From: Paul Eggert <eggert@cs.ucla.edu>
To: Stefan Monnier <monnier@iro.umontreal.ca>
Cc: 12471@debbugs.gnu.org, lekktu@gmail.com
Subject: bug#12471: Avoid some signal-handling races, and simplify.
Date: Sat, 22 Sep 2012 14:55:23 -0700 [thread overview]
Message-ID: <505E33CB.7090709@cs.ucla.edu> (raw)
In-Reply-To: <jwv4nmpu7k6.fsf-monnier+emacs@gnu.org>
[-- Attachment #1: Type: text/plain, Size: 3189 bytes --]
On 09/22/2012 01:28 PM, Stefan Monnier wrote:
> Maybe a better solution is to use `emacs_raise' which can then
> either use `raise' (on POSIX hosts) or something else (on Windows
> hosts).
I still don't see why that helps, but since you and
Eli both seem to prefer that sort of solution I wrote an additional
patch to do it that way, as follows. Updated total
patch (compressed) attached.
=== modified file 'nt/ChangeLog'
--- nt/ChangeLog 2012-09-21 18:10:25 +0000
+++ nt/ChangeLog 2012-09-22 21:44:27 +0000
@@ -1,6 +1,6 @@
-2012-09-21 Paul Eggert <eggert@cs.ucla.edu>
+2012-09-22 Paul Eggert <eggert@cs.ucla.edu>
- * inc/ms-w32.h (raise): New macro.
+ * inc/ms-w32.h (emacs_raise): New macro.
2012-09-18 Eli Zaretskii <eliz@gnu.org>
=== modified file 'nt/inc/ms-w32.h'
--- nt/inc/ms-w32.h 2012-09-21 18:10:25 +0000
+++ nt/inc/ms-w32.h 2012-09-22 21:44:27 +0000
@@ -200,10 +200,12 @@
/* Subprocess calls that are emulated. */
#define spawnve sys_spawnve
#define wait sys_wait
-#define raise sys_raise
#define kill sys_kill
#define signal sys_signal
+/* Internal signals. */
+#define emacs_raise(sig) kill (getpid (), sig)
+
/* termcap.c calls that are emulated. */
#define tputs sys_tputs
#define tgetstr sys_tgetstr
=== modified file 'src/ChangeLog'
--- src/ChangeLog 2012-09-22 21:13:11 +0000
+++ src/ChangeLog 2012-09-22 21:44:27 +0000
@@ -40,7 +40,7 @@
(terminate_due_to_signal): Rename from fatal_error_backtrace, since
it doesn't always backtrace. All uses changed. No need to reset
signal to default, since sigaction and/or die does that for us now.
- Use raise (FOO), not kill (getpid (), FOO).
+ Use emacs_raise (FOO), not kill (getpid (), FOO).
(main): Check more-accurately whether we're dumping.
Move fatal-error setup to sysdep.c
* floatfns.c: Do not include "syssignal.h"; no longer needed.
@@ -124,7 +124,7 @@
(emacs_backtrace): Output backtrace for the appropriate thread,
which is not necessarily the main thread.
* syssignal.h: Include <stdbool.h>.
- * w32proc.c (sys_raise): New function.
+ (emacs_raise): New macro.
* xterm.c (x_connection_signal): Remove; no longer needed
now that we use sigaction.
(x_connection_closed): No need to mess with sigmask now.
=== modified file 'src/emacs.c'
--- src/emacs.c 2012-09-22 21:13:11 +0000
+++ src/emacs.c 2012-09-22 21:44:27 +0000
@@ -311,7 +311,7 @@
}
#endif
- raise (sig);
+ emacs_raise (sig);
/* This shouldn't be executed, but it prevents a warning. */
exit (1);
=== modified file 'src/syssignal.h'
--- src/syssignal.h 2012-09-18 21:00:00 +0000
+++ src/syssignal.h 2012-09-22 21:44:27 +0000
@@ -40,6 +40,10 @@
# define NSIG NSIG_MINIMUM
#endif
+#ifndef emacs_raise
+# define emacs_raise(sig) raise (sig)
+#endif
+
/* On bsd, [man says] kill does not accept a negative number to kill a pgrp.
Must do that using the killpg call. */
#ifdef BSD_SYSTEM
=== modified file 'src/w32proc.c'
--- src/w32proc.c 2012-09-21 18:10:25 +0000
+++ src/w32proc.c 2012-09-22 21:44:27 +0000
@@ -1421,12 +1421,6 @@
}
int
-sys_raise (int sig)
-{
- sys_kill (getpid (), sig);
-}
-
-int
sys_kill (int pid, int sig)
{
child_process *cp;
[-- Attachment #2: syssignaq.txt.gz --]
[-- Type: application/x-gzip, Size: 67921 bytes --]
next prev parent reply other threads:[~2012-09-22 21:55 UTC|newest]
Thread overview: 31+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-09-18 23:39 bug#12471: Avoid some signal-handling races, and simplify Paul Eggert
2012-09-19 16:18 ` Eli Zaretskii
2012-09-20 6:07 ` Paul Eggert
2012-09-20 17:44 ` Eli Zaretskii
2012-09-21 7:42 ` Paul Eggert
2012-09-21 8:31 ` Eli Zaretskii
2012-09-21 17:26 ` Paul Eggert
2012-09-21 17:38 ` Eli Zaretskii
2012-09-21 18:13 ` Paul Eggert
2012-09-21 18:27 ` Eli Zaretskii
2012-09-21 19:59 ` Paul Eggert
2012-09-22 8:02 ` Eli Zaretskii
2012-09-22 8:47 ` Paul Eggert
2012-09-22 9:10 ` Eli Zaretskii
2012-09-22 9:40 ` Paul Eggert
2012-09-22 10:07 ` Eli Zaretskii
2012-09-22 10:55 ` Paul Eggert
2012-09-22 11:16 ` Eli Zaretskii
2012-09-22 20:28 ` Stefan Monnier
2012-09-22 21:55 ` Paul Eggert [this message]
2012-09-23 3:55 ` Eli Zaretskii
2012-09-23 3:52 ` Eli Zaretskii
2012-09-19 16:45 ` Jan Djärv
2012-09-19 19:58 ` Paul Eggert
2012-09-20 6:27 ` Jan Djärv
2012-09-19 21:36 ` Andy Moreton
2012-09-23 9:36 ` bug#12471: installed into trunk Paul Eggert
2012-09-23 15:22 ` Andy Moreton
2012-09-23 16:23 ` Andy Moreton
2012-09-23 17:37 ` Juanma Barranquero
2012-09-23 17:37 ` Eli Zaretskii
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=505E33CB.7090709@cs.ucla.edu \
--to=eggert@cs.ucla.edu \
--cc=12471@debbugs.gnu.org \
--cc=lekktu@gmail.com \
--cc=monnier@iro.umontreal.ca \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
Code repositories for project(s) associated with this external index
https://git.savannah.gnu.org/cgit/emacs.git
https://git.savannah.gnu.org/cgit/emacs/org-mode.git
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.