Small portions of the changes by Fabrice Popineau <fabrice.popineau@supelec.fr>.
src/makefile.w32-in (TAGS-gmake): Don't use $(patsubst ...), as
Nmake barfs on that. Use $(OBJ*_c) variables instead, defined on
nt/gmake.defs.
src/lisp.h (ENUM_BF): New macro, for enumerated types in bitfields,
which are not supported by MSVC.
(Lisp_Symbol, Lisp_Misc_Any, Lisp_Marker, Lisp_Misc_Overlay)
(Lisp_Save_Value, Lisp_Free): Use ENUM_BF for enumerated types in
bitfields.
(Lisp_Object) [USE_LISP_UNION_TYPE]: Use ENUM_BF for enumerated
types in bitfields.
(DEFUN) [_MSC_VER]: Define in a different way for MSVC.
src/w32fns.c [_MSC_VER]: DECLARE_HANDLE for any MSVC version.
src/w32.c: Don't include w32api.h for MSVC.
(init_environment) [_MSC_VER]: Call sys_access, not _access.
src/s/ms-w32.h <sigset_t, ssize_t> [_MSC_VER]: Typedefs for MSVC.
[_MSC_VER]: Include sys/timeb.h, sys/stat.h, and signal.h.
(fstat, stat, utime) [_MSC_VER]: Redirect to their sys_* cousins.
(malloc, free, realloc, calloc) [_MSC_VER]: Always redirect to the
e_* cousins.
(alloca) [_MSC_VER]: Define to _alloca.
src/lisp.h (DECL_ALIGN) [_MSC_VER]: Define for MSVC.
src/regex.c <re_char> [_MSC_VER]: A separate definition for MSVC.
nt/makefile.w32-in (clean-other-dirs-nmake)
(distclean-other-dirs-nmake, maintainer-clean-other-dirs-nmake):
Update for current structure of doc/ subdirectories.
nt/gmake.defs (OBJ0_c, OBJ1_c, OBJ2_c): New variables.
nt/INSTALL: Update for newer versions of MSVC.
lib/makefile.w32-in (FRC): New dummy target.
(TAGS): Depend on FRC.
Fixes: debbugs:9960
* lib/signal.in.h, lib/sigprocmask.c, m4/gl-comp.m4:
* m4/gnulib-common.m4: Merge from gnulib.
Without this, 'configure' would say "func_gl_gnulib_m4code_pathmax:
command not found" on powerpc-apple-darwin9.8.0.
This clarifies and fixes some licensing issues raised by Glenn Morris
<http://lists.gnu.org/archive/html/bug-gnulib/2011-09/msg00397.html>.
It also merges the latest version of texinfo.tex and has some
MSVC-related changes that don't affect Emacs.
* Makefile.in (GNULIB_TOOL_FLAGS): Avoid msvc-inval, msvc-nothrow,
pathmax, and raise, since these are needed only to address
MSVC-related issues that Emacs doesn't have.
* doc/misc/texinfo.tex, lib/dup2.c, lib/gnulib.mk, lib/signal.in.h:
* lib/sigprocmask.c, lib/stat.c, lib/stdio.in.h, lib/sys_stat.in.h:
* lib/unistd.in.h, m4/dup2.m4, m4/getloadavg.m4, m4/gl-comp.m4:
* m4/include_next.m4, m4/signal_h.m4, m4/signalblocking.m4:
* m4/stdint.m4, m4/stdio_h.m4, m4/sys_stat_h.m4, m4/time_h.m4:
* m4/unistd_h.m4: Merge from gnulib.
* lib/lstat.c, lib/stat.c, m4/include_next.m4, m4/nocrash.m4:
* m4/signal_h.m4, m4/time_h.m4, m4/unistd_h.m4:
Merge from gnulib. This fixes a compilation error on Tru64 UNIX
aka OSF/1 5.1 DTK cc. There is also some mingw stuff here that
doesn't affect Emacs.
* m4/largefile.m4: New file, so that Emacs does not mess up when
accessing files with large inode numbers in MacOS X 10.5 and later.
* m4/nocrash.m4: New file, to avoid triggering background debugger
and/or create core dumps during 'configure'.
* build-aux/move-if-change: Renamed from move-if-change.
* build-aux/snippet/arg-nonnull.h: Renamed from arg-nonnull.h.
* build-aux/snippet/c++defs.h: Renamed from c++defs.h.
* build-aux/snippet/warn-on-use.h: Renamed from warn-on-use.h.
* build-aux/snippet/_Noreturn.h: New file, for draft C1X _Noreturn.
* .bzrignore: The autogenerated files compile, config.guess,
config.sub, depcomp, install-sh, and missing are now in build-aux.
* Makefile.in (epaths-force, sync-from-gnulib):
move-if-change is now in build-aux.
(GNULIB_TOOL_FLAGS): Avoid threadlib; this is now a prerequisite
of gnulib's pthread_sigmask module, but Emacs doesn't need it.
(mkdir): install-sh is now in build-aux.
* config.bat: c++defs.h is now in build-aux/snippets.
* configure.in: Specify AC_CONFIG_AUX_DIR with build-aux (the
usual parameter).
* lib/gnulib.mk, m4/gl-comp.m4: Regenerate.
* lib/makefile.w32-in (ARG_NONNULL_H): arg-nonnull.h moved
to build-aux/snippet.
* lib/pthread_sigmask.c, lib/stdlib.in.h, m4/extensions.m4:
* m4/getopt.m4, m4/gnulib-common.m4, m4/pthread_sigmask.m4:
Merge from gnuilib. This fixes porting bugs on Cygwin, Irix, and
Solaris, enables MacOS extensions, and enables nocrash during
'configure'.
* make-dist: Adjust to new build-aux and build-aux/snippit dirs.
* admin/notes/copyright: The files compile, config.guess, config.sub,
depcomp, install-sh, missing, and move-if-change are now in the
new build-aux subdirectory. The files arg-nonnull.h, c++defs.h,
and warn-on-use.h are now in build-aux/snippets. New file
build-aux/snippets/_Noreturn.h.
* leim/Makefile.in (install): install-sh is now in build-aux.
* lib-src/Makefile.in ($(DESTDIR)${archlibdir}): install-sh moved
to build-aux.
* msdos/sedlibmk.inp (CONFIG_CLEAN_VPATH_FILES): Adjust to snippet moves
from top level to build-aux/snippet.
* src/Makefile.in (gl-stamp): move-if-change is now in build-aux.
* Makefile.in (GNULIB_TOOL_FLAGS): Do not avoid sigprocmask.
* configure.in (AC_TYPE_UID_T): New dummy macro.
Configure gnulib after adjusting LIBS,
so that gnulib can assume the libraries in LIBS.
* lib/pthread_sigmask.c, lib/sigprocmask.c, m4/signalblocking.m4:
New files, automatically imported from gnulib.
* lib/gnulib.mk, m4/gl-comp.m4: Regenerate.
* src/Makefile.in (LIB_PTHREAD_SIGMASK): New macro.
(LIBES): Use it.
* Makefile.in (GNULIB_MODULES): Add pthread_sigmask.
(GNULIB_TOOL_FLAGS): Avoid sigprocmask. Emacs does its own
implementation of 'sigprocmask' on Windows, and it assumes
'sigprocmask' on non-Windows hosts, so it doesn't need the
sigprocmask module.
* lib/signal.in.h, m4/pthread_sigmask.m4, m4/signal_h.m4:
New files, automatically imported from gnulib.
* lib/gnulib.mk, m4/gl-comp.m4: Automatically-imported update
due to the above changes.
* .bzrignore: Add lib/signal.h.
* Makefile.in (GNULIB_MODULES): Add dup2.
* configure.in: Do not check for dup2; gnulib does that now.
* lib/dup2.c, m4/dup2.m4: New files, from gnulib.
* src/sysdep.c (dup2) [!HAVE_DUP2]: Remove; gnulib now does this.
* .bzrignore: Add lib/alloca.h.
* Makefile.in (GNULIB_MODULES): Add alloca-opt.
* configure.in (AC_FUNC_ALLOCA): Remove almost all the alloca stuff,
as gnulib now does that for us. Put alloca check after gl_INIT.
* lib/gnulib.mk, m4/gl-comp.m4: Regenerate.
* lib/alloca.in.h, m4/alloca.m4: New files, from gnulib.
fakemail: Remove dependency on ignore-value.
* Makefile.in (GNULIB_MODULES): Add stdio.
* lib/stdio.in.h, m4/stdio_h.m4: New files, automatically
imported from gnulib.
[lib-src/ChangeLog]
fakemail: Remove dependency on ignore-value.
This undoes some of the recent fakemail-related changes.
It is made possible due to recent changes to gnulib's stdio module.
* Makefile.in (fakemail${EXEEXT}): Do not depend on ignore-value.h.
* fakemail.c: Do not include ignore-value.h.
(put_line): Do not use ignore_value.
Ref: http://lists.gnu.org/archive/html/emacs-devel/2011-03/msg00225.html
* autogen/: New directory, to be excluded from releases.
* autogen/copy_autogen, autogen/update_autogen: New scripts.
* autogen/README: New file.
* autogen/aclocal.m4, autogen/config.in, autogen/configure:
* autogen/Makefile.in: Add auto-updated generated files.
* autogen.sh: No longer a no-op, now it tests for autotools
and runs them as necessary.
* configure.in: Defaule maintainer-mode to on.
* aclocal.m4, configure, lib/Makefile.in: Remove files.
* src/config.in: Remove file.
* INSTALL.BZR, admin/make-tarball.txt: Update instructions.
Work around some portability problems with symlinks.
* Makefile.in (GNULIB_MODULES): Add lstat, readlink, symlink.
* configure.in (lstat, HAVE_LSTAT): Remove special hack.
* lib/lstat.c, lib/readlink.c, lib/stat.c, lib/symlink.c:
* m4/dos.m4, m4/lstat.m4, m4/readlink.m4, m4/stat.m4, m4/symlink.m4:
New files, automatically generated from gnulib.
* aclocal.m4, configure, lib/Makefile.in, lib/gnulib.mk:
* lib/stdlib.in.h, m4/gl-comp.m4, m4/stdlib_h.m4: Regenerate.
2011-02-22 Paul Eggert <eggert@cs.ucla.edu>
[src/ChangeLog]
Work around some portability problems with symlinks.
* fileio.c (Frename_file, Fmake_symbolic_link, Ffile_symlink_p):
Simplify the code by assuming that the readlink and symlink calls
exist, even if they always fail on this host.
(Ffile_readable_p): Likewise, for fifos.
* config.in: Regenerate.
* lib/min-max.h: New file, for "min" and "max".
[lib-src/ChangeLog]
New file "lib/min-max.h".
* ebrowse.c (min, max): Define them by including <min-max.h>
instead of defining it ourselves.
* pop.c (min): Likewise.
nt/inc/stdint.h: New file, to support compilation with tool chains
that do not have stdint.h (e.g. MSVC).
lib/makefile.w32-in: ($(BLD)/md5.$(O)): Added dependency on
$(EMACS_ROOT)/nt/inc/stdint.h.
src/makefile.w32-in: Remove md5.$(O).
($(BLD)/md5.$(O)): Remove prerequisites, moved to
lib/makefile.w32-in.
($(BLD)/fns.$(O)): Depend on $(EMACS_ROOT)/lib/md5.h and on stamp_BLD.
lib/makefile.w32-in ($(BLD)/md5.$(O)): New recipe, moved from
src/makefile.w32-in.
* aclocal.m4, configure, lib/Makefile.in, lib/gnulib.mk, m4/gl-comp.m4:
* src/config.in: Regenerate.
* lib/md5.c, lib/md5.h: Regenerate with simpler version, which
assumes C99-style <stdint.h>, supplied by the stdint module.
* lib/stdint.in.h, m4/longlong.m4, m4/stdint.m4: New files,
generated from gnulib.
* src/deps.mk (fns.o): Do not depend on md5.h, fixing a typo in
the earlier patch.
* Makefile.in (MAKEFILE_MODULES): Add crypto/md5.
* admin/notes/copyright: Remove src/md5.c and src/md5.h as
special cases.
* src/Makefile.in (base_obj): Remove md5.o, since this file
is in lib now.
* src/deps.mk (md5.o): Remove.
* aclocal.m4, configure, lib/Makefile.in, lib/gnulib.mk, m4/gl-comp.m4:
Regenerate.
* lib/md5.c: Regenerate. This renames the file from src/md5.c,
and adds some porting improvements from gnulib.
* lib/md5.h: Regenerate, likwise; rename from src/md5.h.
* m4/md5.m4: New file, from gnulib.
2011-02-18 Paul Eggert <eggert@cs.ucla.edu>
This defines a few preprocessor symbols that should not affect Emacs.
* lib/getopt.in.h, lib/time.in.h, lib/unistd.in.h: Regenerate
via "make sync-from-gnulib".
* configure: Regenerate from the following.
2011-01-30 Paul Eggert <eggert@cs.ucla.edu>
mktime: clarify long_int width checking
* lib/mktime.c (long_int_is_wide_enough): Move this assertion to
the top level, to make it clearer that the assumption about
long_int width is being checked. See
<http://lists.gnu.org/archive/html/bug-gnulib/2011-01/msg00554.html>.
2011-01-29 Paul Eggert <eggert@cs.ucla.edu>
TYPE_MAXIMUM: avoid theoretically undefined behavior
* lib/intprops.h (TYPE_MINIMUM, TYPE_MAXIMUM): Do not shift a
negative number, which the C Standard says has undefined behavior.
In practice this is not a problem, but might as well do it by the book.
Reported by Rich Felker and Eric Blake; see
<http://lists.gnu.org/archive/html/bug-gnulib/2011-01/msg00493.html>.
* m4/mktime.m4 (AC_FUNC_MKTIME): Likewise.
* lib/mktime.c (TYPE_MAXIMUM): Redo slightly to match the others.
mktime: #undef mktime before #defining it
* lib/mktime.c (mktime) [DEBUG]: #undef mktime before #defining it.
mktime: systematically normalize tm_isdst comparisons
* lib/mktime.c (isdst_differ): New function.
(__mktime_internal): Use it systematically for all isdst comparisons.
This completes the fix for libc BZ #6723, and removes the need for
normalizing tm_isdst. See
<http://sourceware.org/bugzilla/show_bug.cgi?id=6723>
(not_equal_tm) [DEBUG]: Use isdst_differ here, too.
mktime: fix some integer overflow issues and sidestep the rest
This was prompted by a bug report by Benjamin Lindner for MinGW
<http://lists.gnu.org/archive/html/bug-gnulib/2011-01/msg00472.html>.
His bug is due to signed integer overflow (0 - INT_MIN), and I
I scanned through mktime.c looking for other integer overflow
problems, fixing all the bugs I found.
Although the C Standard says the resulting code is still not safe
in the presence of integer overflow, in practice it should be good
enough for all real-world two's-complement implementations, except
for debugging environments that deliberately trap on integer
overflow (e.g., gcc -ftrapv).
* lib/mktime.c (WRAPV): New macro.
(SHR): Also check that long_int and time_t shift right in the
usual way, before using the fast-but-unportable method.
(TYPE_ONES_COMPLEMENT, TYPE_SIGNED_MAGNITUDE): Remove, no longer
used. The code already assumed two's complement, so there's
no need to test for alternatives. All uses removed.
(TYPE_MAXIMUM): Don't rely here on overflow behavior not defined by
the C standard. Problem reported by Rich Felker in
<http://lists.gnu.org/archive/html/bug-gnulib/2011-01/msg00488.html>.
(twos_complement_arithmetic): Also check long_int and time_t.
(time_t_avg, time_t_add_ok, time_t_int_add_ok): New functions.
(guess_time_tm, ranged_convert, __mktime_internal): Use them.
(__mktime_internal): Avoid integer overflow with unary subtraction
in two instances where -1 - X is an adequate replacement for -X,
since the calculations are approximate.
2011-01-29 Eric Blake <eblake@redhat.com>
mktime: avoid infinite loop
* m4/mktime.m4 (AC_FUNC_MKTIME): Avoid overflow on possibly-signed
type; behavior is still undefined but portable to all known targets.
Reported by Rich Felker.
2011-01-28 Paul Eggert <eggert@cs.ucla.edu>
mktime: avoid problems on NetBSD 5 / i386
* lib/mktime.c (long_int): New type. This works around a problem
on NetBSD 5 / i386, where 'long int' and 'int' are both 32 bits
but time_t is 64 bits, and where I expect the existing code is
wrong in some cases.
(leapyear, ydhms_diff, guess_time_tm, __mktime_internal): Use it.
(ydhms_diff): Bring back the compile-time check for wide-enough
year and yday.
mktime: fix misspelling in comment
* lib/mktime.c (__mktime_internal): Fix misspelling in comment.
This merges all recent glibc changes of importance.
2011-01-28 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
move-if-change: cope with concurrent mv of identical file.
* move-if-change (CMPPROG): Accept environment
variable as an override for `cmp'.
(usage): Document CMPPROG.
Adjust comparison to drop stdout. Cope with failure of mv if
the target file exists and is identical to the source, for
parallel builds.
Report from H.J. Lu against binutils in PR binutils/12283.
