06-28 15:52:36 +00:00
|
|
|
*signgamp = 0;
|
2004-11-21 19:58:34 +00:00
|
|
|
return 1.0 / x;
|
1999-06-28 15:52:36 +00:00
|
|
|
}
|
2011-10-12 15:27:51 +00:00
|
|
|
if (__builtin_expect (hx < 0, 0)
|
2017-08-03 19:55:04 +00:00
|
|
|
&& (uint32_t) hx < 0xfff00000 && __rint (x) == x)
|
Update.
1997-06-19 19:38 Ulrich Drepper <drepper@cygnus.com>
* features.h: Define __STDC_IEC_559__ and _STDC_IEC_559_COMPLEX__.
* elf/dl-minimal.c (__dcgettext): Remove assertion.
* inet/rcmd.c: Correct a few typos. Reported by Eric Troan.
* manual/Makefile (distribute): Add dir.
* manual/dir: New file.
* math/Makefile (libm-support): Rename s_rinttol, s_rinttoll,
s_roundtol, and s_roundtoll to s_lrint, s_llrint, s_lround,
and s_llround respectively.
(libm-calls): Add e_gamma_r.
* math/libm-test.c (check_int_exc): New function.
(signbit_test): Remove test for sign of NaN.
(gamma_test): Clear exception after test of existence.
Correct tests and and epsilons.
(lgamma_test): Likewise.
(ilogb_test): Correct all tests.
(scalb_test): Rewrite.
(rinttol_test): Rename to lrint_test and correct tests.
(rinttoll_test): Likewise.
(roundtol_test): Likewise.
(roundtoll_test): Likewise.
(main): Call lrint/lround functions instead of rinttol/roundtol.
* math/math.h: Change prototypes for rinttol/roundtol.
* math/mathcalls.h: Rearrange prototypes according to ISO C9X draft.
* sysdeps/generic/mathbits.h: Define FP_ILOGB0 and FP_ILOGBNAN.
* sysdeps/i386/mathbits.h: Likewise.
* sysdeps/libm-i387/e_scalb.S: Handle special cases correctly.
* sysdeps/libm-i387/e_scalbf.S: Likewise.
* sysdeps/libm-i387/e_scalbl.S: Likewise.
* sysdeps/libm-i387/s_asinh.S: Handle -inf correctly.
* sysdeps/libm-i387/s_asinhf.S: Likewise.
* sysdeps/libm-i387/s_asinhl.S: Likewise.
* sysdeps/libm-i387/s_ilogb.S: Optimize.
* sysdeps/libm-i387/s_ilogbf.S: Likewise.
* sysdeps/libm-i387/s_ilogbl.S: Likewise.
* sysdeps/libm-i387/s_rinttol.S: Rename to...
* sysdeps/libm-i387/s_lrint.S: ...this.
* sysdeps/libm-i387/s_rinttoll.S: Rename to...
* sysdeps/libm-i387/s_llrint.S: ...this.
* sysdeps/libm-i387/s_remquo.S: Correctly set sign of remainder.
* sysdeps/libm-i387/s_remquof.S: Likewise.
* sysdeps/libm-i387/s_remquol.S: Likewise.
* sysdeps/libm-i387/e_gamma_r.c: New file. Implementation of gamma
function according to ISO C.
* sysdeps/libm-i387/e_gammaf_r.c: New file.
* sysdeps/libm-i387/e_gammal_r.c: New file.
* sysdeps/libm-i387/e_lgamma_r.c: Don't let optimize compile the
generation of exceptions away.
* sysdeps/libm-i387/e_lgammaf_r.c: Likewise.
* sysdeps/libm-i387/k_standard.c: Correct return value for infinity
points of gamma function when not SVID mode.
* sysdeps/libm-i387/s_rinttoll.c: Renamed to...
* sysdeps/libm-i387/s_llrint.c: ...this.
* sysdeps/libm-i387/s_rinttol.c: Renamed to...
* sysdeps/libm-i387/s_lrint.c: ...this.
* sysdeps/libm-i387/s_roundtoll.c: Renamed to...
* sysdeps/libm-i387/s_llround.c: ...this.
* sysdeps/libm-i387/s_roundtol.c: Renamed to...
* sysdeps/libm-i387/s_lround.c: ..this.
* sysdeps/libm-i387/s_scalbn.c: Change second parameter according to
ISO C.
* sysdeps/libm-i387/s_scalbnf.c: Likewise.
* sysdeps/libm-i387/s_scalbnl.c: Likewise.
* sysdeps/libm-i387/w_gamma.c: Call __ieee754_gamma_r if library
mode is not _SVID_.
* sysdeps/libm-i387/w_gammaf.c: Likewise.
* sysdeps/libm-i387/w_gammal.c: Likewise.
* sysdeps/m68k/fpu/__math.h: Rename __rinttol to __lrint and
rinttol to lrint.
* sysdeps/m68k/fpu/s_rinttol.c: Renamed to...
* sysdeps/m68k/fpu/s_lrint.c: ...this.
* sysdeps/m68k/fpu/s_rinttoll.c: Renamed to...
* sysdeps/m68k/fpu/s_llrint.c: ...this.
* md5-crypt/Makefile: Link md5test program with md5.o.
* stdio-common/temptest.c: Don't use __stdio_gen_tempname which
is not exported by the libc.so.
* stdio-common/vfscanf.c: Correct scanning of strings after last
change.
* sysdeps/unix/sysv/linux/i386/sysdep.S: Use .comm to define errno.
1997-06-19 07:37 H.J. Lu <hjl@gnu.ai.mit.edu>
* time/tzfile.c (__tzfile_read): Store getc () return in int.
1997-06-13 Miles Bader <miles@gnu.ai.mit.edu>
* argp/argp-parse.c (argp_version_parser): Include `(PROGRAM ERROR)' in
the no-version error text to indicate that something's fucked.
[!_] (N_): New macro.
(argp_default_options, argp_version_options): Wrap doc strings in N_().
1997-06-12 Miles Bader <miles@gnu.ai.mit.edu>
* argp/argp-parse.c (parser_parse_opt): Detect and report unhandled
options here.
(parser_parse_arg): Handle ARGP_KEY_ARGS here.
Adjust NEXT pointer back if we fail to parse anything.
(parser_parse_next): Simplify arg code. Leave state NEXT frobbing
to parser_parse_arg.
1997-06-11 Miles Bader <miles@gnu.ai.mit.edu>
* argp/argp.h (ARGP_KEY_ARGS, ARGP_KEY_FINI): New macros.
* argp/argp-parse.c (parser_finalize): Do another pass over the
parsers with ARGP_KEY_FINI.
1997-06-18 Miles Bader <miles@gnu.ai.mit.edu>
* string/Makefile (routines): Add argz-replace.
1997-06-16 00:16 Miles Bader <miles@gnu.ai.mit.edu>
* manual/string.texi (Argz Functions): Document argz_replace.
1997-06-12 Miles Bader <miles@gnu.ai.mit.edu>
* string/argz.h (__argz_replace, argz_replace): New declarations.
* string/argz-replace.c: New file.
1997-06-11 Miles Bader <miles@gnu.ai.mit.edu>
* manual/argp.texi (Argp Special Keys): Document ARGP_KEY_ARGS.
1997-06-16 23:08 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* Makeconfig (libc-map): Remove definition.
* Makerules (libc-map): Define it here, using the full name.
(load-map-file): Remove case for empty $(..).
($(common-objpfx)libc.so): Revert last change.
1997-06-17 22:18 Mark Kettenis <kettenis@phys.uva.nl>
* login/programs/utmpd.c (main): Improve signal handling.
* login/programs/request.c (do_setutent, do_updwtmp):
Allow arbitrary length filenames.
* login/programs/utmpd.h (setutent_request, updwtmp_request):
Get rid of fixed length file field.
* login/utmp_daemon.c (do_setutent, do_updwtmp):
Allow arbitrary length filenames.
* login/programs/request.c (do_pututline):
Don't fail if connection->position is -1 on entry.
1997-06-15 16:32 Mark Kettenis <kettenis@phys.uva.nl>
* login/utmp_file.c (updwtmp_file): Use the same method for
appending an entry as in pututline_file.
1997-06-11 18:59 Mark Kettenis <kettenis@phys.uva.nl>
* login/utmp_file.c (getutent_r_file):
Use read lock instead of write lock.
(getutline_r_file, internal_getut_r): Lock utmp file.
(updwtmp_file): Use fcntl to lock file instead of flock.
1997-06-18 00:11 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* stdio-common/vfscanf.c (inchar, ungetc): Don't count EOF as
character read in/put back.
* stdio-common/tstscanf.c: Add test case for this.
1997-06-17 22:17 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* libc.map: Add more symbols.
1997-06-18 12:01 Ulrich Drepper <drepper@cygnus.com>
* manual/Makefile (install): Make sure `dir' file exists if we use
install-info.
1997-06-17 19:32 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* manual/Makefile (info): Depend on dir-add.info.
1997-06-17 17:19 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* elf/ldd.bash.in: Fix spacing in message.
1997-06-17 14:28 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* elf/do-rel.h (elf_dynamic_do_rel): Always use version
information if available.
1997-06-17 11:34 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sunrpc/Makefile ($(rpcsvc:%.x=$(objpfx)rpcsvc/%.h)): Make
command non-empty to force make to recheck modification time.
($(rpcsvc:%.x=$(objpfx)x%.c)): Likewise.
1997-06-17 00:26 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/stub/e_acoshl.c: Set errno to ENOSYS.
* sysdeps/stub/e_acosl.c: Likewise.
* sysdeps/stub/e_asinl.c: Likewise.
* sysdeps/stub/e_atan2l.c: Likewise.
* sysdeps/stub/e_expl.c: Likewise.
* sysdeps/stub/e_fmodl.c: Likewise.
* sysdeps/stub/e_j0l.c: Likewise.
* sysdeps/stub/e_j1l.c: Likewise.
* sysdeps/stub/e_jnl.c: Likewise.
* sysdeps/stub/e_lgammal_r.c: Likewise.
* sysdeps/stub/e_log10l.c: Likewise.
* sysdeps/stub/e_logl.c: Likewise.
* sysdeps/stub/e_powl.c: Likewise.
* sysdeps/stub/e_rem_pio2l.c: Likewise.
* sysdeps/stub/e_sqrtl.c: Likewise.
* sysdeps/stub/k_cosl.c: Likewise.
* sysdeps/stub/k_rem_pio2l.c: Likewise.
* sysdeps/stub/k_sinl.c: Likewise.
* sysdeps/stub/k_tanl.c: Likewise.
* sysdeps/stub/s_atanl.c: Likewise.
* sysdeps/stub/s_erfl.c: Likewise.
* sysdeps/stub/s_exp2.c: Likewise.
* sysdeps/stub/s_exp2f.c: Likewise.
* sysdeps/stub/s_exp2l.c: Likewise.
* sysdeps/stub/s_expm1l.c: Likewise.
* sysdeps/stub/s_log1pl.c: Likewise.
* sysdeps/stub/s_log2l.c: Likewise.
1997-06-18 11:46 Ulrich Drepper <drepper@cygnus.com>
* sysdeps/libm-ieee754/w_gamma.c: If _LIB_VERSION is _SVID_ compute
result as before last change.
* sysdeps/libm-ieee754/w_gammaf.c: Likewise.
* sysdeps/libm-ieee754/w_gammal.c: Likewise.
1997-06-16 23:37 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/libm-ieee754/s_remquo.c: Fix off-by-one when computing
quotient.
* sysdeps/libm-ieee754/s_remquof.c: Likewise.
* sysdeps/libm-ieee754/s_remquol.c: Likewise.
* sysdeps/m68k/fpu/s_remquo.c: Remove FIXME and special case for
quotient.
* sysdeps/libm-ieee754/w_gamma.c: Add missing call to exp
function. Don't use global signgam.
* sysdeps/libm-ieee754/w_gammaf.c: Likewise.
* sysdeps/libm-ieee754/w_gammal.c: Likewise.
* math/Makefile (libm-calls): Remove w_gamma_r.
* sysdeps/libm-ieee754/w_gamma_r.c: Remove file.
* sysdeps/libm-ieee754/w_gammaf_r.c: Remove file.
* sysdeps/libm-ieee754/w_gammal_r.c: Remove file.
* math/libm-test.c (atanh_test): Declare x only if needed.
(signbit_test): Fix typo.
(gamma_test): Check whether function is implemented. Add
epsilons.
(lgamma_test): Likewise.
(fmod_test): Add epsilons.
(exp2_test): Use right function for existence test.
1997-06-07 09:20 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/powerpc/Dist: Add fenv_const.c.
* sysdeps/unix/sysv/linux/Dist: Add net/if_slip.h.
* sysdeps/unix/sysv/linux/powerpc/Dist: Add init-first.h and
syscall.h.
* sysdeps/unix/sysv/linux/sparc/Dist: Add init-first.h.
* string/Makefile (distribute): Add tst-svc.expect.
* nis/Makefile (distribute): Add nis_intern.h and Banner.
* elf/Makefile (distribute): Add dl-hash.h.
* Rules (subdir_echo-distinfo): Add headers from $(distribute).
* login/Makefile (others): Add utmpdump.
* sysdeps/unix/sysv/linux/Makefile (sysdep_headers): Add
net/if_slip.h.
* manual/Makefile (dir-add.texi): Also look in indirectly included
files.
1997-06-16 23:15 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/unix/sysv/linux/configure.in: Remove commands to
generate stdio_lim.h.
* sysdeps/unix/sysv/linux/mk-stdiolim.c: Remove.
* sysdeps/unix/sysv/linux/Makefile: Add rules to generate
stdio_lim.h here.
(common-generated): Add generated files.
(inhibit-stdio_lim): Define.
* sysdeps/posix/Makefile [$(inhibit-stdio_lim)=yes]: Disable rules
to generate stdio_lim.h.
* sysdeps/unix/sysv/linux/Makefile: Suppress inclusion of
dependecy files if no_deps is set.
($(objpfx)syscall-%.d): Add header file as target to dependency
generation.
1997-06-14 19:19 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/unix/sysv/linux/arm/Dist: New file.
1997-06-14 17:59 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* extra-lib.mk (others): Depend on versioned shared library, not
the unversioned one.
* Makerules (build-shlib): Don't make the version link here.
($(common-objpfx)libc.so$(libc.so-version)): New rule for libc
version link.
1997-06-16 03:07 Ulrich Drepper <drepper@cygnus.com>
* sysdeps/libm-ieee754/k_standard.c: Undo change of Tue Aug 6
01:13:56 1996.
* argp/argp-help.c (argp_args_usage): Supply correct argp to
filter_doc.
* argp/argp-help.c (hol_add_cluster): Initialize CL->depth.
* argp/argp-help.c (_help): Supply STATE to argp_args_usage.
* argp/argp.h (ARGP_KEY_HELP_ARGS_DOC): New macro.
* argp/argp-fmtstream.c: Add casts to prevent warnings.
* argp/argp.h (OPTION_NO_USAGE): New macro.
* argp/argp-help.c (usage_long_opt, usage_argful_short_opt,
* argp/argp-fmtstream.c (__argp_fmtstream_update): Account for case
* argp/argp-help.c <stddef.h>: New include.
* argp/argp.h (argp_state_help, __argp_state_help, argp_usage,
* argp/argp.h (argp_program_bug_address): Make const.
* argp/argp-parse.c (argp_default_parser): Set STATE->name for OPT_PROGNAME.
* argp/argp-help.c (__argp_error, __argp_failure, __argp_state_help):
* argp/argp-parse.c (parser_init): Set PARSER->state.flags.
1997-06-19 19:11:22 +00:00
|
|
|
{
|
|
|
|
/* Return value for integer x < 0 is NaN with invalid exception. */
|
1999-06-28 15:52:36 +00:00
|
|
|
*signgamp = 0;
|
Update.
1997-06-19 19:38 Ulrich Drepper <drepper@cygnus.com>
* features.h: Define __STDC_IEC_559__ and _STDC_IEC_559_COMPLEX__.
* elf/dl-minimal.c (__dcgettext): Remove assertion.
* inet/rcmd.c: Correct a few typos. Reported by Eric Troan.
* manual/Makefile (distribute): Add dir.
* manual/dir: New file.
* math/Makefile (libm-support): Rename s_rinttol, s_rinttoll,
s_roundtol, and s_roundtoll to s_lrint, s_llrint, s_lround,
and s_llround respectively.
(libm-calls): Add e_gamma_r.
* math/libm-test.c (check_int_exc): New function.
(signbit_test): Remove test for sign of NaN.
(gamma_test): Clear exception after test of existence.
Correct tests and and epsilons.
(lgamma_test): Likewise.
(ilogb_test): Correct all tests.
(scalb_test): Rewrite.
(rinttol_test): Rename to lrint_test and correct tests.
(rinttoll_test): Likewise.
(roundtol_test): Likewise.
(roundtoll_test): Likewise.
(main): Call lrint/lround functions instead of rinttol/roundtol.
* math/math.h: Change prototypes for rinttol/roundtol.
* math/mathcalls.h: Rearrange prototypes according to ISO C9X draft.
* sysdeps/generic/mathbits.h: Define FP_ILOGB0 and FP_ILOGBNAN.
* sysdeps/i386/mathbits.h: Likewise.
* sysdeps/libm-i387/e_scalb.S: Handle special cases correctly.
* sysdeps/libm-i387/e_scalbf.S: Likewise.
* sysdeps/libm-i387/e_scalbl.S: Likewise.
* sysdeps/libm-i387/s_asinh.S: Handle -inf correctly.
* sysdeps/libm-i387/s_asinhf.S: Likewise.
* sysdeps/libm-i387/s_asinhl.S: Likewise.
* sysdeps/libm-i387/s_ilogb.S: Optimize.
* sysdeps/libm-i387/s_ilogbf.S: Likewise.
* sysdeps/libm-i387/s_ilogbl.S: Likewise.
* sysdeps/libm-i387/s_rinttol.S: Rename to...
* sysdeps/libm-i387/s_lrint.S: ...this.
* sysdeps/libm-i387/s_rinttoll.S: Rename to...
* sysdeps/libm-i387/s_llrint.S: ...this.
* sysdeps/libm-i387/s_remquo.S: Correctly set sign of remainder.
* sysdeps/libm-i387/s_remquof.S: Likewise.
* sysdeps/libm-i387/s_remquol.S: Likewise.
* sysdeps/libm-i387/e_gamma_r.c: New file. Implementation of gamma
function according to ISO C.
* sysdeps/libm-i387/e_gammaf_r.c: New file.
* sysdeps/libm-i387/e_gammal_r.c: New file.
* sysdeps/libm-i387/e_lgamma_r.c: Don't let optimize compile the
generation of exceptions away.
* sysdeps/libm-i387/e_lgammaf_r.c: Likewise.
* sysdeps/libm-i387/k_standard.c: Correct return value for infinity
points of gamma function when not SVID mode.
* sysdeps/libm-i387/s_rinttoll.c: Renamed to...
* sysdeps/libm-i387/s_llrint.c: ...this.
* sysdeps/libm-i387/s_rinttol.c: Renamed to...
* sysdeps/libm-i387/s_lrint.c: ...this.
* sysdeps/libm-i387/s_roundtoll.c: Renamed to...
* sysdeps/libm-i387/s_llround.c: ...this.
* sysdeps/libm-i387/s_roundtol.c: Renamed to...
* sysdeps/libm-i387/s_lround.c: ..this.
* sysdeps/libm-i387/s_scalbn.c: Change second parameter according to
ISO C.
* sysdeps/libm-i387/s_scalbnf.c: Likewise.
* sysdeps/libm-i387/s_scalbnl.c: Likewise.
* sysdeps/libm-i387/w_gamma.c: Call __ieee754_gamma_r if library
mode is not _SVID_.
* sysdeps/libm-i387/w_gammaf.c: Likewise.
* sysdeps/libm-i387/w_gammal.c: Likewise.
* sysdeps/m68k/fpu/__math.h: Rename __rinttol to __lrint and
rinttol to lrint.
* sysdeps/m68k/fpu/s_rinttol.c: Renamed to...
* sysdeps/m68k/fpu/s_lrint.c: ...this.
* sysdeps/m68k/fpu/s_rinttoll.c: Renamed to...
* sysdeps/m68k/fpu/s_llrint.c: ...this.
* md5-crypt/Makefile: Link md5test program with md5.o.
* stdio-common/temptest.c: Don't use __stdio_gen_tempname which
is not exported by the libc.so.
* stdio-common/vfscanf.c: Correct scanning of strings after last
change.
* sysdeps/unix/sysv/linux/i386/sysdep.S: Use .comm to define errno.
1997-06-19 07:37 H.J. Lu <hjl@gnu.ai.mit.edu>
* time/tzfile.c (__tzfile_read): Store getc () return in int.
1997-06-13 Miles Bader <miles@gnu.ai.mit.edu>
* argp/argp-parse.c (argp_version_parser): Include `(PROGRAM ERROR)' in
the no-version error text to indicate that something's fucked.
[!_] (N_): New macro.
(argp_default_options, argp_version_options): Wrap doc strings in N_().
1997-06-12 Miles Bader <miles@gnu.ai.mit.edu>
* argp/argp-parse.c (parser_parse_opt): Detect and report unhandled
options here.
(parser_parse_arg): Handle ARGP_KEY_ARGS here.
Adjust NEXT pointer back if we fail to parse anything.
(parser_parse_next): Simplify arg code. Leave state NEXT frobbing
to parser_parse_arg.
1997-06-11 Miles Bader <miles@gnu.ai.mit.edu>
* argp/argp.h (ARGP_KEY_ARGS, ARGP_KEY_FINI): New macros.
* argp/argp-parse.c (parser_finalize): Do another pass over the
parsers with ARGP_KEY_FINI.
1997-06-18 Miles Bader <miles@gnu.ai.mit.edu>
* string/Makefile (routines): Add argz-replace.
1997-06-16 00:16 Miles Bader <miles@gnu.ai.mit.edu>
* manual/string.texi (Argz Functions): Document argz_replace.
1997-06-12 Miles Bader <miles@gnu.ai.mit.edu>
* string/argz.h (__argz_replace, argz_replace): New declarations.
* string/argz-replace.c: New file.
1997-06-11 Miles Bader <miles@gnu.ai.mit.edu>
* manual/argp.texi (Argp Special Keys): Document ARGP_KEY_ARGS.
1997-06-16 23:08 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* Makeconfig (libc-map): Remove definition.
* Makerules (libc-map): Define it here, using the full name.
(load-map-file): Remove case for empty $(..).
($(common-objpfx)libc.so): Revert last change.
1997-06-17 22:18 Mark Kettenis <kettenis@phys.uva.nl>
* login/programs/utmpd.c (main): Improve signal handling.
* login/programs/request.c (do_setutent, do_updwtmp):
Allow arbitrary length filenames.
* login/programs/utmpd.h (setutent_request, updwtmp_request):
Get rid of fixed length file field.
* login/utmp_daemon.c (do_setutent, do_updwtmp):
Allow arbitrary length filenames.
* login/programs/request.c (do_pututline):
Don't fail if connection->position is -1 on entry.
1997-06-15 16:32 Mark Kettenis <kettenis@phys.uva.nl>
* login/utmp_file.c (updwtmp_file): Use the same method for
appending an entry as in pututline_file.
1997-06-11 18:59 Mark Kettenis <kettenis@phys.uva.nl>
* login/utmp_file.c (getutent_r_file):
Use read lock instead of write lock.
(getutline_r_file, internal_getut_r): Lock utmp file.
(updwtmp_file): Use fcntl to lock file instead of flock.
1997-06-18 00:11 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* stdio-common/vfscanf.c (inchar, ungetc): Don't count EOF as
character read in/put back.
* stdio-common/tstscanf.c: Add test case for this.
1997-06-17 22:17 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* libc.map: Add more symbols.
1997-06-18 12:01 Ulrich Drepper <drepper@cygnus.com>
* manual/Makefile (install): Make sure `dir' file exists if we use
install-info.
1997-06-17 19:32 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* manual/Makefile (info): Depend on dir-add.info.
1997-06-17 17:19 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* elf/ldd.bash.in: Fix spacing in message.
1997-06-17 14:28 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* elf/do-rel.h (elf_dynamic_do_rel): Always use version
information if available.
1997-06-17 11:34 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sunrpc/Makefile ($(rpcsvc:%.x=$(objpfx)rpcsvc/%.h)): Make
command non-empty to force make to recheck modification time.
($(rpcsvc:%.x=$(objpfx)x%.c)): Likewise.
1997-06-17 00:26 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/stub/e_acoshl.c: Set errno to ENOSYS.
* sysdeps/stub/e_acosl.c: Likewise.
* sysdeps/stub/e_asinl.c: Likewise.
* sysdeps/stub/e_atan2l.c: Likewise.
* sysdeps/stub/e_expl.c: Likewise.
* sysdeps/stub/e_fmodl.c: Likewise.
* sysdeps/stub/e_j0l.c: Likewise.
* sysdeps/stub/e_j1l.c: Likewise.
* sysdeps/stub/e_jnl.c: Likewise.
* sysdeps/stub/e_lgammal_r.c: Likewise.
* sysdeps/stub/e_log10l.c: Likewise.
* sysdeps/stub/e_logl.c: Likewise.
* sysdeps/stub/e_powl.c: Likewise.
* sysdeps/stub/e_rem_pio2l.c: Likewise.
* sysdeps/stub/e_sqrtl.c: Likewise.
* sysdeps/stub/k_cosl.c: Likewise.
* sysdeps/stub/k_rem_pio2l.c: Likewise.
* sysdeps/stub/k_sinl.c: Likewise.
* sysdeps/stub/k_tanl.c: Likewise.
* sysdeps/stub/s_atanl.c: Likewise.
* sysdeps/stub/s_erfl.c: Likewise.
* sysdeps/stub/s_exp2.c: Likewise.
* sysdeps/stub/s_exp2f.c: Likewise.
* sysdeps/stub/s_exp2l.c: Likewise.
* sysdeps/stub/s_expm1l.c: Likewise.
* sysdeps/stub/s_log1pl.c: Likewise.
* sysdeps/stub/s_log2l.c: Likewise.
1997-06-18 11:46 Ulrich Drepper <drepper@cygnus.com>
* sysdeps/libm-ieee754/w_gamma.c: If _LIB_VERSION is _SVID_ compute
result as before last change.
* sysdeps/libm-ieee754/w_gammaf.c: Likewise.
* sysdeps/libm-ieee754/w_gammal.c: Likewise.
1997-06-16 23:37 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/libm-ieee754/s_remquo.c: Fix off-by-one when computing
quotient.
* sysdeps/libm-ieee754/s_remquof.c: Likewise.
* sysdeps/libm-ieee754/s_remquol.c: Likewise.
* sysdeps/m68k/fpu/s_remquo.c: Remove FIXME and special case for
quotient.
* sysdeps/libm-ieee754/w_gamma.c: Add missing call to exp
function. Don't use global signgam.
* sysdeps/libm-ieee754/w_gammaf.c: Likewise.
* sysdeps/libm-ieee754/w_gammal.c: Likewise.
* math/Makefile (libm-calls): Remove w_gamma_r.
* sysdeps/libm-ieee754/w_gamma_r.c: Remove file.
* sysdeps/libm-ieee754/w_gammaf_r.c: Remove file.
* sysdeps/libm-ieee754/w_gammal_r.c: Remove file.
* math/libm-test.c (atanh_test): Declare x only if needed.
(signbit_test): Fix typo.
(gamma_test): Check whether function is implemented. Add
epsilons.
(lgamma_test): Likewise.
(fmod_test): Add epsilons.
(exp2_test): Use right function for existence test.
1997-06-07 09:20 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/powerpc/Dist: Add fenv_const.c.
* sysdeps/unix/sysv/linux/Dist: Add net/if_slip.h.
* sysdeps/unix/sysv/linux/powerpc/Dist: Add init-first.h and
syscall.h.
* sysdeps/unix/sysv/linux/sparc/Dist: Add init-first.h.
* string/Makefile (distribute): Add tst-svc.expect.
* nis/Makefile (distribute): Add nis_intern.h and Banner.
* elf/Makefile (distribute): Add dl-hash.h.
* Rules (subdir_echo-distinfo): Add headers from $(distribute).
* login/Makefile (others): Add utmpdump.
* sysdeps/unix/sysv/linux/Makefile (sysdep_headers): Add
net/if_slip.h.
* manual/Makefile (dir-add.texi): Also look in indirectly included
files.
1997-06-16 23:15 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/unix/sysv/linux/configure.in: Remove commands to
generate stdio_lim.h.
* sysdeps/unix/sysv/linux/mk-stdiolim.c: Remove.
* sysdeps/unix/sysv/linux/Makefile: Add rules to generate
stdio_lim.h here.
(common-generated): Add generated files.
(inhibit-stdio_lim): Define.
* sysdeps/posix/Makefile [$(inhibit-stdio_lim)=yes]: Disable rules
to generate stdio_lim.h.
* sysdeps/unix/sysv/linux/Makefile: Suppress inclusion of
dependecy files if no_deps is set.
($(objpfx)syscall-%.d): Add header file as target to dependency
generation.
1997-06-14 19:19 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/unix/sysv/linux/arm/Dist: New file.
1997-06-14 17:59 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* extra-lib.mk (others): Depend on versioned shared library, not
the unversioned one.
* Makerules (build-shlib): Don't make the version link here.
($(common-objpfx)libc.so$(libc.so-version)): New rule for libc
version link.
1997-06-16 03:07 Ulrich Drepper <drepper@cygnus.com>
* sysdeps/libm-ieee754/k_standard.c: Undo change of Tue Aug 6
01:13:56 1996.
* argp/argp-help.c (argp_args_usage): Supply correct argp to
filter_doc.
* argp/argp-help.c (hol_add_cluster): Initialize CL->depth.
* argp/argp-help.c (_help): Supply STATE to argp_args_usage.
* argp/argp.h (ARGP_KEY_HELP_ARGS_DOC): New macro.
* argp/argp-fmtstream.c: Add casts to prevent warnings.
* argp/argp.h (OPTION_NO_USAGE): New macro.
* argp/argp-help.c (usage_long_opt, usage_argful_short_opt,
* argp/argp-fmtstream.c (__argp_fmtstream_update): Account for case
* argp/argp-help.c <stddef.h>: New include.
* argp/argp.h (argp_state_help, __argp_state_help, argp_usage,
* argp/argp.h (argp_program_bug_address): Make const.
* argp/argp-parse.c (argp_default_parser): Set STATE->name for OPT_PROGNAME.
* argp/argp-help.c (__argp_error, __argp_failure, __argp_state_help):
* argp/argp-parse.c (parser_init): Set PARSER->state.flags.
1997-06-19 19:11:22 +00:00
|
|
|
return (x - x) / (x - x);
|
|
|
|
}
|
2014-02-10 13:45:42 +00:00
|
|
|
if (__glibc_unlikely ((unsigned int) hx == 0xfff00000 && lx == 0))
|
2001-03-04 19:25:06 +00:00
|
|
|
{
|
|
|
|
/* x == -Inf. According to ISO this is NaN. */
|
|
|
|
*signgamp = 0;
|
|
|
|
return x - x;
|
|
|
|
}
|
2014-02-10 13:45:42 +00:00
|
|
|
if (__glibc_unlikely ((hx & 0x7ff00000) == 0x7ff00000))
|
2013-05-08 11:58:18 +00:00
|
|
|
{
|
|
|
|
/* Positive infinity (return positive infinity) or NaN (return
|
|
|
|
NaN). */
|
|
|
|
*signgamp = 0;
|
|
|
|
return x + x;
|
|
|
|
}
|
Update.
1997-06-19 19:38 Ulrich Drepper <drepper@cygnus.com>
* features.h: Define __STDC_IEC_559__ and _STDC_IEC_559_COMPLEX__.
* elf/dl-minimal.c (__dcgettext): Remove assertion.
* inet/rcmd.c: Correct a few typos. Reported by Eric Troan.
* manual/Makefile (distribute): Add dir.
* manual/dir: New file.
* math/Makefile (libm-support): Rename s_rinttol, s_rinttoll,
s_roundtol, and s_roundtoll to s_lrint, s_llrint, s_lround,
and s_llround respectively.
(libm-calls): Add e_gamma_r.
* math/libm-test.c (check_int_exc): New function.
(signbit_test): Remove test for sign of NaN.
(gamma_test): Clear exception after test of existence.
Correct tests and and epsilons.
(lgamma_test): Likewise.
(ilogb_test): Correct all tests.
(scalb_test): Rewrite.
(rinttol_test): Rename to lrint_test and correct tests.
(rinttoll_test): Likewise.
(roundtol_test): Likewise.
(roundtoll_test): Likewise.
(main): Call lrint/lround functions instead of rinttol/roundtol.
* math/math.h: Change prototypes for rinttol/roundtol.
* math/mathcalls.h: Rearrange prototypes according to ISO C9X draft.
* sysdeps/generic/mathbits.h: Define FP_ILOGB0 and FP_ILOGBNAN.
* sysdeps/i386/mathbits.h: Likewise.
* sysdeps/libm-i387/e_scalb.S: Handle special cases correctly.
* sysdeps/libm-i387/e_scalbf.S: Likewise.
* sysdeps/libm-i387/e_scalbl.S: Likewise.
* sysdeps/libm-i387/s_asinh.S: Handle -inf correctly.
* sysdeps/libm-i387/s_asinhf.S: Likewise.
* sysdeps/libm-i387/s_asinhl.S: Likewise.
* sysdeps/libm-i387/s_ilogb.S: Optimize.
* sysdeps/libm-i387/s_ilogbf.S: Likewise.
* sysdeps/libm-i387/s_ilogbl.S: Likewise.
* sysdeps/libm-i387/s_rinttol.S: Rename to...
* sysdeps/libm-i387/s_lrint.S: ...this.
* sysdeps/libm-i387/s_rinttoll.S: Rename to...
* sysdeps/libm-i387/s_llrint.S: ...this.
* sysdeps/libm-i387/s_remquo.S: Correctly set sign of remainder.
* sysdeps/libm-i387/s_remquof.S: Likewise.
* sysdeps/libm-i387/s_remquol.S: Likewise.
* sysdeps/libm-i387/e_gamma_r.c: New file. Implementation of gamma
function according to ISO C.
* sysdeps/libm-i387/e_gammaf_r.c: New file.
* sysdeps/libm-i387/e_gammal_r.c: New file.
* sysdeps/libm-i387/e_lgamma_r.c: Don't let optimize compile the
generation of exceptions away.
* sysdeps/libm-i387/e_lgammaf_r.c: Likewise.
* sysdeps/libm-i387/k_standard.c: Correct return value for infinity
points of gamma function when not SVID mode.
* sysdeps/libm-i387/s_rinttoll.c: Renamed to...
* sysdeps/libm-i387/s_llrint.c: ...this.
* sysdeps/libm-i387/s_rinttol.c: Renamed to...
* sysdeps/libm-i387/s_lrint.c: ...this.
* sysdeps/libm-i387/s_roundtoll.c: Renamed to...
* sysdeps/libm-i387/s_llround.c: ...this.
* sysdeps/libm-i387/s_roundtol.c: Renamed to...
* sysdeps/libm-i387/s_lround.c: ..this.
* sysdeps/libm-i387/s_scalbn.c: Change second parameter according to
ISO C.
* sysdeps/libm-i387/s_scalbnf.c: Likewise.
* sysdeps/libm-i387/s_scalbnl.c: Likewise.
* sysdeps/libm-i387/w_gamma.c: Call __ieee754_gamma_r if library
mode is not _SVID_.
* sysdeps/libm-i387/w_gammaf.c: Likewise.
* sysdeps/libm-i387/w_gammal.c: Likewise.
* sysdeps/m68k/fpu/__math.h: Rename __rinttol to __lrint and
rinttol to lrint.
* sysdeps/m68k/fpu/s_rinttol.c: Renamed to...
* sysdeps/m68k/fpu/s_lrint.c: ...this.
* sysdeps/m68k/fpu/s_rinttoll.c: Renamed to...
* sysdeps/m68k/fpu/s_llrint.c: ...this.
* md5-crypt/Makefile: Link md5test program with md5.o.
* stdio-common/temptest.c: Don't use __stdio_gen_tempname which
is not exported by the libc.so.
* stdio-common/vfscanf.c: Correct scanning of strings after last
change.
* sysdeps/unix/sysv/linux/i386/sysdep.S: Use .comm to define errno.
1997-06-19 07:37 H.J. Lu <hjl@gnu.ai.mit.edu>
* time/tzfile.c (__tzfile_read): Store getc () return in int.
1997-06-13 Miles Bader <miles@gnu.ai.mit.edu>
* argp/argp-parse.c (argp_version_parser): Include `(PROGRAM ERROR)' in
the no-version error text to indicate that something's fucked.
[!_] (N_): New macro.
(argp_default_options, argp_version_options): Wrap doc strings in N_().
1997-06-12 Miles Bader <miles@gnu.ai.mit.edu>
* argp/argp-parse.c (parser_parse_opt): Detect and report unhandled
options here.
(parser_parse_arg): Handle ARGP_KEY_ARGS here.
Adjust NEXT pointer back if we fail to parse anything.
(parser_parse_next): Simplify arg code. Leave state NEXT frobbing
to parser_parse_arg.
1997-06-11 Miles Bader <miles@gnu.ai.mit.edu>
* argp/argp.h (ARGP_KEY_ARGS, ARGP_KEY_FINI): New macros.
* argp/argp-parse.c (parser_finalize): Do another pass over the
parsers with ARGP_KEY_FINI.
1997-06-18 Miles Bader <miles@gnu.ai.mit.edu>
* string/Makefile (routines): Add argz-replace.
1997-06-16 00:16 Miles Bader <miles@gnu.ai.mit.edu>
* manual/string.texi (Argz Functions): Document argz_replace.
1997-06-12 Miles Bader <miles@gnu.ai.mit.edu>
* string/argz.h (__argz_replace, argz_replace): New declarations.
* string/argz-replace.c: New file.
1997-06-11 Miles Bader <miles@gnu.ai.mit.edu>
* manual/argp.texi (Argp Special Keys): Document ARGP_KEY_ARGS.
1997-06-16 23:08 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* Makeconfig (libc-map): Remove definition.
* Makerules (libc-map): Define it here, using the full name.
(load-map-file): Remove case for empty $(..).
($(common-objpfx)libc.so): Revert last change.
1997-06-17 22:18 Mark Kettenis <kettenis@phys.uva.nl>
* login/programs/utmpd.c (main): Improve signal handling.
* login/programs/request.c (do_setutent, do_updwtmp):
Allow arbitrary length filenames.
* login/programs/utmpd.h (setutent_request, updwtmp_request):
Get rid of fixed length file field.
* login/utmp_daemon.c (do_setutent, do_updwtmp):
Allow arbitrary length filenames.
* login/programs/request.c (do_pututline):
Don't fail if connection->position is -1 on entry.
1997-06-15 16:32 Mark Kettenis <kettenis@phys.uva.nl>
* login/utmp_file.c (updwtmp_file): Use the same method for
appending an entry as in pututline_file.
1997-06-11 18:59 Mark Kettenis <kettenis@phys.uva.nl>
* login/utmp_file.c (getutent_r_file):
Use read lock instead of write lock.
(getutline_r_file, internal_getut_r): Lock utmp file.
(updwtmp_file): Use fcntl to lock file instead of flock.
1997-06-18 00:11 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* stdio-common/vfscanf.c (inchar, ungetc): Don't count EOF as
character read in/put back.
* stdio-common/tstscanf.c: Add test case for this.
1997-06-17 22:17 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* libc.map: Add more symbols.
1997-06-18 12:01 Ulrich Drepper <drepper@cygnus.com>
* manual/Makefile (install): Make sure `dir' file exists if we use
install-info.
1997-06-17 19:32 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* manual/Makefile (info): Depend on dir-add.info.
1997-06-17 17:19 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* elf/ldd.bash.in: Fix spacing in message.
1997-06-17 14:28 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* elf/do-rel.h (elf_dynamic_do_rel): Always use version
information if available.
1997-06-17 11:34 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sunrpc/Makefile ($(rpcsvc:%.x=$(objpfx)rpcsvc/%.h)): Make
command non-empty to force make to recheck modification time.
($(rpcsvc:%.x=$(objpfx)x%.c)): Likewise.
1997-06-17 00:26 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/stub/e_acoshl.c: Set errno to ENOSYS.
* sysdeps/stub/e_acosl.c: Likewise.
* sysdeps/stub/e_asinl.c: Likewise.
* sysdeps/stub/e_atan2l.c: Likewise.
* sysdeps/stub/e_expl.c: Likewise.
* sysdeps/stub/e_fmodl.c: Likewise.
* sysdeps/stub/e_j0l.c: Likewise.
* sysdeps/stub/e_j1l.c: Likewise.
* sysdeps/stub/e_jnl.c: Likewise.
* sysdeps/stub/e_lgammal_r.c: Likewise.
* sysdeps/stub/e_log10l.c: Likewise.
* sysdeps/stub/e_logl.c: Likewise.
* sysdeps/stub/e_powl.c: Likewise.
* sysdeps/stub/e_rem_pio2l.c: Likewise.
* sysdeps/stub/e_sqrtl.c: Likewise.
* sysdeps/stub/k_cosl.c: Likewise.
* sysdeps/stub/k_rem_pio2l.c: Likewise.
* sysdeps/stub/k_sinl.c: Likewise.
* sysdeps/stub/k_tanl.c: Likewise.
* sysdeps/stub/s_atanl.c: Likewise.
* sysdeps/stub/s_erfl.c: Likewise.
* sysdeps/stub/s_exp2.c: Likewise.
* sysdeps/stub/s_exp2f.c: Likewise.
* sysdeps/stub/s_exp2l.c: Likewise.
* sysdeps/stub/s_expm1l.c: Likewise.
* sysdeps/stub/s_log1pl.c: Likewise.
* sysdeps/stub/s_log2l.c: Likewise.
1997-06-18 11:46 Ulrich Drepper <drepper@cygnus.com>
* sysdeps/libm-ieee754/w_gamma.c: If _LIB_VERSION is _SVID_ compute
result as before last change.
* sysdeps/libm-ieee754/w_gammaf.c: Likewise.
* sysdeps/libm-ieee754/w_gammal.c: Likewise.
1997-06-16 23:37 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/libm-ieee754/s_remquo.c: Fix off-by-one when computing
quotient.
* sysdeps/libm-ieee754/s_remquof.c: Likewise.
* sysdeps/libm-ieee754/s_remquol.c: Likewise.
* sysdeps/m68k/fpu/s_remquo.c: Remove FIXME and special case for
quotient.
* sysdeps/libm-ieee754/w_gamma.c: Add missing call to exp
function. Don't use global signgam.
* sysdeps/libm-ieee754/w_gammaf.c: Likewise.
* sysdeps/libm-ieee754/w_gammal.c: Likewise.
* math/Makefile (libm-calls): Remove w_gamma_r.
* sysdeps/libm-ieee754/w_gamma_r.c: Remove file.
* sysdeps/libm-ieee754/w_gammaf_r.c: Remove file.
* sysdeps/libm-ieee754/w_gammal_r.c: Remove file.
* math/libm-test.c (atanh_test): Declare x only if needed.
(signbit_test): Fix typo.
(gamma_test): Check whether function is implemented. Add
epsilons.
(lgamma_test): Likewise.
(fmod_test): Add epsilons.
(exp2_test): Use right function for existence test.
1997-06-07 09:20 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/powerpc/Dist: Add fenv_const.c.
* sysdeps/unix/sysv/linux/Dist: Add net/if_slip.h.
* sysdeps/unix/sysv/linux/powerpc/Dist: Add init-first.h and
syscall.h.
* sysdeps/unix/sysv/linux/sparc/Dist: Add init-first.h.
* string/Makefile (distribute): Add tst-svc.expect.
* nis/Makefile (distribute): Add nis_intern.h and Banner.
* elf/Makefile (distribute): Add dl-hash.h.
* Rules (subdir_echo-distinfo): Add headers from $(distribute).
* login/Makefile (others): Add utmpdump.
* sysdeps/unix/sysv/linux/Makefile (sysdep_headers): Add
net/if_slip.h.
* manual/Makefile (dir-add.texi): Also look in indirectly included
files.
1997-06-16 23:15 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/unix/sysv/linux/configure.in: Remove commands to
generate stdio_lim.h.
* sysdeps/unix/sysv/linux/mk-stdiolim.c: Remove.
* sysdeps/unix/sysv/linux/Makefile: Add rules to generate
stdio_lim.h here.
(common-generated): Add generated files.
(inhibit-stdio_lim): Define.
* sysdeps/posix/Makefile [$(inhibit-stdio_lim)=yes]: Disable rules
to generate stdio_lim.h.
* sysdeps/unix/sysv/linux/Makefile: Suppress inclusion of
dependecy files if no_deps is set.
($(objpfx)syscall-%.d): Add header file as target to dependency
generation.
1997-06-14 19:19 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/unix/sysv/linux/arm/Dist: New file.
1997-06-14 17:59 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* extra-lib.mk (others): Depend on versioned shared library, not
the unversioned one.
* Makerules (build-shlib): Don't make the version link here.
($(common-objpfx)libc.so$(libc.so-version)): New rule for libc
version link.
1997-06-16 03:07 Ulrich Drepper <drepper@cygnus.com>
* sysdeps/libm-ieee754/k_standard.c: Undo change of Tue Aug 6
01:13:56 1996.
* argp/argp-help.c (argp_args_usage): Supply correct argp to
filter_doc.
* argp/argp-help.c (hol_add_cluster): Initialize CL->depth.
* argp/argp-help.c (_help): Supply STATE to argp_args_usage.
* argp/argp.h (ARGP_KEY_HELP_ARGS_DOC): New macro.
* argp/argp-fmtstream.c: Add casts to prevent warnings.
* argp/argp.h (OPTION_NO_USAGE): New macro.
* argp/argp-help.c (usage_long_opt, usage_argful_short_opt,
* argp/argp-fmtstream.c (__argp_fmtstream_update): Account for case
* argp/argp-help.c <stddef.h>: New include.
* argp/argp.h (argp_state_help, __argp_state_help, argp_usage,
* argp/argp.h (argp_program_bug_address): Make const.
* argp/argp-parse.c (argp_default_parser): Set STATE->name for OPT_PROGNAME.
* argp/argp-help.c (__argp_error, __argp_failure, __argp_state_help):
* argp/argp-parse.c (parser_init): Set PARSER->state.flags.
1997-06-19 19:11:22 +00:00
|
|
|
|
2013-05-08 11:58:18 +00:00
|
|
|
if (x >= 172.0)
|
|
|
|
{
|
|
|
|
/* Overflow. */
|
|
|
|
*signgamp = 0;
|
2015-09-23 18:14:57 +00:00
|
|
|
ret = math_narrow_eval (DBL_MAX * DBL_MAX);
|
Improve tgamma accuracy (bug 18613).
In non-default rounding modes, tgamma can be slightly less accurate
than permitted by glibc's accuracy goals.
Part of the problem is error accumulation, addressed in this patch by
setting round-to-nearest for internal computations. However, there
was also a bug in the code dealing with computing pow (x + n, x + n)
where x + n is not exactly representable, providing another source of
error even in round-to-nearest mode; it was necessary to address both
bugs to get errors for all testcases within glibc's accuracy goals.
Given this second fix, accuracy in round-to-nearest mode is also
improved (hence regeneration of ulps for tgamma should be from scratch
- truncate libm-test-ulps or at least remove existing tgamma entries -
so that the expected ulps can be reduced).
Some additional complications also arose. Certain tgamma tests should
strictly, according to IEEE semantics, overflow or not depending on
the rounding mode; this is beyond the scope of glibc's accuracy goals
for any function without exactly-determined results, but
gen-auto-libm-tests doesn't handle being lax there as it does for
underflow. (libm-test.inc also doesn't handle being lax about whether
the result in cases very close to the overflow threshold is infinity
or a finite value close to overflow, but that doesn't cause problems
in this case though I've seen it cause problems with random test
generation for some functions.) Thus, spurious-overflow markings,
with a comment, are added to auto-libm-test-in (no bug in Bugzilla
because the issue is with the testsuite, not a user-visible bug in
glibc). And on x86, after the patch I saw ERANGE issues as previously
reported by Carlos (see my commentary in
<https://sourceware.org/ml/libc-alpha/2015-01/msg00485.html>), which
needed addressing by ensuring excess range and precision were
eliminated at various points if FLT_EVAL_METHOD != 0.
I also noticed and fixed a cosmetic issue where 1.0f was used in long
double functions and should have been 1.0L.
This completes the move of all functions to testing in all rounding
modes with ALL_RM_TEST, so gen-libm-have-vector-test.sh is updated to
remove the workaround for some functions not using ALL_RM_TEST.
Tested for x86_64, x86, mips64 and powerpc.
[BZ #18613]
* sysdeps/ieee754/dbl-64/e_gamma_r.c (gamma_positive): Take log of
X_ADJ not X when adjusting exponent.
(__ieee754_gamma_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed.
* sysdeps/ieee754/flt-32/e_gammaf_r.c (gammaf_positive): Take log
of X_ADJ not X when adjusting exponent.
(__ieee754_gammaf_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed.
* sysdeps/ieee754/ldbl-128/e_gammal_r.c (gammal_positive): Take
log of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* sysdeps/ieee754/ldbl-128ibm/e_gammal_r.c (gammal_positive): Take
log of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* sysdeps/ieee754/ldbl-96/e_gammal_r.c (gammal_positive): Take log
of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* math/libm-test.inc (tgamma_test_data): Remove one test. Moved
to auto-libm-test-in.
(tgamma_test): Use ALL_RM_TEST.
* math/auto-libm-test-in: Add one test of tgamma. Mark some other
tests of tgamma with spurious-overflow.
* math/auto-libm-test-out: Regenerated.
* math/gen-libm-have-vector-test.sh: Do not check for START.
* sysdeps/i386/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Likewise.
2015-06-29 23:29:35 +00:00
|
|
|
return ret;
|
2013-05-08 11:58:18 +00:00
|
|
|
}
|
Improve tgamma accuracy (bug 18613).
In non-default rounding modes, tgamma can be slightly less accurate
than permitted by glibc's accuracy goals.
Part of the problem is error accumulation, addressed in this patch by
setting round-to-nearest for internal computations. However, there
was also a bug in the code dealing with computing pow (x + n, x + n)
where x + n is not exactly representable, providing another source of
error even in round-to-nearest mode; it was necessary to address both
bugs to get errors for all testcases within glibc's accuracy goals.
Given this second fix, accuracy in round-to-nearest mode is also
improved (hence regeneration of ulps for tgamma should be from scratch
- truncate libm-test-ulps or at least remove existing tgamma entries -
so that the expected ulps can be reduced).
Some additional complications also arose. Certain tgamma tests should
strictly, according to IEEE semantics, overflow or not depending on
the rounding mode; this is beyond the scope of glibc's accuracy goals
for any function without exactly-determined results, but
gen-auto-libm-tests doesn't handle being lax there as it does for
underflow. (libm-test.inc also doesn't handle being lax about whether
the result in cases very close to the overflow threshold is infinity
or a finite value close to overflow, but that doesn't cause problems
in this case though I've seen it cause problems with random test
generation for some functions.) Thus, spurious-overflow markings,
with a comment, are added to auto-libm-test-in (no bug in Bugzilla
because the issue is with the testsuite, not a user-visible bug in
glibc). And on x86, after the patch I saw ERANGE issues as previously
reported by Carlos (see my commentary in
<https://sourceware.org/ml/libc-alpha/2015-01/msg00485.html>), which
needed addressing by ensuring excess range and precision were
eliminated at various points if FLT_EVAL_METHOD != 0.
I also noticed and fixed a cosmetic issue where 1.0f was used in long
double functions and should have been 1.0L.
This completes the move of all functions to testing in all rounding
modes with ALL_RM_TEST, so gen-libm-have-vector-test.sh is updated to
remove the workaround for some functions not using ALL_RM_TEST.
Tested for x86_64, x86, mips64 and powerpc.
[BZ #18613]
* sysdeps/ieee754/dbl-64/e_gamma_r.c (gamma_positive): Take log of
X_ADJ not X when adjusting exponent.
(__ieee754_gamma_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed.
* sysdeps/ieee754/flt-32/e_gammaf_r.c (gammaf_positive): Take log
of X_ADJ not X when adjusting exponent.
(__ieee754_gammaf_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed.
* sysdeps/ieee754/ldbl-128/e_gammal_r.c (gammal_positive): Take
log of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* sysdeps/ieee754/ldbl-128ibm/e_gammal_r.c (gammal_positive): Take
log of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* sysdeps/ieee754/ldbl-96/e_gammal_r.c (gammal_positive): Take log
of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* math/libm-test.inc (tgamma_test_data): Remove one test. Moved
to auto-libm-test-in.
(tgamma_test): Use ALL_RM_TEST.
* math/auto-libm-test-in: Add one test of tgamma. Mark some other
tests of tgamma with spurious-overflow.
* math/auto-libm-test-out: Regenerated.
* math/gen-libm-have-vector-test.sh: Do not check for START.
* sysdeps/i386/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Likewise.
2015-06-29 23:29:35 +00:00
|
|
|
else
|
2013-05-08 11:58:18 +00:00
|
|
|
{
|
Improve tgamma accuracy (bug 18613).
In non-default rounding modes, tgamma can be slightly less accurate
than permitted by glibc's accuracy goals.
Part of the problem is error accumulation, addressed in this patch by
setting round-to-nearest for internal computations. However, there
was also a bug in the code dealing with computing pow (x + n, x + n)
where x + n is not exactly representable, providing another source of
error even in round-to-nearest mode; it was necessary to address both
bugs to get errors for all testcases within glibc's accuracy goals.
Given this second fix, accuracy in round-to-nearest mode is also
improved (hence regeneration of ulps for tgamma should be from scratch
- truncate libm-test-ulps or at least remove existing tgamma entries -
so that the expected ulps can be reduced).
Some additional complications also arose. Certain tgamma tests should
strictly, according to IEEE semantics, overflow or not depending on
the rounding mode; this is beyond the scope of glibc's accuracy goals
for any function without exactly-determined results, but
gen-auto-libm-tests doesn't handle being lax there as it does for
underflow. (libm-test.inc also doesn't handle being lax about whether
the result in cases very close to the overflow threshold is infinity
or a finite value close to overflow, but that doesn't cause problems
in this case though I've seen it cause problems with random test
generation for some functions.) Thus, spurious-overflow markings,
with a comment, are added to auto-libm-test-in (no bug in Bugzilla
because the issue is with the testsuite, not a user-visible bug in
glibc). And on x86, after the patch I saw ERANGE issues as previously
reported by Carlos (see my commentary in
<https://sourceware.org/ml/libc-alpha/2015-01/msg00485.html>), which
needed addressing by ensuring excess range and precision were
eliminated at various points if FLT_EVAL_METHOD != 0.
I also noticed and fixed a cosmetic issue where 1.0f was used in long
double functions and should have been 1.0L.
This completes the move of all functions to testing in all rounding
modes with ALL_RM_TEST, so gen-libm-have-vector-test.sh is updated to
remove the workaround for some functions not using ALL_RM_TEST.
Tested for x86_64, x86, mips64 and powerpc.
[BZ #18613]
* sysdeps/ieee754/dbl-64/e_gamma_r.c (gamma_positive): Take log of
X_ADJ not X when adjusting exponent.
(__ieee754_gamma_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed.
* sysdeps/ieee754/flt-32/e_gammaf_r.c (gammaf_positive): Take log
of X_ADJ not X when adjusting exponent.
(__ieee754_gammaf_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed.
* sysdeps/ieee754/ldbl-128/e_gammal_r.c (gammal_positive): Take
log of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* sysdeps/ieee754/ldbl-128ibm/e_gammal_r.c (gammal_positive): Take
log of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* sysdeps/ieee754/ldbl-96/e_gammal_r.c (gammal_positive): Take log
of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* math/libm-test.inc (tgamma_test_data): Remove one test. Moved
to auto-libm-test-in.
(tgamma_test): Use ALL_RM_TEST.
* math/auto-libm-test-in: Add one test of tgamma. Mark some other
tests of tgamma with spurious-overflow.
* math/auto-libm-test-out: Regenerated.
* math/gen-libm-have-vector-test.sh: Do not check for START.
* sysdeps/i386/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Likewise.
2015-06-29 23:29:35 +00:00
|
|
|
SET_RESTORE_ROUND (FE_TONEAREST);
|
|
|
|
if (x > 0.0)
|
|
|
|
{
|
|
|
|
*signgamp = 0;
|
|
|
|
int exp2_adj;
|
|
|
|
double tret = gamma_positive (x, &exp2_adj);
|
|
|
|
ret = __scalbn (tret, exp2_adj);
|
|
|
|
}
|
|
|
|
else if (x >= -DBL_EPSILON / 4.0)
|
|
|
|
{
|
|
|
|
*signgamp = 0;
|
|
|
|
ret = 1.0 / x;
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
double tx = __trunc (x);
|
|
|
|
*signgamp = (tx == 2.0 * __trunc (tx / 2.0)) ? -1 : 1;
|
|
|
|
if (x <= -184.0)
|
|
|
|
/* Underflow. */
|
|
|
|
ret = DBL_MIN * DBL_MIN;
|
|
|
|
else
|
|
|
|
{
|
|
|
|
double frac = tx - x;
|
|
|
|
if (frac > 0.5)
|
|
|
|
frac = 1.0 - frac;
|
|
|
|
double sinpix = (frac <= 0.25
|
|
|
|
? __sin (M_PI * frac)
|
|
|
|
: __cos (M_PI * (0.5 - frac)));
|
|
|
|
int exp2_adj;
|
|
|
|
double tret = M_PI / (-x * sinpix
|
|
|
|
* gamma_positive (-x, &exp2_adj));
|
|
|
|
ret = __scalbn (tret, -exp2_adj);
|
2015-09-23 22:42:30 +00:00
|
|
|
math_check_force_underflow_nonneg (ret);
|
Improve tgamma accuracy (bug 18613).
In non-default rounding modes, tgamma can be slightly less accurate
than permitted by glibc's accuracy goals.
Part of the problem is error accumulation, addressed in this patch by
setting round-to-nearest for internal computations. However, there
was also a bug in the code dealing with computing pow (x + n, x + n)
where x + n is not exactly representable, providing another source of
error even in round-to-nearest mode; it was necessary to address both
bugs to get errors for all testcases within glibc's accuracy goals.
Given this second fix, accuracy in round-to-nearest mode is also
improved (hence regeneration of ulps for tgamma should be from scratch
- truncate libm-test-ulps or at least remove existing tgamma entries -
so that the expected ulps can be reduced).
Some additional complications also arose. Certain tgamma tests should
strictly, according to IEEE semantics, overflow or not depending on
the rounding mode; this is beyond the scope of glibc's accuracy goals
for any function without exactly-determined results, but
gen-auto-libm-tests doesn't handle being lax there as it does for
underflow. (libm-test.inc also doesn't handle being lax about whether
the result in cases very close to the overflow threshold is infinity
or a finite value close to overflow, but that doesn't cause problems
in this case though I've seen it cause problems with random test
generation for some functions.) Thus, spurious-overflow markings,
with a comment, are added to auto-libm-test-in (no bug in Bugzilla
because the issue is with the testsuite, not a user-visible bug in
glibc). And on x86, after the patch I saw ERANGE issues as previously
reported by Carlos (see my commentary in
<https://sourceware.org/ml/libc-alpha/2015-01/msg00485.html>), which
needed addressing by ensuring excess range and precision were
eliminated at various points if FLT_EVAL_METHOD != 0.
I also noticed and fixed a cosmetic issue where 1.0f was used in long
double functions and should have been 1.0L.
This completes the move of all functions to testing in all rounding
modes with ALL_RM_TEST, so gen-libm-have-vector-test.sh is updated to
remove the workaround for some functions not using ALL_RM_TEST.
Tested for x86_64, x86, mips64 and powerpc.
[BZ #18613]
* sysdeps/ieee754/dbl-64/e_gamma_r.c (gamma_positive): Take log of
X_ADJ not X when adjusting exponent.
(__ieee754_gamma_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed.
* sysdeps/ieee754/flt-32/e_gammaf_r.c (gammaf_positive): Take log
of X_ADJ not X when adjusting exponent.
(__ieee754_gammaf_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed.
* sysdeps/ieee754/ldbl-128/e_gammal_r.c (gammal_positive): Take
log of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* sysdeps/ieee754/ldbl-128ibm/e_gammal_r.c (gammal_positive): Take
log of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* sysdeps/ieee754/ldbl-96/e_gammal_r.c (gammal_positive): Take log
of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* math/libm-test.inc (tgamma_test_data): Remove one test. Moved
to auto-libm-test-in.
(tgamma_test): Use ALL_RM_TEST.
* math/auto-libm-test-in: Add one test of tgamma. Mark some other
tests of tgamma with spurious-overflow.
* math/auto-libm-test-out: Regenerated.
* math/gen-libm-have-vector-test.sh: Do not check for START.
* sysdeps/i386/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Likewise.
2015-06-29 23:29:35 +00:00
|
|
|
}
|
|
|
|
}
|
2015-09-23 18:14:57 +00:00
|
|
|
ret = math_narrow_eval (ret);
|
2013-05-08 11:58:18 +00:00
|
|
|
}
|
Improve tgamma accuracy (bug 18613).
In non-default rounding modes, tgamma can be slightly less accurate
than permitted by glibc's accuracy goals.
Part of the problem is error accumulation, addressed in this patch by
setting round-to-nearest for internal computations. However, there
was also a bug in the code dealing with computing pow (x + n, x + n)
where x + n is not exactly representable, providing another source of
error even in round-to-nearest mode; it was necessary to address both
bugs to get errors for all testcases within glibc's accuracy goals.
Given this second fix, accuracy in round-to-nearest mode is also
improved (hence regeneration of ulps for tgamma should be from scratch
- truncate libm-test-ulps or at least remove existing tgamma entries -
so that the expected ulps can be reduced).
Some additional complications also arose. Certain tgamma tests should
strictly, according to IEEE semantics, overflow or not depending on
the rounding mode; this is beyond the scope of glibc's accuracy goals
for any function without exactly-determined results, but
gen-auto-libm-tests doesn't handle being lax there as it does for
underflow. (libm-test.inc also doesn't handle being lax about whether
the result in cases very close to the overflow threshold is infinity
or a finite value close to overflow, but that doesn't cause problems
in this case though I've seen it cause problems with random test
generation for some functions.) Thus, spurious-overflow markings,
with a comment, are added to auto-libm-test-in (no bug in Bugzilla
because the issue is with the testsuite, not a user-visible bug in
glibc). And on x86, after the patch I saw ERANGE issues as previously
reported by Carlos (see my commentary in
<https://sourceware.org/ml/libc-alpha/2015-01/msg00485.html>), which
needed addressing by ensuring excess range and precision were
eliminated at various points if FLT_EVAL_METHOD != 0.
I also noticed and fixed a cosmetic issue where 1.0f was used in long
double functions and should have been 1.0L.
This completes the move of all functions to testing in all rounding
modes with ALL_RM_TEST, so gen-libm-have-vector-test.sh is updated to
remove the workaround for some functions not using ALL_RM_TEST.
Tested for x86_64, x86, mips64 and powerpc.
[BZ #18613]
* sysdeps/ieee754/dbl-64/e_gamma_r.c (gamma_positive): Take log of
X_ADJ not X when adjusting exponent.
(__ieee754_gamma_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed.
* sysdeps/ieee754/flt-32/e_gammaf_r.c (gammaf_positive): Take log
of X_ADJ not X when adjusting exponent.
(__ieee754_gammaf_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed.
* sysdeps/ieee754/ldbl-128/e_gammal_r.c (gammal_positive): Take
log of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* sysdeps/ieee754/ldbl-128ibm/e_gammal_r.c (gammal_positive): Take
log of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* sysdeps/ieee754/ldbl-96/e_gammal_r.c (gammal_positive): Take log
of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* math/libm-test.inc (tgamma_test_data): Remove one test. Moved
to auto-libm-test-in.
(tgamma_test): Use ALL_RM_TEST.
* math/auto-libm-test-in: Add one test of tgamma. Mark some other
tests of tgamma with spurious-overflow.
* math/auto-libm-test-out: Regenerated.
* math/gen-libm-have-vector-test.sh: Do not check for START.
* sysdeps/i386/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Likewise.
2015-06-29 23:29:35 +00:00
|
|
|
if (isinf (ret) && x != 0)
|
2013-05-08 11:58:18 +00:00
|
|
|
{
|
Improve tgamma accuracy (bug 18613).
In non-default rounding modes, tgamma can be slightly less accurate
than permitted by glibc's accuracy goals.
Part of the problem is error accumulation, addressed in this patch by
setting round-to-nearest for internal computations. However, there
was also a bug in the code dealing with computing pow (x + n, x + n)
where x + n is not exactly representable, providing another source of
error even in round-to-nearest mode; it was necessary to address both
bugs to get errors for all testcases within glibc's accuracy goals.
Given this second fix, accuracy in round-to-nearest mode is also
improved (hence regeneration of ulps for tgamma should be from scratch
- truncate libm-test-ulps or at least remove existing tgamma entries -
so that the expected ulps can be reduced).
Some additional complications also arose. Certain tgamma tests should
strictly, according to IEEE semantics, overflow or not depending on
the rounding mode; this is beyond the scope of glibc's accuracy goals
for any function without exactly-determined results, but
gen-auto-libm-tests doesn't handle being lax there as it does for
underflow. (libm-test.inc also doesn't handle being lax about whether
the result in cases very close to the overflow threshold is infinity
or a finite value close to overflow, but that doesn't cause problems
in this case though I've seen it cause problems with random test
generation for some functions.) Thus, spurious-overflow markings,
with a comment, are added to auto-libm-test-in (no bug in Bugzilla
because the issue is with the testsuite, not a user-visible bug in
glibc). And on x86, after the patch I saw ERANGE issues as previously
reported by Carlos (see my commentary in
<https://sourceware.org/ml/libc-alpha/2015-01/msg00485.html>), which
needed addressing by ensuring excess range and precision were
eliminated at various points if FLT_EVAL_METHOD != 0.
I also noticed and fixed a cosmetic issue where 1.0f was used in long
double functions and should have been 1.0L.
This completes the move of all functions to testing in all rounding
modes with ALL_RM_TEST, so gen-libm-have-vector-test.sh is updated to
remove the workaround for some functions not using ALL_RM_TEST.
Tested for x86_64, x86, mips64 and powerpc.
[BZ #18613]
* sysdeps/ieee754/dbl-64/e_gamma_r.c (gamma_positive): Take log of
X_ADJ not X when adjusting exponent.
(__ieee754_gamma_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed.
* sysdeps/ieee754/flt-32/e_gammaf_r.c (gammaf_positive): Take log
of X_ADJ not X when adjusting exponent.
(__ieee754_gammaf_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed.
* sysdeps/ieee754/ldbl-128/e_gammal_r.c (gammal_positive): Take
log of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* sysdeps/ieee754/ldbl-128ibm/e_gammal_r.c (gammal_positive): Take
log of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* sysdeps/ieee754/ldbl-96/e_gammal_r.c (gammal_positive): Take log
of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* math/libm-test.inc (tgamma_test_data): Remove one test. Moved
to auto-libm-test-in.
(tgamma_test): Use ALL_RM_TEST.
* math/auto-libm-test-in: Add one test of tgamma. Mark some other
tests of tgamma with spurious-overflow.
* math/auto-libm-test-out: Regenerated.
* math/gen-libm-have-vector-test.sh: Do not check for START.
* sysdeps/i386/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Likewise.
2015-06-29 23:29:35 +00:00
|
|
|
if (*signgamp < 0)
|
|
|
|
{
|
2015-09-23 18:14:57 +00:00
|
|
|
ret = math_narrow_eval (-__copysign (DBL_MAX, ret) * DBL_MAX);
|
Improve tgamma accuracy (bug 18613).
In non-default rounding modes, tgamma can be slightly less accurate
than permitted by glibc's accuracy goals.
Part of the problem is error accumulation, addressed in this patch by
setting round-to-nearest for internal computations. However, there
was also a bug in the code dealing with computing pow (x + n, x + n)
where x + n is not exactly representable, providing another source of
error even in round-to-nearest mode; it was necessary to address both
bugs to get errors for all testcases within glibc's accuracy goals.
Given this second fix, accuracy in round-to-nearest mode is also
improved (hence regeneration of ulps for tgamma should be from scratch
- truncate libm-test-ulps or at least remove existing tgamma entries -
so that the expected ulps can be reduced).
Some additional complications also arose. Certain tgamma tests should
strictly, according to IEEE semantics, overflow or not depending on
the rounding mode; this is beyond the scope of glibc's accuracy goals
for any function without exactly-determined results, but
gen-auto-libm-tests doesn't handle being lax there as it does for
underflow. (libm-test.inc also doesn't handle being lax about whether
the result in cases very close to the overflow threshold is infinity
or a finite value close to overflow, but that doesn't cause problems
in this case though I've seen it cause problems with random test
generation for some functions.) Thus, spurious-overflow markings,
with a comment, are added to auto-libm-test-in (no bug in Bugzilla
because the issue is with the testsuite, not a user-visible bug in
glibc). And on x86, after the patch I saw ERANGE issues as previously
reported by Carlos (see my commentary in
<https://sourceware.org/ml/libc-alpha/2015-01/msg00485.html>), which
needed addressing by ensuring excess range and precision were
eliminated at various points if FLT_EVAL_METHOD != 0.
I also noticed and fixed a cosmetic issue where 1.0f was used in long
double functions and should have been 1.0L.
This completes the move of all functions to testing in all rounding
modes with ALL_RM_TEST, so gen-libm-have-vector-test.sh is updated to
remove the workaround for some functions not using ALL_RM_TEST.
Tested for x86_64, x86, mips64 and powerpc.
[BZ #18613]
* sysdeps/ieee754/dbl-64/e_gamma_r.c (gamma_positive): Take log of
X_ADJ not X when adjusting exponent.
(__ieee754_gamma_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed.
* sysdeps/ieee754/flt-32/e_gammaf_r.c (gammaf_positive): Take log
of X_ADJ not X when adjusting exponent.
(__ieee754_gammaf_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed.
* sysdeps/ieee754/ldbl-128/e_gammal_r.c (gammal_positive): Take
log of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* sysdeps/ieee754/ldbl-128ibm/e_gammal_r.c (gammal_positive): Take
log of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* sysdeps/ieee754/ldbl-96/e_gammal_r.c (gammal_positive): Take log
of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* math/libm-test.inc (tgamma_test_data): Remove one test. Moved
to auto-libm-test-in.
(tgamma_test): Use ALL_RM_TEST.
* math/auto-libm-test-in: Add one test of tgamma. Mark some other
tests of tgamma with spurious-overflow.
* math/auto-libm-test-out: Regenerated.
* math/gen-libm-have-vector-test.sh: Do not check for START.
* sysdeps/i386/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Likewise.
2015-06-29 23:29:35 +00:00
|
|
|
ret = -ret;
|
|
|
|
}
|
|
|
|
else
|
2015-09-23 18:14:57 +00:00
|
|
|
ret = math_narrow_eval (__copysign (DBL_MAX, ret) * DBL_MAX);
|
Improve tgamma accuracy (bug 18613).
In non-default rounding modes, tgamma can be slightly less accurate
than permitted by glibc's accuracy goals.
Part of the problem is error accumulation, addressed in this patch by
setting round-to-nearest for internal computations. However, there
was also a bug in the code dealing with computing pow (x + n, x + n)
where x + n is not exactly representable, providing another source of
error even in round-to-nearest mode; it was necessary to address both
bugs to get errors for all testcases within glibc's accuracy goals.
Given this second fix, accuracy in round-to-nearest mode is also
improved (hence regeneration of ulps for tgamma should be from scratch
- truncate libm-test-ulps or at least remove existing tgamma entries -
so that the expected ulps can be reduced).
Some additional complications also arose. Certain tgamma tests should
strictly, according to IEEE semantics, overflow or not depending on
the rounding mode; this is beyond the scope of glibc's accuracy goals
for any function without exactly-determined results, but
gen-auto-libm-tests doesn't handle being lax there as it does for
underflow. (libm-test.inc also doesn't handle being lax about whether
the result in cases very close to the overflow threshold is infinity
or a finite value close to overflow, but that doesn't cause problems
in this case though I've seen it cause problems with random test
generation for some functions.) Thus, spurious-overflow markings,
with a comment, are added to auto-libm-test-in (no bug in Bugzilla
because the issue is with the testsuite, not a user-visible bug in
glibc). And on x86, after the patch I saw ERANGE issues as previously
reported by Carlos (see my commentary in
<https://sourceware.org/ml/libc-alpha/2015-01/msg00485.html>), which
needed addressing by ensuring excess range and precision were
eliminated at various points if FLT_EVAL_METHOD != 0.
I also noticed and fixed a cosmetic issue where 1.0f was used in long
double functions and should have been 1.0L.
This completes the move of all functions to testing in all rounding
modes with ALL_RM_TEST, so gen-libm-have-vector-test.sh is updated to
remove the workaround for some functions not using ALL_RM_TEST.
Tested for x86_64, x86, mips64 and powerpc.
[BZ #18613]
* sysdeps/ieee754/dbl-64/e_gamma_r.c (gamma_positive): Take log of
X_ADJ not X when adjusting exponent.
(__ieee754_gamma_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed.
* sysdeps/ieee754/flt-32/e_gammaf_r.c (gammaf_positive): Take log
of X_ADJ not X when adjusting exponent.
(__ieee754_gammaf_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed.
* sysdeps/ieee754/ldbl-128/e_gammal_r.c (gammal_positive): Take
log of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* sysdeps/ieee754/ldbl-128ibm/e_gammal_r.c (gammal_positive): Take
log of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* sysdeps/ieee754/ldbl-96/e_gammal_r.c (gammal_positive): Take log
of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* math/libm-test.inc (tgamma_test_data): Remove one test. Moved
to auto-libm-test-in.
(tgamma_test): Use ALL_RM_TEST.
* math/auto-libm-test-in: Add one test of tgamma. Mark some other
tests of tgamma with spurious-overflow.
* math/auto-libm-test-out: Regenerated.
* math/gen-libm-have-vector-test.sh: Do not check for START.
* sysdeps/i386/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Likewise.
2015-06-29 23:29:35 +00:00
|
|
|
return ret;
|
2013-05-08 11:58:18 +00:00
|
|
|
}
|
Improve tgamma accuracy (bug 18613).
In non-default rounding modes, tgamma can be slightly less accurate
than permitted by glibc's accuracy goals.
Part of the problem is error accumulation, addressed in this patch by
setting round-to-nearest for internal computations. However, there
was also a bug in the code dealing with computing pow (x + n, x + n)
where x + n is not exactly representable, providing another source of
error even in round-to-nearest mode; it was necessary to address both
bugs to get errors for all testcases within glibc's accuracy goals.
Given this second fix, accuracy in round-to-nearest mode is also
improved (hence regeneration of ulps for tgamma should be from scratch
- truncate libm-test-ulps or at least remove existing tgamma entries -
so that the expected ulps can be reduced).
Some additional complications also arose. Certain tgamma tests should
strictly, according to IEEE semantics, overflow or not depending on
the rounding mode; this is beyond the scope of glibc's accuracy goals
for any function without exactly-determined results, but
gen-auto-libm-tests doesn't handle being lax there as it does for
underflow. (libm-test.inc also doesn't handle being lax about whether
the result in cases very close to the overflow threshold is infinity
or a finite value close to overflow, but that doesn't cause problems
in this case though I've seen it cause problems with random test
generation for some functions.) Thus, spurious-overflow markings,
with a comment, are added to auto-libm-test-in (no bug in Bugzilla
because the issue is with the testsuite, not a user-visible bug in
glibc). And on x86, after the patch I saw ERANGE issues as previously
reported by Carlos (see my commentary in
<https://sourceware.org/ml/libc-alpha/2015-01/msg00485.html>), which
needed addressing by ensuring excess range and precision were
eliminated at various points if FLT_EVAL_METHOD != 0.
I also noticed and fixed a cosmetic issue where 1.0f was used in long
double functions and should have been 1.0L.
This completes the move of all functions to testing in all rounding
modes with ALL_RM_TEST, so gen-libm-have-vector-test.sh is updated to
remove the workaround for some functions not using ALL_RM_TEST.
Tested for x86_64, x86, mips64 and powerpc.
[BZ #18613]
* sysdeps/ieee754/dbl-64/e_gamma_r.c (gamma_positive): Take log of
X_ADJ not X when adjusting exponent.
(__ieee754_gamma_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed.
* sysdeps/ieee754/flt-32/e_gammaf_r.c (gammaf_positive): Take log
of X_ADJ not X when adjusting exponent.
(__ieee754_gammaf_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed.
* sysdeps/ieee754/ldbl-128/e_gammal_r.c (gammal_positive): Take
log of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* sysdeps/ieee754/ldbl-128ibm/e_gammal_r.c (gammal_positive): Take
log of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* sysdeps/ieee754/ldbl-96/e_gammal_r.c (gammal_positive): Take log
of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* math/libm-test.inc (tgamma_test_data): Remove one test. Moved
to auto-libm-test-in.
(tgamma_test): Use ALL_RM_TEST.
* math/auto-libm-test-in: Add one test of tgamma. Mark some other
tests of tgamma with spurious-overflow.
* math/auto-libm-test-out: Regenerated.
* math/gen-libm-have-vector-test.sh: Do not check for START.
* sysdeps/i386/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Likewise.
2015-06-29 23:29:35 +00:00
|
|
|
else if (ret == 0)
|
2013-05-08 11:58:18 +00:00
|
|
|
{
|
Improve tgamma accuracy (bug 18613).
In non-default rounding modes, tgamma can be slightly less accurate
than permitted by glibc's accuracy goals.
Part of the problem is error accumulation, addressed in this patch by
setting round-to-nearest for internal computations. However, there
was also a bug in the code dealing with computing pow (x + n, x + n)
where x + n is not exactly representable, providing another source of
error even in round-to-nearest mode; it was necessary to address both
bugs to get errors for all testcases within glibc's accuracy goals.
Given this second fix, accuracy in round-to-nearest mode is also
improved (hence regeneration of ulps for tgamma should be from scratch
- truncate libm-test-ulps or at least remove existing tgamma entries -
so that the expected ulps can be reduced).
Some additional complications also arose. Certain tgamma tests should
strictly, according to IEEE semantics, overflow or not depending on
the rounding mode; this is beyond the scope of glibc's accuracy goals
for any function without exactly-determined results, but
gen-auto-libm-tests doesn't handle being lax there as it does for
underflow. (libm-test.inc also doesn't handle being lax about whether
the result in cases very close to the overflow threshold is infinity
or a finite value close to overflow, but that doesn't cause problems
in this case though I've seen it cause problems with random test
generation for some functions.) Thus, spurious-overflow markings,
with a comment, are added to auto-libm-test-in (no bug in Bugzilla
because the issue is with the testsuite, not a user-visible bug in
glibc). And on x86, after the patch I saw ERANGE issues as previously
reported by Carlos (see my commentary in
<https://sourceware.org/ml/libc-alpha/2015-01/msg00485.html>), which
needed addressing by ensuring excess range and precision were
eliminated at various points if FLT_EVAL_METHOD != 0.
I also noticed and fixed a cosmetic issue where 1.0f was used in long
double functions and should have been 1.0L.
This completes the move of all functions to testing in all rounding
modes with ALL_RM_TEST, so gen-libm-have-vector-test.sh is updated to
remove the workaround for some functions not using ALL_RM_TEST.
Tested for x86_64, x86, mips64 and powerpc.
[BZ #18613]
* sysdeps/ieee754/dbl-64/e_gamma_r.c (gamma_positive): Take log of
X_ADJ not X when adjusting exponent.
(__ieee754_gamma_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed.
* sysdeps/ieee754/flt-32/e_gammaf_r.c (gammaf_positive): Take log
of X_ADJ not X when adjusting exponent.
(__ieee754_gammaf_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed.
* sysdeps/ieee754/ldbl-128/e_gammal_r.c (gammal_positive): Take
log of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* sysdeps/ieee754/ldbl-128ibm/e_gammal_r.c (gammal_positive): Take
log of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* sysdeps/ieee754/ldbl-96/e_gammal_r.c (gammal_positive): Take log
of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* math/libm-test.inc (tgamma_test_data): Remove one test. Moved
to auto-libm-test-in.
(tgamma_test): Use ALL_RM_TEST.
* math/auto-libm-test-in: Add one test of tgamma. Mark some other
tests of tgamma with spurious-overflow.
* math/auto-libm-test-out: Regenerated.
* math/gen-libm-have-vector-test.sh: Do not check for START.
* sysdeps/i386/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Likewise.
2015-06-29 23:29:35 +00:00
|
|
|
if (*signgamp < 0)
|
|
|
|
{
|
2015-09-23 18:14:57 +00:00
|
|
|
ret = math_narrow_eval (-__copysign (DBL_MIN, ret) * DBL_MIN);
|
Improve tgamma accuracy (bug 18613).
In non-default rounding modes, tgamma can be slightly less accurate
than permitted by glibc's accuracy goals.
Part of the problem is error accumulation, addressed in this patch by
setting round-to-nearest for internal computations. However, there
was also a bug in the code dealing with computing pow (x + n, x + n)
where x + n is not exactly representable, providing another source of
error even in round-to-nearest mode; it was necessary to address both
bugs to get errors for all testcases within glibc's accuracy goals.
Given this second fix, accuracy in round-to-nearest mode is also
improved (hence regeneration of ulps for tgamma should be from scratch
- truncate libm-test-ulps or at least remove existing tgamma entries -
so that the expected ulps can be reduced).
Some additional complications also arose. Certain tgamma tests should
strictly, according to IEEE semantics, overflow or not depending on
the rounding mode; this is beyond the scope of glibc's accuracy goals
for any function without exactly-determined results, but
gen-auto-libm-tests doesn't handle being lax there as it does for
underflow. (libm-test.inc also doesn't handle being lax about whether
the result in cases very close to the overflow threshold is infinity
or a finite value close to overflow, but that doesn't cause problems
in this case though I've seen it cause problems with random test
generation for some functions.) Thus, spurious-overflow markings,
with a comment, are added to auto-libm-test-in (no bug in Bugzilla
because the issue is with the testsuite, not a user-visible bug in
glibc). And on x86, after the patch I saw ERANGE issues as previously
reported by Carlos (see my commentary in
<https://sourceware.org/ml/libc-alpha/2015-01/msg00485.html>), which
needed addressing by ensuring excess range and precision were
eliminated at various points if FLT_EVAL_METHOD != 0.
I also noticed and fixed a cosmetic issue where 1.0f was used in long
double functions and should have been 1.0L.
This completes the move of all functions to testing in all rounding
modes with ALL_RM_TEST, so gen-libm-have-vector-test.sh is updated to
remove the workaround for some functions not using ALL_RM_TEST.
Tested for x86_64, x86, mips64 and powerpc.
[BZ #18613]
* sysdeps/ieee754/dbl-64/e_gamma_r.c (gamma_positive): Take log of
X_ADJ not X when adjusting exponent.
(__ieee754_gamma_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed.
* sysdeps/ieee754/flt-32/e_gammaf_r.c (gammaf_positive): Take log
of X_ADJ not X when adjusting exponent.
(__ieee754_gammaf_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed.
* sysdeps/ieee754/ldbl-128/e_gammal_r.c (gammal_positive): Take
log of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* sysdeps/ieee754/ldbl-128ibm/e_gammal_r.c (gammal_positive): Take
log of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* sysdeps/ieee754/ldbl-96/e_gammal_r.c (gammal_positive): Take log
of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* math/libm-test.inc (tgamma_test_data): Remove one test. Moved
to auto-libm-test-in.
(tgamma_test): Use ALL_RM_TEST.
* math/auto-libm-test-in: Add one test of tgamma. Mark some other
tests of tgamma with spurious-overflow.
* math/auto-libm-test-out: Regenerated.
* math/gen-libm-have-vector-test.sh: Do not check for START.
* sysdeps/i386/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Likewise.
2015-06-29 23:29:35 +00:00
|
|
|
ret = -ret;
|
|
|
|
}
|
|
|
|
else
|
2015-09-23 18:14:57 +00:00
|
|
|
ret = math_narrow_eval (__copysign (DBL_MIN, ret) * DBL_MIN);
|
Improve tgamma accuracy (bug 18613).
In non-default rounding modes, tgamma can be slightly less accurate
than permitted by glibc's accuracy goals.
Part of the problem is error accumulation, addressed in this patch by
setting round-to-nearest for internal computations. However, there
was also a bug in the code dealing with computing pow (x + n, x + n)
where x + n is not exactly representable, providing another source of
error even in round-to-nearest mode; it was necessary to address both
bugs to get errors for all testcases within glibc's accuracy goals.
Given this second fix, accuracy in round-to-nearest mode is also
improved (hence regeneration of ulps for tgamma should be from scratch
- truncate libm-test-ulps or at least remove existing tgamma entries -
so that the expected ulps can be reduced).
Some additional complications also arose. Certain tgamma tests should
strictly, according to IEEE semantics, overflow or not depending on
the rounding mode; this is beyond the scope of glibc's accuracy goals
for any function without exactly-determined results, but
gen-auto-libm-tests doesn't handle being lax there as it does for
underflow. (libm-test.inc also doesn't handle being lax about whether
the result in cases very close to the overflow threshold is infinity
or a finite value close to overflow, but that doesn't cause problems
in this case though I've seen it cause problems with random test
generation for some functions.) Thus, spurious-overflow markings,
with a comment, are added to auto-libm-test-in (no bug in Bugzilla
because the issue is with the testsuite, not a user-visible bug in
glibc). And on x86, after the patch I saw ERANGE issues as previously
reported by Carlos (see my commentary in
<https://sourceware.org/ml/libc-alpha/2015-01/msg00485.html>), which
needed addressing by ensuring excess range and precision were
eliminated at various points if FLT_EVAL_METHOD != 0.
I also noticed and fixed a cosmetic issue where 1.0f was used in long
double functions and should have been 1.0L.
This completes the move of all functions to testing in all rounding
modes with ALL_RM_TEST, so gen-libm-have-vector-test.sh is updated to
remove the workaround for some functions not using ALL_RM_TEST.
Tested for x86_64, x86, mips64 and powerpc.
[BZ #18613]
* sysdeps/ieee754/dbl-64/e_gamma_r.c (gamma_positive): Take log of
X_ADJ not X when adjusting exponent.
(__ieee754_gamma_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed.
* sysdeps/ieee754/flt-32/e_gammaf_r.c (gammaf_positive): Take log
of X_ADJ not X when adjusting exponent.
(__ieee754_gammaf_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed.
* sysdeps/ieee754/ldbl-128/e_gammal_r.c (gammal_positive): Take
log of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* sysdeps/ieee754/ldbl-128ibm/e_gammal_r.c (gammal_positive): Take
log of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* sysdeps/ieee754/ldbl-96/e_gammal_r.c (gammal_positive): Take log
of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* math/libm-test.inc (tgamma_test_data): Remove one test. Moved
to auto-libm-test-in.
(tgamma_test): Use ALL_RM_TEST.
* math/auto-libm-test-in: Add one test of tgamma. Mark some other
tests of tgamma with spurious-overflow.
* math/auto-libm-test-out: Regenerated.
* math/gen-libm-have-vector-test.sh: Do not check for START.
* sysdeps/i386/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Likewise.
2015-06-29 23:29:35 +00:00
|
|
|
return ret;
|
2013-05-08 11:58:18 +00:00
|
|
|
}
|
Improve tgamma accuracy (bug 18613).
In non-default rounding modes, tgamma can be slightly less accurate
than permitted by glibc's accuracy goals.
Part of the problem is error accumulation, addressed in this patch by
setting round-to-nearest for internal computations. However, there
was also a bug in the code dealing with computing pow (x + n, x + n)
where x + n is not exactly representable, providing another source of
error even in round-to-nearest mode; it was necessary to address both
bugs to get errors for all testcases within glibc's accuracy goals.
Given this second fix, accuracy in round-to-nearest mode is also
improved (hence regeneration of ulps for tgamma should be from scratch
- truncate libm-test-ulps or at least remove existing tgamma entries -
so that the expected ulps can be reduced).
Some additional complications also arose. Certain tgamma tests should
strictly, according to IEEE semantics, overflow or not depending on
the rounding mode; this is beyond the scope of glibc's accuracy goals
for any function without exactly-determined results, but
gen-auto-libm-tests doesn't handle being lax there as it does for
underflow. (libm-test.inc also doesn't handle being lax about whether
the result in cases very close to the overflow threshold is infinity
or a finite value close to overflow, but that doesn't cause problems
in this case though I've seen it cause problems with random test
generation for some functions.) Thus, spurious-overflow markings,
with a comment, are added to auto-libm-test-in (no bug in Bugzilla
because the issue is with the testsuite, not a user-visible bug in
glibc). And on x86, after the patch I saw ERANGE issues as previously
reported by Carlos (see my commentary in
<https://sourceware.org/ml/libc-alpha/2015-01/msg00485.html>), which
needed addressing by ensuring excess range and precision were
eliminated at various points if FLT_EVAL_METHOD != 0.
I also noticed and fixed a cosmetic issue where 1.0f was used in long
double functions and should have been 1.0L.
This completes the move of all functions to testing in all rounding
modes with ALL_RM_TEST, so gen-libm-have-vector-test.sh is updated to
remove the workaround for some functions not using ALL_RM_TEST.
Tested for x86_64, x86, mips64 and powerpc.
[BZ #18613]
* sysdeps/ieee754/dbl-64/e_gamma_r.c (gamma_positive): Take log of
X_ADJ not X when adjusting exponent.
(__ieee754_gamma_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed.
* sysdeps/ieee754/flt-32/e_gammaf_r.c (gammaf_positive): Take log
of X_ADJ not X when adjusting exponent.
(__ieee754_gammaf_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed.
* sysdeps/ieee754/ldbl-128/e_gammal_r.c (gammal_positive): Take
log of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* sysdeps/ieee754/ldbl-128ibm/e_gammal_r.c (gammal_positive): Take
log of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* sysdeps/ieee754/ldbl-96/e_gammal_r.c (gammal_positive): Take log
of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* math/libm-test.inc (tgamma_test_data): Remove one test. Moved
to auto-libm-test-in.
(tgamma_test): Use ALL_RM_TEST.
* math/auto-libm-test-in: Add one test of tgamma. Mark some other
tests of tgamma with spurious-overflow.
* math/auto-libm-test-out: Regenerated.
* math/gen-libm-have-vector-test.sh: Do not check for START.
* sysdeps/i386/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Likewise.
2015-06-29 23:29:35 +00:00
|
|
|
else
|
|
|
|
return ret;
|
Update.
1997-06-19 19:38 Ulrich Drepper <drepper@cygnus.com>
* features.h: Define __STDC_IEC_559__ and _STDC_IEC_559_COMPLEX__.
* elf/dl-minimal.c (__dcgettext): Remove assertion.
* inet/rcmd.c: Correct a few typos. Reported by Eric Troan.
* manual/Makefile (distribute): Add dir.
* manual/dir: New file.
* math/Makefile (libm-support): Rename s_rinttol, s_rinttoll,
s_roundtol, and s_roundtoll to s_lrint, s_llrint, s_lround,
and s_llround respectively.
(libm-calls): Add e_gamma_r.
* math/libm-test.c (check_int_exc): New function.
(signbit_test): Remove test for sign of NaN.
(gamma_test): Clear exception after test of existence.
Correct tests and and epsilons.
(lgamma_test): Likewise.
(ilogb_test): Correct all tests.
(scalb_test): Rewrite.
(rinttol_test): Rename to lrint_test and correct tests.
(rinttoll_test): Likewise.
(roundtol_test): Likewise.
(roundtoll_test): Likewise.
(main): Call lrint/lround functions instead of rinttol/roundtol.
* math/math.h: Change prototypes for rinttol/roundtol.
* math/mathcalls.h: Rearrange prototypes according to ISO C9X draft.
* sysdeps/generic/mathbits.h: Define FP_ILOGB0 and FP_ILOGBNAN.
* sysdeps/i386/mathbits.h: Likewise.
* sysdeps/libm-i387/e_scalb.S: Handle special cases correctly.
* sysdeps/libm-i387/e_scalbf.S: Likewise.
* sysdeps/libm-i387/e_scalbl.S: Likewise.
* sysdeps/libm-i387/s_asinh.S: Handle -inf correctly.
* sysdeps/libm-i387/s_asinhf.S: Likewise.
* sysdeps/libm-i387/s_asinhl.S: Likewise.
* sysdeps/libm-i387/s_ilogb.S: Optimize.
* sysdeps/libm-i387/s_ilogbf.S: Likewise.
* sysdeps/libm-i387/s_ilogbl.S: Likewise.
* sysdeps/libm-i387/s_rinttol.S: Rename to...
* sysdeps/libm-i387/s_lrint.S: ...this.
* sysdeps/libm-i387/s_rinttoll.S: Rename to...
* sysdeps/libm-i387/s_llrint.S: ...this.
* sysdeps/libm-i387/s_remquo.S: Correctly set sign of remainder.
* sysdeps/libm-i387/s_remquof.S: Likewise.
* sysdeps/libm-i387/s_remquol.S: Likewise.
* sysdeps/libm-i387/e_gamma_r.c: New file. Implementation of gamma
function according to ISO C.
* sysdeps/libm-i387/e_gammaf_r.c: New file.
* sysdeps/libm-i387/e_gammal_r.c: New file.
* sysdeps/libm-i387/e_lgamma_r.c: Don't let optimize compile the
generation of exceptions away.
* sysdeps/libm-i387/e_lgammaf_r.c: Likewise.
* sysdeps/libm-i387/k_standard.c: Correct return value for infinity
points of gamma function when not SVID mode.
* sysdeps/libm-i387/s_rinttoll.c: Renamed to...
* sysdeps/libm-i387/s_llrint.c: ...this.
* sysdeps/libm-i387/s_rinttol.c: Renamed to...
* sysdeps/libm-i387/s_lrint.c: ...this.
* sysdeps/libm-i387/s_roundtoll.c: Renamed to...
* sysdeps/libm-i387/s_llround.c: ...this.
* sysdeps/libm-i387/s_roundtol.c: Renamed to...
* sysdeps/libm-i387/s_lround.c: ..this.
* sysdeps/libm-i387/s_scalbn.c: Change second parameter according to
ISO C.
* sysdeps/libm-i387/s_scalbnf.c: Likewise.
* sysdeps/libm-i387/s_scalbnl.c: Likewise.
* sysdeps/libm-i387/w_gamma.c: Call __ieee754_gamma_r if library
mode is not _SVID_.
* sysdeps/libm-i387/w_gammaf.c: Likewise.
* sysdeps/libm-i387/w_gammal.c: Likewise.
* sysdeps/m68k/fpu/__math.h: Rename __rinttol to __lrint and
rinttol to lrint.
* sysdeps/m68k/fpu/s_rinttol.c: Renamed to...
* sysdeps/m68k/fpu/s_lrint.c: ...this.
* sysdeps/m68k/fpu/s_rinttoll.c: Renamed to...
* sysdeps/m68k/fpu/s_llrint.c: ...this.
* md5-crypt/Makefile: Link md5test program with md5.o.
* stdio-common/temptest.c: Don't use __stdio_gen_tempname which
is not exported by the libc.so.
* stdio-common/vfscanf.c: Correct scanning of strings after last
change.
* sysdeps/unix/sysv/linux/i386/sysdep.S: Use .comm to define errno.
1997-06-19 07:37 H.J. Lu <hjl@gnu.ai.mit.edu>
* time/tzfile.c (__tzfile_read): Store getc () return in int.
1997-06-13 Miles Bader <miles@gnu.ai.mit.edu>
* argp/argp-parse.c (argp_version_parser): Include `(PROGRAM ERROR)' in
the no-version error text to indicate that something's fucked.
[!_] (N_): New macro.
(argp_default_options, argp_version_options): Wrap doc strings in N_().
1997-06-12 Miles Bader <miles@gnu.ai.mit.edu>
* argp/argp-parse.c (parser_parse_opt): Detect and report unhandled
options here.
(parser_parse_arg): Handle ARGP_KEY_ARGS here.
Adjust NEXT pointer back if we fail to parse anything.
(parser_parse_next): Simplify arg code. Leave state NEXT frobbing
to parser_parse_arg.
1997-06-11 Miles Bader <miles@gnu.ai.mit.edu>
* argp/argp.h (ARGP_KEY_ARGS, ARGP_KEY_FINI): New macros.
* argp/argp-parse.c (parser_finalize): Do another pass over the
parsers with ARGP_KEY_FINI.
1997-06-18 Miles Bader <miles@gnu.ai.mit.edu>
* string/Makefile (routines): Add argz-replace.
1997-06-16 00:16 Miles Bader <miles@gnu.ai.mit.edu>
* manual/string.texi (Argz Functions): Document argz_replace.
1997-06-12 Miles Bader <miles@gnu.ai.mit.edu>
* string/argz.h (__argz_replace, argz_replace): New declarations.
* string/argz-replace.c: New file.
1997-06-11 Miles Bader <miles@gnu.ai.mit.edu>
* manual/argp.texi (Argp Special Keys): Document ARGP_KEY_ARGS.
1997-06-16 23:08 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* Makeconfig (libc-map): Remove definition.
* Makerules (libc-map): Define it here, using the full name.
(load-map-file): Remove case for empty $(..).
($(common-objpfx)libc.so): Revert last change.
1997-06-17 22:18 Mark Kettenis <kettenis@phys.uva.nl>
* login/programs/utmpd.c (main): Improve signal handling.
* login/programs/request.c (do_setutent, do_updwtmp):
Allow arbitrary length filenames.
* login/programs/utmpd.h (setutent_request, updwtmp_request):
Get rid of fixed length file field.
* login/utmp_daemon.c (do_setutent, do_updwtmp):
Allow arbitrary length filenames.
* login/programs/request.c (do_pututline):
Don't fail if connection->position is -1 on entry.
1997-06-15 16:32 Mark Kettenis <kettenis@phys.uva.nl>
* login/utmp_file.c (updwtmp_file): Use the same method for
appending an entry as in pututline_file.
1997-06-11 18:59 Mark Kettenis <kettenis@phys.uva.nl>
* login/utmp_file.c (getutent_r_file):
Use read lock instead of write lock.
(getutline_r_file, internal_getut_r): Lock utmp file.
(updwtmp_file): Use fcntl to lock file instead of flock.
1997-06-18 00:11 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* stdio-common/vfscanf.c (inchar, ungetc): Don't count EOF as
character read in/put back.
* stdio-common/tstscanf.c: Add test case for this.
1997-06-17 22:17 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* libc.map: Add more symbols.
1997-06-18 12:01 Ulrich Drepper <drepper@cygnus.com>
* manual/Makefile (install): Make sure `dir' file exists if we use
install-info.
1997-06-17 19:32 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* manual/Makefile (info): Depend on dir-add.info.
1997-06-17 17:19 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* elf/ldd.bash.in: Fix spacing in message.
1997-06-17 14:28 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* elf/do-rel.h (elf_dynamic_do_rel): Always use version
information if available.
1997-06-17 11:34 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sunrpc/Makefile ($(rpcsvc:%.x=$(objpfx)rpcsvc/%.h)): Make
command non-empty to force make to recheck modification time.
($(rpcsvc:%.x=$(objpfx)x%.c)): Likewise.
1997-06-17 00:26 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/stub/e_acoshl.c: Set errno to ENOSYS.
* sysdeps/stub/e_acosl.c: Likewise.
* sysdeps/stub/e_asinl.c: Likewise.
* sysdeps/stub/e_atan2l.c: Likewise.
* sysdeps/stub/e_expl.c: Likewise.
* sysdeps/stub/e_fmodl.c: Likewise.
* sysdeps/stub/e_j0l.c: Likewise.
* sysdeps/stub/e_j1l.c: Likewise.
* sysdeps/stub/e_jnl.c: Likewise.
* sysdeps/stub/e_lgammal_r.c: Likewise.
* sysdeps/stub/e_log10l.c: Likewise.
* sysdeps/stub/e_logl.c: Likewise.
* sysdeps/stub/e_powl.c: Likewise.
* sysdeps/stub/e_rem_pio2l.c: Likewise.
* sysdeps/stub/e_sqrtl.c: Likewise.
* sysdeps/stub/k_cosl.c: Likewise.
* sysdeps/stub/k_rem_pio2l.c: Likewise.
* sysdeps/stub/k_sinl.c: Likewise.
* sysdeps/stub/k_tanl.c: Likewise.
* sysdeps/stub/s_atanl.c: Likewise.
* sysdeps/stub/s_erfl.c: Likewise.
* sysdeps/stub/s_exp2.c: Likewise.
* sysdeps/stub/s_exp2f.c: Likewise.
* sysdeps/stub/s_exp2l.c: Likewise.
* sysdeps/stub/s_expm1l.c: Likewise.
* sysdeps/stub/s_log1pl.c: Likewise.
* sysdeps/stub/s_log2l.c: Likewise.
1997-06-18 11:46 Ulrich Drepper <drepper@cygnus.com>
* sysdeps/libm-ieee754/w_gamma.c: If _LIB_VERSION is _SVID_ compute
result as before last change.
* sysdeps/libm-ieee754/w_gammaf.c: Likewise.
* sysdeps/libm-ieee754/w_gammal.c: Likewise.
1997-06-16 23:37 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/libm-ieee754/s_remquo.c: Fix off-by-one when computing
quotient.
* sysdeps/libm-ieee754/s_remquof.c: Likewise.
* sysdeps/libm-ieee754/s_remquol.c: Likewise.
* sysdeps/m68k/fpu/s_remquo.c: Remove FIXME and special case for
quotient.
* sysdeps/libm-ieee754/w_gamma.c: Add missing call to exp
function. Don't use global signgam.
* sysdeps/libm-ieee754/w_gammaf.c: Likewise.
* sysdeps/libm-ieee754/w_gammal.c: Likewise.
* math/Makefile (libm-calls): Remove w_gamma_r.
* sysdeps/libm-ieee754/w_gamma_r.c: Remove file.
* sysdeps/libm-ieee754/w_gammaf_r.c: Remove file.
* sysdeps/libm-ieee754/w_gammal_r.c: Remove file.
* math/libm-test.c (atanh_test): Declare x only if needed.
(signbit_test): Fix typo.
(gamma_test): Check whether function is implemented. Add
epsilons.
(lgamma_test): Likewise.
(fmod_test): Add epsilons.
(exp2_test): Use right function for existence test.
1997-06-07 09:20 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/powerpc/Dist: Add fenv_const.c.
* sysdeps/unix/sysv/linux/Dist: Add net/if_slip.h.
* sysdeps/unix/sysv/linux/powerpc/Dist: Add init-first.h and
syscall.h.
* sysdeps/unix/sysv/linux/sparc/Dist: Add init-first.h.
* string/Makefile (distribute): Add tst-svc.expect.
* nis/Makefile (distribute): Add nis_intern.h and Banner.
* elf/Makefile (distribute): Add dl-hash.h.
* Rules (subdir_echo-distinfo): Add headers from $(distribute).
* login/Makefile (others): Add utmpdump.
* sysdeps/unix/sysv/linux/Makefile (sysdep_headers): Add
net/if_slip.h.
* manual/Makefile (dir-add.texi): Also look in indirectly included
files.
1997-06-16 23:15 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/unix/sysv/linux/configure.in: Remove commands to
generate stdio_lim.h.
* sysdeps/unix/sysv/linux/mk-stdiolim.c: Remove.
* sysdeps/unix/sysv/linux/Makefile: Add rules to generate
stdio_lim.h here.
(common-generated): Add generated files.
(inhibit-stdio_lim): Define.
* sysdeps/posix/Makefile [$(inhibit-stdio_lim)=yes]: Disable rules
to generate stdio_lim.h.
* sysdeps/unix/sysv/linux/Makefile: Suppress inclusion of
dependecy files if no_deps is set.
($(objpfx)syscall-%.d): Add header file as target to dependency
generation.
1997-06-14 19:19 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* sysdeps/unix/sysv/linux/arm/Dist: New file.
1997-06-14 17:59 Andreas Schwab <schwab@issan.informatik.uni-dortmund.de>
* extra-lib.mk (others): Depend on versioned shared library, not
the unversioned one.
* Makerules (build-shlib): Don't make the version link here.
($(common-objpfx)libc.so$(libc.so-version)): New rule for libc
version link.
1997-06-16 03:07 Ulrich Drepper <drepper@cygnus.com>
* sysdeps/libm-ieee754/k_standard.c: Undo change of Tue Aug 6
01:13:56 1996.
* argp/argp-help.c (argp_args_usage): Supply correct argp to
filter_doc.
* argp/argp-help.c (hol_add_cluster): Initialize CL->depth.
* argp/argp-help.c (_help): Supply STATE to argp_args_usage.
* argp/argp.h (ARGP_KEY_HELP_ARGS_DOC): New macro.
* argp/argp-fmtstream.c: Add casts to prevent warnings.
* argp/argp.h (OPTION_NO_USAGE): New macro.
* argp/argp-help.c (usage_long_opt, usage_argful_short_opt,
* argp/argp-fmtstream.c (__argp_fmtstream_update): Account for case
* argp/argp-help.c <stddef.h>: New include.
* argp/argp.h (argp_state_help, __argp_state_help, argp_usage,
* argp/argp.h (argp_program_bug_address): Make const.
* argp/argp-parse.c (argp_default_parser): Set STATE->name for OPT_PROGNAME.
* argp/argp-help.c (__argp_error, __argp_failure, __argp_state_help):
* argp/argp-parse.c (parser_init): Set PARSER->state.flags.
1997-06-19 19:11:22 +00:00
|
|
|
}
|
2011-10-12 15:27:51 +00:00
|
|
|
strong_alias (__ieee754_gamma_r, __gamma_r_finite)
|