From: Paul Eggert <eggert@cs.ucla.edu>
To: Eli Zaretskii <eliz@gnu.org>
Cc: jim@meyering.net, emacs-devel@gnu.org
Subject: Re: building/using address-sanitizer-enabled emacs?
Date: Wed, 17 May 2017 13:05:29 -0700 [thread overview]
Message-ID: <3ff64f5d-492a-3368-5cb0-3cd7ffb55fbc@cs.ucla.edu> (raw)
In-Reply-To: <83efvn5uxq.fsf@gnu.org>
[-- Attachment #1: Type: text/plain, Size: 636 bytes --]
On 05/17/2017 09:06 AM, Eli Zaretskii wrote:
> I think we might be talking about 2 different reports. I meant the
> one here:
>
> http://lists.gnu.org/archive/html/emacs-devel/2017-05/msg00246.html
Ah, that came after his original report
<http://lists.gnu.org/archive/html/emacs-devel/2017-05/msg00182.html>,
which is the one I was referring to. I reproduced the later problem, and
as near as I can make it out it is due to an incompatibility between
vfork and -fsanitize=address. My guess is that the vforked child
corrupts the parent's shadow memory. I worked around the problem by
installing the attached patch into Emacs.
[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: 0001-Work-around-AddressSanitizer-bug-with-vfork.patch --]
[-- Type: text/x-patch; name="0001-Work-around-AddressSanitizer-bug-with-vfork.patch", Size: 1353 bytes --]
From 709259dcc501ef991991a35a6ffb2aef02a62c60 Mon Sep 17 00:00:00 2001
From: Paul Eggert <eggert@cs.ucla.edu>
Date: Wed, 17 May 2017 10:58:11 -0700
Subject: [PATCH] Work around AddressSanitizer bug with vfork
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Problem reported by Jim Meyering in:
http://lists.gnu.org/archive/html/emacs-devel/2017-05/msg00246.html
* src/conf_post.h (vfork) [ADDRESS_SANITIZER]: Define to fork.
Unfortunately with the AddressSanitizer in Fedora 25 x86-64, the
vforked child messes up the parent’s shadow memory. This is too
bad, as we’d rather have AddressSanitizer catch memory-access bugs
related to vfork.
---
src/conf_post.h | 6 ++++++
1 file changed, 6 insertions(+)
diff --git a/src/conf_post.h b/src/conf_post.h
index 4fc0428..1462bd1 100644
--- a/src/conf_post.h
+++ b/src/conf_post.h
@@ -302,6 +302,12 @@ extern int emacs_setenv_TZ (char const *);
# define ATTRIBUTE_NO_SANITIZE_ADDRESS
#endif
+/* gcc -fsanitize=address does not work with vfork in Fedora 25 x86-64.
+ For now, assume that this problem occurs on all platforms. */
+#if ADDRESS_SANITIZER && !defined vfork
+# define vfork fork
+#endif
+
/* Some versions of GNU/Linux define noinline in their headers. */
#ifdef noinline
#undef noinline
--
2.9.4
next prev parent reply other threads:[~2017-05-17 20:05 UTC|newest]
Thread overview: 41+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-05-07 3:40 building/using address-sanitizer-enabled emacs? Jim Meyering
2017-05-07 19:54 ` Paul Eggert
2017-05-07 21:44 ` Jim Meyering
2017-05-08 2:36 ` Eli Zaretskii
2017-05-08 5:42 ` Paul Eggert
2017-05-08 14:39 ` Eli Zaretskii
2017-05-08 14:46 ` Paul Eggert
2017-05-08 16:04 ` Eli Zaretskii
2017-05-09 5:48 ` Jim Meyering
2017-05-09 15:18 ` Eli Zaretskii
2017-05-09 17:06 ` Jim Meyering
2017-05-09 17:45 ` Eli Zaretskii
2017-05-09 19:22 ` Paul Eggert
2017-05-09 22:49 ` Jim Meyering
2017-05-10 2:41 ` Eli Zaretskii
2017-05-16 21:49 ` Paul Eggert
2017-05-17 2:24 ` Eli Zaretskii
2017-05-17 14:46 ` Paul Eggert
2017-05-17 16:06 ` Eli Zaretskii
2017-05-17 20:05 ` Paul Eggert [this message]
2017-05-18 4:15 ` Eli Zaretskii
2017-05-09 23:15 ` Philipp Stephani
2017-05-10 2:42 ` Eli Zaretskii
2017-05-10 22:24 ` Philipp Stephani
2017-05-13 8:02 ` Eli Zaretskii
2017-05-13 15:08 ` [PATCH] Fix use of sockaddr_in Philipp Stephani
2017-05-13 16:52 ` Eli Zaretskii
2017-05-13 19:14 ` Andreas Schwab
2017-05-13 19:29 ` Eli Zaretskii
2017-05-13 20:05 ` Andreas Schwab
2017-05-14 2:32 ` Eli Zaretskii
2017-05-14 6:11 ` Andreas Schwab
2017-05-14 14:20 ` Eli Zaretskii
2017-05-15 6:15 ` Paul Eggert
2017-05-15 9:04 ` Philipp Stephani
2017-05-17 20:38 ` Paul Eggert
2017-05-27 11:35 ` Philipp Stephani
2017-05-17 15:16 ` Eli Zaretskii
2017-05-17 20:15 ` Paul Eggert
2017-05-14 10:28 ` Lars Ingebrigtsen
2017-05-14 19:06 ` Philipp Stephani
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
List information: https://www.gnu.org/software/emacs/
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=3ff64f5d-492a-3368-5cb0-3cd7ffb55fbc@cs.ucla.edu \
--to=eggert@cs.ucla.edu \
--cc=eliz@gnu.org \
--cc=emacs-devel@gnu.org \
--cc=jim@meyering.net \
/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 public inbox
https://git.savannah.gnu.org/cgit/emacs.git
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).