From: Carlos O'Donell <carlos@redhat.com>
To: Florian Weimer <fweimer@redhat.com>,
Aurelien Jarno <aurelien@aurel32.net>
Cc: libc-stable@sourceware.org
Subject: Re: Backporting CVE-2016-10739
Date: Tue, 01 Jan 2019 00:00:00 -0000 [thread overview]
Message-ID: <0a9daa70-7ea9-1ebd-8690-04b6ff2acd88@redhat.com> (raw)
In-Reply-To: <87r2cno9qq.fsf@oldenburg2.str.redhat.com>
On 2/4/19 10:26 AM, Florian Weimer wrote:
> That didn't work due to linking failures in the test suite.
>
> Here's something less ambitious. It retains the ABI for testing, but
> patches nscd to use its own private copy.
>
> (I have not put this on the branch due to Bugzilla spam this would
> cause.)
This was a non-causal v2 of your first patch.
It arrived while I was reviewing the first one, and I like this one
much more.
> Thanks,
> Florian
>
> nscd: Do not use __inet_aton_exact@GLIBC_PRIVATE [BZ #20018]
OK.
>
> This commit avoids referencing the __inet_aton_exact@GLIBC_PRIVATE
> symbol from nscd. In master, the separately-compiled getaddrinfo
> implementation in nscd needs it, however such an internal ABI change
> is not desirable on a release branch if it can be avoided easily.
OK. Good note.
> 2019-02-04 Florian Weimer <fweimer@redhat.com>
>
> [BZ #20018]
> nscd: Do not rely on new GLIBC_PRIVATE ABI after CVE-2016-10739 fix.
> * nscd/nscd-inet_addr.c: New file. Build resolv/inet_addr.c for
> nscd, without public symbols.
> * nscd/Makefile (nscd-modules): Add it.
>
> diff --git a/nscd/Makefile b/nscd/Makefile
> index b713a84c49..eb23c01a39 100644
> --- a/nscd/Makefile
> +++ b/nscd/Makefile
> @@ -36,7 +36,7 @@ nscd-modules := nscd connections pwdcache getpwnam_r getpwuid_r grpcache \
> getsrvbynm_r getsrvbypt_r servicescache \
> dbg_log nscd_conf nscd_stat cache mem nscd_setup_thread \
> xmalloc xstrdup aicache initgrcache gai res_hconf \
> - netgroupcache
> + netgroupcache nscd-inet_addr
OK. Add a new object to nscd.
>
> ifeq ($(build-nscd)$(have-thread-library),yesyes)
>
> diff --git a/nscd/nscd-inet_addr.c b/nscd/nscd-inet_addr.c
> new file mode 100644
> index 0000000000..ce42ba3ea8
> --- /dev/null
> +++ b/nscd/nscd-inet_addr.c
> @@ -0,0 +1,30 @@
> +/* Legacy IPv4 text-to-address functions. Version for nscd.
> + Copyright (C) 2019 Free Software Foundation, Inc.
> + This file is part of the GNU C Library.
> +
> + The GNU C Library is free software; you can redistribute it and/or
> + 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.
> +
> + 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
> + Lesser General Public License for more details.
> +
> + You should have received a copy of the GNU Lesser General Public
> + License along with the GNU C Library; if not, see
> + <http://www.gnu.org/licenses/>. */
> +
> +#include <arpa/inet.h>
> +
Please add a comment explaining why this is here.
> +/* Declare __inet_aton_exact as hidden, so that it does not get
> + exported from nscd. */
> +__typeof__ (__inet_aton_exact) __inet_aton_exact attribute_hidden;
> +
> +/* Do not provide definitions of the public symbols exported from
> + libc. */
> +#undef weak_alias
> +#define weak_alias(from, to)
> +
> +#include <resolv/inet_addr.c>
>
Can we kill the prototype from the public header and use an internal
header? It seems messy to leave that prototype for the GLIBC_PRIVATE
symbol in the public header. It might tempt people to workaround the
linkage protection.
--
Cheers,
Carlos.
next prev parent reply other threads:[~2019-02-04 15:36 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-01-01 0:00 Aurelien Jarno
2019-01-01 0:00 ` Florian Weimer
2019-01-01 0:00 ` Aurelien Jarno
2019-01-01 0:00 ` Florian Weimer
2019-01-01 0:00 ` Aurelien Jarno
2019-01-01 0:00 ` Florian Weimer
2019-01-01 0:00 ` Florian Weimer
2019-01-01 0:00 ` Carlos O'Donell [this message]
2019-01-01 0:00 ` Florian Weimer
2019-01-01 0:00 ` Carlos O'Donell
2019-01-01 0:00 ` Florian Weimer
2019-01-01 0:00 ` Carlos O'Donell
2019-01-01 0:00 ` Florian Weimer
2019-01-01 0:00 ` Carlos O'Donell
2019-01-01 0:00 ` Florian Weimer
2019-01-01 0:00 ` Carlos O'Donell
2019-01-01 0:00 ` Aurelien Jarno
2019-01-01 0:00 ` Carlos O'Donell
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=0a9daa70-7ea9-1ebd-8690-04b6ff2acd88@redhat.com \
--to=carlos@redhat.com \
--cc=aurelien@aurel32.net \
--cc=fweimer@redhat.com \
--cc=libc-stable@sourceware.org \
/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.
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).