[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
bug#9772: bug#10155: OSX breakage
From: |
Eli Zaretskii |
Subject: |
bug#9772: bug#10155: OSX breakage |
Date: |
Tue, 29 Nov 2011 19:30:30 +0200 |
> Date: Tue, 29 Nov 2011 08:58:34 -0800
> From: Paul Eggert <eggert@cs.ucla.edu>
> CC: Chong Yidong <cyd@gnu.org>, mario@lassnig.net, 9772@debbugs.gnu.org,
> 9960@debbugs.gnu.org, 10155@debbugs.gnu.org,
> Eli Zaretskii <eliz@gnu.org>
>
> You're right, on further thought Emacs itself needn't use stdalign,
> and I've prepared a simpler patch (below) that omits all changes to
> src/*. Gnulib now uses stdalign, though, and Emacs uses
> multiple Gnulib modules such as sha256 that use stdalign, so
> this patch still brings in the stdalign module for Gnulib's own
> purposes.
I understand that the only part needed for solving the problems with
the missing declaration of `environ' is this:
> === added file 'm4/environ.m4'
> --- m4/environ.m4 1970-01-01 00:00:00 +0000
> +++ m4/environ.m4 2011-11-29 16:38:21 +0000
> @@ -0,0 +1,47 @@
> +# environ.m4 serial 6
> +dnl Copyright (C) 2001-2004, 2006-2011 Free Software Foundation, Inc.
> +dnl This file is free software; the Free Software Foundation
> +dnl gives unlimited permission to copy and/or distribute it,
> +dnl with or without modifications, as long as this notice is preserved.
> +
> +AC_DEFUN_ONCE([gl_ENVIRON],
> +[
> + AC_REQUIRE([gl_UNISTD_H_DEFAULTS])
> + dnl Persuade glibc <unistd.h> to declare environ.
> + AC_REQUIRE([gl_USE_SYSTEM_EXTENSIONS])
> +
> + AC_CHECK_HEADERS_ONCE([unistd.h])
> + gt_CHECK_VAR_DECL(
> + [#if HAVE_UNISTD_H
> + #include <unistd.h>
> + #endif
> + /* mingw, BeOS, Haiku declare environ in <stdlib.h>, not in <unistd.h>.
> */
> + #include <stdlib.h>
> + ],
> + [environ])
> + if test $gt_cv_var_environ_declaration != yes; then
> + HAVE_DECL_ENVIRON=0
> + fi
> +])
IOW, the reason why lib/unistd.in.h didn't do its job is that
HAVE_DECL_ENVIRON was not defined to zero.
If that is true, I wonder if we could have a much smaller change that
fixes just that single problem.
- bug#9772: OSX breakage, Paul Eggert, 2011/11/29
- bug#10155: OSX breakage, Stefan Monnier, 2011/11/29
- bug#10155: OSX breakage, Paul Eggert, 2011/11/29
- bug#9772: bug#10155: OSX breakage,
Eli Zaretskii <=
- bug#9960: bug#10155: OSX breakage, Paul Eggert, 2011/11/29
- bug#10155: bug#9772: bug#10155: OSX breakage, Chong Yidong, 2011/11/29
- bug#10155: bug#9772: bug#10155: OSX breakage, Eli Zaretskii, 2011/11/29
- bug#10155: bug#9772: bug#10155: OSX breakage, Chong Yidong, 2011/11/30
- bug#10155: bug#9772: bug#10155: OSX breakage, Jan Djärv, 2011/11/30
- bug#10155: bug#9772: bug#10155: OSX breakage, Mario Lassnig, 2011/11/30
bug#9772: OSX breakage, Eli Zaretskii, 2011/11/29