linux-hardening - Re: [PATCH] security: don't treat structure as an array of struct hlist

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Date:   Tue, 12 Apr 2022 10:26:04 -0700
From:   Kees Cook <keescook@...omium.org>
To:     Bill Wendling <morbo@...gle.com>
Cc:     James Morris <jmorris@...ei.org>,
        "Serge E. Hallyn" <serge@...lyn.com>,
        linux-hardening@...r.kernel.org, linux-kernel@...r.kernel.org,
        linux-security-module@...r.kernel.org
Subject: Re: [PATCH] security: don't treat structure as an array of struct
 hlist_head

On Thu, Apr 07, 2022 at 10:59:30AM -0700, Bill Wendling wrote:
> The initialization of "security_hook_heads" is done by casting it to
> another structure pointer type, and treating it as an array of "struct
> hlist_head" objects. This requires an exception be made in "randstruct",
> because otherwise it will emit an error, reducing the effectiveness of
> the hardening technique.
> 
> Instead of using a cast, initialize the individual struct hlist_head
> elements in security_hook_heads explicitly. This removes the need for
> the cast and randstruct exception.
> 
> Signed-off-by: Bill Wendling <morbo@...gle.com>
> Cc: Kees Cook <keescook@...omium.org>

Ah! Yes, thanks. This is a good solution for this. I'd rather not have
any exceptions in the randstruct plugin. :) I'll take this via the
hardening tree unless anyone objects.

-Kees

-- 
Kees Cook